Solar Power - A Bodger's Guide

2012 - All Change!

2012-02-09T22:47:00.000Z

Well, we're back from our adventures in Japan and while out there, I made an important decision...

I hate lead acid batteries. They're rubbish for solar applications.

So I took the plunge and put in an order for some batteries I've been eyeing for a few months now. They get used a lot in the DIY electric car (EV) arena because they are light, powerful, small and can withstand deep discharges daily. What's more, they are based on lithium ion technology and so do not need or even particularly like being kept at full charge all the time, as lead acid batteries do.

In short, they are much superior to lead acid batteries in many respects (but not all).

Of the makes available, the only one that was available in Europe (let alone the UK) was the Winston Battery Company make. These are a new sub-type of lithium ion cells that use yttrium (a rare earth metal) to significantly improve the reliability and safety of these cells, compared to the previous generation of cobalt based lithium ion cells. These could explode violently in a ball of flames (hence the stories in the papers about exploding mobile phones and laptop batteries).

Over the next few days and weeks, I'll be getting to grips with these new cells in order to build them into a 24V nominal 400Ah battery pack for my solar system.

You can find them at the GWL EV Power shop

[photo coming soon]

Trouble at t'mill

2011-12-27T20:00:00.000Z

[Retrospective posting again - It's actually February 2012]

Hmmm... Lead acid batteries... You gotta love 'em. Not.

12/12/2011
The new setup initially seemed fine, but I've noticed a nasty gassy smell when charging on sunny days. The block balancers are working fine and the two 12V halves of the banks are in balance, but there's some venting of gas under charge somewhere. Quite a lot of gas - as I had to open all the doors and windows to clear my nausea.

Not good behaviour for a set of "sealed" batteries.

With the Ritar batteries replacing the gel ones, I had increased the absorption charge Voltage to 2.375Vpc (14.25V / 28.50V at 25C with 4mV per degree C compensation). The Marathon bank is still attached and doesn't like to go up to much more than this but the Ritar blocks actually specify a higher Voltage of up to 2.433Vpc (14.60V at 25C). So it's unlikely that the Ritars are the problem.

14/12/2011
So I reduced the absorption charge Voltage down a bit to 14.20V and set the temperature compensation slope to stop at 17C. That way the compensated Voltage never goes above 14.40V. The absolute limit for the Marathons. Not normally a problem but lately it has been 16C in the mornings by the battery bank (the patio door double glazing isn't what it could be...).

Still no good. Maybe the Marathons can't handle higher than gel Voltage at all. So I threw the breaker on the Marathons and ran with the new Ritar battery bank only to see if that would isolate the problem (to the Marathons).

20/12/2011
No good. After a few days of charging and chopping and changing the Marathons around I discovered that it's one or both of the new Ritar batteries that are making the gas. They seem to be in balance electrically and deliver the power ok and charge ok but I'm getting poisoned while working downstairs!

I even resorted to ripping out the Marathons and sticking them in the garage outside to totally eliminate the possibility that they were gassing while idle in the room.

22/12/2011
So now I have had to swap over again. The new Ritars are definitely dodgy in some way. Even quite gentle charging above float level causes a degree of noxious behaviour.

The Ritars are now the ones outside in the garden and I brought the Marathons back in but discovered that one of the 6 packs is nearly dead. It doesn't accept charge very well at all. The fact that it was in parallel with all the others and they used active balancing had masked the fact that one of the packs was dead.

So now I'm down to just two pairs of the Marathons that are probably quite long in the tooth and my new batteries are pants...

Have a VRLA Merry Christmas!

2011-12-24T12:15:00.000Z

Ok, it wasn't the best pun in the world.

But for your Christmas entertainment, I present the dissection of a dead VRLA 7Ah battery :D

Been wondering what exactly the "Valve Regulated" bit of a Valve Regulated Lead Acid battery is? Wait no more...

Here's a dead battery I've been using as a door stop for a while...

The lid is spot glued down so levering it off breaks these. It could be super-glued back down afterwards but this little guy is destined for the recycling centre so no worries there.

Under that "sealed for life" exterior we have the cell tops with their "valves"...

Which turn out to be nothing more than rubber caps that are pressed on the cell tops with a bit of grease to make a gas tight seal. The cells have gone to vacuum a bit and you can see the rubber caps sucked in a bit. When I prised this fella off, it made a little hissing noise (but sucking air in, not blowing gas out).

The top of the cell vent has some slots in it to allow gas out and above some random pressure below 6psi, the grease and rubber combo lets the gas out and then re-seals to stop air getting in (the oxygen would poison the cell). The gas can then escape along the shallow channels to the gaps in the lid (which, at first sight, looked to be just a gap for jamming my screwdriver into, in order to get this guy opened up... silly me :D ).

In the final picture, you can see the top cover was pressing lightly on the rubber caps, to stop them popping off under pressure. You can also see where cells 2, 3, 4 and 6 had vented under overcharge and left some acid residue on the lid.

Well, I'm off to see the family, watch bad TV, and eat too much for a few days :D. See you on the other side!

Un-FIT for Purpose

2011-12-18T11:26:00.000Z

Another epic fail spotted today on our travels.

This was taken at 10.25 in the morning today... Where's the Sun in this picture, children?

Yep, that's right... it's at right angles to the solar array that some PV company has talked these folks into installing on an almost west facing (250 degrees) roof with the promise of massive returns on investment in the rush to beat the FIT rate downgrade last week.

This array was installed back in early November, just after the announcement that the rates were being slashed to 21p per kWh.

The one thing that should be applauded from the "rent-a-roof" PV mob is that they never agree to install on a east-west roof because you just don't make the money on such an array...

Other companies are happy to take your cash and install wherever they can. The fact that it will take 20 years to pay back instead of 10 isn't their problem.

Danger UXB - Uber-eXploding Battery!

2011-12-16T18:17:00.001Z

The other day there was a funny smell in the living room... and come to think of it, I heard some unaccounted for popping noises.

Couldn't trace it - inspected the solar batteries and couldn't figure it out. Had to open the window and vent the smell.

Today I found the culprit... And it was an unlikely suspect... Some cheap Tronic Energy Eco 850mAh AAA NiMH cells I bought from Lidl a while ago to use in the multimeters on the wall. I'd put two of these cells on charge and one exploded in the charger! Well, the end cap had blown and leaked electrolyte all over the charger. The third cell that had been on charge at the same time (a Sony one) was fine.

I wondered if it was just the one duff cell, so I put three more in the charger and set it running. It finished in the usual time and the lights went out for each cell. I felt the cells and they didn't seem unusually warm. Huh... maybe it was just one duff cell after all.

But just as I was saying that to myself, all three cells erupted in a fizzing hissy fit! Nope, all of them are rubbish.

Can't be that they don't like fast charging, as Lidl also sells them in a pack with a fast charger.

At least my new cheap 180Ah solar batteries haven't exploded yet...

Bad Installs

2011-12-11T18:23:00.001Z

As PV has been more about a "rush for gold" than producing useful energy, sometimes the allure of adverts and salesmen promising future returns of 10-15% in exchange for £10,000 of your hard earned cash today has lead to some questionable installs.

What's wrong with these installs, anybody?

Can't see the PV for the trees...

This one doesn't look so bad...

But what's that on the roof in the middle of the array?

So when the trees aren't shading the array in the afternoon, the sewer vent is shading part of it all day. They've installed the worlds most expensive sun dial... At least the pipe will cast its shadow in the gap between the panels at 1pm (high solar noon in BST time) for the maximum power. It will just reduce power output from the panels left and right of it from dawn to 12.55 and from 13.05 to sunset.

New Battery Time (Again)

2011-12-09T22:50:00.001Z

Well, the Deka Solar gel batteries finally caved in. They've been in service daily since about October 2009 and they were second hand when I picked them up. After another 2 years and 1 month (some 760 charge cycles), they're way down on capacity and have gotten very high internal resistance but have more than met their expected life.

When pushed hard (say by the 2.2kW heater in the washing machine), they could still pump out 20-30 Amps with the other set of batteries but they took a long time to recharge, only managing to absorb 9 Amps of charge. The other bank will greedily suck up nearly 40 Amps, even through its deliberately under-sized long wiring.

I'd considered making a leap to lithium iron phosphate (LiFePO4) cells but having seen Jack Rickard nearly burn down his workshop on his electric vehicle web show - EVtv a couple of weeks ago, I was not certain that I'd have the charge control issues all figured out at the first go on big 400Ah cells. I may do a trial with some much smaller (cheaper if killed) 60Ah lithium cells with my spare Morningstar controller.

In the meantime, I found a good deal on some new 180Ah AGM deep cycle batteries from China by an outfit called Ritar. These guys usually make VRLA batteries for computer backup supplies (UPS). But these batteries are usually for "standby" use and not too amenable to being discharged daily. Although, the Marathon 105Ah packs are also UPS batteries and seem to still be going strong after 1 year and 5 months of daily cycling.

Anyway, this big seller of used UPSs on eBay gets old UPSs and puts new batteries in them (not OEM batteries but these Ritar ones) and then sells the UPSs. But he had a side line in big deep cycle and EV batteries for toy electric cars and disabled mobility cars and so on. Ritar RA12-180D packs are slightly different from the RA12-180 packs (which are the UPS "standby" type, whereas the "D" type are supposed to be for "cyclic deep discharge" use). We'll see...

Very good price: £178 for 180Ah at the 10 hour rate (most makers quote the more inflated 20 or even 100 hour discharge rates - better makes quote the 10 hour or even 5 hour rate). Under £1 per Ah. And that included free next day TNT delivery on a pallet. I had to order two for my 24V system, and they turned up the next morning even though I ordered them at just after 4pm the previous day.

The batteries needed to be given an initial charge to wake them up from their months of storage in the warehouse and on the ship from China. Thankfully, AGM batteries do not self-discharge that much (2-3% per month). Even so, the Ritar technical guide recommended charging them for up to 24 hours before first use. So I set this up on a mains powered charger and monitored the Voltage and current until neither changed over time. One battery took the full 24 hours to reach this state; the other was a bit faster at about 18 hours (it read a higher initial Voltage when I first opened the box).

Luckily, it was sunny for a couple of days and so I disconnected the defunct gel batteries and turned off the inverter to let the Marathons come up to something like full charge.

After the new Ritars had both been charged up, I connected them in where the gel battery bank had been and let the packs equalise overnight. I had to reprogramme the charge controllers a bit as the Ritars like higher Voltages than the gel ones did (although this conflicts a bit with the Marathons that also liked gel Voltages). I may have to disconnect the Marathons when I run periodic equalisation charges on the Ritars, as the Marathons start to vent gas above 14.40V charge and the Ritars want to go up to 14.60V on an equalisation charge.

Yesterday was one of the windiest days in the UK for over 10 years and gloomy with it, but bright enough to gently charge the new combo and allow some tweaking of the charge controllers.

I mounted the active battery balancer on the distribution board to make things a bit neater. This can get a bit warm when working hard so I mounted it on stand-off washers; raising the heat sink off the wood. It will give a bit of extra air flow on the back of the device.

Today was the first use of the whole pack "in anger". It was wall to wall sunny and we used solar power to heat water (some 2.4kWh went into the water tank). We even managed to do a load of laundry with solar power again. The Ritar bank was more highly charged than the Marathons and it did more work initially. I'll have to see how they balance out over the next few days.

One interesting thing I've noticed is that the very low impedance of the Ritars has really cut down the "twittering" noise the inverter used to make. I always assumed that it was a "design feature" of the inverter but seems it was a feature of the gel batteries. Maybe if it starts up again as the Ritars get older, I'll plug my super capacitors into the circuit and see if that helps.

New LED Energy Saving Lamp

2011-11-07T21:09:00.000Z

The darkness... Tell me about it. Today was even darker than yesterday.

An unbelievably tiny 390Wh generated by my whole solar array in the whole "day". The "sunset" at 4.35pm was eclipsed by "nuclear winter" at about 3.30 today, blacking out the Sun so much that the street lights came on early!

Lucky then that I've got a new LED lamp that consumes only 5W of energy.

It's about as bright as a 8W CFL lamp (or a 40W tungsten lamp) but starts up immediately at full brightness, lasts maybe 4x longer and doesn't hum annoyingly in the plastic Ikea uplighter on my dining table.

It's a Civilight GLS 5W type and has a wide even light, a frosted dome, and pleasant mid-warm-cool colour; not too blue and not too red white light.

A bit dimmer than the 11W CFL it replaced, but I'd wanted a lower light, as it's over the TV.

At just under half the power consumption of the old CFL lamp, it will also extend the PV battery life... A jolly good thing in these dark days.

Best of all, my mate in the trade gave it to me for helping him out at some eco fairs over the last few weeks. He's on a evangelical crusade to get people to use LED lights instead of mercury containing CFLs.

Lummy, It's Dark!

2011-11-06T23:30:00.000Z

As Victor Meldrew would say, "I don't beeleeve it!".

It's been so dark the last two days... I mean properly night time dark at mid-day.

839Wh generated yesterday.
Only a measly 712Wh today!

From a 2.5kWp PV array!!!

I've not had to turn the inverter off since maybe February, but today the battery bank crashed to just 26% and everything had to go back on the mains and the inverter was off - at lunchtime. Not even the night... Lunchtime!

My base load on PV is pretty high though. I put a clamp-on Current Cost meter (kindly donated by a friend) on the inverter output and all the network gear and DECT phones and bits and bobs around the house add up to about 80W continuous. Quite surprising really that the battery can run all of that for a couple of days. But the inverter has an idle power of 35W all on its lonesome, so that bumps the total DC load up to about 105W. That's 2.5kWh per day. And with only 1.55kWh charged in the last two days, and only 3.37kWh charged in the previous two days, it was just a question of time before the battery reached the end of its tether.

Didn't help the BBC 24 hour weather forecast being completely wrong for two days in a row, so I used too much battery power on Thursday and Friday on the promise of "sunny intervals" for Saturday and Sunday.

If things don't brighten up tomorrow, I may even have to break out the mains powered battery charger from the loft.

Charles Hendry & Me Brill Idea

2011-11-05T11:00:00.000Z

I met with Charles Hendry (the Minister of State for the Department of Energy) at an eco fair in Forest Row (East Sussex) today.

He was delivering his bombshell - a snap FIT reduction for PV generation, arguing the point in a presentation that the FIT had to be reduced as demand has far outstripped cash supply (from the green tax levy on energy companies). If they did not shut the door hard, they would have had such a rush before April 2012 that the entire budget for the next 3 years for FITs would have been allocated by next April.

Here he is, on the left, being verbally beaten up by an angry audience of commercial solar installers (one farmer had just canned a 40kW install, as it could not be done before the Dec12th FIT deadline) and assorted solar traders (notably Jeremy Leggett, founder of Solar Century, on the right). The dude in the middle was refereeing the bout.

They were all upset that they could not peddle a guaranteed 10-12% index linked and government backed investment opportunity (they're not selling PV, y'know - judging by the advertising copy that usually reads "MAKE FREE MONEY FOR 25 YEARS!!!").

After the debate officially ended, I talked briefly to the minister about alternatives. I suggested that the government should "do it's bit" in enabling me to "do my bit" by making all energy saving and generating equipment and services zero VAT rated. It's odd that my utility bills attract only 5% VAT but the LED lights I buy to save energy attract 20% VAT.

It's also odd that the more energy I use, the cheaper it gets. The first 900 kWhs are are 24p for me and then it drops to 10.5p. Should be the other way round, if we want to curb excessive use, no? It would also mean that the units I save first with my PV generated ones are the more expensive ones.

I also suggested the government extend the "invest as you earn" (IAYE) scheme that already exists (for buying bicycles to work and pension contributions, for example) to other eco products and services (say PV or ASHP or a new A rated boiler or windows or insulation...).

These IAYE investments are deducted by your employer through the PAYE model from your salary prior to income tax is calculated, making the buying of a bike or a pension (or PV or ASHP, etc) income tax free.

You don't get the cash. You get a voucher to buy the prescribed product or service. That way people can't scam the system to dodge tax and blow it on something else.

Think how much more you could invest in saving your skin (eco investment is about saving yourself) if you didn't have to pay 10-40% income tax and another 5-20% VAT for the pleasure.

Charles rubbed his chin and said "That's an intriguing idea...". He'll probably file it under "cranky ideas that reduce government revenue..." and black Mercedes "G cars" will be parked outside my house from now on. [sigh].

Micro Battery Off Grid System

2011-09-20T17:49:00.000+01:00

Sorry for the lack of posts... No excuses, but I have been busy over the summer [Damn. That sounded like an excuse...]

Remember the Big Capacitors of a few posts ago? Well they've found a use :D

They're really cool, as it happens. I charged them up and left them with a Volt meter on them, to see how long they could hold a charge for. Kinda expected they would self discharge in a few minutes, if left alone.

Time ticked on, and they didn't seem to discharge (now that the LED Voltmeter was not connected to the tops). In fact, a couple of weeks later they had only discharged down to a few Volts less than when they were full! Extremely low leakage current. When I shorted them out by accident, they still packed a terminal blackening punch :D

I got a couple of big 230Wp panels to test from a friend, and got to wondering about how I could use them. I had a spare 1kW inverter and a spare MPPT charge controller. Having a big 500Ah battery bank on the main house, I wondered about the opposite. How small a battery could you use for an off grid system..?

Luck came in the form of a bargain bin at a DIY superstore. They were selling off loads of drill batteries for just £3 each. [rant mode ON] The cunning manufacturers change the drills and batteries to be slightly different each year so that your old batteries can't be used on their new drills... (built in obsolescence, again). It also mops up those pesky consumers who think that they can keep their old drill until it breaks... Oh, no. They change the shape of the new "spare" batteries so that they don't fit the old drills - forcing you to buy a whole new drill set. [rant mode OFF]

Anyhoo... it means experimenters can pick up brand new 12V and 14.4V NiCd battery packs for less than the cost of a single cell at an electronics store. These cells are also rated for high current applications - A cordless hammer drill can pull 20A when working hard.

So, I had a pair of 12V drill batteries wired in series to make a titchy (for off grid use) 24V, 1.5Ah battery pack. But it can deliver 20A easy (480W). Good for a steady load but not enough to start something big, like a fridge or big TV... Enter the big capacitors!

The capacitors can pack a punch of maybe a few hundred Amps for a second or two and the NiCd batteries can deliver 20 Amps for about four and a half minutes. NiCd cells are good in that very small ones can sustain high drain currents without their effective capacity being degraded. Small lead acid batteries would be useless, as firstly their capacity drops as you drain them faster and draining them until they are empty is a sure way to kill them. NiCd cells don't mind being drained flat - it's what happens when they're used in a drill.

So what I had was a 460W solar array with a 480W four minute "buffer". You can use this system to power quite big stuff on a sunny day without the bulk and expense of a big lead acid battery bank and it will keep your load going when small clouds pass in front of the array. It only takes about 15 minutes of charge for the battery to get to full from empty.

Some drill batteries (the ones designed for a fast charger) even come with a temperature sensor in the pack, so the drill charger won't over charge the pack and cook it. Unfortunately, the 12V packs I bought didn't, but the Morningstar MPPT-15 charge controller has one. I super-glued it to the top of one of the packs. This charge controller is also fully programmable. So, despite being primarily for lead acid batteries, it can easily be programmed to charge NiCd batteries. You just have to set the Voltages differently (a topping charge of 30V and then a float of 28.5V) and the timers much shorter (as the pack is so small). I programmed the temperature compensation to do nothing below 25 degrees C, but then start an aggressive negative slope above that. So, if the cells are heating up from high charge or discharge currents, the charge Voltage would be reduced. The charger itself is limited to 15 Amps charge current.

So far it's worked ok. I managed to run a 140W 28" LCD TV on it for a whole afternoon with only a couple of power failures (when it started to cloud over in the afternoon). The fully charged battery could run the set for about 12 minutes without direct sunlight on the array. The capacitors provide the oomph to start the inverter and the TV without the Voltage dropping. At the "float" Voltage of 28.5V, the two 1F capacitors store 28.5² x 0.5 Farads x 0.5 = 203 Joules of energy - or enough for 203W for 1s or 406W for 0.5s or 812W for 0.25s (although that would require draining the capacitor flat).

Huge battery banks are hugely expensive in off grid systems. This alternative method allows quite moderate loads to be run in good weather, during the day, with only a tiny cheap battery. If you planned it right, you could use the inverter in the day to do power hungry things, and then just use the NiCd battery pack at night to run ultra low power 12V DC LED lights... Or watch a very short TV show :D

The Big Build (Part 2)

2011-07-19T11:00:00.000+01:00

Well, it's been one of those weeks...

With the to-ing and fro-ing from the sellers for wrong parts, it meant that the plan to get the project done this week rather went to pot.

But some progress was made, we got the tower up and started to figure out how to mount the rails.

My friends Peter and James helped massively here, with three brains being better than one and two pairs of hands on the roof being safer than one.

The first problem was that the tiles were arranged so that the vertical wooden rafters were not in line with the troughs in the tile shapes but the peaks (at the edges of the tiles). You can see this in the picture below; the hooks are sitting in the join between tiles.

This meant I needed to make up blocks to raise the hooks. I could have used hooks with lateral adjuster plates or made up wooden blocks between the rafters but this seemed the easiest bodge to do.

Conveniently, I had some treated roof batten wood that was almost exactly the right thickness. It wasn't quite square 25mm, so we could make some blocks with a thicker profile than others (to suit the particular wonkiness of the rafters) and make up the difference with thick galvanised washers as shims.

Of course now the rafter screws that I'd ordered were a bit too short for comfort, so I had to buy longer ones.

There are two pairs of rails for two rows of solar panels... Oh, yeah... The plan? To mount all the 80W panels from the patio and garage roof on two rows; one of 10 panels on top and 8 panels on the bottom. The lower row is fewer because there's a sewer pipe vent poking up though the roof.

I needed two lengths of rail for each row, as the rails are 3.2m long and I have about a 6.0m roof span (allowing some margin around the edge). The lengths are joined together with a U bar insert that bolts them together.

A special T bolt goes into the rail mounting channel and then bolts the rail to the hooks.

A good tip: Don't do this on a really sunny day as you'll burn your hands on the hot concrete tiles :D

I used 12 hooks on the top row and 10 on the shorter bottom row. In the picture above, you can see I've bolted a sawn-off bit of rail to the top two hooks to use as a make-shift hand rail.

Once you get some hooks in the roof and bolt some rail to it, you can move about the roof much more safely, as you have something to stand on other than the tiles, and you can't slip off the roof without the rails stopping your slide.

Helps too to have a tower where the deck is at the exact height of the gutter so that you are effectively stepping on to the roof at "ground" height, rather than clambering up. I now have much less fear of the height. This can be a bad thing though, as fear can keep you alert.

In the sequence below, you can see us moving the tower back and forth and adding the rails from the bottom up, using the first "rung" to make it easy to then add the upper rows.

Finally, the panels just clamp on to the rails using special plate nuts that clip into the rail and can slide along it. Stainless steel hex bolts hold down the panels with end clamps that have to be the same height as the panel frame height. In-between the panels, a U shaped clamp holds adjacent panels to the rails.

You can see these end clamps and inter-panel clamps here:

Note how the clamp on the left is taller than the one on the right because the panels have different frame heights. If I'd been planning ahead, I'd have bought panels with all the same frame height. :D

The inter-panel clamps are universal because they don't actually extend down to the rail. They just grab the panels and then you torque the bolt so that it holds the panel tightly down. Don't go mad with the wrench or else you might crush the panel frame wall and crack the glass!! This is more of a risk on cheap panels that don't have strong frames.

Now, the eagle-eyed among you may have spotted that the two panels in the bottom left are not in line with the rest of the panels in that row. This is an annoying consequence of the old panels on the left having junction boxes that are too big! They stick out below the height of the frame and so when I came to mount them, the junction boxes fouled the rail... GRRRR. There was nothing for it but to mount those two panels lower on the rail, to clear the junction boxes. If I'd checked this before starting, I could have set the rails closer together...

Hiiiigh...Hooooo.... High ho, hi ho, it's off to bodge we go...

The Big Build (Part 1)

2011-07-15T11:00:00.000+01:00

It was about time... About time I got these solar panels up on the roof, where they belonged.

So began two weeks of major bodgery on the house roof.

I'd started collecting materials for the project a few weeks ago. Proper solar cable - 50m of 6mmsq. MC4 plug sets from a Chinese maker called Lensun sold widely on eBay.

I'd also done some research into mounting kit. Peter, a friend from across town, had mounted some 100W and 185W panels using glavanised steel 41mm cable conduit and spring channel nuts.

I ordered some of the same rail but it was out of stock. They'd also stopped selling the right spring nuts to go in the channels so I ordered these from another company. The nuts turned up quickly enough but only half of them were galvanised! A call the supplier sorted this out and they sent me a whole load more at no extra charge.

After 5 days of waiting for the rails to turn up, the electrical supplier admitted they weren't stocking them any more. Useless. Now I had two loads of channel nuts for non-existent rails... Back to the drawing board.

Peter had recommended a PV installers supply shop on the web (www.solarseller.co.uk). They specialise in low cost aluminium rails and clamps and the roof brackets. Not as complicated an aluminium profile as some makes, but cheap and it does the job. The rails come in 3m lengths with U bar joiners to make longer rails. I ordered 8 rails to make the two rows of modules to go on the roof (one a bit shorter than the other as their is a sewer vent chimney sticking out of the roof at the bottom corner).

They also sold the matching Hilti stainless steel channel nuts for these rails.

Unfortunately, this supplier also screwed up and sent me a box of wrong stainless steel bolts instead of the roof joist screws for the roof hooks, but a phone call had the right screws delivered the next day at no extra cost.

One thing that made this install more expensive was the fact that all my PV modules are different heights, ranging from 30mm to 50mm tall. This meant buying different sized end clamps to match and installing the modules in blocks that were the same height. Luckily, they sell "universal" in-between clamps that are shorter than 30mm deep to allow them to hold down any sized PV module.

Now all I needed was some scaffold or an access tower to get up to the roof... And a head for heights!

Renting scaffold turned out to cost a fortune (especially if you didn't know how long you would need it for). In theory, you'd only need it for a couple of days to do a solar install, but when did any of your DIY ever go to plan?

Then I found the answer... It was cheaper to buy an access tower than to rent one! These guys at www.laddersandscaffoldtowers.co.uk sell a well made DIY 7m tower kit that flat packs for storage. It came with big outrigger legs to keep the tower steady even when me and Peter were up at the top. In the end it was a good buy as I used it for over two weeks, making it break even compared to renting a tower. If I use it again, it will have saved me money.

Measuring Dirt

2011-05-25T19:10:00.002+01:00

A friend asked me if I could quantify the dirt on my solar panels from the previous post... o_O

I'd never tried to measure the effect of dirt before, but the conditions were right today to give it a go. Not completely scientific (I am a bodger after all), but here's what I did:

Firstly, there was the pair of panels from the pictures in the previous post. Now, they were very dirty, and I'd cleaned half of them to show the dirt. These modules are easy to measure, as they have a chocblock in the link between the pair. So, I crawled under them (taking the opportunity to clear out the weeds growing under them!), and inserted an Ammeter in the circuit.

I manually recorded some values for the current at 15 second intervals; before cleaning, whilst cleaning, and post cleaning. Obviously, while cleaning the things, I didn't actually record any data in my scrap of paper, so missing values are just the last known value repeated (to fill the gap - Excel wouldn't draw a line unless every cell had a number in it...). The modules were still wired into the array, and the battery was at bulk mode, so was drawing all the current it could.

Afterwards, I took a copy of the logger file that had the time and other solar system data (like the solar sensor data that tells you how strong the sun was as a percentage of the maximum it could be), and input the manual data against the same time records. A bit of fiddling in Excel chart maker and...

If you click on the graph, it will open up bigger.

You can see that when I washed the pair of modules (half of which had been washed yesterday for the photos), the output current jumped up. The solar input power remained within 1% of flat the whole time, so it was definitely the effect of the dirt on the module that made the difference. I've noted on the chart where the modules were wet, and I dried them with a cloth, and then you can see the panels output drop off a bit, as they warm up again in the sun. PV cell power drops off a bit, as they get hot. I used warm water to wash them, so as to prevent any thermal shock to the glass, and not cool them down too much.

Then I thought, that seeing as I'd got all set up for washing these panels, I might as well do some of the others too. The trace below shows the power output of the TriStar array controller while I cleaned most of the panels attached to it. This time, I just let the modules dry in the sun.

I excluded the power figures from the SunSaver array, as those panels are completely separate, and I didn't clean them. I also ran out of water before being able to clean the bunch of panels at the front of the garage. In truth, I couldn't be bothered to do them, and I was supposed to be doing some work anyway :D.

Again, you can see a marked improvement in output from not having a load of grunge on the array...

Muddy Solar Panels

2011-05-24T20:18:00.000+01:00

One problem that solar panels face in such long dry spells of weather as we're having now is the accumulation of dust, pollen and other bird grunge on their surface.

It doesn't help that we've had some very light rain over the last day that has, in itself, contained a lot of mud.

This can degrade the output of the array until a really good downpour washes the panels, or you get a telescopic window cleaner out...

Before:

After:

Some Fav Quotes by Thomas Edison

2011-05-23T19:03:00.000+01:00

"To invent, you need a good imagination and a pile of junk."

Amen to that. Bodgineering is at the heart of invention. Once you've bodged something together from bits of this and that, you can see how to do it "properly".

"We are like tenant farmers chopping down the fence around our house for fuel when we should be using Nature's inexhaustible sources of energy; sun, wind, and tide.... I'd put my money on the sun and solar energy. What a source of power! I hope we don't have to wait until oil and coal run out before we tackle that."

That was his hope in 1903. He'd be shocked that we're still "working on it" over 100 years later...

"Hell, there are no rules here - we're trying to accomplish something."

OFGEM, take note.

1 MWh on the Clock

2011-05-19T23:01:00.020+01:00

A bit of a milestone today. Since fitting the "proper" OFGEM generation meter back in March last year, the AC power generated (consumed) from my solar panels has been ticking up.

Today, it passed the 1000 kWh (1 MWh) mark. But amazingly, a full 153kWh of that total was generated just last month! May is shaping up pretty well too, having made 114kWh this May, compared to only 92kWh last May.

Big Capacitors!

2011-05-11T21:42:00.000+01:00

Get a load of these babies!

Giant 1 Farad capacitors. Used originally on super loud car audio systems.

I found them at a car boot sale, going for a tenner for the pair. They'd cost about £30 each when new.

They have these so called protection circuits on the top. I thought that it was an anti-surge device for when they have power applied. They have extremely low internal resistance (maybe as low as 0.002 Ohms) so can draw a huge surge current when connected to a source of current. But no, the circuit turned out to be just a buzzer that sounds when they are reverse connected by accident, or when subjected to over-Voltage. They have a quite small maximum working Voltage of 16V. Given the near explosion that occurred when I over-charged the baby capacitor on the bike generator, these would pose quite a risk, if mal-treated.

The display reads Volts and the two other LEDs are really just for show (they're for car audio "enthusiasts" after all)... Although, they do serve a useful purpose as discharge loads, to make the capacitors safe in a short time after the power is removed.

Luckily, the tops just unbolt, and you can remove the circuit board. It's just held on the capacitor terminals with press contacts. Then the bolts can screw back into the terminals directly.

The Voltmeter circuits are possibly useful later for something. On their own, they're worth £5 each, so the capacitors were actually "free".

I thought that as I had a pair, I could wire them in series to allow them to be charged from one of the left-over small amorphous solar panels. Wiring them in series increases the Voltage they can handle, but reduces the capacitance, so you get a 0.5F capacitor that is safe up to 32V. If you wire them in parallel, you get a 16V 2F capacitor, and mind bending amounts of current (the internal resistance is halved again to 0.001 Ohms, giving a short circuit current of up to 16,000 Amps!).

The panel puts out about 23V, open circuit, in the sun, but you can't use that Voltage, as the current is so low. But you can charge a 0.5F capacitor from that current, over a minute or so. The capacitor, unlike a battery, does not seek to limit the Voltage it charges to. So, eventually, it will store a unit of energy at the highest Voltage attached to it, and can later deliver that energy with massive current.

I disconnected the charge source, and started measuring the Voltage across the capacitors, thinking the charge would leak away in a few minutes. To my surprise, only about 10% of the energy had leaked away after 24 hours! The energy level of a capacitor is exactly proportional to the Voltage it reads, unlike batteries that can read 12V, but be completely flat.

But what use are giant capacitors..? I'll let you know when I think of something :D They're just cool toys right now. A bit dangerous though. I shorted the wires of the pair when charging it from the small solar panel, and the entire lug at the end of one wire completely disappeared in a green flash of copper vapour discharge light.

Recycling Grey Water for the Garden

2011-05-02T18:41:00.000+01:00

The other thing the Met Office said about April was that it was also the driest April for many years, with some parts of the UK receiving only 2mm of rain.

Despite having bought a second water butt for the garden last year to collect rain water, it still wasn't enough this month.

We've previously filled the water butt using saved bath water, but it was a troublesome process, and the cheap drill powered pump I had wasn't very good. In short, it was a... pain in the butt :D.

With more dry weather forecast, it was time to do some bodging...

First of all, you have to collect your grey water. This is water that isn't full of bits (mud, rocks, etc) or full of chemicals or organic matter. So water from the kitchen is no good (contains fats and food bits and so on). Washing machine water can be ok, but often has too much soap, fabric conditioner or even bleach in it. Used bath water is fine. Just don't use any bubble bath, bath salt, or go mad with the shower gel and shampoo or conditioner. The plants don't mind a little bit of soap.

Then you need to catch all the hair and bits of fluff in the water, as the pump won't like it (even with an inlet filter) and you don't want to clog the water butt or hoses either.

The missus had this handy tool, just for the job.

Dunno if you can get them here, as she bought this one at a 100 Yen shop in Japan, where they are very common bathroom accessories. You can easily make one with a coat hanger and a bit of old net curtain though.

Incidentally, in Japan, it's actually quite common to recycle bath water into the washing machine for the wash cycle, as the soap powder will suspend the dirt in the water anyway. Clean water is only needed for the rinse cycle.

Next, you need a water pump. I'd been looking around for something better than the drill pump (a water pump you attach to a cordless electric drill). I found this new one by the garden hose makers, Hozelock.

It's designed to be used in a water butt to power hose pipes and sprinklers and so on. It also says you can use it for emptying small paddling pools, and recycling bath water. It seems to retail for £70, but with a bit of shopping around, I managed to pick it up for £52.

Water is sucked up from the bottom of the unit through a removable foam in-let filter that catches medium sized (1mm) particles and hair. In the screw-on hose adaptor at the top, it has a finer mesh outlet filter to catch particles as small as maybe 0.2mm in size. You need to use the fine filter if you're using the company's miniature irrigation system that drip feeds plants, or a hose fitting that has small spray holes that could get clogged. If you're using ordinary hose pipe fittings then you can take the fine filter out to get more flow.

The water in our water butt has bits in it anyway, and is only used with a watering can tap or jet hose, so I took the fine filter out.

I rigged up the hose outside, cable tying it to some big electrical cable clips I hammered into the wall. I then used a bit of string to hang the coiled up end loop outside the bathroom window, for when we're not using the hose.

The other end is just coiled up under one of the solar panels at the base of the wall.

When we need to pump water, I just un-roll the bottom end and stick it in the water butt, securing the pipe with a brick... useful things bricks. :D

Then I pull up the other end (using the string from inside), un-roll, and plug it into the pump, which sits in the bath. The pump uses normal Hozelock click-fit connectors, so you can use any normal hose.

The pump is designed to be completely or partially submerged, with the 10m water proof rubber cable going off to the mains remotely.

Make sure you use a RCD breaker on the socket! Don't sit in the bath (or pool) while using the pump, just in case... O_o

Then it's just a case of remotely turning the power on. It draws about 300W, so I ran it from the solar power in the computer room.

The pump proved to be very efficient, quiet and fast. It emptied the bath of about 100 litres of water in 2-3 minutes, and it had enough pressure (1.1 bar / 16 psi) to ensure that the old hose, that has some kinks in it, opened up and worked beautifully.

The only thing that was a bit annoying was that the pump has a small hole on the side of the body, which sprays a jet of water out sideways, while working. This is normal as it's a "vent" to allow air to escape from the pump (part of its self-priming function). That's ok, when it's in a water butt, but it splashed water around the top of the bath tub when the water level got below the height of the hole. Some fettling may be required... A small cowl, or summat, to tame the (quite violent) jet that squirts out.

The pump has 2mm high "feet", and this allows it to suck up water right to the last couple of millimetres of water in the tub, especially if you position it over the bath plug. Very good. They make another pump which is intended for clearing floods, but it can't work in less than about 75mm of water... No good for draining a bath tub!

This water butt pump doesn't have an automatic float cut-off, so you need to watch it, and turn it off as soon as the water has run out, as running the pump dry will damage it. Ideally, you should turn it off just as it is about to start "gurgling" at the bottom of the bath.

If using it in a water butt, they recommend standing the pump on a couple of bricks at the bottom of the tank, so that it is clear of the worst of the sediment that collects at the bottom.

Given that we have grey water available every day, and the pump only cost a few Pounds more than buying yet another 150 litre water butt, it seemed good value. And we get the most out of our metered water.

April 2011 - The Warmest (Sunniest) on Record?

2011-05-01T09:17:00.000+01:00

The Met Office were reporting yesterday that this April may have been the warmest and driest in the UK on record. Judging by the record amount of electricity I harvested, it may also have been the sunniest.

Now this time last year I'd admit that I had less PV installed. The last lot of BP panels were installed in May and I had about 1.50kWp installed in total. But the fact that I have just over 2kWp installed this April doesn't account for the power harvest being 2.9 times higher than the previous year.

The upshot has been that my solar system has offset just over 53% of my total electricity demand for the month, and up to 66% of weekly demand in the best week of the month (the last week). On a daily basis, the best day was the 28th, with over 85% of electrical demand served by solar power.

Quite astonishing, given that I didn't install the upgraded capacity from the recent haul of solar panels until the middle of he month.

The newly upgraded immersion heater controller is also maybe having an impact, as it can more effectively use the spare power available in the system.

Dual Power Immersion Heater

2011-04-27T22:27:00.018+01:00

The automatic load controller for my immersion heater has been working pretty well. It turns on and off with the varying power of the Sun. But it left something to be desired when the battery was starting the absorb cycle.

The battery consumed quite a lot of power, but not all of it. The immersion heater needed 650W to run, so couldn't. The result was a period of under utilised solar power in the late mornings, with a trace that looks like this:

Click to view bigger

By modifying the step down transformer supply for the heater, I created a dual power heater. I step down the AC Voltage from the solar inverter with a 4kVA "tool transformer". It outputs 110V AC from the 230V AC input. This runs the 3kW heater element at just 650W.

Under software control from the PC load manager, the first relay turns the heater on and off, while a new second relay selects the power level of the heater, depending on the solar power available.

I added a pair of 6A diodes in parallel (for power handling, as the peak current when the heater is on is about 8.4A). This converts the 110V AC into half-wave rectified DC. The diodes are rated at 600V so they are pretty bullet proof. This has the effect of reducing the power consumption of the heater from 650W to just 350W; measured with an AC plug-in power meter at the 230V AC input to the transformer.

The thermostat switch on the heater ordinarily wouldn't like DC power, as it would cause arcing when the contacts open, and this would soon destroy the thermostat. But as this is half-wave DC, it still has the periods of zero Voltage in each 50Hz cycle, so the thermostat contacts can open and close as normal without arcing.

I then had to modify the control software to take advantage of the new dual power heater.

I decided to ditch the purely light level & system power level threshold system, for one that attempts to estimate the array power available for driving the heater loads (at two power levels).

Click to view bigger

It has a seed value that is the expected array power (the "system size"). It then applies a self tuning modifier to that base value (plus/minus 200W) and then multiplies that by the measured light strength from the new sensor (as a percentage). This gives me the estimated "array power". The "system power" is the real measured output power from the Morningstar charge controller log data. This includes all loads: battery charge load, other loads (e.g. the fridge), as well as the heater load (if it happens to be on).

By comparing the real power output during times when the battery is likely to be fully loading the array with the estimated "array power", the self tuning parameter adjusts the estimate up or down, so that the estimate gets better. It has some limits set in the routine, so that it does not tune the estimate at very low light levels that would never be enough to drive the heater load. It also has some fuzziness in the tuning so that if it is within 1% of the real power, it stops hunting. If the tuning parameter gets bigger than 200W variance, it starts to modify the base assumption about the "system size", saving the change in a config file for next time the program runs.

Most of the estimate tuning happens in the MPPT bulk charge phase, as that is when the battery will absorb all the power the array can muster, and so the "system power" should equal the estimated "array power". The idea is that the tuning parameter will compensate for the distributed orientation of the panels (some are East-West, some are South, some are at steep angles, some at shallow angles). The system will also "learn" how dirty the array is (if there has been no rain for a while, and dust has collected on the panels).

The final step is that when the battery enters the absorption phase, the program looks at the estimated array power and subtracts the current "system power", which includes all non-heater loads plus charge demand, and calculates the "available power" for running the heater load. If the "available power" is greater than the low (350W) heater setting, but less than the high (650W) heater setting, it turns the heater on, and selects "low power" mode (the diode bypass relay is energised and the normally closed contacts change to be open). If the "available power" is higher than the high power setting, the diode bypass relay is de-energised, the contacts revert to the normally closed position, and the heater receives the full 110 V AC power.

The resulting power utilisation is more even, with the heater able to use low levels of available power and maintain the tank temperature. 350W is enough to very slowly heat the water, or at least compensate for losses through the insulation. It all helps. When it's sunny enough, and the other loads are low enough, the heater can run at full power (650W).

Click to view bigger

The above trace also shows the water heater interacting with the cyclic load of the fridge freezer. While the recorded system power varies considerably, note that the battery is given priority in attaining full charge and holding a steady float Voltage for as long as possible.

During the absorption and float stages of battery charge, the heater decision process also includes some "array power" estimate tuning. If the heater repeatedly has to reduce power to low power, the tuning parameter slowly drifts downwards. If the heater has to be shut off due to low power, the parameter decreases more quickly. As a last resort, if the battery Voltage actually drops below the float threshold set in the load controller, then a much more severe adjustment of the parameter occurs. This behaviour means that on clear sunny days, the heater is given priority and has a tendency to stay on. On days with very changeable weather, the heater progressively errs on the side of caution, becoming less and less likely to turn on and more likely to turn off or remain in low power mode. This favours maintaining the battery charge level.

Use of a half-wave rectifier at high power (350W is a significant load) is normally frowned upon, as it presents a very non-linear AC power load (only half the cycle is used). The plug-in AC meter did show a very bad power factor (PF = 0.5). This would result in power being wasted in the wiring and generator as reactive power (current out of phase with the Voltage). But the 3kW inverter is stable into any power factor load (inductive or capacitive), and in the end, the source of the power is a DC battery or solar panel. With the very large capacitors in the inverter input (for surge delivery), the DC source is not aware of the non-linear AC load, and merely sees a useful reduction in load. The "bad" AC load does consume more of the available VA capacity of the inverter than a good power factor load would, but provided the total VA load is less than the permissible load, no harm is done.

Some More Piccys

2011-04-17T17:13:00.055+01:00

A couple more piccys of my upgraded PV arrays.

As well as the movable ones, pointing largely West to extend the generating day almost to sunset, I also rotated the Sharp array on the garage roof to the S-SW a bit. In the early morning, it didn't generate much, as there is some shading from trees to the West. In the late afternoon, the power would drop off too, as the sun soon goes round to an oblique angle. So, rotating it S-SW a bit flattens out the mid-day power curve of the whole system, and extends the peak power generating time longer into the afternoon.

Breaking the 2kWp Barrier

2011-04-16T10:10:00.044+01:00

Some nice weather this week meant a spot of gardening and some more woodwork...

The other 48Wp amorphous array has been upgraded to 160Wp, using a couple of the new Klearskies panels. That leaves four more lolling about in the living room, waiting for a place to go. :D

The nearer one is the new one, and I move it around in the day, depending on how low the battery was from the previous night, or if the weather is forecast to be sunny only in the morning, or only the afternoon. This brings the total installed capacity up to 2080Wp.

I can't actually use all of it at the same time as the Morningstar TriStar MPPT controller is only rated for 60A at 24V, or some 1600W (they seem to rate it for the float Voltage of 27V on a 24V system). I've got 1740Wp installed (would generate just over 64A), but it points in different directions, so will never make the total power at the same time. Orienting the panels in different directions just extends the time that the whole system runs without using the battery, and gives the battery the best chance of reaching full charge.

It's been incredibly sunny the last few days, and I've been able to move the 160W mobile array to track the sun from the SE in the morning, to the setting Westerly light, further increasing the "battery free" run time.

New Mobile 170Wp Array

2011-04-03T14:45:00.012+01:00

Starting to deploy the new panels I bought... :D

First up is replacing the old 12Wp panels on my mobile array.

Which now looks like this...

The panels are mounted high on the frame so that they aren't shaded by plants growing up at the base or by long shadows from other plant pots on the path.

This new array points W-SW, so that it starts to come into prime generation when parts of the main S facing array are beginning to be shaded by the neighbours conservatory and the garden fence. For now, the other 48Wp Topray array is still on the lawn but now pointing SE to catch the early morning light. The 30Wp plastic pair of Topray panels has been disconnected and I'll probably sell them at a car boot.

To accommodate the extra power generation, I doubled the feed wires (up to about 2.6mmsq). I've also ordered some proper MC4 waterproof connectors. I'm beginning to collect the materials that will be needed to install some of the panels on the roof, where it will be too dangerous to be messing about with chockblock connectors. Much easier just to plug the modules together. I'll also have to buy some proper solar cable that can stand the weather for a long time.

Generating Boost!

2011-04-01T23:37:00.001+01:00

After missing out at the police auction in Bristol that was selling loads of recovered stolen PV dirt cheap, and missing out on some other auctions on line, I got lucky this week...

Eight 80-85W mono panels (650Wp) for £940... £117.50 each!
Two are 85W models (a BP and a very old looking GB Sol) and the rest are pretty new Chinese "Klearskies 2000" 80W models. One had a load of garden fence paint splashed on it but an hour with some water and the Japanese "magic sponge" soon got rid of that. Quite a feat, considering the glass is textured.

Now all I have to do is figure out where to mount them. :D

Strange as it may seem, I'm thinking of selling one of them, as that way I'll have an even number of 80W panels overall. I'd only gone on to eBay to look for one panel to match up to the odd panel I have on the garage (I bought 5 panels last time). Although I could use all eight, it would leave me with an odd panel again and it's too much of a temptation to buy "just one more" to make it an even number.

If I sell one, then I won't have that itch to scratch.

The guy was selling them to pay for his MCS / Part P training and then he'll do his house with a proper grid tied array as his exam install to get certified. So now I'm sponsoring PV installers to get through college and do things properly by bodging my own system... There's an irony.

He donated the suspicious 600W Chinese GTI that he'd used with this lot at the back of his garden.

I had a quick look at the innards and was glad that I did. A capacitor on the output display power supply had swelled up and burst.

If magic smoke hadn't already issued forth, it soon would have. As its only the Watt meter on the output and not part of the inverter itself, I bodged on a similar capacitor (but alas too small to make the meter work, but safe enough to make it not work reliably, if that makes sense).

An initial play with it in the garden yesterday showed that the inverter works, sort of. I hooked up two 80W panels in series (the inverter works from 28-52V DC) and it made some noise and a plug in AC meter said something like 40W output (it wasn't sunny). But then it sort of stopped altogether and restarted if I turned the DC off and on again. Not sure if this thing works properly at all (or ever did...). But as he gave it to me as a freebie, it can't hurt to play with it a bit. On the other hand, he did mention that he used to have a 1kW one, but it exploded.

It has some appeal though... plugging in maybe just 80W of grid tied power to offset the base load that I can't get off grid. I could use the old pair of 40W Kyocera panels on the garage and just put the GTI in the garage, plugged into the spur there. I've some leeches that can't go on solar power (like the clock in the electric cooker, the central heating boiler, and so on). With a little bit of grid tied power I could null most of that energy import out without losing power to the grid (and I can't get paid for or worse, the import meter may count up even when power is flowing out of the house).

Of course the next thing will be arguing with the FIT bods about having expanded my DIY system. They may throw their toys out of the pram and kick me off the FIT scheme altogether (fine by me) or they might agree to pro-rate the kWh payments to exclude the additional capacity. I think they are already calculating it wrong for my existing system; I got my second payment a few weeks ago (only the second statement though as I didn't get approved until December) and it doesn't look quite right.

I'll also now have too much PV for the 60A Tristar controller so I'm going to use some of these panels on a E-W system. I'll get rid of the mickey mouse 12W & 15W amorphous panels and use the new ones to boost morning and late afternoon production, as off-grid needs a long flat power curve rather than lining all 2400Wp up at the noon sun and over loading the controllers and battery. I can't even use 2400W as the immersion heater is geared down to just 650W. But it will mean that I can use that 650W for longer than before and run other things like the fridge longer without hitting the battery at all.

It might even be time to bite the bullet and get a roof rail kit and talk to the BC bods about allowing me to install this kit on the roof, where it belongs.

New Solar Sensor (again...)

2011-03-23T22:59:00.049Z

Lately, the tupperware solar sensor that is supposed to measure the available solar power to control my solar water heater has been playing up. Sometimes reading zero in full sunlight or other wonky values that didn't seem right. It also was too directionally sensitive - a feature of the old solar panel used that has micro lenses on it that focus the light - but only when it shines square on to the panel. Also the box itself was possibly causing some shading or variation in light getting through. On closer inspection, the UV had destroyed the tupperware, making it crack and go brittle.

So the search was on for a small sensor that will be weather proof and less directionally sensitive. The local Robert Dyas had the perfect bit of bodgineering raw material... a £1.49 solar garden path LED lantern spike thingy.

It had a little amorphous solar panel on the top and all the bits came apart easily to leave the plastic and stainless steel capsule on its own. I could just rip out the little circuit board with the LED and battery, and connect the wires from the solar panel across a 100 Ohm resistor load in a chocblock (so that the device measures solar power). Then I just had to hot glue up the holes on the base to prevent water getting in. The hole where the LED came out was just the size to fill with a rubber blanking grommet.

Some of the existing holes, that were used for the clear lantern bit to snap on to, were also handy for threading the plastic cable tie through. I mounted the new capsule sensor on a handy hanging basket bracket that was already on the wall when we moved into the house. This, by sheer coincidence, is at the same angle as the main solar panels.

The amorphous panel has no outer window or lens to restrict the angle of light acceptance and amorphous panels are less sensitive to direction anyway, so it seems to give a good reading through the day.

The output Voltage was a little lower than the old sensor, so I had to recalibrate the measuring software on the load controller to get a proper 100% reading in full sunlight. In the end, I decided it would be easier to modify the software to have a user parameter for the sensor scaling, avoiding the need for recompiling the program just to change the value. Amorphous panels put out up to 20% more power when new, but quickly settle down to their usual power when exposed to the sun for a few weeks, so it made sense to change the software to allow tuning the sensor. I even added new today and yesterday counters for the heater run time and estimated kWh of DC power generated.

I've even got the (still working) white LED, solar controller chip, NiMH coin cell and on/off switch gubbins that I can play with. Most of that would be worth at least £2-3 if I'd bought it from Maplins. So the solar panel was actually "free".

Oh, yeah... Like my new multimeter? It's an antique Micronta (Tandy) analogue (but with FET inputs) 1980's test meter that I picked up at a car boot the other week. Nowhere as accurate or convenient as a DMM but looks "retro-cool".

The new sensor has been working better than I expected, and now the water heater works really well. Getting good heating yields but not cycling the battery much at all.

Here you can see the results of a totally blue sky day on the right of the graphs... Even on the not so good day, the battery Voltage holds up pretty well with the more accurate solar power availability estimate. Click on the chart below to see it in full size.

Today's trace (on the right) was the first full solar hot water day of the year. You can just see there were no clouds at all by the totally smooth sensor trace. No gas used today, and 48'C water in the tank :D.

Bicycle Generator (part 2)

2011-02-09T20:37:00.101Z

So, with the problems of the Mark II bike, I had to adapt it again so that I could make some useful power to charge the house battery.

The Mark III bike uses a different approach to powering the stator field coil. I got some old NiCd batteries from a dead cordless drill and laid them out on a bit of wood.

I bought and cannibalised a car power extension lead so that I could connect the battery to the generator, using a plug and socket. The plug was handy as it has a fuse in it (so I didn't need the breaker in the photo above), and it also has a green LED that gives you some idea of when the battery is getting discharged, glowing dimly when the battery is low. I just tied the plug to the bike handlebars, making it easy to plug in and switch on.

By charging the batteries from the solar power, and then using the stored fixed Voltage to power the stator coils, the generator can start by itself, and you get a current and Voltage output from it that is now directly proportional to how fast / hard you pedal.

At first, I tried using a 12V battery, but the stator winding is only something like 1.8 Ohms, so it took too much current. By using a string of six NiCd cells, I could get an ideal 7.2V at about 4A going through the stator coil. The NiCd battery lasts about 15 minutes at that rate, and this also is about as long as I can cycle continuously :D.

The bike can make anywhere from 30 to about 70V, if you pedal really fast. The Morningstar Tristar MPPT solar charge controller can accept any Voltage up to 150V DC to charge the 24V battery, and it will convert higher Voltages to the right Voltage, stepping up the current in the process. This means that I can pedal fast, and the bike will generate maybe 45-55V at a a low current of around 1.0-1.5A. Most of the time I pedal more slowly and it makes about 32-35V at about 2A. This keeps the motor brushes and windings cool, reduces the losses on the long wire run to the house and the battery still gets all the power generated (well about 95% as some is lost in the conversion).

I was a bit worried that the MPPT charge controller is not designed to work with generators (like wind turbines or bikes) as it periodically unloads the input and does a sweep to measure the maximum power Voltage of the solar array. This could cause a wind turbine, or my bike generator, to suddenly become unloaded and over spin (with the result of me falling off the bike again :/ ). Happily, when I tried it out, the sweeps don't happen very often, and are very short on this controller (just a fraction of a second). The motor does over spin a little, but not so much as to be a problem because the bike has a heavy fly wheel that limits the acceleration in that split second.

To smooth out any Voltage spikes, I set a pair of output diodes and a high Voltage capacitor in the circuit so that charge from the motor is pumped into the capacitor. Then the wiring from the generator just joins the rest of the solar panel wiring network in the garden.

The first capacitor I used was a spare 50V one that was laying about in my "lab"; It got over loaded and exploded during a "sprint" :D. So I bought a new 100V rated one, and (so far) that one has survived, but now I also have a Volt meter to keep an eye on the output level.

Here is the first power trace on the main system from my bike generator. The Morningstar MSView software logger clearly shows me pedalling and generating a steady 60W of power.

To provide a nice finishing touch, I bought a couple of analogue meters from eBay.

One is a 0-3A moving iron Ammeter for school labs and the other is actually an antique Bakelite 0-7 Volt meter. I just worked out what extra series resistance the meter needed in order to convert it to read 0-70V instead. This scale gives a good indication of the bike output, which is in the 30-55V range, but can surge to about 65V if the charge controller does a measurement sweep while I'm pedalling hard and the generator over spins a bit.

Now all I need is some decent leg muscles...

Bicycle Generator (part 1)

2011-02-08T18:26:00.002Z

A while ago, I started playing with making an exercise bicycle generator.

I got an unwanted and unloved exercise bike from a car boot sale for £5 and a car alternator from another car boot sale for £10. The result was the Mark I generator bike.

The problem with this design was that it only worked at 12V and house battery is 24V, so it was only good for charging little 12V batteries or running a light in the garage. Alternators are also designed to be powered by a car engine and not a bicycle so they need to spin quite fast before they generate any power. When they do start to generate, they use quite a lot of the power to create the magnetic field required and so you waste quite a lot of your energy just making the thing "charge up". The alternator also had a Voltage regulator that limited the output to 14.5V, which was a bit too low to charge a battery at the end of a long wire, that suffered a Voltage drop from the wire resistance.

Later, at another car boot sale, I found an old washing machine motor and matching drive belt for 50p. This is then the basis of the Mark II bike.

The washing machine motor works at 230V but is a 16 pole DC motor, controlled by a microprocessor in the washing machine. Using the motor as a generator, the rotor, with its 16 pole winding on the rotor and a big stator coil generates fairly smooth DC power. It also has a smaller pulley on the rotor than the alternator, which increased the gear ratio to about 30:1, meaning I didn't have to pedal so fast to make the generator turn at the same speed.

I discovered that the laminated iron core retained some weak magnetism and this was enough to generate a small current in the rotor winding. By connecting the rotor and stator windings in series with the load, there is a positive feedback loop.

The small residual magnetic field in the core is enough to start a tiny current flowing in the rotor. This current passes though the stator coil and the load. The current flowing in the stator coil causes a stronger magnetic field to be created, which in turn causes more rotor current to be generated. The system has positive feedback, and if the output of the motor generator has a low resistance load, it can quickly cause very large currents and high Voltages to be produced. The only limit would be the input mechanical power and whether the load or the rotor windings and carbon brushes catch fire!

Below is a diagram of the circuit.

I use a push switch to start the generator by connecting a 21W 12V lamp or a 4 Ohm power resistor directly across the output and start to pedal. When the generated Voltage gets above the main battery Voltage plus the diode drop of 0.8V, the current starts to charge the battery and I can release the push button. The current going to the battery now sustains the generator stator field, and the generator will make as much power as you can supply. The generator will easily make 60-70V, so long wiring to the battery is not an issue.

The only problem is that the power and Voltage output can vary wildly, and the effort to pedal quickly gets too hard. But if you slow down, even for a second, the battery current may stop (if the generator Voltage falls below the battery Voltage) and then the generator suddenly stops working and as you are pedalling hard, the load is suddenly removed and you fall off the bike! :D It also runs the risk of a Voltage spike damaging anything else attached to the battery (like my inverter)... efficient but dangerous.

Active Battery Balancing

2011-01-29T17:14:00.033Z

A while back, I had a problem with the AGM battery bank getting out of balance. This would have damaged it by having one 12V battery under charge all the time, while the other one over charged and gassed (eventually drying up and being killed).

Luckily, I found a fairly cheap (£29) battery balancer that has fixed this.

This one is from Rapid Electronics, made by a British company called Camden Boss.
You connect the balancer to the 24V + terminal and the 0V - terminal and the yellow wire to the 12V middle point. It then works when each battery is over 12.8V (when being charged) and shunts up to 1A of current across either battery, to keep the mid-point at exactly half the full terminal Voltage, ensuring that the batteries charge evenly and fully.

I actually ordered two of these but one hasn't turned up yet, on back order from the maker.

It gets a bit warm when it's working but as the batteries get to be equal in charge, it gradually stops taking power and cools down. So it wastes a bit of solar power by dumping it as heat, but that's better than a fried battery :/

When you've discharging the batteries, they are below 12.8V each and so the balancer does not work and draws no power.

In order for the balancer to work across all the batteries in the AGM bank, I installed the equalisation network that I said I was going to install months ago (but was too lazy to actually do :D ).

It has fuses in the links so that if there is a problem with one string of batteries or the balancing current is too big, the fuse will prevent a melt-down. It doesn't matter so much that the wires are not going to a star point as the current across the links should be zero (or close to it over time). The battery balancing module only moves 1A so that doesn't cause any real Voltage drop either.

The whole thing seems to be behaving itself perfectly and the six batteries now charge much more evenly and fully than before.

No International Battery. Hello GWL!

2011-01-28T15:58:00.004Z

Well the International Battery guys totally ignored me, so I'll return the favour as a lost customer.

But later I discovered a EU stockist of the Chinese large EV batteries that Thundersky (now renamed Winston Battery) and they were happy to answer questions. They reckon that the Moriningstar controller, suitably programmed, can safely charge their lithium iron yttrium phosphate (LiFeYPO4) cells. They also have a web shop with all the prices and ordering for any user (not just big car manufacturers).

http://www.ev-power.eu

Fuelly... Track and share your fuel figures

2011-01-15T18:05:00.000Z

Like the new gadget on my page?

Click on it and you'll be taken to the Fuelly.com web site where you can sign up for a free fuel economy tracking database. You can also compare how well your driving and car fare against other cars and drivers.

Can you beat my high-score?

2010 10:10 Challenge Results

2011-01-07T18:19:00.000Z

So 2010 came to a close. How did I do on my 10:10 challenge?

.                  2009 kWh  Demand %  2010 kWh  Demand %
Electricity (grid) 4379      94.2%     3448      83.0%
Electricity (solar) 269       5.8%      705      17.0%
Total Demand        4648                4153

So, from 2009 to 2010, I reduced my total demand for electricity by 10.6%. This was due to better efficiency in use of power (turning things off at the wall to prevent standby power leeching, running a laptop as a server instead of a desktop PC, using the coldest setting on the washing machine, etc.).

Looking at grid power only, I reduced my consumption by over 21%, compared to 2009 levels. This was largely driven by steadily increasing solar power capacity over late 2009 and early 2010, as I bought more panels, bigger charge controllers and bigger batteries to capture and store more power.

But that's only half the story. We use gas to heat the house and provide hot water. In 2010, I started using spare solar power in the Summer to pre-heat the water with the immersion heater; reducing the need for gas in the evenings. But even with the additional wall and roof insulation, the record-breaking severe cold weather we had in early and late 2010 pushed the total demand for gas up by 16.8% (10,295kWh in 2009, to 12,025kWh in 2010). So there's still work to be done.

Our windows aren't very good. The old double glazing isn't very good quality (old types have thin air gaps) and there are three small port-hole windows in the upstairs landing and bathroom that are only single glazed. Our boiler is very old and should be upgraded to a condensing type. We could build a front door porch so that when we open the front door, there's an 'air lock' to prevent cold air blowing into the hall and through the house.

More projects for 2011...

Slightly Niffy Battery...

2010-11-30T23:13:00.030Z

What's that nasty niff?

Oh... That's the unmistakeble smell of battery gas. Not good when you're running a bunch of "sealed for life" gel and AGM batteries.

Closer inspection revealed two of the AGM Marathon batteries were way out of balance under charge. I'd been cutting back on usage to get the bank fully charged (as you never know when it will be able to charge again in these gloomy and now snowy days). The pair in question have some kind of lost capacity problem. One is getting full or has a higher internal resistance than the other and so the two 12V batteries are out of whack, with one sitting at 14.4V; boiling up and gurgling (never a good sound from an AGM pack), and the other is languishing at 13.2V; not even making the float Voltage... When operating at partial charge, they bumble along without one gassing, but the partial charge will soon destroy all the batteries in the bank; not just these misbehaving twins.

Probably the only thing for it is to take them out of service (dropping my bank capacity from 495Ah to 395Ah), and use them as spares for a 12V system, where they can run in parallel (and not suffer the series Voltage imbalance stress), or be used just one at a time. I took them out of service for one afternoon to load test and, individually, they seem ok. They just can't "get along" as a series pair.

The only other alternative would be to invent some kind of active pack balancing electronics to bypass some of the current on one or t'other battery when charging; a-la lithium ion packs that have individual cell balancing. Lead acid batteries (especially sealed ones) would last a lot longer if the makers introduced similar Battery Management Systems (BMS) that makers of more flamable cells are forced to.

Heaven only knows how you manage to keep a 48V AGM pack from self-destructing from imbalance. Most fork lift packs suvive because the passive balancing they employ is exactly allowing some cells to overcharge and "blow off a little steam", so to speak.

On the upside, I've just started making enquiries about a new 24V 160Ah lithium iron phosphate battery module from an American outfit that makes large format lithium cells. Other makers put together packs using small (3-10Ah) cylindrical cells in big blocks, with a BMS to keep them in check. International Battery make these new HUGE 160Ah-200Ah cells, so the BMS is simpler and the construction is simpler (and cheaper), as it only has 8 cells connected in series to give a 24V pack. A lithium pack should offer much more usable power, as you can run it in partial charge mode all the time, and it won't die like a lead acid pack. In fact, lithium chemistry prefers not to be kept fully charged for long periods. Ideal for opportunist solar charging. Operating in partial charge mode also means not wasting solar power when your battery has gotten full and can't accept any more charge. If your battery gets full every day, it's too small for the array. But lead acid chemistry demands that you keep the battery as close to fully charged as possible all the time.

The specs for the pack look promising. Min working Voltage is 20.0V (the inverter cuts out at 21.0V). Max working Voltage during charge is 29.0V and the inverter can work up to 30.5V before cutting out.

I'm hoping they'll say that I can use my existing Morningstar charge controllers as they can be programmed to provide Voltage phases (as in bulk, absorption and float phases with different Voltage limits) or even a single Voltage limited output (float only) with a current limit. Over-current shouldn't be an issue though as the cells are rated for 1C discharge (160A) and C/2 charge (80A). The inverter load at 3kW tops out at 125A and the most I've ever seen from my solar array was 73A charge.

Check it out here:
http://www.internationalbattery.com/news_nov_01_2010.php

Timber!

2010-10-21T18:12:00.005+01:00

Trees. Good for the environment but no good for your solar power.

In the winter, my panels used to get shaded by a bunch of trees near by, as the Sun made its low 15 degree arc above the horizon.

Well, now a big gap has opened up in the horizon! For years, there has been this nearly dead fir tree in someone's garden, on the other side of the courtyard, blotting out the late morning sun in our garden. But yesterday, unannounced, a van turned up with some tree surgeons and a massive shredder.

It took about a day and half, but the dead tree was gradually chopped up and shredded, leaving a big new window on the horizon - and no doubt a few extra welcome Watt-hours on the clock this winter.

Time lapse over two days

Dodgy Plastic Panels...

2010-10-03T11:30:00.000+01:00

Hmmm... I'm glad I sold off most of the plastic framed amorphous panels. One of the two remaining ones broke the other day. It suddenly went open circuit. Rather than bin it, I decided to take it to bits to see how they were put together.

The black caps on the back hide a load of small screws that hold the thing together. Luckily, the sealant that was under the cover wasn't glue so I could easily take the back off of the panel. Under the silicone sealant on the positive lead, the end contact on the glass had broken off and the red wire came away with the sealant it was embedded in. This was possibly from thermal stress as I remember hearing the frames creaking in the sun and shade.

There was a thin stub of copper poking out from under the laminate. The panel seems to be made from a sandwich of two plates of glass (presumably with one having the amorphous cells evaporated on to it). I just about managed to solder a thin wire on to this and then joined it to a new thicker wire and potted the whole thing in silicone sealant again before screwing the back on. How long will it last? Who knows...

To save propping up the thing on a garden chair, I made up a rear leg, just fixed on with the usual brass hinge.

Now that it's getting darker in the days, it's actually better to have the amorphous panels at a shallow angle as they produce more power from just "seeing" more diffuse light from a bigger proportion of the sky than if you aim them at the non-existent Sun.

A Nice Solar Powered Office

2010-09-17T21:56:00.000+01:00

Coming back home from work out in Green Park on the edge of Reading, I noticed one of the new office buildings that they have recently completed. No one has moved in yet but it seems finished.

The outside wall has a bunch of solar panels integrated into the design. Looks good but in mid-summer when the sun is directly overhead, it won't make much power using these polysilicon modules. Amorphous would have been better. Or forego the "look at me!" architects design and put the panels unobtrusively (and more efficiently) on the roof.

Still, they meant well.

You can see my Honda Insight parked outside the "Eco-Office" for added effect... Now all I need is a plug-in hybrid and somewhere to plug it into that free electricity the building is making...

Save Fuel, Drive Smart

2010-09-14T19:11:00.000+01:00

One of the biggest consumers of energy is our cars.

The best thing would be to not have one but that isn't practical in my line of work, having to visit clients and far flung offices at odd hours of the day and night.

Still, I try not to use my car if possible. I spend quite a lot of time working at home. No fuel used for days at a time. But I do often have to go to an office in Reading, some 65 miles away.

I'd normally go there by car but recently I had cause to go there by public transport (as some low life crashed into my car at the Reading M4 junction and ran off without stopping - GRRR!!!!). Two trains and a bus and over 2 hours each way weren't ideal (especially when I discovered the Reading night bus doesn't accept the return leg of your train "Plus Bus" ticket and you have to buy another ticket).
Any way, I digress...
I normally drive a Honda Insight, a cheaper cousin to the Toyota Prius, in that it is a hybrid petrol/electric car. This car has a driver feedback system that teaches you to drive more efficiently by showing you a colour change from green to blue in different shades as you waste more fuel by "lead-footing". But the lessons learned by driving the Insight are applicable to any car.

While my car was in the workshop, the dealer loaned me a little Honda Jazz. To my surprise, I found that it could get very good mileage by careful driving. It's a smaller car than the Insight and has a smaller 1.2L engine, which helped in traffic jams (less fuel wasted when idling). I could get it up to about 62mpg on the run to Reading. This is diesel territory normally and excellent for a petrol car. The Insight can do 72mpg on the same run.

The main lessons that need to be applied to get these sorts of returns are:

Don't overload the car - remove all junk from the boot, travel light. It costs fuel to accelerate a mass. The heavier the mass the more fuel. So ditch the weight to save the fuel. If your favourite petrol station is on your route, consider filling up only half a tank twice a week rather than a full tank once a week. Carrying a full tank of fuel around weighs a lot and wastes fuel.

Look ahead MUCH furter than you're used to and try to drive without using the accelerator or brake. An odd suggestion, but one that saves fuel and makes you a safer driver.

Don't accelerate agressively - use a light touch and change speed gradually. Don't accelerate to a fast speed if you can see that you'll have to stop or slow down seconds later.

Don't brake unnecessarily - stopping wastes your stored kinetic energy in the moving car and then you have to waste fuel to speed up again. If you can see far enough ahead, let the car slow naturally and with luck the lights will change before you get there. Accelerating from a rolling start uses less fuel than starting from a stop.

When driving a hybrid, the last point is a big factor as the electric motor acts as a generator when slowing down or braking. It recovers some of the kinetic energy that would have been lost in the brakes and stores it in the battery for use when speeding up again. In a non-hybrid car it still makes sense to preserve the kinetic energy you have, rather than waste it in heating the brakes.

Drive slowly - no, really... I mean it. Unless you are in a hurry, drive at the most efficient speed for the engine and gear box. If you are always in a hurry, leave earlier! Change up gear as soon as possible, keeping the RPM counter below 2500. The Jazz had a shift light for the manual gears to prompt you to change up and it's earlier than you'd think. The Jazz did well for such a small engined car - 53mph at 2300 RPM. The Insight has "long legs" with the CVT only making 1600 RPM at 53mph.

Air resistance is much lower at slower speeds too - An average car only needs 10hp to drive at 50mph but at 100mph it requires 80hp just to overcome the air resistance. I've noticed that Tesco lorries have reduced their regulated speed from the usual 56mph to 52mph. This must be to save the company fuel costs as the frontal area of a truck is huge and the aerodymanics are bad so the air resistance is very high.

Consider the trip average speed. It's meaningless to drive round the M25 in the morning at 70mph between traffic jams where you travel at 15mph. You'll just get to the next traffic jam faster. For instance, I know there's no point in driving at 70mph until I get past the M3 going round the M25 so I drive at 53mph max until then (and sit in the traffic jams 5 cars behind the guy who sped past me earlier). Once past the jams I'll drive at up to 70mph, as I know it will actually make a difference to my arrival time.

If you get through the jams more quickly than usual, drive more slowly afterwards. A sat nav will tell you your estimated time of arrival and "range to target". Use that information to adjust your speed accordingly!

Drive up steep hills more slowly than when on the flat or going down hill (especially on motorways)... Stressing the engine to maintain high speed on a hill wastes a lot of fuel with little saving in time.

Chose a motorway route if available. Motorways allow for constant speed driving, increasing efficiency. Use the cruise control if you have one. It maintains your speed with much greater accuracy than your right foot can and saves fuel. You can't use cruise control on a twisting A road and you have to stop for round abouts and so on. Plus motorways allow other faster traffic to go harmlessly around you. A roads suffer from "caravan syndrome". Even if a motorway route is slightly longer it may be actually more fuel efficient.

Don't put anything on the roof! Roof bars, boxes and so on increase air resistance significantly.

Inflate your tyres properly. Under inflated tyres waste energy as the rubber side walls flex and generate heat and can be dangerous at motorway speeds, increasing the risk of a blow-out.

Have your car serviced - make sure fuel and air filters are replaced to avoid wasting fuel from poor air-fuel mixture ratio. Use the thinnest grade of engine oil suitable to the climate you live in (the manual usually suggests a range of different oils for different climates). Thinner oil increases engine efficiency. Change the oil regularly - old oil gets thick with goo and doesn't protect the engine well.

Where's the solar power relevence of this post then? The Insight is big enough to ferry my panels from their various original homes to their new one... using the least fuel possible.

Updated Monitoring Network

2010-08-15T02:09:00.000+01:00

I finally got round to completing the Morningstar monitoring network. The problems with the Toshiba laptop and Vista and the clunkiness of having to start up the machine and configure the logger and so on each day (to save wasting too much power at night) got to me. Below is an example of the type of connection that can be made between Morningstar products. I don't have a Relay Driver but it shows that you can connect a bunch of things to the EIA485 data bus.

The Morningstar SunSaver MPPT charge controller can talk to the TriStar controller if it has the right adapters. I already had a RS232 adapter to connect the SunSaver to the laptop but you need an EIA485 converter to connect it to the same port on the TriStar.

Although EIA485 isn't common in day to day PC networking, it is common in industrial telemetry as it can work over a four wire bus and transmit up to 1.2km without repeaters. RS232 is only good for a few meters and even Ethernet runs out of steam at 100m. It's a bit of a pfaff when you only want to connect two things together by a 30cm bit of string though...

The instructions suggest using Cat 5 Ethernet cable for the 4 wire bus. Two are used for +12V and Ground power lines and the other two are the A and B serial data lines. They suggested using Ethernet cables because they have twisted pairs to eliminate interference. But I ignored that and just used 4 core flat telephone wire as it was such a short cable I was making. They also say that you're supposed to terminate the A and B wires at each end of the bus with a 100 Ohm resistor between them. I sort of did this by putting a 100 Ohm resistor at one end (inside the Tristar wiring box where it would be safe from being knocked). I didn't have two 100 Ohm resistors in my spares box so I didn't bother with the one at the other end... Seems to work ok without it. I've been receiving data from the SunSaver without problem so again it's probably only important for long wire runs.

With the SunSaver connected to the TriStar, you can use the MSView software to talk to any device on the EIA485 bus (up to 128 devices) using IP. The TriStar can be set to bridge the Ethernet and EIA485 networks (it just forwards the MODBUS packets to the devices on the EIA485 network). It's especially handy because with the exception of the TriStar MPPT-60, no other Morningstar products are IP enabled but now they can be. It opens up the possibility of using cheap and commonly available networking products to move the data around. For instance, I don't have Ethernet cable run around the house. I have a wi-fi transmitter and the TriStar is actually connected to the computer upstairs via an Ethernet switch and a pair of Netgear Ethernet over AC power adapters. This was more reliable than the wi-fi (which has a dead spot in that part of the house) and the CCTV data also travels over that switch and link.

So now my laptop upstairs can fetch performance data from the Tristar controller and also directly from the SunSaver controller. The two controllers share the same IP address on the Ethernet but have different MODBUS IDs so each controller responds only to it's own commands.

You have to run the custom settings wizard for each controller to change the default MODBUS ID (all Morningstar devices are preset to "1"). So I changed the SunSaver controller to be ID "2" and left the TriStar on its default of "1".

The EIA485 network needs external power of 12V DC so I connected the two power pins on the EIA485 connector plug to the variable lab power supply that I use to run the AA battery charger and mobile phone chargers. This converts the battery 24V to the needed 12V, although the EIA485 adapter isn't fussy and will work on any voltage between 8 and 16V, so you could just connect it to a 12V solar battery directly. But I'm running a 24V system so I have to use a DC-DC converter. The pair of adapters (RS232 and EIA485) together only consume about 20mA (or 0.5W power) so I can leave them running 24 hours (unlike the Toshiba laptop that consumed about 11W.

10:10 - Mission Accomplished

2010-08-03T00:55:00.003+01:00

It's taken a while, but I've reached a milestone. I signed up for the 10:10 challenge to reduce my electricity use by 10% in 2010.

June and July were bumper harvest months, with my solar system making a record 45% of my total demand for July. The running total - the Lifetime Contribution (blue line) that my evolving solar system has made since I started this experiment is now just over 10% of total demand (and climbing fast now). The green columns are daily solar kWh consumed, showing a good run of sunny days this summer with many days delivering 5-6 kWh.

Make Windows 98 See Vista / Win7 Shares

2010-07-20T18:09:00.000+01:00

Sneaky Micro$oft... In the move to Vista and Windows 7, they broke connectivity to Windows 98 clients without telling anyone.

When you install your file shares on a new Vista or Windows 7 machine you'll discover a weird error message on your Windows 98 machines when trying to connect to file shares on the new machines. It just keeps saying that the password for a share called IPC$ is no good.

My charge controller data logger is a Windows 98SE Toshiba laptop that can't be upgraded to XP. Although I could use a newer machine, I like the Tosh because it only uses 11W of power to do its work and doesn't need a fan to keep cool so it will work quietly in the corner of the living room.

Fear not. I've found out how to get around the problem.

The basic problem is that after XP, MS changed the logon authentication method to improve security. Windows 98 uses LM authentication. XP uses NTLM1 and LM authentication so it was backwards compatible. Vista only uses NTLM2 authentication (but it was ok because XP machines got upgraded automatically to NTLM2 by automatic update). Poor old Win98 got left behind.

However, what you need is the Directory Services Client package from the Windows 2000 Server (from a server or the server CD or as a download from Microsoft). This allows Win98 machines to log on to Windows 2000 domains but also includes the updates to support NTLM2 authentication. Download DSClient.exe and install the package by running it. Make sure you get the Win9x version (there was a NT4 version with the same file name).

That's the first step. After installing it you need to enable NTLM2 on the Win98 machine. To do this you need to edit the registry to add a new value to a key.

Run regedit and navigate to:

HKEY_LOCAL_MACHINE\System\CurrentControlSet\Control\LSA

Add a new value called LMCompatibility of type DWORD and value 3

Setting this value to 3 tells Windows to use NTLM2 authentication. Reboot.

Now, Windows 98 can't log on to shares using a different user name (XP introduced the "connect using different user credentials" option) so you need to make sure your Win98 user and password is the same as a local account on the Vista machine that you've given permission to use the shared folder. This is the same as normal share set-up.

On the Vista machine you also need to edit the registry to change the hashing of LM credentials. Otherwise, you'll be able to see the share but not be able to access any file from the Win98 machine (if you try, you'll wait for a few minutes and Windows 98 will say something like "the network resource is no longer available").

Run regedit on the Vista machine and navigate to:

HKEY_LOCAL_MACHINE\System\CurrentControlSet\Control\LSA

There should already be a value called nolmhash set to a value of 1. Change it to 0.

This will make Vista file shares less secure but only to the same degree that XP was.

The final step is that on the Vista machine you have to reset the password for the local user account that you will be using on the Win98 machine to access the share. You can set the new password to be the same as the old one but by going through the change process, it changes the way the new password is decoded by Vista (to the old method used by XP). Now you should be able to browse the network neighbourhood to the Vista machine from the Win98 machine and access files on the share.

I still had some weirdness after this in that I could write new files and rename existing files on the share but not open existing files for reading or copy a file from the share to my Win98 local disk. Some kind of permissions problem on the Vista machine but in my case I only needed to write logged data to the share and rename yesterday's file (and I could do this) so I wasn't going to fret too much about not being able to read/copy files.

There were some bugs in the Win9x DSClient package originally and these did apparently get fixed in a Windows 2000 Server service pack but you can't download this updated DSClient package separately for some reason (unless you have a Windows 2000 Server and get it in the service pack).

So for now (at least), my Windows 98 laptop lives on to fight another day!

Saved From the Dustbin

2010-07-19T11:33:00.000+01:00

In a move to reduce the power consumption of my "server" room upstairs, I replaced the XP desktop machine that was doing file serving, CCTV recording and mail filtering with a Vista laptop I picked up at a car boot sale for £20. It was a bit busted, the power supply was lost along with the battery, and the OS was trashed but as a server it doesn't need all the keys on the keyboard to work, doesn't need a battery as it will live exclusively on the mains and I found a matching power supply at the same car boot sale for £1. Best of all, the hidden recovery partition was still in tact on the hard disk so I could re-image the machine and now it works fine. It consumes a mere 25W of power compared to the 110W of the old desktop, greatly extending the time the server can run on solar power.

My wife's laptop recently expired from overheating - a design defect in the HP G60 series :( but the memory from it was compatible with the car boot machine so it now has the salvaged RAM from the dead HP.

The HP may yet live though. In the spirit of "make do and mend" to reduce the waste of throwing things away that can be fixed, I found a company that specialises in repairing the G60. It usually dies because the graphics processor had a poorly designed heat sink. It gets too hot and the the solder joints on the chip fail. The company resolders the chip on the motherboard and then fits a new customised copper heat sink and an upgraded BIOS that makes the fan run faster all the time to prevent the problem recurring.

The operation runs from a garage in Thatcham and I was shocked to be met at the door of an ordinary suburban house by a man in a white coat! He showed me the "lab" and the stack of machines awaiting repair along with a couple on soak test after being fixed. At £50, it will be an economic repair (if it works) and I'll have saved a machine from becoming unecessary land fill fodder. If it doesn't work, they only make a charge for return postage so I'll only have lost £10. I'll scavenge the hard disk for the new server (as the HP one is twice as big) and sell off the carcase to repairers who can use the LCD screen and plastic body parts to refurbish other G60's. There's a market for them on eBay...

"Waste not, want not..."

She Cannae Take the Power Cap'n!

2010-06-22T22:19:00.000+01:00

Hurrumph....

Automotive blade fuses... PAH!

Don't use 'em.

I had my new batteries connected to the main bank with an in-line blade fuse holder and a 30A fuse. Not a cheap-o rubbish fuse holder but an RS one. It was the day after the longest day today and the sun was beating down after I'd more than half drained the batteries watching movies last night. Pulled 6.4kWh from the system :D.

Today though, the secondary bank was very greedy and spent quite some time sat at about 25-30A charging. The poor little blade fuses couldn't hack it. Two of them melted (the plastic holders - not the fuse wire!).

So I replaced the in-line fuse holder with an old 32 Amp MCB breaker that I got at a car boot sale a while ago. It's better that way any way as it can disconnect the battery if there's a fault and it also serves as an isolation switch.

The rest of the afternoon's sun was soaked up without further melting or burning of plastic / rubber / fingers...

More on Balancing Batteries

2010-06-16T20:35:00.000+01:00

Googling around the last couple of days I discovered that Elecsol (well known maker of leisure batteries) has released a new range of sealed VRLA AGM batteries primarily aimed at solar storage.

They make some impressive claims like that these batteries can deliver 1,100 100% discharges and 1,400 80% discharges. They offer a 7 year "unlimited" warranty on them.

You can read the blurb on them here: http://www.elecsolbatteries.com/literature/

The tech book / brochure did have an interesting addition to the traditional star pattern series-parallel wiring scheme. It had not occurred to me to put balancer cables between the mid-points of the series strings to allow internal equalisation of the bank. I'd had a problem with one of the six new batteries in my bank being low (out of balance with the other one in series) but I cured it by taking that pair out of circuit and slow charging the weak one to bring it up and then put them back in circuit. The whole bank appears to be behaving ok now but I periodically measure the volt differences on each block.

I might put a star equaliser network in (connecting the mid points of all three pairs to a common point so they will equalise er... equally).

The red/black lines are the power lines I have on my bank and the green ones would be the equalisation network that would allow all the weak batteries in pairs to charge up more without over charging the stronger ones. In my "bad" pair they showed 28.2V across the pair but one was 13.8 and the other was 14.4. With the equaliser network added, the weak "bottom side" battery might continue to charge by "finding" another weak "top side" battery to pass current through (13.8 + 13.8 is only 27.6 so that pair would continue to charge without over charging the others sitting at 14.1V). It wouldn't help in every scenario though - if all the bottom side batteries were weak and all the top side batteries were strong then the equaliser network wouldn't help.

Maybe it's just as well to rotate batteries in strings (like tyres on a car)?

Meanwhile, it's been a great couple of days harvesting. In those partial cloud surges that I mentioned before, I've seen surges up to 1.98kW from my 1.8kW array (over 109% of rated power). The Sharp ND170 340Wp string stole the show though, putting out 409W (120% of rated power). They might have managed more but the charge controller capped the output at 15 Amps! The other charge controller was also close to capping its output to the 60 Amp limit, as it surged to over 57 Amps - the pair pumping an eye-watering 72 Amps into the battery bank.

Balancing Batteries

2010-06-11T21:33:00.000+01:00

When connecting batteries in parallel to get more capacity, you have to be careful to balance the individual strings of batteries so that they all do the same amount of work. Otherwise, some will become more discharged than others and suffer damage from hard lead sulphate forming on the plates that cannot be dissolved by charging.

Having batteries of different sizes is also not a good idea as the small ones will discharge to a greater extent than the bigger ones and so suffer again.

So what did I do? I bought a bunch of different sizes and types of battery and connected them all up together... Go figure.

So, how do we bodge this so that the batteries all stand a chance of surviving?

The main block of batteries are gel types and 180Ah in size. They have a rating of 1,000 cycles at 50% depth of discharge (the deeper you discharge a battery, the fewer times you can do it). The Marathon ones I've just bought are AGM types and 104Ah in size. They are intended for computer uninterruptible power supplies (UPS) and as such are not expected to be discharged every day (if ever). They probably have a rating of 250 cycles to 50% discharge. But if discharged by only say 25%, they might last nearly 1,000 cycles.

So, to balance the different batteries lifespans, you have to try to balance the amount of work each does. To do this your main tools are size of battery and the wiring that connects them. Thinner wires resist the flow of current and so large loads will drain batteries connected by fat wires faster than ones connected by thin ones.

In the diagram above you can see how I've wired the batteries together. The big gel ones are directly connected to the solar chargers and the AC inverter by very heavy gauge 35mmsq cables. This lets the gel bank do the bulk of the work when under high load. The weaker Marathon batteries are connected in 3 groups that use a star wiring pattern. This means that each group is connected by wires that are individually quite thin (6mmsq). But the wires are deliberately a bit long and are all the exact same length (hard to draw so take my word for it). The other important fact is that they are all connected together at one point (I soldered them together after weaving the ends into a sort of knot. This means that, as far as possible, the wire resistance for each branch is the same and so each group of batteries will do the same work.

The whole group of six batteries is then connected to the main bank by a single pair of 6mmsq cables with a 30A fuse. The fuse is important as 6mmsq cable can only carry about 50A and a fault could cause a few hundred Amps to flow. If the main battery dried up or got a short circuit somewhere inside it, a very large current could flow and start a fire. Actually, each of the groups should have a fuse but adding fuses in every leg makes it difficult to balance the resistances as the fuses and connectors introduce variances.

So, by having the Marathon battery bank 1.7 times the size of the gel one and using thin wires to connect it, I hope to keep the workload low enough on the Marathon bank to make it live for as many cycles as the gel ones.

When running loads, I've noticed that the gel bank does do more work (delivering about 2/3rds of the current). But as the gel bank starts to run down the Marathons start to assist more. This is because gel batteries are designed to resist deep discharge by being acid starved. They don't have enough acid to fully convert all the lead in the plates. As the acid runs out, the gel battery increases its internal resistance and so does less work until the output voltage falls to a point where the protection in inverters causes them to turn off.

Because the resistance of the gel battery increases as it gets discharged, the balance of resistance between the gel bank and the Marathons will change. At some point, the Marathons will have the lower resistance path and so assist the gel bank by doing more work. So it's more important that before for me to stop discharging the bank before it gets much below 50% as beyond that point, the Marathons will start to be drained quite quickly. But even then, that shouldn't be a big problem as they are 1.7 times bigger than the gel bank.

Tonight we're at 50% as the last couple of days has been very gloomy (it rained all day today). Hopefully tomorrow will bring some sun.

I'll have to watch the charge in the morning as the reverse is true for charge balancing. Because the gel bank is low and has high resistance, the Marathons will accept more charge current than the gel battery (at first). The charger can deliver up to 73A and could cause the 30A fuse on the link wire to blow...

Hello Darkness My Old Friend

2010-06-10T21:39:00.000+01:00

Dark days have indeed returned. The last couple of days have been a steady deterioration from the wall to wall sunshine of last weekend through the record breaking partial cloud of Monday (when I briefly saw the 1.8kW array spike to over 1.94kW) to today's wall to wall gloom.

Solar cells are temperature sensitive and when in full sun they get hot and lose some power. But on partially cloudy and cold days, the cells get cold when there's a cloud but when the sun peeks out they put out more power for a short time before they heat up. Also, as the sun peeks out from behind a cloud a sort of focusing effect happens to the light and for a few seconds it can be stronger than in a clear sky. Under these circumstances it is not uncommon for solar panels to put out as much as 115% of their rated power for a few seconds.

The new larger battery bank is capable of absorbing a lot more power now and had no trouble lapping up the 73 Amps that the solar chargers blasted out. Sadly, although these events produce very high rates of charge, they only last a few seconds so don't actually charge the batteries much. So over the last couple of days the batteries have been slowly fighting a losing battle.

Having increased the capacity from 180Ah to 492Ah has made a big difference though, extending the time that we can survive without a decent sunny day from just one day to maybe two or three, depending on how reckless I am in the evenings. But tonight after a particularly gloomy day where the PV output never crept above 375W and with the forecast for tomorrow to be non-stop heavy rain all day, the battery gauge is sitting at 69% tonight and likely to fall to below 50% tomorrow night. We might have to switch the fridge freezer back to mains power at night until Saturday when the forecast is for sunny intervals again.

We'll see.

How to Bodge Ammeters

2010-06-08T18:56:00.000+01:00

One of the big problems you have when putting together a battery based solar system is measuring where the juice is going. You need to know whether it's going in the battery or to the inverter and how fast it's going.

For this you need low loss ammeters. You need ammeters that can measure maybe 100-200A without causing much voltage loss. You can buy shunts that you can connect in series with cables to read the current but this means more joins in the cable and each join adds resistance that causes loss. Shunts are also mostly of the type that create a Voltage of 50mV per 500 Amps passing through them which is quite a small signal when most of the time you'll be measuring a few 10's of Amps. They're also quite expensive...

Enter the In-Line Bodger's Shunt... :D

Solar battery systems by necessity need to be connected together by big fat cables to carry the heavy current without losing Volts ('coz we're working at maybe only 12V or 24V). Any bit of copper cable (no matter how thick) will have a resistance and so make a voltage across it's length when a current flows. Even the cheapest digital multimeter will read down to 0.1mV (they all have a 200mV range). So all you need to do to measure the current in a cable is attach some sensor wires to the cable at a distance apart that causes some useful Voltage to be developed relative to the current.

I have two types of in line shunt in my system. One uses a long spacing (about 50cm on a 10mmsq cable) to give a reading of 1mV per Amp. On my big group of four 6V batteries, I made one of the huge 35mmsq link cables that join the blocks a bit longer than the others - just enough to make 0.1mV per Amp when I soldered the sensor wires at the lug ends. So that DMM reads 0.1 for 1A and 1.0 for 10A - you just have to ignore the decimal point on the display.

This one's a shorty as it's on a bit of 6mmsq cable.

It should go without saying (but I'll say it anyway) that you should only do this on low voltage DC power cables. Don't try it on mains AC lines, as a) it won't work and b) you'll probably get electrocuted and die.

All you need to do is connect another DMM with a 10A current range in series with the cable in question and arrange for a few Amps to pass through it. Then solder a sensor wire to the fat cable and using a pin, poke the cable at different places until you find the distance that gives you 0.1mV per Amp reading on the new meter. Then you can either cut the insulation away at that point and solder the other sensor wire in there or sometimes on big 35mmsq cables you can just solder the sensor wire to the head of a drawing pin and just stick the pin into the cable at the spot. Tape it up with electrical tape and run the sensor wires to where you want to see the meter. It can be some distance away (like a couple of meters) as the sensor wires carry no current so won't lose any volts on the way (the DMM input is usually about 10MOhms so the fraction of one Ohm on the sensor wire makes no difference).

Here you can see my three bodged ammeters. The top and left ones read net battery current in or out of the two main banks at 0.1mV per A (so showing 3A charge each) and the third one reads the current from the smaller Morningstar MPPT15 charge controller at 1mV per A (so showing 2.0A charge). The big Morningstar TriStar has a built in meter so I didn't need to bodge a fourth meter. The meters are arranged so that positive readings mean charge current and negative readings means discharge current.

Ok, these shunts aren't going to be super accurate and the short (or fat) ones will only read to a resolution of 1A but mostly that's all you need to know - whether the battery is charging or discharging and by roughly how much (some Amps or lots of Amps).

Maybe most importantly, it can tell you when there are no Amps flowing as that tells you two things: a) something is busted or b) the battery has finished absorbing power and is full. In a system like mine with several batteries, it also tells me how well (or otherwise) the batteries are sharing the load or charge... More on this in the next post!

Attack of the Killer Bees

2010-06-07T20:18:00.000+01:00

Woke up this morning and noticed that the end pair of panels on the patio were being partially shaded by new growth from a tall bush by the fence.

No problem, I thought. I'll just get the loppers out and trim a bit off the top so that the morning light can reach the panel. Snip... snip...snip. Hmm quite a lot of bees on these flowers... You can see all the blue flowers all over the branches. No worries, they'll just fly off. Snip...snip...snip... what the..? Hey! Ouch! Wahhh!!!

Retired to a safe distance to see the top of the bush a mass of bees. I thought one had "bumped" into my hand rather hard. Then it started to come up in a red bump. Oh oh...

Luckily, we have some anti-histamine tablets in the house as I get hayfever and we had some sting neutralising liquid (ammonia solution). If you put it on quickly after a sting the bump goes away completely and you're as good as new.

Back in the garden, I took a closer look at what I was cutting... Hidden inside the thickest part of the bush was a bees (or is it a wasp? Never quite sure...) nest. Quite beautifully crafted from paper and always has a couple of "sentries" guarding the entrance at the bottom and others making the next layer of the shell from chewed up wood.

It's an ideal spot for them as the bush has a lot of flowers and we grow roses and all sorts of flowers plus there's a pile of rotting wood next to the bush so no shortage of building materials for them.

Oh yeah... I managed to half clear the offending branches that were shading the last pair of panels on the array so I guess I bought a few more Watt hours in the morning but I doubt I'll be doing any more hacking and slashing this summer.

Even More Batteries!

2010-06-06T00:24:00.000+01:00

Well, the first two Marathon batteries I bought a couple of weeks ago worked out quite well. And flush with the money from the car boot sale of my old solar bits and bobs, I went back to the same UPS battery guy and bought up four more of the same batteries for £50 a pop.

The first pair I bought certainly hold a decent charge (although I haven't measured them on a discharge cycle) and assist the gel battery bank quite well.

Here you can see my cinema...

And nestling under and behind the sofa...

492Ah of 24V batteries!

Ought to be "enough" for now ;)

Sowing Some Seeds

2010-06-04T11:27:00.000+01:00

With the upgrade of all my plastic 15W amorphous panels to "proper" high power BP and Sharp solar panels complete, there remained the question of what to do with all those spare panels gathering dust in the garage.

We decided to take part in a car boot sale.

We had a load of other unwanted things to unload as well (the usual bric-a-brac) so we saddled up and headed out for the fields of Sussex.

I had the original four wet batteries and eight of the 12 solar panels to sell. I also had a bunch of little 7Ah AGM batteries that a Hungarian dude gave to me for free when I bought a 40W panel from him last year.

Business was a bit slow at first but by the middle of the afternoon the big batteries had all gone for like £8 each and even some of the 7Ah batteries had gone for £3 each.

Once I got my solar demonstration working (a panel plugged straight into a 12V fan) people were drawn to our pitch and were curious to know more about solar power. By the end of the afternoon, I'd sold all eight of the panels for between £15 and £20 each and even had someone come round to my house to collect another two that I'd not had time to unscrew from the wooden frame before the sale. I'd also sold some of the LED lights that were part of the kits and two of the charge controllers.

Some kids were amazed by solar power, asking where the electricity was coming from and exclaiming "Cool!!" when I showed them that it was coming from the sun, making the fan go slower or faster by having them stand in front of the panel so that they cast a shadow on it.

Hopefully these kids will have been as impressed as I was as a kid when I was given a solar power electronic experiment kit from Tandy (remember those "10 in1" and "30 in 1" experiment kits?).

My other secret plan was to get adults hooked on solar power. As I've discovered, collecting solar power is addictive. You start off with one or two small panels and before you know it you've got loads of the things and are making serious amounts of power. By selling off these seeds at a low price, I hope that some more people will experiment with solar power and eventually take the plunge and get a full blown system.

Made in Japan (part 3)

2010-06-02T21:55:00.000+01:00

Back in November 2008, while out sightseeing at the dammed lake reservoir near Miharu, we stopped at a hotel spa place that has a very good tofu cafe that sells all sorts of yummy food entirely (almost) made from tofu or derivatives (even the donuts!).

Anyway, while walking about I noticed that somehow they had also built an observatory (it gets very dark up here in the mountains away from any town).

Next to the observatory was what at first glance appeared to be a bus shelter (err.. because it had buses parked under it). In fact it was a sizeable solar array!

It consists of 8 rows of 14 x 167W 48 cell (23.2V MPP) Kyocera panels (I could read the info plates on the undersides). From the date stamps on the panels it seems to have been in operation since 2002. The nominal output is 18.7kW - maybe enough to offset most of the power used by the small hotel, observatory and cafe.

The electronics were all in a sort of phone box shaped hut next to the back row.

Travelling from the remote town of Takayama through the mountains to Matsumoto I saw quite a few new looking home based PV systems but many more old looking solar hot water systems. These types of evacuated tube or flat plate collectors have been popular for decades with the Japanese.

Made in Japan (part 2)

2010-05-31T23:27:00.000+01:00

This time we're in Miharu in the prefecture of Fukushima to the NNE of Tokyo by some 200km.

If you keep hitting (+) on the map it will zoom in and show Miharu (eventually!).

My father-in-law's house has two arrays on the South and West facing roof slopes and luckily even though it is in a quite steep valley, the valley runs East to West and his house is on the North side so it gets a fair bit of sun even in the winter.

The panels are made by Hitachi and are their newest hybrid crystalline amorphous type that have layers that give excellent wide spectrum responsiveness on cloudy days but the high output of crystalline panels on sunny days. The Sharp ones used on the house in Tokyo were a common crystalline type. On the South roof there are 18 panels and another 13 on the West roof.

All the power conversion kit is made by Sanyo.

As the array is split into unequal strings of 18 and 13 panels, they work at two different DC voltages. So outside on the wall is a voltage converter / combiner that converts the two different DC supplies into one at a common voltage that then goes into the inverter indoors.

A single DC cable goes to the power conditioner in the house where it is converted to mains. It's a less elaborate unit than the Sharp one and just has a single display that toggles between current kW output and the lifetime kWh generated. The information plate says it is rated for 70-380V DC input and 5.5kW output at 200VAC with an efficiency of 95%.

Outside, the utility company has fitted two digital electric meters, again one for power used and another for surplus power sold to the grid.

The inverter appears to be putting out 200V rather than the house 100V, so either it's designed to be split phase (100-0-100) relative to earth or it actually doesn't feed the house directly but just the grid through the separate meter so that the household usage bill is offset by the generating credits.

My father-in-law said that in the summer and up until September it generates a net surplus to the household requirements but by November it doesn't quite make enough to completely offset their usage. On the clear sunny day shown here, the inverter reported making 3.0kW for a short while around mid-day but 2.1 to 2.6kW for the bulk of the day. The sun was quite low in the sky even at noon and as it moved round to the West, the output declined a bit. In summer it was probably ok with the sun much higher in the sky and on cloudy days it probably doesn't make much difference either having part of the array facing west.

Another novel feature of their house is a Sanyo air source heat pump (ASHP) that heats water by extracting heat from the outside air (usually hot and humid). Using a sort of reverse air conditioner it pumps the heat into an insulated hot water tank.

It uses about 60-70% less electricity than a resistive heater. The Japanese call it an Eco-Cute. It doesn't mean "cute" though... "kyu" is a kanji in Japanese than means "to heat water". The difference between an ordinary air conditioner compressor and these water heaters is that the refrigerant is at a much higher pressure in this system to more efficiently extract heat and it's only in recent years that they've been made small enough and cheaply enough for home use.

Inside the house is the large hot water tank and compressor. A digital display shows the water temperature and how "full" the hot tank is as a bar graph.

The system also controls the heating of the bath to a constant temperature - very important for Japanese. You take a shower and wash and then sit in the tub just to relax and get warm while the system cycles the water. You could sit in it all evening.

He also had the kitchen refitted and they changed from a gas hob for cooking to electric induction. This wastes a lot less energy (95% of the energy goes into heating the pan rather than the hob or the air) but did have the drawback that they had to replace most of their cookware. Induction hobs only work on ferrous metal pans, so the typical aluminium or stainless steel with copper bottoms don't work well. As they generate quite a large surplus of electricity, it made sense to change from gas to electric for cooking.

Made in Japan (part 1)

2010-05-29T13:27:00.000+01:00

It's not just me in my family that has solar power, no sireee. It was actually my in-laws that started things off.

My wife is Japanese and her sister lives in the suburbs of Tokyo. Back in 2008 they installed a grid tied 3.07kWp Sun Vista system by Sharp.

Sharp call the inverter a "power conditioner" and it has a control and info panel near the utility breaker panel indoors. It lights up different colours during the day to indicate the power level. Blue means the system is generating a surplus and is selling electricity to the grid and yellow means it is generating but grid power is being used as well.

The panels are mounted on both sides of the roof but the slopes face sort of east and west so one half of the array is always sub-optimal. The sun doesn't make such a low angle in winter in Tokyo when compared to the south of the UK so the variation of day hours are less extreme and Japanese Autumns and Winters are characterised by clear sunny days (very unlike the UK). In the summer the sun goes mostly directly overhead so it's not such a problem.

Their house is the one at the back of the white one - you can just see the panels on the roof. The houses are terraced back to back.

The installers mounted the inverter on the outside wall and then ran a wire to the remote control panel inside.

The mains is fed to two meters outside that the utility company installed. The digital one on the left measures grid power drawn and the old school mechanical one on the right measures solar power sold to the grid.

The remote panel in the house displays an estimate of power being generated as well as the proportion being used or sold and the lifetime kWh. Flicking through the menus you can pull up generating history graphs for the last hour, day, week, or month.

It's about 11am in this shot and it's just started to make a surplus now that we are done with breakfast and have stopped using the kettle. The sun is a bit hazy today.

They tend to sell a lot of power during the weekdays when they are out at work but then buy it back at the weekends when they use the air conditioning and clothes dryer.

The whole system is a kit made by Sharp but a local installation company fitted it.

PV in Germany

2010-05-26T21:08:00.000+01:00

I knew that solar was more popular in Germany than the UK but this set of pictures gives you some idea just how far ahead they are compared to us in the UK...

I wasn't intending to go on holiday in Germany but fate (or rather volcanic ash) interceded and we had our flights to Japan cancelled. So, not wanting to spend the whole of the four weeks I'd booked off sitting in our back yard, we caught the ferry to France and drove across to Germany. We went to Seiffen near the Czech border and up to Berlin and across to the Nurburgring (but I couldn't drive on it as my insurance excludes it by name!).

Along the way, we were amazed by how many houses had solar PV or thermal (sometimes both) and here is a sample of what we saw. The biggest array by far was one in the "back yard" of Amazon's warehouse in the middle of nowhere in the middle of Germany. Absolutely huge it was.... a whole field of trackers.

Some like this one are communal, shared somehow by the residents of a block of flats. This one has 121 panels and at a typical 180Wp per panel that's about 21.7kWp

Many businesses install very big arrays on their warehouses or factories.

This was one pair we saw from the window of our castle hotel at Neuhausen near Seiffen that specialises in wooden toys and other intricate wooden decorations. You can see the 88 panel array on the house but also a bigger one on the barn just peeking over the hills behind. This house had in the region of 15.8kWp installed. And this is quite common in Germany.

These two we saw in another castle town, Wernigerode, where we stayed at a ski hotel. This house must get through a lot of hot water! It has a massive thermal solar panel.

This one not only has panels all over the roof but also on angles all over two garages / sheds to the left of the house.

Driving in the scenic valleys to the North of Kassel, we saw a solar showroom. A big ex-supermarket type building with a large car park. Could solar be so popular that they have something akin to a car showroom for selling installations? A helpful woman enquired if I needed help. I looked at a selection of example panels they had in different powers and styles (including black framed ones and big amorphous ones). They stocked some brands but she pointed out that they also made their own brand too.

Driving on through that town we started to see more and more installed PV systems... Obviously, the showroom had been doing plenty of deals for the 'Burghers.

Driving along the 80 towards the fairytale town of Hameln (as in the Pied Piper of), we were stopped in our tracks by the sight of this village (Oedelsheim) with its numerous houses with big PV arrays.

The farmhouse on the far left having some 264 modules installed on the upper and lower roof! I did notice on my travels that mostly the Germans seem to go for poly panels rather than mono types. Maybe they're a lot cheaper or perhaps the German made ones are predominantly poly types?

The neighbours were keen to keep up with the Johans in some arms race to see who could have the biggest array.

Hameln Banhof (train station) has this bus station with see-though panels specially made to make power and act as a sun shade. They are not so different to normal panels just the cells are spaced out a bit and a clear backing used instead of the usual white / blue EVA.

Treysa in Schwalmstadt to the SE of Kassel has a walled old town and from the ruin of the church you can see this school complex with a massive PV array on several buildings.

This office, visible from the old town wall has 198 panels (~35.6kWp) installed. By now this looks "average" as installations go.

Even some of the old timbered houses in the town have been kitted out. Not sure massive solar arrays on quaint English Tudor houses would sit well with the parish council but the Germans don't seem to mind.

Just outside the nearby castle village of Ziegenhain (supposedly the town of Red Riding Hood) some nutter has put a 5x5 module tracker on his shed...

Ferienhof Todenhausen wins the award for the most solar systems installed in one village though... It's not that big a place but it has a crazy amount of PV installed.

I counted 14 PV / thermal installations on different houses in this close-up shot alone.

And someone went mad on the far right of this shot where they installed PV all over the two enormous grey sheds with specially designed sloping roofs!

All the excitement went to me head and I bought the used 50W panel and 5x new 80W panels off of eBay for me home brew setup when we came back home.

On the way to Yorkshire to pick up the 80W panels, I noticed plenty of large buildings (offices, warehouses, farm buildings, homes) that were all ideal candidates for massive PV arrays that would almost certainly had one if they were in Germany... Not a sausage here. Pitiful.

Last of the New BP Panels

2010-05-25T10:08:00.000+01:00

The weather was hot, the sun blazed but I persevered.

When I bought the last five panels, I'd only thought about using four of them as I have to use pairs of panels to make 35V strings for charging 24V batteries. I had thought that I might sell the spare one or wait until I found another to go with it.

But then it dawned on me that I already had 2x 40W Kyocera panels on the garage roof. By connecting these in parallel to make an 80W "panel" and then the spare 80W panel in series, I could use the last BP panel and get rid of the last of the plastic 15W panels.

With this, my system increased marginally to 1,798Wp but probably makes much closer to this figure now as the 15W panels only seemed to put out about 80% of their official rating after about 6 months in the sun. Amorphous panels do degrade as they are first used but makers are supposed to factor this into their rating so that they declare a panel as being 15W after the stabilisation period. Some unscrupulous makers only declare the "Day One" power output though...

With this lot I've been making up to 63% of all my electricity needs each day (well at least 50% for the last three days in a row).

The main benefit has been free hot water each day with my automated water heater dump load absorbing all the spare solar power at lunchtime.

I'm not sure I'd want my panels on the house roof as they keep getting covered in dust and so I wash them about once a week with a pressure sprayer. It seems it's not actually just dust but loads of yellow pollen from trees nearby that is settling on them. Bird poo is also a problem as a large seagull one can cover a whole cell and then stop an entire pair of panels from charging the batteries.

Back in April, I did notice some of that volcanic ash collecting on them too. Not a lot but you could see some specks of what looked like burnt paper that were so fine they turned into a grey smudge if you touched them.

More Batteries

2010-05-22T12:30:00.000+01:00

The solar gel batteries I have are fine... I haven't managed to kill them (yet).

But they aren't terribly big. Only 180Ah. If I watch a couple of movies on the video projector (that electrically resembles a fan heater!) the batteries soon run down to 50% or less.

Another issue was that I've now got so much solar power available that it's actually a bit too much for the gel battery bank to absorb all at once. On a good day, I've seen charge rates as high as 56 Amps and the makers of the solar batteries recommended that I keep it under 30 Amps... More batteries to share the charge current and to add some spare capacity to "help out" when the main batteries were under the cosh were the answer.

I had a load of old tired wet batteries from a couple of years ago and these sort of helped but only when the gel bank was getting quite low. They also couldn't help much as they had lost a lot of capacity. I measured them on a load a couple of months ago and when new they were 110Ah but now could only muster about 35Ah to flat...

The other problem with them was that (as usual) when charging, they made gas that is explosive and not very pleasant to sniff.

I decided to try some used uninterruptable power supply (UPS) batteries from a regular seller on eBay. He gets various types of battery from small 38Ah 12V packs to 1600Ah 2V cells. Data centres usually replace their UPS batteries on a fixed schedule to avoid taking any risks with them failing when called upon to work in a power cut. Some of the batteries he gets are from companies that have gone bankrupt.

The only problem is size and weight and collecting them. Luckily, last Monday I was in Wales for a meeting and on Tuesday on the way home I had to pass within a few miles of his place. So I decided to try a couple of 105Ah 12V AGM packs. These were £50 each and only a couple of years old so presumably came from one of the many companies that have gone bust in these times or from one that had to upgrade to a bigger power supply.

They seem to be ok and after a conditioning charge to bring them up to full after having been in a shed for a few months, I connected them up to the gel batteries via a 30A fuse. They are much more willing to assist the gel bank when loaded and also absorb solar charge much more greedily than the old wet batteries.

Being sealed batteries, they also don't release any gas and can be used on their sides without the acid spilling out as it is held in place by a compressed glass mat "sponge" between the lead plates. Their flat profile also means that I can turn them on their side and keep them under the sofa :D.

If they work out, I might buy some more.

The Story So Far... (part 3)

2010-05-20T19:33:00.000+01:00

Having gotten bored of running up the stairs to turn on and off the water heater, I started looking around for a way to do this automatically.

The answer came in the form of a Velleman kit from Maplins called snappily a K8055 USB experiment interface. It allows a PC to read switches and sensors and then output signals or turn on / off relay switches. You can get them either pre-made or in kit form and you have to solder it together yourself. I chose the kit version as it's £10 cheaper.

After putting it all together, it looks something like this...

The inputs are the terminal blocks on the left (two analogue inputs of 0-5V and four inputs for switches). The kit actually included little press buttons to test it with but I didn't bother installing them. On the right are the outputs (8 on/off outputs with LEDs and two analogue outputs of either 0-5V or PWM of 0-100%).

The latter could be used to dim a light or even a heater by varying the duty cycle of power to the load and using thyristor to chop the power.

A CD comes with the board that has the Windows DLL that makes it work and an example program in various languages. I had an old copy of Visual Basic 6 sitting around and it works fine with this. To make my program to control the heater, I just took the demo program and modified it a bit.

In order to control the heater so that it only used spare solar power and not battery power, I needed to measure several things:

The voltage of the battery bank.
The mode of the charge controllers (whether they are charging or finished).
How much power the system is generating for loads (including the battery bank).
How bright the sunlight is (to estimate how much solar power is available).

Most of this information is available via the Morningstar charge controllers. They record this information and it can be read as data via their communications port. The TriStar MPPT-60 additionally has an Ethernet port so it can be networked. The main computer in the house is upstairs (conveniently near the water heater) but the solar stuff is downstairs on the other side of the house.

I didn't want to run Ethernet cable all over the house so I used a pair of Ethernet over Power plugs. These just plug into spare power sockets and then carry data over the house wiring. Very handy. This then plugs into an old 10Mbit hub as I also have a couple of computers in the living room and an IP CCTV camera outside.

One of the computers is an old Toshiba laptop that I bought at a car boot sale a couple of years ago for £5. It works and with a new battery bought off the internet, it even runs better than new. It only has Windows 98 but I managed to find an old Ethernet card for it - would you believe from another pair of laptops I bought at a car boot sale for £5. Those laptops were a bit broken but work now and the Ethernet card was free. The Toshiba can't use modern cards as it isn't Cardbus compatible so I had to find an antique card to use in it.

The Toshiba is plugged into the two charge controllers. The Ethernet of the TriStar and directly into the serial port on the MPPT-15. The Morningstar data logger software then records the output and battery state from both controllers and saves it in a CSV file on a network file share (on the PC upstairs).

This data file is read by the program I wrote using the K8055 demo software. I combine the two sets of output figures for charge Amps from the two controllers and then multiply that by the battery voltage to give a figure for total power generated. The system records these values every 15 seconds so a graph can be drawn (again by the Morningstar logger software) like this:

The load manager program then takes an input from a light sensor to measure how strong the sunlight is. The power graphs above only tell you how much power was used, not how much was available.

At first I tried using a light dependent resistor to measure the light but it was too sensitive (they are commonly used to measure how dark it is for dusk to dawn lights).

I tried to cut down the light reaching the sensor by putting it inside a sea shell as these are white and nearly opaque so that they cut out most of the light. I sealed the sensor inside two such shells with waterproof polyfilla.

Unfortunately, after a couple of weeks, the sensor broke and stopped working so I had to come up with another idea...

This time I used an old miniature solar panel from a AA battery charger and put it in a waterproof box. I connected it to a 100 Ohm resistor as a load so that the USB A/D input would read the voltage produced as a measure of the power produced and so the strength of the sunshine.

Finally, the K8055 program reads the charge state from the charge controllers (MPPT, Absorption, Float, Equalize, or Night) and uses all this information to try and estimate when there is enough solar power available to charge and keep the batteries full while turning on the water heater.

Here you can see all the information about the solar input (sun strength), power produced by both charge controllers combined and the battery status, along with the information that the water heater has been running for a total time of just over two hours today (it was a bit cloudy).

The output of the K8055 interface card drives a miniature 12V mains relay that just fitted inside an extension lead socket so that the heater (via its 230V to 115V transformer) just plugs into the controller.

I installed a digital thermometer on the water tank (cutting though the foam insulation to glue the sensor on the tank itself about 1/3rd the way down from the top). For every 10 minutes the water heater runs on solar power, it conveniently increases the tank temperature by 1'C. Some days in May, the heater can run for almost 6 hours, raising the water from 15'C to over 50'C.

The Story So Far... (part 2)

2010-05-18T20:32:00.000+01:00

Originally, I just wanted to run my work laptop on solar power. I thought, "How hard can it be to power a 30W load for 8 hours a day, every day?". Turns out the answer is, "pretty hard".

The problem is that solar power is very variable and unreliable. Some days you get loads other days you get nearly nothing. The batteries help but they are fragile beasties and if not kept charged up fully will soon die (like in a matter of days if left completely flat). So you are always balancing using the stored solar power in the batteries with not using so much that you kill them.

One thing this leads to is any big constant loads like the water heater are very difficult to run. Even if you've got enough solar panels installed to run it, you will almost never know if you've got enough solar power coming in to run it. If not, then the batteries start to drain and you risk not having enough power for the night.

The first problem is that water heaters use an enormous rate of power consumption (typically 3kW). I've only got 1.8kW of solar panels. The answer was to reduce the power used by the heater. The easiest way to do this is by reducing the voltage fed to it. At 230V, the heater will use 3kW of power. At 110V, it will only use about 650W. It will of course take much longer to heat a tank of water at this rate but in the summer, you can count on up to 6 hours of solar power on a good day... More than enough to heat a tank of water to 55'C.

I found an old tool transformer at a car boot sale (I love car boot sales - why buy new when you can buy randomly old?). Conveniently, this does just what I need; converting 230V solar power to 110V. So now I only need 650W of reliable solar power to run the water heater.

It wasn't as bad as it looks... Honest. Just flaky paint and some external rust. The guy let me have it for £15, not bad considering it was a custom wound 4kVA unit - more than man enough to run a water heater continuously without getting hot or catching fire or 'owt :)

With a bit of Hammerite, things were looking a lot more ship-shape (or at least transformer-shape).

And then off it went to its new home in the airing cupboard.

That leaves you with problem number two... You could turn the water heater on and off by hand while watching the power levels and the window for clouds to appear and spoil your fun. I did this at first but quickly grew tired of running up the stairs to the airing cupboard. So what was needed was an automatic way to measure the solar power, the condition of the batteries and then turn the water heater on and off so that you keep the batteries charged up and only use incoming spare solar power to heat the water. In off-grid living, this is called a dump load controller. Something that takes all the spare energy and dumps it somewhere useful (water heaters are the usual dump load of choice).

Coming up in Part 3... A whole lot of bodgery involving a kit from Maplins, a doorstop laptop, some retro 90's software development, seashells and polyfilla...

The Story So Far... (part 1)

2010-05-14T12:10:00.000+01:00

Welcome to my blog about my attempts to do away with utility power and make all my own electricity for "free". Of course it's not free (it costs a fortune to generate) but you get a warm feeling inside when you make your own power :D

Let's start at the beginning... I work at home. At first this was just a convenient option to the office but back in late 2008 my company started a policy of home working to replace the offices they had. So I found myself kicked out of my office and forced to work at home (better than being kicked out of the company mind you...)

I work in IT so all I need for my job is my laptop and the internet. I wondered if it would be possible to generate the power for my laptop from solar power. At the time, the electronics store Maplins was having a sale on some cheap 12W solar panels. So I bought a couple for a laugh to play with...

As special offers came up now and then, I added a few more. They started selling a kit of four with slightly more powerful 15W panels (the ones on the ground in the picture above). Our garden would never be the same!

For you see, I'm a bit of a collector. And collectors like to collect. If you're lucky, you collect something with a finite number of things in the set. If you're unlucky, you collect something that is never ending and it takes over your house and life. Solar power is one of those things you can keep on collecting forever. You can't have too much power. It's addictive.

I'll skip over a lot of the detail in how I got to where I am today with my system but if you want a blow by blow history you can check out my threads at:

http://www.mrsharkey.com for older posts
http://www.navitron.org.uk for newer posts and lots more. This forum is UK centric and covers many aspects of green living and technology for bodgers and professionals alike.

From those humble beginnings my system grew and grew until today it looks like this:

But that's not all...

There's more on the garage....

The system as it stands now has the following statistics:

System standard: 24V DC nominal

Installed PV: 1,778Wp

Charge Control: 1x Morningstar MPPT-15 (15Amps)

1x Morningstar TriStar MPPT-60 (60Amps)

Batteries: 24V bank comprising

4x Deka Solar Gel 6V 180Ah +

2x 110Ah leisure batts.

Inverter: 3kW Antares (Cotek) SK3000 pure sine.

Indoors, on the wall are the charge controllers and meters.

And on the floor below, the batteries hide under a comfy chair. The power is distributed on mains extensions that plug into the inverter. Being off-grid, my system is not connected to the house mains and so I have to swap appliances from mains to solar manually.

Using batteries means I can use the collected solar power at night or when it's cloudy and also comes in handy for when there's a power cut. Ironically, those who elect to install grid-connected solar power can't use it if there's a power cut as those systems are designed to only work when the mains is on!

We live sort of in the country and the mains isn't as reliable as when we lived in London, so it's been useful for keeping the fridge-freezer running in a power cut.

Stay tuned for more on my system as it develops and lots of other stuff about solar power!