Where do I start with solar?

[paulca]

Charge controller and battery arrived.  Battery arrived with 12.38V so it's not broken.  I don't have a load capable of capacity testing it properly, but I'm not that fussed as long as it does the job.

Wired it up, partially anyway, and the charge controller powers up and while it says it's daylight due to indoor lighting, it obviously isn't getting enough to put anything into the battery.

I'm not going to run it until the weekend when I'm present and it might take me a bit of time to trust it to run while I'm out at work.

I have a metal cased fridge I don't use.  I'm tempted to put the charge controller and battery inside it as some minimal fire protection.  How worried should I be?

[metrologist]

I have not taken any precautions with my batteries. One is for motorcycle and the other for a typical family car. Since they are wet cells, I would not put them in the house, but on the bench in the garage seems fine. The e-bike Li-ION batteries get moved around as shrink-wrapped battery packs would.

I would not even worry about SLA - most backup power supplies use them, so just don't short it out.

I used some automotive aux lights to test my system - something that would drain more than the system could output. I should run that all day to see what the system could actually cope with.

[paulca]

I have got a 5A DC load I built, but it would take nearly 5 hours to drain the SLA (26Ah).  I also have some 100W power resistors, but they are 1Ohm and 0.1Ohm, so that's what on 12V...  The 1 Ohm would draw 12 amps / 144 watts.  I might have two of them, in series they might make it, 6Amps, 72 watts shared between them.  Still going to take 5 hours and not something I would leave unattended.  The 0.1Ohm would make fire.

Anyway, I wired a load up to it and it's charged it's first LIon pack (via a Turnigy charger mind).

So... do I put the panel in the window and risk leaving it when I go to work?  I'm thinking, no.  I can't confirm the MPPT portion does not have a fault that will let smoke out while I'm at work.  I don't want to come home to a broken in door and bill from the fire service (as it will set off the building wide fire alarm) ... or much worse.

Saturday then.  Which will give me time to discharge the SLA a bit and give the panel something to do Saturday morning.

[mtdoc]

I have a metal cased fridge I don't use.  I'm tempted to put the charge controller and battery inside it as some minimal fire protection.  How worried should I be?

With a SLA battery you have nothing to worry about. No need for metal enclosure. Even with FLA batteries - you only have to worry about significant amounts of hydrogen gas venting during equalization and then it's only an issue if you have a large battery bank (> kWh worth).  Large FLA battery banks require enclosures with venting to outside.

Since they are wet cells, I would not put them in the house, but on the bench in the garage seems fine.

In the house shouldn't be an issue unless you plan on spilling electrolyte on your carpet...    Of course as above, a large FLA battery bank needs an enclosure and venting to the outdoors.

[paulca]

I have solar power!  Woohoo!

Put the panel in the window yesterday, but it was the darkest, wettest, miserable of days.  The CC did register charging, but 0.0A.

This morning however it's blue skies (for now) and the sun isn't up above the houses across the street and I have registered my first amps to battery, 0.1A.

Will see how far that gets when the sun is up properly before the weather closes in later again.

EDIT:  And when the sun came out from the behind the houses I have 1.1Amps.  Oddly the charge controller has not switched to PPT mode, it's sitting at 14V, the max power point of the panel is 17V.  I read this about the Tracer, at low power levels it remains in PWM mode and just sustains a charge voltage for the battery.  Maybe if the sun comes up out of the horizon haze it will improve.

I'm not expecting the full panel 50W behind a window, but 25W would be nice to see in direct sun light.  Any more would be a bonus.

[paulca]

Our weather sucks.  There has been virtually no sun this week.  I started with a battery voltage of 12.2V and I now have a voltage of 12.6V.  I've taken no load off it.  I have also taken to switching the battery off at night and back on in the morning to save the little the charge controller is pulling.

I did manage to get it all fused and switched up properly.  Makes me feel a bit safer leaving it unattended.

I bought an RS485 to USB adapter and played with it's software, but I want to log the data.  So in the post is a MAX485 board and another ESP8266.  It can send it's data to my existing data logging set up (originally for home heating but now generic enough that it will accept any data over the network).  It will create RRA archives and graphs for the panel volts/amp/power and battery/volts/amps.

I also plan to sleep the ESP overnight, once the panel voltage drops below 5V or so it can deep sleep and wake up periodically to check for dawn.  It will be running of the charge controllers RS485 5V.

Interestingly(?) I was going to add a "light level" sensor to my data logging system, but monitoring panel power seems a lot easier than messing about with calibrating LDRs.

[bazza]

* 40w 12v ebay panel ( make your own stand out of wood ).
* 10 A  pwm or mppt charge controller ( EPSOLAR or EPEVER brands are decent if going Chinese)
* ~20 Ah to ~30 Ah 12v AGM lead acid battery
* loads of common ( cheap) elecrical cable stripped for use in solar (running off the regulator's load terminals)
* several lovely DC-DC step-up and step-down DC-DC converters on ebay to run lots of DC gadgets of varying voltages
* several spring lever terminal blocks on ebay, so you can hook up several 12v items at once
* automotive fuses for safety, or din-rail mounted circuit breakers if you wish
* cigarette lighter sockets ( plug in cheap cigarette plug USB chargers into these )
* not forgetting a bunch of 2.1mm  x 5.5 mm screw- down plugs and sockets
* not forgetting universal laptop plug sets 2.1mm available on ebay, which will help you in your quest to hook up a bunch of stuff to any DC gadget with a removable ac adapter plug ( in conjunction with the Dc-Dc converter with correct voltage dialled in)

Years of fun. Best education is to learn by doing.

[Lord of nothing]

I found that Croation one: https://www.ebay.de/itm/Solarmodul-Photovoltaik-SolarPanel-300W-24V-Monokristalline-wohnmobil-hause/253647307997?hash=item3b0e8ec4dd:g:RJ0AAOSwbA5bCDaP
How about that?

[Red Squirrel]

If you want to do a solar project for fun but that still has utility value, look at building a portable solar power pack for camping or emergency.  I did this so I can play around with solar.  Basically it's a box with a battery, charge controller and inverter, and a separate folding panel that plugs in.  A microcontroller (basically at atmega running arduino code) monitors the voltage so if it goes too low it cuts power to not drain battery too much. There is also a LCD.  I can console in with a laptop to adjust calibration etc too. 





Now using it for my shed until I get a proper setup and roof top panels. It's surprisingly hard to buy big solar panels here due to shipping cost.  I have not found a source yet where I don't need to take out a 2nd mortgage for the shipping.   I may end up having to build my own.  Just not sure what materials to use, as I only know how to work with wood and expansion/contraction would probably crack the glass.

[mtdoc]

Don’t try to  build your own PV panels unless it’s just for fun and educational. You will not produce anything even close to the cheapest commercially produced panels. Many have tried and failed.

As far a buying single large panels on the cheap, try Solar Blvd.. Fedex shipping on individual panels in the US is $30 and I believe it’s the same for at least some parts of Canada (Vancouver, BC for example).

[metrologist]

I just jumped in with no objective. A 100W panel and a few charge controllers for different applications.

I've been thinking of small systems as this, such as to take my fish/reptile tanks off-grid. The reptile tanks are ideal because they really only need to run when the sun is out (to recreate the sun with artificial lights and heat ), the fish tanks need some heat and circulation during the night, however, and the battery makes it not worth it in the long run, unless just for experimentation.

The one application that has paid back the most is the MPPT controller set up to charge my e-bike. I cannot quantify that, but the payback is not just an obscure financial calculation, but the objective enjoyment that all that power and acceleration I'm experiencing is basically free - after paying for the system, of course.

Let me run a number... that system cost $130. @ $0.20/kWh, it will be 650kWh to break even. The e-bike uses about 160Wh to go 5 miles, with some assist, so about 4000 charges or 20,000 miles to break even. I'm going to need more tires.

[Seekonk]

It is hard to do anything with solar with less than 500W. One of the best paybacks is heating water for those who have water tanks. This is supplemental heating which means that almost 100% of potential PV energy is used.  That is a far better payback than oter uses for PV.  For $340 you can get two 280W panels out the door locally. $10 to build a little controller and you have a system that out performs a heat pump water heater in payback. Plus you have  a little system to provide power in an outage.  Yes, most wall warts work fine on 60V DC.  This laptop right now is. I'm off grid and get all my hot water from excess PV that would normally be just wasted. People with solar are the greatest energy wasters in the world. A working system usually requires that by design.

[NiHaoMike]

Don’t try to  build your own PV panels unless it’s just for fun and educational. You will not produce anything even close to the cheapest commercially produced panels. Many have tried and failed.

Great Scott noted that there are a few times where DIY solar panels make sense - small panels where the cost per watt is high and panels that have to be unusual dimensions.

I would like to see someone do a "DIY or Buy" for MPPT controllers. In my experience, the cheap ones are not very good so DIY should easily win, especially after someone makes open source code for it.

[mtdoc]

Don’t try to  build your own PV panels unless it’s just for fun and educational. You will not produce anything even close to the cheapest commercially produced panels. Many have tried and failed.

Great Scott noted that there are a few times where DIY solar panels make sense - small panels where the cost per watt is high and panels that have to be unusual dimensions.

Yeah it cost him 140 Euros in materials and 8 hours labor to produce a 100W panel- about twice the cost of a commercial panel. Worse, the durability is highly suspect which has been the case with other DIYers.

I would like to see someone do a "DIY or Buy" for MPPT controllers. In my experience, the cheap ones are not very good so DIY should easily win, especially after someone makes open source code for it.

That’s probably true for the low power ones. There’s been low power Arduino MPPT plans and code around on the interwebs for years now.  But where MPPT really makes a difference is with higher power systems ( say > 500W or so). In that case, best to stay away from the cheap ones and DIY unless you really know power electronics well.

[Scrts]

Just wondering: did anyone make any analysis on where it's worth getting the solar system in terms or earth latitude? I understand that equator countries will obviously be good, but where is the cut-off line? Where statistically there's less sun for the system to be effective enough and pay off in reasonable amount of time?

[mtdoc]

There’s no latitude cut off. It is situational dependent. Obviously there will be huge seasonality issues, but even in Alaska and nothern Canada there are situations where solar PV makes sense - especially if grid power is unavailable or prohibitively expensive.

If you’re asking about adding solar to a locale that already has reliable grid power, then payback time will depend on local solar insolation (not just latitude dependent), how much grid power costs, how much space one has to install how large of an array, how much it will cost to do the installation, etc, etc.

[Seekonk]

There is a $5 72V buck converter that I love because it is very easy to modify to stand alone and or be controlled by a uno etc as a charger. I have several in parallel in my system with 60V grid tie panels.  Saves a lot of effort and searching for parts.  Nice when you want to supplement a 12V PWM system with grid tie panels.