What happened to the DIY in solar!

[sokoloff]

By all means, let's not have a discussion about the relative tradeoffs of other engineering solutions to a problem.

We should only discuss the specific route a poster has chosen, nothing else...

[fourtytwo42]

By all means, let's not have a discussion about the relative tradeoffs of other engineering solutions to a problem.
We should only discuss the specific route a poster has chosen, nothing else...

You are quite right, my language was intemperate BUT I have recently suffered mightily from trolls elsewhere.

[Red Squirrel]

The idea of going DIY on the electronics is intriguing and something I've considered.  Need to learn more about SMPS design and get to know all the ins and outs first, especially on the safety and failsafe side of things.  Ex: if it fails you don't want it to take out a bunch of other stuff with it.  Especially at the inverter you would not want it to fail in a way where it presents destructive voltage/frequencies to what is connected to it.  Or even the charge controller, you would not want it to kill the battery bank.   

What would also be interesting is DIY battery cells.  Seems lead acid is not THAT complicated, though the issue is getting the raw materials and the acid itself.  But if one can come up with a design and a way to manufacture it consistently you could whip out cells made from like 5 gal pails or something and have crazy reserve time for way cheaper than buying the equivalent batteries.    Though I guess it might make more sense to go lithium ion at this point if you were to DIY on the batteries.  Issue is they have a more complex charging profile.  Not sure how you'd charge a battery bank while it's also in use, which is essentially what happens in a solar setup.   I'm sure there's a way to do it though.

[fourtytwo42]

The idea of going DIY on the electronics is intriguing and something I've considered.  Need to learn more about SMPS design and get to know all the ins and outs first, especially on the safety and failsafe side of things.  Ex: if it fails you don't want it to take out a bunch of other stuff with it.  Especially at the inverter you would not want it to fail in a way where it presents destructive voltage/frequencies to what is connected to it.  Or even the charge controller, you would not want it to kill the battery bank.   

What would also be interesting is DIY battery cells.  Seems lead acid is not THAT complicated, though the issue is getting the raw materials and the acid itself.  But if one can come up with a design and a way to manufacture it consistently you could whip out cells made from like 5 gal pails or something and have crazy reserve time for way cheaper than buying the equivalent batteries.    Though I guess it might make more sense to go lithium ion at this point if you were to DIY on the batteries.  Issue is they have a more complex charging profile.  Not sure how you'd charge a battery bank while it's also in use, which is essentially what happens in a solar setup.   I'm sure there's a way to do it though.

The reason I do it is to keep the grey matter ticking over as much as anything but also I am intrigued with all the different possibilities and by doing DIY I can wring many idea's/answers from one setup, nothing is ever finished, it's always work in progress to the next trial/improvement
I am fortunate to have some professional experience in SMSP/Converter design so I have a starting place, as you say boiling batteries or fire are serious consequences of careless design. There are unfortunately a lot of dangerous designs floating around on the web so it is not a good thing to copy. One of the best routes to getting started is manufacturers application notes that often include details of magnetics etc, TI & ON semi are good (Natsemi got swallowed by TI and Fairchild by ON).
Fortunately PV is fairly safe as the input current is self limited by the cells to ~10A, High voltages can be avoided by limiting the length of strings employed and converter topology (avoid boost and flyback). Certainly no inexperienced person should attempt grid-tie as both the voltages and potential energies are high. If you want to avoid boiling batteries a transformer coupled design is good as if a mosfet fails the result is no output!
I have seen home built batteries in oil drums on the web somewhere, intriguing but I am not sure where to get hold of large quantities of electrode materials. I think about Lithium and my mind immediately springs to fire so I have not been tempted even though I have several applications, again I have seen many DIYers using Chevy Volt packs and similar, if I did something I think it would be in a shed down the bottom of the garden

[Codebird]

If you are building your own batteries, perhaps nickel-iron would be a better option? As for acid, well... I got two litres of concentrated sulfuric on eBay once.