diy solar

diy solar

House burned down

They provide the same protection as a T class?

I'm sure you've mentioned it somewhere in this thread but I didn't see it. In your use case, which specific Midnite DC breaker are you using? Link if you're feeling generous! :)
Its an MNEDC 250


It has an AIC rating of 50,000A at 125V. This particular one has a delay of 16:

Screenshot_20240429_191628_Samsung Notes.jpg
 
You're allowed to have a fuse and a breaker. Breakers are very handy for taking a battery offline if required.
That's the route I took and I'm very glad I did.
Note, this rack pictured will be rebuilt with steel and skinned in cement board when I expand and add another 690Ah.

I'm of the opinion, and I'm in the minority, that breakers are so bad at the over current part that it's actually more dangerous to have them in the mix. I'll take a disconnect and a fuse. I don't want the disconnect trying to do the fuse's job.
Have you looked at the specs for Midnite / Carling Tech breakers? MNEDC
"MidNite's Breakers Are Rated To Break The Full Rated Load At The Rated Voltage Repeatedly, With NO DAMAGE. Always Use A Properly Sized Breaker For Disconnecting.

All MidNite Circuit Breakers Can Be Used At 100% Rated Current When Mounted In A MidNite Enclosure."
Link if you're feeling generous! :)
 

Attachments

  • Screenshot_20240429-191824.jpg
    Screenshot_20240429-191824.jpg
    146.8 KB · Views: 61
Have you looked at the specs for Midnite / Carling Tech breakers? MNEDC
"MidNite's Breakers Are Rated To Break The Full Rated Load At The Rated Voltage Repeatedly, With NO DAMAGE. Always Use A Properly Sized Breaker For Disconnecting.
I have more respect for Midnite breakers than any other. Those are vetted for the purpose.

But they're just as expensive or more expensive than fuses, so you see less people actually using them, and I'm not personally tempted.

And I don't quite buy the claim of no damage. Every contact suffers damage when breaking DC under load, it's just a matter of how many times they can do it. I like that my contactors are straightforward about that and it's in the data sheet.
 
I'd like to see a dielectric object ("insulating busbar") between breaker lug and shunt. Or a least a longer negative jumper.
Right now they can pivot to touch. Are those unprotected, straight from battery?



How about a panel?

Ideally, square knock-out in metal, with holes for the screws.
Maybe just two rectangular pieces, one for top and one for bottom. After mounting, cover with two more rectangles to keep fingers out.

Typical breaker panel bezels have twist-out pieces leaving a square hole, probably not the right dimensions.

I haven't used any bezel mount yet, only plug-on or DIN rail.
Some reviews showed what folks came up with for mounting solutions:


94927_319_b7f4bc8a_f6f8_4588_bb94_cd301f7fa57e (1).jpg

94927_319_22c7e8f0_b978_4019_9b46_af9a4366ab7b (2).jpg
 
I have more respect for Midnite breakers than any other. Those are vetted for the purpose.

But they're just as expensive or more expensive than fuses, so you see less people actually using them, and I'm not personally tempted.
Alte store had them on sale for $47 each so I picked up a few, less than a class T when you take the fuse holder into account.
 
I believe both reported fires share a suspicion of a lack of separator.

I'm taking all three conclusions:

1. Class T Fuses
2. Cells in metal boxes
3. 5 side FR4 separation
I could not agree more, that's why I did all three.
Having the whole electric room a safe distance from the House is the way to cover the unknown possibilities.
 
I could not agree more, that's why I did all three.
Having the whole electric room a safe distance from the House is the way to cover the unknown possibilities.
Power shed is definitely the ultimate goal and the hardest to attain. When I ran my ground mount conduit I also ran a 1"conduit half way as preparation to build one there some day. Then I can dig up my PV conduit and bring it up into the shed, and send power to the house through the 1".
 
I am skeptical of the fuse being the initial cause, based on the fact that this happened at night with very low currents.

My though is, it could have been a cell that internally shorted, caught on fire and the fuse blew after the fire had already started melting things, one thing falls on top of another, etc.

Everyone is running out, pointing fingers, and changing out their fuses...the thing I'd like to point out is that every devastating fire I've seen on this forum was using raw cells. Not a single one I can recall use pre-built batteries that were fully contained in metal boxes holding smaller quantities of cells within. I'd argue that the metal boxes are what slow down and often times smother out the fires - Seplos makes some and I think they just make sense if you insist on DIYing your batteries.


suggested this earlier

 
My though is, it could have been a cell that internally shorted, caught on fire and the fuse blew after the fire had already started melting things, one thing falls on top of another, etc.
The only issue I have with that analysis is that spontaneous combustion for what ever reason is highly unlikely with LiFePO4 chemistry. It it more likely with NMC or LIPO. I agree with the thought that we do not know what the reason for the fuse to melt and that is the big question? This system did have 100 kWhs of capacity so that is a lot of energy accross that kind of fuse.
 
Think he said it was one of the fuse-per-string.

Could be backfed, don't know how much voltage drops for a bad cell (several bad cells?) or how much current they will accept - is that enough to blow a fuse?

If the fuse blew while batteries were venting, that explains it.

How the fire re-ignited after being extinguished, I don't know. Thought that was only a thing for the other chemistries.
 
I'm seriously reconsidering using a DIY wood ended box. On all metal shelves that I intend on putting metal doors on, but... hm.
When I got serious about running my home with batteries, I dumped the wood boxes and went metal case.
Also there are perfectly good breakers for DC, you just can't buy them on Amazon for $20.
 
So,

Reviewing this thread the OP had 7 strings of 16s 280ah cells completed in 2020

Still using the batrium BMS with 112 top of battery balancers. Using multiplus II 3k in 3-phase configuration.

Using Mega 200amp fuses per the victron manual. I assume 70mm^2 cables also per the manual.

Shelves were glassy smooth per the OP.


Earlier when I mentioned venting followed by over current followed spark of any sort either from burning fuse holder or by the hydrogen or other by product reaching combustion temperature. This is pure speculation on my part but I honestly don't think we will ever know the actual cause unless the victron data shows something.
 
Regular 1/2 sheet gypsum board has a fire rating of about an hour.
You can get thicker boards and boards with 4 hour ratings.

I think it might be wise to look into that for a battery room in a wood structure and its not that much more expensive than regular dry wall...

An Ontario electrical code regulation thats since been removed from the code called for 1.2 normal dry wall between wood and electrical services in homes.
I thought this was a good rule.

Its also something to consider, its not a requirement but dry wall is so cheap.
 
Not a single one I can recall use pre-built batteries that were fully contained in metal boxes holding smaller quantities of cells within. I'd argue that the metal boxes are what slow down and often times smother out the fires - Seplos makes some and I think they just make sense if you insist on DIYing your batteries.
I think the important part is that with the pre-built batteries from a reputable manufacturer you also automatically get rid of a lot of the questionable practices that people have done when building raw cell DIY batteries on a shelf. I also wonder about assuming the metal boxes would snuff out the fires when we now have pre-built batteries with fire extinguishers within the metal cases. Lots of unknowns here.
 
  • Like
Reactions: DPC
It looks to me like this system was build better than most .... and yet his house burned down .... kind of a punch in the gut for DIYers.
Of course Walmart kept having fires in their commercial solar systems too.

I'm glad everyone was OK and also glad the OP posted this .... I should result in a lot of soul searching a renewed emphasis on safety.
 
Someone recommend me a case for Eve 230Ah cells plz... heh
Seplos, apexium, luyuan, EEL, are some common ones. Apexium are made by Docan. I have EEL that are supposed to be Seplos clones. Some will let you customize to spec with breakers, fuses, etc. I believe @Steve_S got some real nice luyuan boxes done
 
Last edited:
Back
Top