House burned down

[wattmatters]

That's my takeaway from both fires so far.

Both DIY batteries.

Not clear to me from the photos I've seen on the OP's build thread whether the individual cells had an isolation barrier between them. It does not appear so:

I noted they reported leaking cell a couple of years ago. Perhaps a portent of other issues.

LFP Powerwall

newbie here. why does it matter whether the p or s comes first Mechanically and electrically they are different and that has implications on BMS strategies. From the inverter standpoint it doesn't matter because the total voltage is the same.

diysolarforum.com

Just glad no one was killed/injured but such an experience is traumatic. I wish them well with their home recovery.

 

[robbob2112]

Class T fuses are not cheap. And they are not instant, either. When I was doing my research on this (on account of the high price tag for Class T), I discovered that whereas I had thought they were almost instant, they also have a delay before blowing. Another fuse, at a much cheaper price, which is said to have a similar delay, is what I ultimately went with for my system. The Class T fuses, for my locality, seemed uncommon, whereas the NH fuses are in common use.

Here's one example: https://www.solar-europe.co.za/product/dc-200a-fuse-link-500v-fl200a/

I have not seen much discussion on the forums about the NH fuses. Apparently, they have a lag time on par with Class T, and some sites I have found from users in Europe appeared more confident in the NH fuse than in a Class T fuse. I'm interested in learning more on this comparison, if anyone here has intimate knowledge of both, their advantages and disadvantages. The main advantage to the NH1 fuse for me was the price. The fuse itself is rather bulky, and occupies roughly the lower third of my breaker box--virtually an entire row to itself under two rows of breakers.

Tidbit about the NH fuses - this makes picking the right fuseable link difficult. You have to derate by half the ampacity and also adjust the time curve.

DC applications
Eaton’s Bussmann series
NH fuse links can be used
on DC applications. In all
cases the fuse links can be
used at half of their AC rating
with a time constant of no
more than 10mS. The time
constant is the rate of rise of
fault current and should be as
close to a 50Hz AC half cycle
as possible.

https://www.eaton.com/content/dam/eaton/products/electrical-circuit-protection/bussmann-iec-high-speed-semi-conductors-fuses/bussmann-iec-battery-energy-storage-fuses/eaton-bussmann-series-nh-battery-storage-fuses-datasheet-135001en.pdf

 

[Skypower]

Been thinking of the middle of the night aspect. Pure speculation, but that’s when mice are most likely to be chewing on insulation. Still…an appropriate fuse close to the battery should have prevented this.

You hit upon one of my fears of rodent damage. Years ago voles(like a mouse but bigger) chewed through the cords of my favorite power tools, random spots in a full 250’ roll of romex and for good measure almost though the hose of a full outboard motor’s portable fuel tank. Ever since I sealed all the door gaps and twice a year put out new glue traps. Never found evidence of any new voles in the traps, just bugs.

 

[robbob2112]

That's why I also mentioned BS88 fuses.

In principle, you want either aR/gR or gS class fuses. The first ones are very fast acting, semiconductor protection fuses. The latter one is for battery protection. NH type fuses are available in these classes.

Actually, this looks most promising for time curves and interupt current.. not as good as a class T in several respects, but better than others we commonly use. And DC rated and available from mouser.

https://www.eaton.com/sg/en-us/skuPage.PV-200A-1XL-15.html

 

[ricardocello]

Actually, this looks most promising for time curves and interupt current.. not as good as a class T in several respects

Some interesting numbers highlighted by me.

 

[seneysolar]

What type of shelves are these?

I worry about compressed cells on a shelf that sags in the middle, it squeezes the cells at the top like an accordion.

Was there ever a determination of the previous cell failure in the other post?

 

[AntronX]

I worry about compressed cells on a shelf that sags in the middle, it squeezes the cells at the top like an accordion.

I've seen this effect with 100Ah cell stack. Compression was at least 2x on top with compressible foam sheets in between cells. Without compliant material between cells they must have been getting crushed at top corners. I also doubt that fuse started the fire. It could have blown as one cell stack developed internal short and sunk current via fuses from another battery stacks.

 

[robbob2112]

Some interesting numbers highlighted by me.

What voltage was that at and how does it compare to the class T. I only looked at the aic and time curves

 

[ricardocello]

What voltage was that at and how does it compare to the class T. I only looked at the aic and time curves

I’ve never seen milliohm or wattage specified for Class T.
Looks like Eaton measures at the rated current and 0.7 rated current.

 

[robbob2112]

I’ve never seen milliohm or wattage specified for Class T.
Looks like Eaton measures at the rated current and 0.7 rated current.

I just happen to have a class T fuse and a 4 wire milliohm meter downstairs. I'll check the resistance in the morning to see what it says.

 

[ChrisG]

The class T fuse is rated at 300 amps but is connected to the positive post of the battery with 12” of 2 gauge welding cable and the fuse holder is screwed to a steel plate on the shelf.

@Brett V do you happen to have a photo of this? I was thinking with something similar off the positive post.

 

[Brett V]

@Brett V do you happen to have a photo of this? I was thinking with something similar off the positive post.

Went to the garage and got a few shots. Hope this answers your question.

 

[ChrisG]

Went to the garage and got a few shots. Hope this answers your question.

@Brett V , Perfect. Thank you. I have a Lynx power-in converted to use 58v 125amp mega fuses for my 100ah batteries. Want to add a layer of additional protection like this.

 

[koppted]

Yes I saw that, I always worry about liquid oozing out and weakening fiber based shelves:

View attachment 212088

This makes me think: One cell leaked/shorted, the resulting fuse blow from excess current arced and ignited the gasses from the cell leak and that's what caused the fire, not the fuse itself.

 

[Alkaline]

Glad ya'll safe, but after seeing that leaky cell I'm thinking this is a runway grade B disaster, that liquid is highly flammable:

so if one of your cells puffed and spilled its goo, it would be an easy source for an ignition.

 

[SparkyJJO]

I still think the biggest unknown is WHY did the fuse blow in the first place? What failure started the whole thing? A leaky cell won't blow a fuse, just provide fuel.

 

[Brucey]

So potentially lack of insulating material between/underneath cells caused friction during expansion/contraction and wore thru the thin film around the cell causing a short.

 

[Q-Dog]

We may never know, but I see a lot of sloppy heat shrink on/under some of these lugs.

 

[Bluedog225]

I was under the impression they (LiFePO4) would not do this. Too bad.

 

[Tulex]

I've got lugs that look like that with heat shrink. Not unlikely that he did as me, cut it away on the back side.

 

[Brucey]

I was under the impression they (LiFePO4) would not do this. Too bad.

On the positive side if that had been a lithium ion battery bank everyone would prob be dead from the intense fireball

 

[Mattb4]

We may never know, but I see a lot of sloppy heat shrink on/under some of these lugs.

It seems odd to see the black insulated wires coming from the fuse block. One other oddity I notice is the large washer on the second to bottom left lug. Though the battery failing due to leakage causing a short seems a more likely possibility.

 

[Mike 134]

So sorry for the folks who lost their home. Are you a friend to have this much time to post? I can't imagine going online and posting after having lost my house.

 

[Hedges]

I was under the impression they (LiFePO4) would not do this. Too bad.

No more than a leaking propane tank would.

LiFePO4 is different from chemistry in laptops, Chevy Bolt etc. in that it doesn't "Vent with Fire"; it vents below the auto-ignition temperature. A spark is needed to ignite it. The others, if they vent they are burning.

Unlike a propane tank, these might just release combustible liquid, not flammable gas. But if overheated, e.g. because damage causes internal short or there is an external short or over-charge, it heats up and gives off gas.

 

[Ampster]

the 100 kwh lifepo4 battery caught fire at night and burned our house down. Luckily our family just made in out on time.

I am sorry for your loss. Just to clarify for other readers and emphasize what others have said, the cause of the fire was the arcing from the fuse not the LFP battery.