diy solar

diy solar

Monitoring pack pressure for safety

fafrd

Solar Wizard
Joined
Aug 11, 2020
Messages
4,188
I’m concerned enough about the 2-3 reported fires here on the forum that I’m considering adding a pressure sensor to my 16-cell LiFePO4 battery pack.

I’ve already got my single-line of cells in a 300kgf clamping fixture which I’m setting to ~12psi @ 50% by counting turns on my 8 calibrated springs so adding a pressure sensor should be straightforward and inexpensive: https://www.amazon.com/dp/B07S9YJH8C

I’ll probably just do this and monitor the pressure range I measure under ‘normal’ operation before deciding what I want to do with this pressure signal and how to use it for an extra level of protection.

Has anyone else done anything like this?

Once I know what the max pressure is at 100% SOC, I’m thinking I’ll run the pressure signal into a threshold circuit controlling an alarm and perhaps eventually a fire extinguisher / preventer.

Any thoughts & advice appreciated…
 
Last edited:
I’m concerned enough about the 2-3 reported fires here on the forum that I’m considering adding a pressure sensor to my 16-cell LiFePO4 battery pack.

I’ve already got my single-line of cells in a 300kgf clamping fixture which I’m setting to ~12psi @ 50% by counting turns on my 8 calibrated springs so adding a pressure sensor should be straightforward and inexpensive: https://www.amazon.com/gp/aw/d/B07S...rd_i=B07S9YJH8Cp13NParams&smid=A3KMW5XZ660KGT

I’ll probably just do this and monitor the pressure range I measure under ‘normal’ operation before deciding what I want to do with this pressure signal and how to use it for an extra level of protection.

Has anyone else done anything like this?

Once I know what the max pressure is at 100% SOC, I’m thinking I’ll run the pressure signal into a threshold circuit controlling an alarm and perhaps eventually a fire extinguisher / preventer.

Any thoughts & advice appreciated…

Good luck with that.
 
Would not recommend triggering a fire extinguisher based on pressure alone. Sounds like a nightmare to clean up on the first false trip.
Appreciate the thought, but first, if I can trade off the (small but catastrophic) risk of burning down my house go the hassle of cleaning up after a false alarm, I think that’ll be a good trade-off.

And second, since I won’t do anything until I’ve got months if not a full year’s worth of ‘normal’ pressure data, the risk of a false alarm can be reduced close to zero by setting the threshold high enough (>2X normal maximum pressure, for example).

But more importantly, what would you suggest as the best use of a early warning of a possible combustion event?

I can shut down the whole system, but that’s not going to do anything to stop an internally-shorted cell from continuing to heat up…
 
Good luck with that.
Can’t tell if you mean this is a stupid / impossible / impractical idea (in which case I’d appreciate details) or if you’re wishing me well in thus new initiative…
 
Can’t tell if you mean this is a stupid / impossible / impractical idea (in which case I’d appreciate details) or if you’re wishing me well in thus new initiative…
I'm pretty much in agreement with @HighTechLab.

If you want to do a fire protection system, I think a more conventional Heat / smoke detector system would be more reliable.
Do we have any data to indicate there is a risk of a cell going bad in such a way that it first increases pressure and then starts a fire?
 
I'm pretty much in agreement with @HighTechLab.

If you want to do a fire protection system, I think a more conventional Heat / smoke detector system would be more reliable.
Do we have any data to indicate there is a risk of a cell going bad in such a way that it first increases pressure and then starts a fire?
Gotcha, thanks.

If there are any threads on the conventional Heat / smoke detector system I’d appreciate a pointer.

That’s always an option and I’ll probably add it in any case, though by the time you detect smoke it’s certainly ‘too late’ and while heat at the individual cell level should be a much earlier indicator, heat at y try he battery level may not be.

I’ve not seen any data regarding excess pressure being a reliable precursor of a failure resulting in combustion - that’s why I’ve decided to be the guinea pig ;).

Looks easy and inexpensive to start collecting / monitoring pressure data from my 300kgf fixture which I think will be interesting in any case.

What would be fantastic would be if anyone has heated up cells while monitoring pressure. If heated cells bloat and increase pack pressure, as several suspect, that would make pressure a reliable proxy for increased cell temperature.

I’ve got a couple spare cells and I suspect one or two are failing, so I may resurrect my temperature-controlled chamber, add my new pressure sensor to it, and sacrifice one of my spare cells for science to see what happens to bloat / pressure when temps increase beyond normal operating range (eventually, other priorities for now).
 
You are many many many more times likely to start a fire from a bad electrical connection than an overpressure burst of a LFP cell.

An LFP burst spits out electrolyte surrounded by CO2 gas. Solvent in electrolyte is about as flammable as fuel oil. The most likely way to get it to ignite is to put an open flame about 2 feet away from the direction of burst port to give enough distance for the CO2 envelope to disperse leaving just the misted electrolyte solvent to hit the open flame.

A fire around a battery pack is likely started by a bad electrical connection that starts the fire which overheats the cells to bursting point.
Cell pressure would likely be a lagging indicator of a fire already started.
 
Last edited:
Gotcha, thanks.

If there are any threads on the conventional Heat / smoke detector system I’d appreciate a pointer.

That’s always an option and I’ll probably add it in any case, though by the time you detect smoke it’s certainly ‘too late’ and while heat at the individual cell level should be a much earlier indicator, heat at y try he battery level may not be.

I’ve not seen any data regarding excess pressure being a reliable precursor of a failure resulting in combustion - that’s why I’ve decided to be the guinea pig ;).

Looks easy and inexpensive to start collecting / monitoring pressure data from my 300kgf fixture which I think will be interesting in any case.

What would be fantastic would be if anyone has heated up cells while monitoring pressure. If heated cells bloat and increase pack pressure, as several suspect, that would make pressure a reliable proxy for increased cell temperature.

I’ve got a couple spare cells and I suspect one or two are failing, so I may resurrect my temperature-controlled chamber, add my new pressure sensor to it, and sacrifice one of my spare cells for science to see what happens to bloat / pressure when temps increase beyond normal operating range (eventually, other priorities for now).

I would be interested in seeing how you do the pressure monitoring .... and some of the data about how it tracks with SOC, temperature, etc
I think an easy to implement pressure monitor would be nice to have to experiment with.
Especially since you have a spring fixture.
 
You are many many many more times likely to start a fire from a bad electrical connection than an overpressure burst of a LFP cell.

An LFP burst spits out electrolyte surrounded by CO2 gas. Solvent in electrolyte is about as flammable as fuel oil. The most likely way to get it to ignite is to put an open flame about 2 feet away from the direction of burst port to give enough distance for the CO2 envelope to disperse leaving just the misted electrolyte solvent to hit the open flame.

A fire around a battery pack is likely started by a bad electrical connection that starts the fire which overheats the cells to bursting point.
Cell pressure would likely be a lagging indicator of a fire already started.
Appreciate all of this insight (and it’s exactly what I was hoping for when I started the thread).

As far as a bad electrical connection being a far higher risk of starting a fire than a shorted cell, we are in total agreement.

As far as a fire starting from a bursting cell, that’s not really the failure I was worried about.

I’ll lay out my working hypotheses and would appreciate your insight:

1/ a cell shorts (possibly from wiring failure, possibly from internal terminal failure due to mechanical stress, possibly due to some other mechanism such as dendrite formation possibly combined with a manufacturing defect (Chevy-Bolt-like, and yes, I understand the chemistry is different).

2/ high inter-cell discharge generates a excess heating of the generator me shorted cell.

3/ as the shorted cell heats up, it heats up the adjacent cells above normal operating temps.

4/ as the adjacent cells heat up above normal operating temps, they bloat beyond normal operating levels, resulting in pack pressure exceeding normal operating ranges.

So it was pack pressure exceeding normal operating range that I was interested in, in the hope that the pack pressure might increase from that effect before the primary shorted cell can reach the temps needed to start combustion.

I have no idea whether cell pressure/bloat even increases with increasing temps, let alone whether any such effect would be ‘fast enough’ to be an early indication of a shorted cell heating up y to k the point it could start a fire.

But it was not the risk of a bursting cell I was going after as much as the risk of an individual cell which becomes shorted (for whatever reason).
 
I would be interested in seeing how you do the pressure monitoring .... and some of the data about how it tracks.
Cool - you’ve given me extra motivation to follow through and I’m going to go ahead and invest $10 in one of those Amazon sensors…
 
Treat the box like a fireplace, so it's not touching combustibles and its vented to the outside
Interesting, so metal box on brick or ceramic with a vent….

Couldn’t you even build a complete enclosure out of brick / pavers? (With vent, if needed).

What is the purpose of the vent? To assure the gasses emitting from any bursting cells get vented to the outside rather rust reaching the living space?
 
Thermally blocking between cells will help prevent a cascading meltdown. Durock board (shower wall material) may be good enough.
 
Treat the box like a fireplace, so it's not touching combustibles and its vented to the outside
I'm building a separate enclosure that is not inside my shop. Should something happen, I lose my batteries and Growatt, not my shed and all that is in it. I have 16 more batteries on the way. Once I build them, (no matter how awesome I think I have wiring, fuses, breakers, etc..) I'll at least have that piece of mind. The compression should not be an issue at low C-rates and good quality cells. I do compress mine and have had no issues thus far.
 
I'm building a separate enclosure that is not inside my shop. Should something happen, I lose my batteries and Growatt, not my shed and all that is in it. I have 16 more batteries on the way. Once I build them, (no matter how awesome I think I have wiring, fuses, breakers, etc..) I'll at least have that piece of mind. The compression should not be an issue at low C-rates and good quality cells. I do compress mine and have had no issues thus far.
Pretty sure I replied to the wrong person!!!
 
I built my pack in an aluminum truck toolbox. It seals up well and fires don’t burn well without air. I trust the temperature sensors on the BMS to halt any charge or discharge long before things heat up to a dangerous level and a class T fuse to protect against any sort of catastrophic short.
 
Would not recommend triggering a fire extinguisher based on pressure alone. Sounds like a nightmare to clean up on the first false trip.
can you ever extinguish battery fire with fire extinguisher? Some of them will burn internally without oxygen. You could also use a CO2 or halon system to flood the room in case of fire (then you have no mess to cleanup, besides the fire mess itself)
 
Back
Top