diy solar

diy solar

24 volt 280ah compressed..isolated studs...fire

lithium batteries do NOT contain lithium metals...its a lithium salt so there is no worry about active lithium.

the problem with lifepo4 is that it is a high energy object and using something like inert gases or water to "oxygen starve" the energy release does absolutely no good at all...
most fire suppression is to stop "burning" which is the oxidation of the material...burning wood/plastic/whatever...you get rid of the oxygen and the fire stops.
batteries would "burn" in a vacuum because they are dumping stored energy!

you are not going to put it out, your goal is to isolate and minimize damage to surrounding areas.
Right.

But the cells themselves are not where the fires start. Will has a bit of an open bounty on evidence of lifepo4 cells actually catching fire themselves, rather than surrounding wiring or components. Overcharging or overdischarging or physically puncturing the cell just offgasses without catching on fire.

The other things around the cells is what you need to stop before the cells are destroyed and start contributing their electrolyte to the burn.
 
By the time you have melted through your sand bags the damage has long been done because it will have been on *fire* for a while.

Not sure about that.
What is the exact moment you would start to extinguish the fire ?


I have also an another approach to the problem ....
Measuring the expansion of the cells, or the force it needs to compress them.
That is a very good indicator that here will something bad happen ...
 
Not sure about that.
What is the exact moment you would start to extinguish the fire ?


I have also an another approach to the problem ....
Measuring the expansion of the cells, or the force it needs to compress them.
That is a very good indicator that here will something bad happen ...
excellent video find!!! the idea about monitoring cell pressure seems like a really good idea.

just found this on amazon, I think I may give it a try!!
 
excellent video find!!! the idea about monitoring cell pressure seems like a really good idea.

just found this on amazon, I think I may give it a try!!

That sensor seems nice. But I do not know if it is conductive so it needs some separator too.
After a RasPi and a relay to disconnect pack ... That could stop a bloating in early state. Also more heat sensor.
 
Not sure about that.
What is the exact moment you would start to extinguish the fire ?


I have also an another approach to the problem ....
Measuring the expansion of the cells, or the force it needs to compress them.
That is a very good indicator that here will something bad happen ...
The exact moment I'd start to extinguish the lifepo4 fire is right after something other than the cell starts on fire.

Because I'm not sure if you noticed but your own video shows it not starting on fire.

Since the cell(s) is not going to be the cause of the fire, I'll put out the stuff around it first. A couple of series fusible links above the cell wiring to a spring loaded c02 canister and normally open relay with the coils ran through couple non-resetting thermal disconnects in series (rated for say 200F) and the problem vanishes in a hurry.

Put the links and thermal switches right above the problem areas. 2 each can cover a 4s battery for example.

By the time the c02 has dissipated, the burning wiring will be out and the wiring cooled down.

I've actually built such a system and it works well for disconnecting a resistive heater then flooding the cabinet to rapidly cool everything off.

Stopping it when the cells bulge is nice to protect the cells I guess, if the bms fails that is. However that won't stop anything else from catching fire - which is where the problem lies with lifepo4.



Of course all of this is solved by properly setting up your battery pack in the first place.
 
The exact moment I'd start to extinguish the lifepo4 fire is right after something other than the cell starts on fire.

Because I'm not sure if you noticed but your own video shows it not starting on fire.

Since the cell(s) is not going to be the cause of the fire, I'll put out the stuff around it first. A couple of series fusible links above the cell wiring to a spring loaded c02 canister and normally open relay with the coils ran through couple non-resetting thermal disconnects in series (rated for say 200F) and the problem vanishes in a hurry.

Put the links and thermal switches right above the problem areas. 2 each can cover a 4s battery for example.

By the time the c02 has dissipated, the burning wiring will be out and the wiring cooled down.

I've actually built such a system and it works well for disconnecting a resistive heater then flooding the cabinet to rapidly cool everything off.

Stopping it when the cells bulge is nice to protect the cells I guess, if the bms fails that is. However that won't stop anything else from catching fire - which is where the problem lies with lifepo4.



Of course all of this is solved by properly setting up your battery pack in the first place.

Lets see the events in order
  1. something bad happens: overvoltage, or one cell inner tear
  2. Cell starts to bloat
  3. Cell starts to heat up, more bloating
  4. Depending on compression:
    1. If compressed properly then overpressure valve opens on top blasting hot corrosive gases
    2. If not then extreme bloating and only later overpressure valve
  5. Cell heat up even more ... inner glow and/or burning.
  6. Cell heats up nearby cells enough that they starts bloating
So cooling could be a good thing with CO2, but would need a big CO2 tank.

Also as you can see LFP can start fire
 
Last edited:
Lets see the events in order
  1. something bad happens: overvoltage, or one cell inner tear
  2. Cell starts to bloat
  3. Cell starts to heat up, more bloating
  4. Depending on compression:
    1. If compressed properly then overpressure valve opens on top blasting hot corrosive gases
    2. If not then extreme bloating and only later overpressure valve
  5. Cell heat up even more ... inner glow and/or burning.
  6. Cell heats up nearby cells enough that they starts bloating
So cooling could be a good thing with CO2, but would need a big CO2 tank.

Also as you can see LFP can start fire
Fair enough on the puncture.

But it won't start on fire on its own. Everything else needs to catch fire first. This includes overcharging. They bloat, rupture, and offgas heavily without fire.

Everything else can be stopped with proper electrical safeguards.
 
The exact moment I'd start to extinguish the lifepo4 fire is right after something other than the cell starts on fire.

Because I'm not sure if you noticed but your own video shows it not starting on fire.

Since the cell(s) is not going to be the cause of the fire, I'll put out the stuff around it first. A couple of series fusible links above the cell wiring to a spring loaded c02 canister and normally open relay with the coils ran through couple non-resetting thermal disconnects in series (rated for say 200F) and the problem vanishes in a hurry.

Put the links and thermal switches right above the problem areas. 2 each can cover a 4s battery for example.

By the time the c02 has dissipated, the burning wiring will be out and the wiring cooled down.

I've actually built such a system and it works well for disconnecting a resistive heater then flooding the cabinet to rapidly cool everything off.

Stopping it when the cells bulge is nice to protect the cells I guess, if the bms fails that is. However that won't stop anything else from catching fire - which is where the problem lies with lifepo4.



Of course all of this is solved by properly setting up your battery pack in the first place.
Could you post some pictures of this once you get it built?

I agree the the LFP cells do not self combust and if a fire starts it is likely the wire insulation that will be where it starts.
 
Could you post some pictures of this once you get it built?

I agree the the LFP cells do not self combust and if a fire starts it is likely the wire insulation that will be where it starts.
I have no intention of building anything like that.

The problem is solved with proper wiring practices, understanding your cells and settings on the bms, and ensuring your battery pack is well protected and well built. If possible, without a metal case.
 
I think it is good and bad in the same time :)

Compression is good.
But there is no electric separator between the cells and the metal frame. On the side and on the bottom.
Maybe it is not clear in the images I provided but there are gray electrical insulation sheets at the end and sides. The bottom is wood. So there is no contact between the frame and the cells.. the cells themselves are separated by rubber strips. With about 3/4 gaps between them. This allowed some air flow,. Temp sensors and water to get between the cells..

The water channels between the cells is not to put the fire out , impossible, but only to remove the heat so that adjacent cell do not get into a runaway as well.

Most of the abuse and videos are for uncompressed cells. It seems that because they are allowed to bloat that the actual safety vent does not open.

Regarding flat surfaces and multiple contact surfaces.

The serrated washers used in grounding wires are all about multiple contacts that cut thru the oxidation to insure contact.

Yes braided shields are often made and used with solid ends. It looks neat but the problem moves just to the next layer...how is the contact between the solid end and the braided shield.

In the busbar literature there is a lot of discussion about contact pressure and different coefficient of expansion of aluminum , copper and steel bolts leading to variation in contact pressure..
Indeed Belleville washers are discussed for keeping the contact pressure more constant over a range of temperatures.

Johan
 
A temp sensor between the cells will serve as a pressure point during expansion which may cause the film to break or worse, the case. Of course the severity depends on the physical construction of the sensor itself. Most are a small junction which is effectively a tiny bead.

Makes for one hell of a way to puncture things.

Best place to measure the pack temperature, worst place to have a temperature sensor.

As for the safety vent... it should never open. Compressed or not. If it has, the cell should be scrapped.

And I still maintain that water should not be anywhere near the equation under any circumstances other than "the whole structure is already burning down".
 
Maybe it is not clear in the images I provided but there are gray electrical insulation sheets at the end and sides. The bottom is wood. So there is no contact between the frame and the cells..
That is good. Sorry I did not see it in the pic :)
So on the cells edge that is this silver insulator too and not a metal plate? It fooled me.


the cells themselves are separated by rubber strips. With about 3/4 gaps between them. This allowed some air flow,. Temp sensors and water to get between the cells..
Not a good idea. The cell bodies will bloat. The main reason for the fixture/compression to deny it.
Also it will crush anything between them.


The water channels between the cells is not to put the fire out , impossible, but only to remove the heat so that adjacent cell do not get into a runaway as well.
I was thinking about water cooling. Tesla has water pipes around all the cylindrical cells.
And now they use prismatic cells (chinese Tesla Model 3).
I am curious and waiting for a Munro disassembly to see how Tesla did it :)
 
Last edited:
Yes insulation all around.

Not a good idea. The cell bodies will bloat. The main reason for the fixture/compression to deny it.
Also it will crush anything between them.

Can you show a little more substantiated reasoning / calculation / experience why or if indeed the the solid rubber strips between the cells are going to be crushed and or why the 3/4" wide unsupported casing would expand sofar that the cells would touch each other?

Regarding fire..
Is there documented reasoning why water would be a bad idea to cool cells down and prevent a thermal runaway from affecting other cells?

Sofar:

-Airlines use water
-Tesla says use a lot of water
-The Netherlands drops a smoking BMW in tank of water.

I am hoping for arguments less opinions.

Johan
 
Can you show a little more substantiated reasoning / calculation / experience why or if indeed the the solid rubber strips between the cells are going to be crushed and or why the 3/4" wide unsupported casing would expand sofar that the cells would touch each other?
As you could see in the test videos the side of the cells gets bloated.
I can not find the pic ... there was a 4s or 8s with maybe 100-200Ah cells and weak fixture ... that opened up like a flower.


Regarding fire..
Is there documented reasoning why water would be a bad idea to cool cells down and prevent a thermal runaway from affecting other cells?
Water direct contact to cells is a bad idea. Short.

Sofar:

-Airlines use water
-Tesla says use a lot of water
-The Netherlands drops a smoking BMW in tank of water.

I am hoping for arguments less opinions.
Tesla uses water in water pipes. That is a good solution. The pipe is a long wide strip going around all cells.
 
Yes insulation all around.



Can you show a little more substantiated reasoning / calculation / experience why or if indeed the the solid rubber strips between the cells are going to be crushed and or why the 3/4" wide unsupported casing would expand sofar that the cells would touch each other?

Regarding fire..
Is there documented reasoning why water would be a bad idea to cool cells down and prevent a thermal runaway from affecting other cells?

Sofar:

-Airlines use water
-Tesla says use a lot of water
-The Netherlands drops a smoking BMW in tank of water.

I am hoping for arguments less opinions.

Johan
They flood it with water after the entire thing is already up in flames. At that point its no longer an electrical fire so much as a chemical one.

If your entire pack and trailer or shed or whatever is up in flames then you're not going to have enough water to put it out.

The cells themselves will be the absolute last thing to catch fire unless they've been punctured. This is easy to prevent. Don't stab your battery pack. Therefore the fire is like 99.9% certainly going to be from your wiring.

If the wiring is the source of the fire then putting water on it will only make it worse. You need to cut power and extinguish the burning insulation then cool it down.

If you dump water onto the works you'll short the terminals and most likely rupture the cells.

If you are "only flooding the body" or whatever your plan is then you've wasted time and energy when the wiring fire is what needed attention, which you can't put out with water because of the previously mentioned shorting issue.
 
Ok, thanks for all the help.

Why would the wiring be on fire, that is what breakers and fuses are for. The only potential short that we do not control is inside the cel.

The lishens cel has a 16 mm aluminum post. Industrial Aluminum busbars of that size are rated between 0.5and 1.5 amp/ mm2 depending on the temperature rise you allow.. that ends up being 36 to a 100 amp. The delivered copper busbar has about the same spec.

Now we still have to add the extra resistance because of the bolted connections.
Depending on the environmental temp and airflow this could easily become a hot spot with the cel itself being the cooling element.

Then there is the internal resistance adding to the heat.

With the current going out of the battery being under our control( fuses , breakers) I need to concentrate on possible problem inside the cel. Caused by vibration, poor factory assembly, increased internal resistance , chemical decay, old age.

Short shot,. I have no clue what you are talking about..dis you look up the examples I provided..
Notice that Tesla and bmw have very high voltages.. we are only playing with 12 - 24 and sometimes 48 volt.. so shorting because of water is not even in the picture. With high voltage and some conductivity you could be nervous about getting shocked but 24v is not going to do it.

You folks put opinions out there with such conviction. Shorting because if water. Every normal car fire with a 12v 100 amp battery is put out with water.
Rupture cells because water on the outside???? Just like that, no reasoning, links.?


I indeed wasted my time in the battery section. The issue is preventing escalation of a thermal event with cooling down the battery.
Anybody done it over here? If not , give a well thought out theory why not with links research paper otherwise you just make empty posts and waste everybody time.

Johan
 
You have no idea because you're refusing to understand that once the battery cells are on fire it's no longer exclusively an electrical fire.

You don't understand what you're asking and thus don't understand why it's wrong.

You're just stuck on this idea that water is fine because that's how they put out a fully engulfed pack/vehicle, but you're trying to do so before it ever reaches that point.


For about the 5th time: stop with the water. You're not doing what you think you're doing with it.


This isn't rocket surgery. You don't need to concentrate on what's going on with the cell because by the time that's an issue everything around it is already on fire.

I'm not sure how else to explain that. Unless you puncture the cell, the cell itself will not be the cause of the fire.

You say "why would the wiring be on fire that's what fuses are for".

Exactly!

You prevent the cell from being on fire by preventing everything around it from being on fire because the cell itself will not cause a fire unless you physically stab a hole through it.

If properly built and properly wired and properly protected from damage the worst you can get out of the cell is a bunch of smoke unless you toss it into something else that's on fire.

There's literally no other way to explain it. You're wasting your time trying to do this with water.
 
The issue is preventing escalation of a thermal event with cooling down the battery.
Conversely I would need to be convinced cooling would prevent or stop a thermal event or thermal runaway. If a cell decides to go into a thermal runaway event then dry ice might not help because of the chemical reaction inside the cell. I would rather use sand to put out a fire or a proper fire extinguisher.

Cooling is not necessary. The cells do not heat up enough to justify it. If the cells are used in high temps then Peltier cooling or something properly implemented might be a good solution.

Many manufacturers recommend cell compression. Check with the manufacturer of your cells.
 
Last edited:
I indeed wasted my time in the battery section.
A good healthy fudge factor avoids problems. A 175 motorcycle at full throttle vs a 750 cycle at 1/4 throttle. Over build and throttle it back. As for the details, they've been dealt with on every build since the first. "Geniuses" cause most problems.

Your battery looks fine to me. I'd lose the grounding strap, most bus bars have bows for any needed give. I came across 2mm recycled rubber sheets that truckers use to prevent load shifting. 2ft x 4ft+- a ft and dirt cheap. Great for insulation and vibration. I use 4 layers of it as insoles of my running shoes.
 
Last edited:
Why would the wiring be on fire, that is what breakers and fuses are for. The only potential short that we do not control is inside the cel.

Johan
The number 1 cause of heat / fire in any electrical system is a bad connection.
Over time with heating, cooling and vibration and oxidation .... and even galvanic oxidation can cause a bad connection. Any time there is aluminum in electrical wiring, there is a greater chance of a bad connection developing over time. A bad connection + current flow = heat, and that can be run away heat that can cause a fire.
The thread Fhorst posted where his pack caught fire resulted in nothing but disagreement about what may have caused the fire .... given that we have never seen it happen before, I would have a hard time believing it was thermal runaway of a cell.

We had a situation once where a frequency drive we were controlling caught fire and almost burned the building down. Of course they tried to claim it was because we weren't controlling it properly .... which was ridiculous .... It was obvious that the main power terminals was where the fire started and it was a bad connection .... but we had to hire an expert to prove that to them.

There was a thread recently where a guy with a marine installation had a fuse holder melt down and he caught it just before it broke into flames.

People have intentionally mistreated these LFP cells and there is only the one situation I have seen where the cell was repeatedly stabbed with a metal object that a fire broke out one of the holes that had already been stabbed. That fire may very well have been caused by a spark from the metal object igniting the cell electrolyte that had already spilled out ......So, the cell itself just does not have a fire risk .... or a risk so low as to be almost insignificant ... in the case of LFP cells.

In the early days of this forum, there were opinions that there needed to be some circulation between cells, but as people have built more and more packs and observed pack temperatures under different circumstances, the general opinion has changed and this is generally not recommended any more.
Since we started getting an examining the EVE spec sheets, it has become apparent that controlling the compression of a pack is much more important over all.
I could see situations where the pack was always in a hot environment that the priority might change, because higher temperatures will also cause quicker degradation .... but, in that case, it would probably be better to figure out a way to control the environment.
 
Last edited:
Back
Top