Lithium batteries v/s NiMH batteries the Debate! Battery technologies

Lets stage a test: Use similar energy size/power capability propulsion packs, and heat them to the point of ignition, then see which one makes the worst fire, most heat, etc? Based on performance in my Vectrix scooter, I think that using 1 gallon of gasoline, and 40 ampere-hour batteries, at a nominal 125 volts (5.0 KWH Battery) would be a reasonable comparison, as similar range, at 50-60 mph, could be expected in a similar vehicle powered by either of these sources. I would think that burning a Ni-Mh battery would cause release of the HYDROGEN stored inside, and as we know, hydrogen is a good fuel, readily combustible in air. Lithium also will burn, the question is which battery, if both store identical electrical energy, will burn more violently, releasing the most THERMAL energy? How will the thermal energy released by battery destruction compare to the thermal energy released by burning 1 gallon of gasoline? How toxic are the fumes and byproducts? How difficult would it be to EXTINGUISH the respective fires? How much heat is required to START the respective fires? These are issues that seem most relevant to usage hazards. (Testing the respective batteries at a continuous 10 C discharge, and noting battery temperature rise/or other effects, under this abusive, rapid discharge, should also be noted)--Bob Curry

To make the test more affordable one could scale it down. Let's say we use the following:

- a single 40Ah LiFePO4 cell at 3.2V nominal = 128Wh
- 128/5000 * 1 gallon of fuel
- an equivalent 128Wh NiMH set up (could be a 2S4P of 10Ah cells?)

If anyone wants to do this comparison test I will donate up to three (3) 40 Ah cells. Any potential takers? Note, you'll need to be (a) independent (Mik, I'd be happy to send you the cells if you had any interest in this); (b) be prepared to submit a test plan for public peer review; (c) be prepared to video all tests and test prep and post on YouTube (or similar) & (d) be prepared to publish all test results for peer review.

Any takers?

p.s. three is the initial number of cells. If that proves insufficient but you can prove good progress I'll provide more cells as required.

For improved battery safety/reliability, I believe that any battery management system (BMS) should monitor every cell for under or over voltage and temperature, and all leads to all cells shold be fused, or otherwise protected from overcurrent, in event of physical damage or shorting. The BMS should also detect any OPEN or shorted sampling leads. The charger should be disabled/suspended upon any out of tolerance indication, such as overvoltage or extreme temperature, hot or cold. I have seen installations using unfused cell sampling leads to the BMS, and some situations could cause this wiring to ignite, causing a fire! Unfused/unprotected battery wiring is a BAD idea! Installation of "Polyswitch" current limiters, fuses, or current limiting/sacrificial resistors at each cell connection sample point should be done. The MAIN power leads should be fused, at BOTH positive and negative sides of the battery, for improved protection in case of accident. All solid-state devices/circuits should be properly R.F. (Radio Frequency) shielded, filtered, and bypassed, to eliminate interference to or from other devices.Lack of such protection can result in solid-state devices switching on or off unexpectedly, or failing to properly respond to an input. Many manufacturers do NOT adequately address this R.F. filtering and shielding issue! (Some simply advise against installing 2-way radio equipment in the vehicle, but that gives NO protection to you from the vehicle NEXT to you with the 1,000 watt 2-way radio, nor does it protect you when driving past a 5,000,000 watt TV transmitter tower, or a 50,000 watt radio station transmitter site. (Some unprotected circuits have been known to malfunction/self-destruct from nearby lightning strikes, which also generate R.F. energy.)-Bob

For improved battery safety/reliability, I believe that any battery management system (BMS) should monitor every cell for under or over voltage and temperature, and all leads to all cells shold be fused, or otherwise protected from overcurrent, in event of physical damage or shorting. The BMS should also detect any OPEN or shorted sampling leads. The charger should be disabled/suspended upon any out of tolerance indication, such as overvoltage or extreme temperature, hot or cold. I have seen installations using unfused cell sampling leads to the BMS, and some situations could cause this wiring to ignite, causing a fire! Unfused/unprotected battery wiring is a BAD idea! Installation of "Polyswitch" current limiters, fuses, or current limiting/sacrificial resistors at each cell connection sample point should be done. The MAIN power leads should be fused, at BOTH positive and negative sides of the battery, for improved protection in case of accident. All solid-state devices/circuits should be properly R.F. (Radio Frequency) shielded, filtered, and bypassed, to eliminate interference to or from other devices.Lack of such protection can result in solid-state devices switching on or off unexpectedly, or failing to properly respond to an input. Many manufacturers do NOT adequately address this R.F. filtering and shielding issue! (Some simply advise against installing 2-way radio equipment in the vehicle, but that gives NO protection to you from the vehicle NEXT to you with the 1,000 watt 2-way radio, nor does it protect you when driving past a 5,000,000 watt TV transmitter tower, or a 50,000 watt radio station transmitter site. (Some unprotected circuits have been known to malfunction/self-destruct from nearby lightning strikes, which also generate R.F. energy.)-Bob

I agree with some, but not all, of this:
- per cell voltage monitoring - yep, agreed.
- per cell temperature - I think this is excessive. It's certainly the most safe and is independent on pack design. However, for a contiguous pack of cells and given that the temperature required to start combustion is far higher than the temperature at which cell damage would occur I believe it is adequate to have at least two sensors (for redundancy as well as improved fidelity) in strategic places within the pack. The heat would spread and the temperature gradient would be such that we'd shut down before combustion occurred because we'd have detected cell damaging heat. To test the adequacy of the sensor placement one could simulate a cell furthest from a sensor becoming hot with an alternative heat source (i.e. non-destructive testing could be used to determine temperature sensor adequacy).
- charger cut off based on out of tolerance behavior - yep, agreed.

Before I address the rest I'm going to check with Erik (Chief Engineer at CuMoCo).

...
I don't remember the details - I do remember that we discussed this at CuMoCo around the time that thread was started (or maybe it was a couple of months later when we discovered it - point being we're aware of it).

...
Thank you, excellent! I'm not saying the rapids of the river are impossible to master, but I want awareness! And on VisforVoltage, open communication about such awareness!

Here's hoping that none of us ever see our own battery BBQ!

Amen!
I experienced the "Vectrix Battery Recall" at a time when it seemed patently clear to me that Vectrix Australia was not competent to perform any work on a motorbike which I would like to ride...
And I still have 3 smoke detectors mounted and maintained above the Vectux (and the workbench with all those bits and pieces, mainly NiMH...); and that is after I did the rework myself!

But, if my house was not arranged in such a way that there would be an escape route, even if the garage was on fire, then the blasted Vectux would have to live outside! Battery recall or not!

.

We are at the cutting (or bleeding?) edge here sometimes - you have to be honest about the risks, then take them, nevertheless!

For me, it always seems to come down to surfing analogies...but since the blasted Vectux, I have hardly surfed!
.
If you want to catch a wave, you need to paddle as hard as you can, in the right direction, at the right time... and then, when you (with or without a surfboard) have gone over the edge, when there is absolutely no chance of turning back, then you need to paddle very hard for a little longer, although disaster is already "inevitable": That means you are accelerating down the face of the wave, heading for certain disaster, unless you take initiative and take control and start to steer, somehow.

Sorry about the rant, but I miss those days sometimes, when EV's and VisforVoltage did not matter to me and I chose to access those magic moments around sunrise, when dolphins show you how to do it, and the rainbows are 360degrees double circles...sigh...

Ooops - the last few post of you guys were dropped in whilst I was reminiscing about times gone by...

Life is just getting too fast these days - whilst you dream about a few waves, others write 4 new posts before you know it!

Haha!

...

I agree with some, but not all, of this:
- per cell voltage monitoring - yep, agreed.
- per cell temperature - I think this is excessive.

I think I have some news in that regard!

You can monitor the temperature of all cells (without n + 1 cables!), as long as you don't need to know which cell is the one with the highest temperature! Look at the Prius NHW10 battery pack - temperature sensor strips along all cells. It cannot tell where an error occurs, but it can tell if an (overtemperature) error occurs!

Temperature, as regards overheating, can be monitored by placing a cheap silicon diode in thermal contact with each cell, or mounting the diode to a copper sheet inserted between a pair of cells, and all such diodes could be wired in parallel, to only TWO wires, going to the battery management system(BMS)-this would provide "Overtemperature alarm" to be used to protect the battery, but would not identify WHICH cell was at fault. (A forward biased junction of a silicon diode exhibits a DECREASE in voltage when heated)-An alternative metod, using SERIES connection, is to install normally closed "Bimetallic thermal sensors" on each cell, or between each cell pair, and connect the two end wires to the BMS. Under temperature could be sensed in a similar manner, but fewer sensors would be required for sensing LOW temperature events.-Bob

Anyway, at least you agree that the energy source for the fire melting the gearbox casing and the windows in the Fiat must have been the LiFePO4 battery!

Sorry Mik - I actually don't. I'm simply saying that I don't automatically jump to the battery chemistry being the cause. The couple of PHEV and/or EV fires that I'm aware of that have been investigated to root cause pointed to wiring faults and loose connections as the problem. Except for abused LiPo or lab tests, I've yet to find an EV fire that was caused by the battery pack itself.

What I was trying to suggest in the post is that I fully expect that if a piece of wire is going to get hot enough to cause a problem, the one with more energy behind it should be more of a problem. Look at the car that burned. It was winter. The car was outside. The charger was in the car. Extension cord running to the garage. It could have been something as mundane as a frozen relay. One other factor that comes to mind about the DIY EV that burned - it was a Thunder Sky pack...a bunch of cells wrapped in a lot of plastic...which will sustain a very hot fire once burning... So no - I don't expect the problem has anything to do with LiFePo4 - I expect the fire might have been completely different if the EVer used PSI or headway cells in the pack...

Different point, Mik. I've been hip-deep in LiFePo4 BMS building for quite some some. When I scan the results of a Google search, I can spot LiFePo4 or lithium quickly. I'm just not in a NiMh, sodium sulfur, or NiCd frame of mind. You, on the other hand, have been doing some amazing long-term work with NiMh. I'm not surprised at all that you're thinking NiMh. And because we're 'describing the elephant' from different places... :-)

Have you got a link to some discussion of the Prius fire? I'd like to read about it, I find it important to be informed about such events. But I don't have time to hunt it down myself right now.

Start here, Mate. This one's in Oz. http://townhall-talk.edmunds.com/direct/view/.f12cc22/60#MSG60
"We got the fire out before it progressed too far and the firefighters said the fire started where heat had melted a hole in the far right corner of the protective metal casing for the Prius Hybrid battery pack. "

Bob and John,

I think the fire test would have to include other LiFePO4 cells - some without the plastic...

Lithium certainly will burn, but that's lithium metal. It's my understanding that there's no metalic lithium in a charged LiFePO4 cell. I recall reading something about there being significant metal in a LiPo/LiCo cell. I can't find the document...where DID I read that?

I sliced into a fully charged PSI cylindrical cell and shorted the cell with the cut-off wheel. It boiled off it's electrolyte. It got too hot to touch, and melted the soft plastic 'mar-proof' jaws holding the cell in the vise. But it didnt' get hot enough to ignite anything.

Bob - I agree with fuses 100%!

Andy

Wow! I never heard of Prii going up in flames before, but there seem to be numerous reports!

Here are some more photos of another one: http://www.carspace.com/recycler5/.5c9b8f0c/

I think I'll ban the Vectux from the garage!

Bob and John,

I think the fire test would have to include other LiFePO4 cells - some without the plastic...

Andy

Fair point. I'll be donating 40Ah Thundersky cells of course. But introducing other LiFePO4 style of cells would be good as well. I'd recommend only testing products easily available to the home-builder in their delivered sate.

However, I put the chance of someone stepping up to do this somewhere between slim and none. That's not a complaint (I don't see this test as a high enough priority for us to do here, thus I can't complain if others don't want to do it). However, it does highlight one of the problems with message board debates - one tends to get a lot of "arm-chair hypothesizing" from folks with lots of different agendas (I'm definitely guilty of this!) and very few real world, at least somewhat rigorous tests. I think it's Richard Fechter over on endless-sphere that says "a test is worth a thousand hypotheses" (or something like that).

Andy - tell us more about your BMS. Is it time to start another couple of threads? (1) Combustibility testing of popular BEV cells & (2) What makes the best BMS the best?

BTW, as far as (1) goes - does anyone know a Fire Chief or a high-school or college science lecturer? Someone who might want to either give a small donation to make the test happen or someone who might want to do the test for educational porpoises?

Lets stage a test:
...
(Testing the respective batteries at a continuous 10 C discharge, and noting battery temperature rise/or other effects, under this abusive, rapid discharge, should also be noted)--Bob Curry

Maybe I need to review the testing procedure for the NHW10 sticks!
I test them at 15C (100A) for 10 seconds, then discharge at about 60A (9-10C) to a cutoff voltage of 6V (1V/cell).

The weak cells in a stick show up as warmer than the others in the minutes following the test.

I still have not found a really bad stick, it remains to be seen how hot they get!

The question is though, do i damage the sticks with the 60A discharge test? They stay below 40-45 degC (from memory) during the test.

https://www.endless-sphere.com/forums/viewtopic.php?f=14&t=12764

Lets stage a test:
...
(Testing the respective batteries at a continuous 10 C discharge, and noting battery temperature rise/or other effects, under this abusive, rapid discharge, should also be noted)--Bob Curry

Maybe I need to review the testing procedure for the NHW10 sticks!
I test them at 15C (100A) for 10 seconds, then discharge at about 60A (9-10C) to a cutoff voltage of 6V (1V/cell).

The weak cells in a stick show up as warmer than the others in the minutes following the test.

I still have not found a really bad stick, it remains to be seen how hot they get!

The question is though, do i damage the sticks with the 60A discharge test? They stay below 40-45 degC (from memory) during the test.

https://www.endless-sphere.com/forums/viewtopic.php?f=14&t=12764

I am new to this group but not new to EV's. I built an EV in 1960!! It used NIFE cells from trolley buses. No need for BMS or worry as they were almost indestructible.. Just huge and heavy. Then tried lead acid.. reliable and easy to use just huge and heavy!! Then NiCd not so huge and heavy but can't find large cells so nightmare with series parallel system, then NiMH plenty of power easy to charge but again can't find large cells. With large numbers of cells the probability that one will fail well before the rest is 100% so you have to have multiple systems to make the pack usable under all the millions of failure modes possible. Have a look at the battery nightmares in the space program I forgot to mention I have tried AgZn nice high capacity but developed shorts for no good reason.. then Zn-air.. not much good there either.
NOW along come my dream battery the LiFePO$ ( funny I typed $ instead of a 4 but may be that is about it too!
Just found that my BMS supplied by a Chinese battery maker were no good from the start and the company will not stand by their xrap!! Result is I have loads of small cells that are good and a few that are dead.. and a hole in my wallet big enough to drop them all through!
Bottom line is that although LiFePO4 has promise it is much like the "promise " that some women offer to their men friends . Easy to offer and only works if you don't make good. Meanwhile I am still seduced by the hope that I can get Li packs to work as I have now a great car ready for packs!! need 400 volts 20 Ah.in easy to handle packs of 50 to 72 volts. any offers out there? . Need cell by cell cutoff 2.5 volts, overcharge protection, capable of charging all cells at one time .. If I find a source I'll let you know.
If you want to know the name of the Chinese company you'll have to wait as I am still hoping for some honor and trust in this new battery madness>
If you want to know about motors then that's another story and one I do know about..I also believe in a Battery Exchange system as being the best solution for EV's. No ownership of a battery needed!! Read it all at www.members.cox.net/rdoctors/evs.html
Themotorman