Lipo fire

[Erfolg]

Woolstenflyers video was most interesting.

It gives a clue as what to expect from a fire.

Perhaps just as important it is also relevant to know how the condition was reached. It is reminiscent of the sort of thing that happened with Nicads, when individuals used unsuitable charging equipment. So the question is, is it (sensibly) possible to achieve the events seen when using suitable charging systems to get to the necessary conditions for the events seen, to take place.

It did seem that even when the Lipo was puffing, electrical energy was being pushed in. Would the process have continued without continuing the process? Would a functioning Lipo charger do this?

Anyway the fire, as I saw it.

• initially the wrapper split
• then the cells vented, what would be very interesting to know what were the gasses. Similar to PVC when burnt, is the volume significant?
• after some period, the cells ignited, was the initiator the current being pumped in. Without this energy input, would the energy humphave  been reached and gone over? Also interesting to know.
• Then there was a spectacular, although short lived fire. It looked spectacular, but what we need to know, was the equivalent to a thimble full of boiling water, or a bath full.
• This initial fierce fire was over in a matter of seconds, a secondary fire then became dominant. It was of far lower apparent intensity. I suspect this was the polymers making up the wrappers etc. could be nasty gasses, in a confined space.
• The whole process of venting to self extinguishment was circa one minute

The video does raise a number of questions, the first would be, are the events seen credible from normal handling, then can they arise from say a shorting incident. I have seen a metal tent peg driven through a lipo, stupid thing to do I thought, although, it just sat there. Were the events seen the equivalent of using a non suitable charger on a Nicad or Nimh, ot putting a naked flame to Pb acid battery.

Given it seems prudent to use fireproof bag, would it be enough, even if it is improbable event?

If the events can be initiated from shorting, then the issues change completely. As most shorting incidents would occur whilst placing in the model or some such non charging event.

My last thoughts are, the fire shown was spectacular, it would suggest to me that a bucket full of water would put out the residual fire as it seems the lithium is all consumed in a matter of 2-3 seconds. Do not get me wrong, I would not want a fire of the type shown in my home. It would suggest that the damage could be confined to a few hundred pounds of resultant damage, not a crater in the ground.

There does seem to be safety issues, which need addressing a controlled, considered manner, rather than a hysterical, approach of all charging of lipos must be done in a concrete silo or bunker, several miles from human habitation.

We have controlled methods of operating with Methanol based and in some ways with a more hazardous petrol fuels, which both limit risks, whilst maintaining a sensible operating and storage regime based on the specific characteristics of the media. We just need to adopt a balanced methodology for Lipos, resisting those who see the opportunity to make life awkward, or just a reflection of their own personal objectives and prejudices.

To reach Nirvana we need a better knowledge base than I certainly am privy to, and I strongly suspect some who are seeking to impose operating procedures.

 

Edited By Erfolg on 21/01/2014 16:46:54

Edited By Erfolg on 21/01/2014 16:48:46

Edited By Erfolg on 21/01/2014 16:50:51

[Peter Beeney]

If you try and take into account the total usage of lithium batteries the overall picture must point to to a fairly incident free environment. When you consider how many units use lithium power nowadays, if there were as many cases of malfunction resulting in fires from this gross figure pro rata as modellers seem to suffer from, I’m sure this would very quickly be highlighted, big time. There have been anomalies, the fiery Nokia phones, for instance, but these have been fully explained, and of course it’s not always lithium-ion polymer cells that are at fault, the Dreamliner cells are standard lithium-ion, with a hard container. I did read that the engineers said that after investigation they could find no faults with the cells, or the charging regime, either. So what does that leave, then…. ?

I think the reason for this level of security is because generally the lithium power systems are regulated, the charging and discharging is totally controlled and the operator cannot interfere with this. Whereas within the aeromodeller’s control this is totally unregulated, the battery can now be subjected to any and every form of misuse known to man. The average consumer using their phone will probably not have the slightest knowledge, or indeed interest, of what powers their must-have communication gadget, nor do they need to.

So from this I would tend to suspect that the modeller may in some cases be misusing their lithium packs, so maybe a better system of semi-failsafe charging at least may be an answer. One idea might be this. Construct a charger that can only charge lithium polymer cells for starters; I think it’s possible that inadvertently selecting the wrong battery type is the biggest cause of these spectacular disasters. Then make it such that it will only fire up in the event of the balance lead being connected, this is so that a cell count can be carried out and confirmed, preventing the wrong setting of number of cells. Then also make the charge rate a semi-auto setting, so ensuring that the cell(s) cannot be charged at a rate that is too high, at least at default startup. Not impossible either, my Schulze has been doing this for the past 20 odd years. Finally, the charger being made to some fairly stringent quality specifications, such as regarding voltage limits and safety cut off points etc., thus guaranteeing within reason that the charger itself doesn’t cause any problems. On the discharge, not so much control here, if say a motor seizes and the resultant max current flow caused the pack to go on fire at least it’s likely to be out in the open and hopefully will not do too much damage. One solution may be is that for the ESC to allow the name-plate current to flow, plus a bit, 20% maybe, but not unlimited current; this might just be sufficient. Crash damage also difficult to guard against, again the only consolation being that it’s most likely in a situation where it can do least harm; I’ve seen and examined quite a number of damaged lipos, but I’ve never seen one go on fire.

I personally think the videos are just controlled vandalism. It certainly demonstrates what can happen when you subject these things to conditions that are strictly advised against. I’d be much more interested in a situation where this occurred when the lipo was being charged as per the book; if this result were to happen when normally charging on a regular basis then I’m sure that the quality control would certainly stand a good dollop of looking at!

Also I have to say that when I was trying to explode nicads I was unable to do it with a standard charger, I had to use a 12 volt car battery and jump leads to be able to get the resistance low enough for the necessary current flow. And this was bashing just a single cell. This doesn’t mean that a charger won’t blow cells/batteries, of course, but I’m pretty sure it can’t be done on an everyday basis. In fact, I used to frequently charge nimh flight packs directly off a 12 volt car lead weight, but that’s another story…

PB

[Martin Whybrow]

Interesting quote from a document on LiPoly battery safety from Texas A&M university:

"While having proven to be safe and reliable under proper operating conditions, Lithium Polymer batteries do present significant safety challenges if mishandled.
The batteries have the potential, because of their chemistry, to ignite through a process known as thermal runaway. This can occur if the battery has taken physical damage that ruptures internal components of a cell, enabling a catalytic reaction to occur involving the cathode and the absorbed lithium.
3
Although quite rare, ignition can occur as a direct result of physical damage alone. This will occur within fifteen minutes of the damaging incident. If the battery does not ignite in this time span and is not charged or used after such damage has occurred, then the damaged cell or cells will simply swell as out-gassing occurs from internal reactions, and ignition will not occur.
Proper procedures for handling and storing LiPo batteries can effectively mitigate this risk.
Ignition can result from charging using improper equipment such as chargers not specifically designed for LiPo batteries, or from charging damaged battery packs. Strict adherence to inspection and charging procedures can very effectively mitigate this risk.
Ignition can result from shorting a LiPo battery pack. Care must be taken when soldering battery connectors onto the battery leads to ensure that the battery is NEVER shorted in the process.
Proper handling and maintenance of the packs can effectively mitigate this risk.
Another risk is over-discharging the battery pack. If the pack is discharged below 3.3 V per cell, cell damage or reversal can occur. In such cases the battery should be considered unusable and disposed of. Careful monitoring of pack voltages, use of proper equipment, and strict adherence to inspection and storage procedures can very effectively mitigate this risk."

The full document can be found here

Edited By Martin Whybrow on 21/01/2014 20:32:43

Edited By Martin Whybrow on 21/01/2014 23:37:09

[Erfolg]

I am not sure that the Dreamliner scenario is necessarily relevant to modellers. Dependant on where th Lipos are situated, the temperatures could be as low -30c region. At the other end of the scale they could be mind boggling hot, when sat on the runway on the equator.

The charger issue could be very relevant. Even my cheap charger, checks the cells relative to my cell input, the cut off supposably is 3.2v per cell. The actual measured voltage via a multimeter is lower, from memory about 3.1, although the cell variation was minimal. Any unexpected or unacceptable setting or occurrence is alarmed, to protect you from your own incompetence. So a reasonable charger does not need to cost a fortune now, God bless the Chinese.

When I think back, my original lipo charger did none of these things.

My original MFA Nicad charger was just a primitive timer and a resistance wire. No wonder there were accidents.

The same situation seems to be true with ESCs, where current devices have built in safty features, as does my Tx, to do there best to protect me, from any stupidity.

Perhaps we all need to kick out all our equipment that does not have a minimum level of security built in. As it is seems many of the scare stories with respect to electric models, is either, poor operating procedures or equipment which is an accident waiting to happen by design.

Incidentally, I have burnt out a inrunner in the air, the battery did not seem to be affected, the ESC shut down for a few seconds, then the BEC unit continued to run. It was on 35.

I do believe that there is a place for education on the handling of Lipos. The scare stories seem more to do with an attitude, that is, "do as I say" as it seems a good idea. As for doing stupid things deliberately , I personally would refrain from doing them, that is unless the conditions were controlled, with respect safety, and it was clear in what I was trying to do and it was a good reason.

[Chuck Plains]

Here's one that I'm still using regularly with no apparent loss in performance, though I am conservative in my discharge levels and always balance them.

[Stuart C]

Here is everything you need to know about lipo fires. www.nfpa.org

Type "Lipo" in the search box, and spend a few days absorbing the knowledge.

[Bob Burton]

I did as you suggested and got one hit from the search. Opening the link I was taken to a 112 page document entitled "Lithium-Ion Batteries Hazard and Use Assessment "

It is not clear to me how relevant it is to Lipos. What is the relationship between Lion and Lipo batteries and are the dangers the same ?

[Stuart C]

Lipos are a type of Lion that utilise a gel type electrollite within a polymer "pouch". This makes them much lighter, so we use them. You will find in the text (if you have the patience to read it all), references to lipos that we use and the manufactures that we recognise.

What I found to be of interest was the causes of thermal runaway, and the predictable time of such events. The number of factory fires and storage fires is also well documented, along with the Dreamliner saga.

Sadly, we do not seem to have an answer to effective extinguishing, though the latter part of the study reveals a type of Halon (1031, I think) to be the best option, followed by a water cloud!!

It is a heavy report, and I need to read it all again, and again. So far I have learned not to transport my lipos in the boot of my black car on a sunny day.

[Dane Crosby]

I have seen lead acid car batteries brew up in to a foaming steaming mass. As a kid I blew up carbon zinc batteries by dead shorting them (idiot). A rare but not unknown event on our Sea King is for the (wet type) NiCad battery to enter thermal runaway. Very messy. Land ASAP. Put on fireman's gloves, unbolt from the nose bay and throw away on to the ground! Then run away etc.

I have managed to flame a Mercury Magna (I Know, what on earth is that?) by messily refuelling the diesel tank behind a very hot engine (DC Merlin). I was in a hurry to launch again before a rain cloud when 14 years old. Around the same time I was hit hard in the face when launching a large rubber powered model (type unknown) when the rubber broke free from the prop driver and whistled back destroying the fus and denting my cheek in the process.

It all boils down to treating all power sources well within their design limits and instructions for use. learn and follow the basics and we will be OK

Anything can go wrong so don't be put off using whatever power source you like but take care and expect the unexpected!

Edited By Dane Crosby on 22/01/2014 12:56:06

Edited By Dane Crosby on 22/01/2014 15:06:11

[Erfolg]

Stuart

I have gone to the link, I cannot find anything relevant on the page. Can you give a pointer to where to look.

[Pete B]

Posted by Peter Beeney on 21/01/2014 19:54:48:

"I personally think the videos are just controlled vandalism"

"Also I have to say that when I was trying to explode nicads I was unable to do it with a standard charger, I had to use a 12 volt car battery and jump leads to be able to get the resistance low enough for the necessary current flow. And this was bashing just a single cell"

"In fact, I used to frequently charge nimh flight packs directly off a 12 volt car lead weight, but that’s another story…"

Priceless, Peter!......

Now for the serious bit, relating to charging our Li-Po's....

A while back I followed a similar thread on another forum, where it became clear that some folk, when doing a straight Li-Po charge, do not connect the balance lead, for a variety of reasons. It seems that some feel that it is only necessary when carrying out a balance charge or that the balance lead isn't long enough when using a Li-Po bag.

I was quite aghast when I read this for, IMHO, nothing could be further from the truth. Whether my chargers are set to the charge, balance or storage program, monitoring of individual cells is clearly taking place at all stages of the charge. I'd hazard a guess that would be the same for most chargers on the market, too.

If the balance lead is not connected, then surely that gives ample opportunity for one cell of the pack to be over-charged, as there is no individual voltage detectable on any cell, only on the sum of the cells?

I'll accept that balance leads can be a little short which makes using a Li-Po bag more difficult but extension leads are readily available and should be a part of your charging kit if you use a bag - there's no excuse.

If you really want a belt-and-braces approach, many chargers will accept a temperature monitor which can be attached to the pack and which will shut off the charge if an excessive rise in temperature during the charge is detected.

Personally, I have never yet - and never intend to - charge a muti-cell Li-Po pack without the balance lead attached. I think it is just asking for trouble to do so.....and, so far, I haven't had a charging incident in 5 years or so of use....

Pete

[Bob Burton]

What do you perceive as the danger of doing a non balance charge without the balance lead connected ?

[Chuck Plains]

I would say never do that.

Not monitoring the cells at all is asking for a failure.

You could at best get it charged as long as the charger monitors the pack ok. But it's likely to be at least slightly unbalanced, and at worst ruined and on fire.

I check my cells individually via the charger screen at the start of any charge. If I'm in a hurry and if they are all within 0.03 of a volt I will even fast charge them and monitor the individual readouts often. They often look well balanced right up to full charge. But if any cell begins to wander, I will stop and finish the pack with a balance charge.

[Pete B]

First of all, I'm far from being a leccy expert, so for all I know I may be talking out of the back of my hat here, but I tend to look at things by applying the KISS principle and a bit of commonsense.....

Let's take a 3S pack as an example. The aim is to charge each cell to 4.2V, with a total of 12.6V for a full charge.You have a discharged pack which, unknown to you, is out of balance. It may even be that the internal resistance of the cells varies widely, a cell may be physically damaged or a solder joint failed.

I've had packs which on checking have shown a zero voltage on one cell. If I put them on the charger with the balance lead, the problem is apparent. As far as I recall, and it was a while back, the charger indicated a connection break.

You connect the pack to the charger, without the balance lead, and commence charging. Two cells reach 3.9V - what is to stop the charger continuing to charge the third cell until it measures 4.8V? It is only looking for a total charge of 12.6V and cannot differentiate between cells, after all.

At least with the balance lead fitted, the charger is getting feedback from each cell and will stop the charge when a cell reaches 4.2V.

The risk, obviously, is that the over-charged cell may reach a temperature where it starts to puff and go into thermal runaway, leading to possible bursting and a subsequent fire.

Pete

[Bob Burton]

The original problem was presented as the user doing a straight charge, not a balance one.

If that is the case, then as far as I can determine, the charger will not use the information obtained from the balance connection and will not stop the charge to an individual cell when it reaches 4.2V, only when the total voltage reaches that expected for the whole pack. Stopping the charge to an individual cell only happens when an explicit balance charge is being done otherwise why have separate Charge and Balance Charge options as on the common '4 button' chargers ?

The Fast Charge option introduces a further complication, and dangers, because the charge current is maintained at a higher level throughout charging process and it is not reduced as much during the constant current phase at the end of the charge,

[PatMc]

How many people check that the individual voltages from a balance display themselves are balanced & accurate ?

[John Privett]

Posted by Dane Crosby on 22/01/2014 12:54:43:

I have managed to flame a Mercury Magna (I Know, what on earth is that?)...

In case anybody else was wondering...

(That's from my 1970s KK Handbook, so don't take too much notice of the price!)

[Pete B]

Whilst I accept that there is no 'balancing' taking place in Charge mode, Bob, my experience is that the voltage of any individual cell does not exceed 4.2V, regardless of the voltage of the other cells, certainly by observation. I have assumed therefore that the charger is recognising that voltage and preventing it being exceeded.

The cell monitoring screen on the charger shows the voltage of each cell in both charge and balance modes. In each mode they appear to behave identically, with activity showing on each cell as required during charging. If the charger is detecting the cell voltage should it not hold each cell to 4.2V?

You seem to be suggesting that I'm living in blissful ignorance of the inability of the balance lead to control cell voltage in Charge mode. If that's the case, then It seems a fundamentally dangerous process to charge in anything other than Balance mode, surely? That also begs the question as to why we are offered a Charge mode....

I'm always willing to learn...

Pete

[PatMc]

Pete, for a while I used a charger that took ages to charge at 1c in balance mode - sometimes cutting off when the time limit had been reached - even when the battery was only partialy discharged. To make matters worse the individual cell voltage readings were inaccurate so that the batteries were unbalanced after a supposed balance charge. I didn't realise this until I goosed about 3 batteries, due I believe to some cells being overcharged.
Charging at 1c, without balance was always completed within the expected time scale & would probably have been less damaging.

After having binned that charger I now have one that trust (& check periodicaly) which doesn't take very much longer over a balance charge compared with non-balanced. I do normaly balance charge but if I'm in a hurry I'm happy to occasionaly charge without balance. (always at 1c max)

[Bob Burton]

I find it disappointing that more information is not available about what my chargers actually do in each mode. A quote from the manual for one of them

Charge – a basic charge process without balancing.
Balance – a much better process involving both the charging of the pack and the active
balancing of the individual cells.
Fast – a process designed to allow charge rates at above 1C with the current reducing towards
the end of the charge. This process reduces the charging time but also slightly reduces the
final capacity of the pack. It may also reduce the cyclic life of the pack.

That is the sum total of the information given. This is a '4 button' charger with what appears to be the generic firmware included in many/all of them.

[Martin Harris - Moderator]

Posted by John Privett on 19/01/2014 18:40:41:

I am not a chemist, though I do know that the Lithium in a LiPo battery is not in metallic form. So comparisons to the behaviour of metallic lithium are not necessarily valid.

Otherwise I think we'd be worried about pumping large quantities of a hydrogen compound through pipes into our homes. Or having a little container of sodium on our dinner tables and sprinkling it on our food...

I'm certainly no chemist either (and I apologise if this has been stated in any of the other threads in the thread) but I just came across this (to me) interesting information in my charger instructions:

"If cells are short-circuited or severely overcharged elemental lithium may be deposited internally" - and goes on to highlight that if the casing is ruptured this Lithium can react violently with the oxygen in the air.

As this elemental lithium isn't produced without mistreatment they state that there is no likelihood of explosion or fire in normal use - probably the reason why model shops don't have a fire engine parked outside on stand-by!

[Peter Beeney]

The Dreamliner saga is probably only relevant to modellers in the sense that despite the immense amount of lithium knowledge that must be accruing almost on a daily basis it seems they still occasionally have the ability to constantly confound the head honchos. I very much suspect that it’s not a temperature problem either, surely these have to be situated in a conditioned environment, all batteries lose capacity in the cold, seriously so in the very cold. Although this is only an ancillary pack within the airliner, I understand, I still don’t think they would want it to fail in the air; and I believe it’s very cold at high altitude. However, I’m sure that in the main lithium batteries are not to be considered any more dangerous than any other, I certainly don’t anyway. The vast number in use all the time says there will be isolated incidents, but I’m sure that instant lithium battery fires are not to be seen that frequently. As an aside, just exactly how many folks really do charge their phone, laptop or music player et al in a fireproof bag with a fire extinguisher handy?

As I understand it, the ‘polymer’ bit in lithium-ion polymer cells actually refers the the electrolyte material, this being a solid although flexible polymer. In some cases this is a gel-ionic material, but which is also a type of polymer, I believe. The packaging is a laminated foil.

Isn’t charging the cells is all a matter of personal preference? I have a friend that I’ve mentioned before, he’s been using lipos from day one and he’s never bothered with balancing batteries. For one thing, his first charger, one of the early one, didn’t have a balancing facility anyway. He likes small fast models, so the packs do get a bit of stick, but he has no more problems than anyone else. Certainly nothing that ever gets close to going on fire. I’ve only started using lithium power packs in a regular way over the last eighteen months, and I only connect the balance lead on the odd occasion. Just purely out of interest to get a handle on this I took a 3S pack out of the car earlier, I do most of the charging in the back of the car, because that’s where the power supply is; the charger had completed the charge with the (charger) terminal voltage reading 12.60 volts. Checking it with my voltmeter the pack voltage was 12.540 volts and the individual cells read 4.184, 4.192 and 4.186 volts. That’s 0.008V or 8 millivolts difference across the three cells. This has been used twice today in a Riot, and I don’t know when it was last balanced; and on that showing I’m not overly concerned about putting it back on the balancer overnight anyway.

Adding yet another ingredient to the stewing pot, I’ve never under stood the reasoning behind the assumption that a pack can become overcharged when it’s warmed up; to my mind the the opposite is true, it will become undercharged. This pack was at the outside ambient temperature, fairly cold. After doing the readings I put it on a radiator for a while and it warmed through quite nicely, at a pure guess a rise of something like 15 degrees C. So now the voltages should have also risen because it’s now overcharged. However, the pack was actually reading 5.505 volts and the individuals were 4.159, 4.166 and 4.161 volts. Thus we can see that it has in fact become slightly undercharged and now it’s 7 millivolts difference across the three cells. In truth, it may well be that temperature will not start to have really a noticeable effect until the cells get down to about 0 °C and beyond.

Pete, I can enlarge a bit on what I said earlier if you wish. With regard to the 3S fire video, anyone can apply a vast amount of power to a pack and get that result, we’ve known this for a long while. Just a waste of (probably) of a perfectly good power pack. I’d have been much more impressed if he’d done that trick by just connecting to a standard charger on the correct settings. Watching the burning lithium metal film, I have a suitable gas torch, a junior hacksaw and a few old nimhs lying around, I might have a go at at a similar sort of barbecue one day; just to see what happens; toasted nickel, though, as an alternative dish. The other items relating to nimhs could in fact be tried by anyone else, it’s all quite straight forward; and all perfectly true!

I think it’s possible that we can read into the lithium battery saga as much, or as little, as we wish.

PB 22/1/14

Edited By Peter Beeney on 23/01/2014 11:19:17

[Peter Beeney]

Addendum - … the pack was actually reading 5.505 volts… should in fact read !2.505 volts…

Well, it was late at night, and I’d not had my cocoa…

PB

[Bob Burton]

" !2.505 volts"

You need even more cocoa mate !

[Pete B]

Not being an electrical engineer like most of us, I imagine, I can only go by what I read in respect of the capabilities and processes of chargers. I have two of the generic 4-button chargers and whilst the manuals are fairly comprehensive there do appear to be gaps in the information provided.

The iCharger 208B manual is rather more comprehensive than most. Quoting two sections from the manual:

Page 13: The iCharger provides a number of Lithium programs including Balance, Normal, Fast and Storage. Only the Balance mode requires the balance lead connected. However, the other modes will provide additional
per-cell over-voltage protection if the balance lead is connected compared to running them without it, where
they can only utilize the total pack voltage.

Further on, same page:

Any time the battery‘s balance lead is connected the charger will monitor and display the cell voltages.

The balancer is only active during a BALANCE charge (not during the CHARGE or FAST programs)

and

While the CHARGE and FAST modes do not include any balancing action it is still safer to have the
balance lead connected since then the charger will provide per-cell over-voltage protection as described
below.

In all kinds of charge cycle, if the balance lead is connected and if any cell exceeds the allowable per-cell
peak voltage for the configured chemistry, the charge current will be reduced to ensure the voltage does
not rise any further.

That has been my understanding and expectation all along. The question is, are we right to assume that the generic 4-button chargers have fundamentally the same programming or is this only available at the top end? I'd like to think that the average manual just fails to mention the information. Have we any way of knowing?

Meanwhile, I'll continue to charge with the balance lead connected every time.

Pat, I must admit I only balance charge about 1 time in 10 or thereabouts, or if I see any imbalance, as a result of the chargers topping off all cells within 0.02V during the normal 'Charge' process. I also charge at 1C as the norm, even my 5C packs - time isn't usually an issue for me. I do find that balance charging does prolong the process time significantly.

Pete