I am SO used to keeping things cool I didn’t think of that.
This is a very interesting experiment! It seems to suggest that the pressure buildup from continuously running a sealed light might not be chemical outgassing, but thermally induced expansion. (Recall PV=nRT…) I certainly observe this in a Q8+ with rubber charging port cover. But I don’t think the pressure is enough to blow a flashlight head across the deck…wonder what happened.
4 D-cells are massively larger?
Did you observe anything unusual about the cells after the explosion? (Or did you throw everything away immediately, which is arguably the more sensible thing to do.)
The greatest potential for heat would be while draining the first half of the battery charge.
Plenty of really water tight lights with lithium ion batteries can get really hot and we don’t see flashlights exploding or having over pressurization ruptures due to heat.
After talking with AI, it appears that hydrogen gas is created when over discharging these batteries at temperatures below freezing.
When overcharging they can produce hydrogen and oxygen.
Maybe the threads are better on modern flashlights than they were years ago?
And it appears that pelican recalled at least two different models of flashlights in 2019 because of explosions (actually they were over pressurization ruptures) while charging NiMH batteries and the lights were not waterproof/dunk proof.
1 Thank
I recall this being discussed a long time ago, potentially on the other forum, if I remember correctly it is the build-up of gas (hydrogen) on charging, I assume this is mostly a problem for lights with multiple cells and those with in-light charging.
Also, I recall that simple thermal expansion of the air inside a light doesn’t have sufficient capability of ballooning tailcap switches like some users were reporting.
An easy experiment would be to freeze a flashlight without a cell and with the cap off, assemble while still cold and allow to heat up in an oven set to low (50C), this would be absolutely worse case scenario for air expansion.
I might actually try this on a few lights to test water resistance, it’s like a very cheap version of those watch pressure test rigs, freeze the light, assemble while cold, dunk in warmish water and look for bubbles as the air inside warms up and expands.
1 Thank
Throw away? No! They looked fine and still worked fine since there was no flooding involved.
I knew almost nothing about NiMh/NiCad back then except they generally were lousy, frustrating, and usually dead when you wanted them. These were the first good NiMh I ever got (Radio Shack, 4500mAh D-dated 9/02!!!). Take THAT Eneloop. They don’t make them that good anymore.
I continue use them fairly routinely. Last capacity check at 2.0A they all were still 3000mAh+.
2 Thanks
This is also what my intuition suggests, I’m just confused by the experiment result of @flydiver: with an unsealed light within a sealed bag, such an amount of offgassing should be visible.
That’s what I thought. That’s why I tried the experiment.
What caused the screw-on head of my wife’s scuba light to blow off, I think now I might more reasonably ascribe to thermal heating and expansion rather than off gassing. There wasn’t a lot of excess space in that light. High performing scuba lights can get pretty hot and some require water immersion to stay cool.
I do not profess be an expert in NiMh charging chemistry, but my hypothesis for these occurrences has been that the hydrogen gas is generated far in abundance to normal operational conditions, such as a reverse charging event or something similar. If it is an unusual event, it would explain why the tailcaps on all my 2xAA NiMH lights don’t balloon up like some reports show.
While thinking about the bag test, is polythene permeable to hydrogen?
Hmm. had to research that:
[Ziploc bags are primarily made from low-density polyethylene (LDPE) and linear low-density polyethylene (LLDPE)]
So…threw that at an Ai for this info:
[LDPE allows the passage of hydrogen at rates that may be acceptable for some low-pressure applications but generally not for scenarios requiring high hydrogen containment, such as pressurized storage or transfer, especially if minimal hydrogen loss is required.]
I would guess, since this was a relatively short duration experiment, in a shallow bowl of water, that permeability would not have been a big factor.
In hindsight there were a lot of uncontrolled variables. So much for being clever.
2 Thanks
At the same time, this means there isn’t a lot of air to expand and blow off the head to start with, so I still think that thermal expansion does not fully explain the incident. The ideal gas law states that under a fixed volume and mass, gas pressure is proportional to temperature–to even double the pressure would require heating the gas to 300+ degC, which likely did not happen as the walls of the light would have cooled it to a sensible temperature.
This explanation (freak accident), although unsatisfying, seems like the only one thus far that does not contradict existing evidence.
It is reasonable to suspect that hydrogen might percolate due to low molecular size, but I’d assume this amount of hydrogen would have caused bubbling all over the place.
TLDR: Yes, the concern is gas venting, but if you follow the best practices for using NiMH batteries, it is highly unlikely to be an issue. I’ve never had problems.
Longer response:
Although hydrogen and oxygen are released alternately from one electrode or the other during charging or discharging, if charge or discharge is not happening beyond the recommended currents, it is absorbed by the other electrode, and it is a cycle that should be continuously consuming the same amount of hydrogen. If it was instead normally off-gassing, it would be impossible to get anywhere even remotely close to 2000 cycles with these.
I have not noticed any signs of pressurization of my AA/AAA-powered lights, and will continue using Eneloops in them. I think the most important thing is to take care of the batteries: avoid significantly overcharging them, or deeply over-discharging them, or using mismatched batteries together, especially in series.
Something confirmed by Sanyo researchers in discussions by @ChibiM on his Eneloop101 site is that NiMH batteries are relatively tolerant of minor to moderate overcharging. In fact, the standard IEC test for determining cycle life of NiMH batteries requires overcharging the batteries every cycle by about 20% at 0.25C, and a capacity test every 50 cycles that starts by overcharging it by 60% at 0.1C. Again, Eneloops achieve 2000 cycles under those conditions, so clearly they’re not losing hydrogen by off-gassing.
Over-discharging is more likely to be an issue, but my understanding, which unfortunately I have not found explicitly confirmed by the manufacturers, is that single cell lights can’t really over-discharge enough to be a major concern. Even battery vampires like the old Fenix E01 and clones seem to struggle to drain a battery below 0.6V. This is lower than typically recommended and will reduce cycle life, but I’m pretty confident will not cause meaningful off-gassing. I’ve run an Eneloop this low in a Sofirn C01 a couple times, and the light didn’t seem to have any pressure built up when I opened it immediately afterwards.
Where the risk is most likely is in lights that use the cells in series, particularly if you use unmatched cells. What happens is once one cell is completely empty, the other batteries can continue to push current through the empty battery. This causes the voltage of that battery to be reversed, and this starts to first harm the positive electrode, which occurs at a very small reversed voltage. Push it even further or harder into the negative, and the reverse voltage will jump to around -1V (per Energizer), the negative electrode will start to be damaged, and it will start releasing hydrogen. I think this is unlikely in a 2s configuration, but the more cells you have in series, the more likely a voltage reversal could be caused in any weak cells.
Because it is a known possibility, it makes some sense for the manufacturers to provide a warning against the risk. They don’t want some company with a safety critical application, like use in explosive environments, to be unaware of the possibility when they do their safety assessments and potentially have an accident. I would argue that in more normal use cases, we have enough collective experience to be comfortable using them in waterproof flashlights.
Sources (So you don’t have take my word for it):
Chibim’s notes from the discussion with Eneloop technicians that I mentioned:
Both Eneloop and Energizer have handbooks they publish with technical information about their NiMH batteries, and both have similar warnings about off-gassing. Energizer provides a little more detail, so I’ll quote them:
5 Thanks
Great information. Thank you.
Again, for my applications under the conditions I operate my devices, I am not concerned at all.
1 Thank
It seems like either thermal expansion, or a voltage reversal that caused enough off-gassing to over-pressurize the light are plausible explanations, but I agree with QReciprocity42 that thermal expansion doesn’t seem as likely.
Since the batteries continued to work afterwards, obviously it wasn’t a serious reversal, but 4 cells in series could do it. If the light needs 3V to work, and a weaker cell is pushed to just below 0V, the other cells remaining above 1V each would be enough to do it.
I suppose it’s even possible that just enough hydrogen off-gassed to ignite and pop the head off, but not enough to cause serious damage or even create enough overall heat to leave signs of burning.