Hello, I am relatively new to electronics and I am building a RC rover. The brain of the rover in an Arduino Mega board with two stacked Adafruit v2.3 Motorshields. To power the rover I purchased 18650 3.7V LiIon batteries which I am using in a three pack so the overall voltage remains under 12V which is the max voltage for the motors and motor shield. In all of my testing the batteries preformed as expected. It was only after removing the battery pack from the drone and setting it on the table did the holder begin to glow and smoke - within 5 seconds! I yanked all of the batteries and there was no visible damage to the batteries, but the spring in the battery holder connected to the negative wire was blackened and brittle and snapped off when I pulled on it. The spring was pushed together tightly and it was glowing. The battery pack was wired in series and the batteries were correctly installed (it had been working a few seconds earlier for many minutes without heating up or being a problem). Has this happened to anyone else? Could the battery have pushed back on the spring and created a short within the wire coils? Is that even possible (I have the image of a car cigarette lighter in mind which looked similar to what the battery coil looked like)?

 

What kind of table did you set it on? Metal? Was there anything on the the table that could have shorted the pack terminals?

 

It was a wooden table. I have the same thing happen in my classroom three other times, but with students using them. It is always the negative terminal and the spring always looks like it is pressed flat which also means the positive terminal is not touching. I didn't mention the other times because I couldn't be certain of the conditions because students were involved. They might have flipped the batteries or done something else. I know what I did, so that is what I am going off of. Also, the positive and negative wires were about four inches apart and didn't touch each other. Could the battery have shifted in the holder, smashing the negative spring together and caused it heat up? Is that even possible? The wire was red hot in a matter of seconds and so brittle once it cooled down that is broke off when I touched it. I am at a loss for the cause, but on Monday I have to go back to class and tell the students we can't use the battery packs, so their drone projects are over. Unless I can find a reason and a way to fix it. Is there a better way to work with these three 18650 batteries? Like combining them into one like they do for RC cars?

 

It certainly sounds like a short circuit happened leading to a lot of current flowing. The weakest part being the thin wire on the negative terminal spring then overheated. If the wires didn't touch and it was a wooden table then it would seem that the short was within the battery holder/pack.
Can you post some photos of the holder?

 

It was a wooden table. I have the same thing happen in my classroom three other times, but with students using them. It is always the negative terminal and the spring always looks like it is pressed flat which also means the positive terminal is not touching. I didn't mention the other times because I couldn't be certain of the conditions because students were involved. They might have flipped the batteries or done something else. I know what I did, so that is what I am going off of. Also, the positive and negative wires were about four inches apart and didn't touch each other. Could the battery have shifted in the holder, smashing the negative spring together and caused it heat up? Is that even possible? The wire was red hot in a matter of seconds and so brittle once it cooled down that is broke off when I touched it. I am at a loss for the cause, but on Monday I have to go back to class and tell the students we can't use the battery packs, so their drone projects are over. Unless I can find a reason and a way to fix it. Is there a better way to work with these three 18650 batteries? Like combining them into one like they do for RC cars?

Some pictures of the battery holder, both with and without the batteries, would be helpful.

What I find interesting is that this only seems to happen when the pack is out of the drone. If there is something that can cause the batteries to short with it just sitting on a wooden table, I would expect it to happen far more frequently when in the drone being bounced around.

 

This image shows the negative wire that snapped off. This image shows the battery packs and the positive and negative wire positions (why they couldn't have crossed when I set it on the wooden table in the background). I followed an instructable to convert the packs from four cells to three cells. The red tape represents where not to put a battery for my students. While my incident occurred with this setup, the other three my students experienced were before the conversion.


This image shows one side of the battery pack where the modification took place.


This image shows where the negative spring was attached.

I know that modifying the holder might seem to be an easy thing to place blame on, but the holder is a very simple thing. It is wired to provide serial power with the last cell excluded. Also, this same thing happened on three previous occasions with my students in unconverted holders with four batteries. I do not have the knowledge to diagnose this problem. Does anyone see a reason why the negative spring would heat up?

 

I'm at a loss to explain it. I don't see any smoking gun (no pun intended) and the modification you made doesn't explain it.

The only thing I can think of is that it has nothing to do with the holder, but rather how the batteries were used. LiPo's are pretty notorious for going into a slow-burn thermal runaway if they are abused and it can take awhile before the reaction gets far enough to get out of hand. We had a team of cadets out at China Lake about ten years ago that were charging some LiPo's in their motel room and they took them off the charger before they went to a team meeting in another room. About a half hour later the batteries erupted in flame and set the room on fire.

 

The batteries I am using are straight out of the package and have never been charged. I used them for a few minutes to upload the code to make my motors run - 6 wheels - forwards, backwards and forwards again, three or four times. The meltdown happened all four times to the negative spring after similar kinds of use, yet the majority of my other students used the batteries and similar packs in the same way with no meltdown. I haven't been able to find anything like this on the internet. LiIon batteries can explode if heated - correct? Also, what would be the best way to store them? I have over 100 of them and I am keeping them in a plastic tupperware container. They are stacked in rows, all facing the same direction with cardboard walls between the rows to prevent contact of the terminals - does that sound safe?

 

The military guidelines (from ten years ago) is that they are supposed to be charged in a metal, fireproof container and left in the container for some length of time after charging was complete -- a couple hours, IIRC. Specific requirements weren't made, so a common approach was to put a couple inches of sand in the bottom of an ammo can and put the batteries in that. Some folks used metal waste cans similarly. I don't recall the requirements for actual storage since I never had to deal with that -- I borrowed several batteries when I went out to China Lake and it was convenient for me just to leave them in the ammo cans when not in use.

That was what had a number of people very concerned -- those procedures were in place when the motel room fire happened and the cadets had set the batteries on the desk after they disconnected them from the chargers. The motel manage took everything in stride (it's a military town and so, on the one hand, you don't want to get blackballed for military business and, on the other, you know you are going to get paid for the damage); he just moved the cadets into another room -- but asked that the batteries be left outside from then on. The OIC (officer in charge) of the cadets got a fairly pointed call from the Dean that evening and the next morning, but beyond that everything was pretty low-key.

 

Thanks for the info. Coincidentally, I noticed your Eagle Keeper badge and my dad was in the Air Force stationed at Luke AFB when the Eagles were first deployed. He worked as a graphic designer and worked on the media side of the event. As an airplane nut my dad got me a seat in the audience to watch the roll out and a media folder that gave data sheets on the F-15 and the Soviet planes in the Russian arsenal. Did you have a connection to the F-15?

 

Thanks for the info. Coincidentally, I noticed your Eagle Keeper badge and my dad was in the Air Force stationed at Luke AFB when the Eagles were first deployed. He worked as a graphic designer and worked on the media side of the event. As an airplane nut my dad got me a seat in the audience to watch the roll out and a media folder that gave data sheets on the F-15 and the Soviet planes in the Russian arsenal. Did you have a connection to the F-15?

Yep. I was an Eagle Keeper. I worked flight line and shop level maintenance mostly of F-15's. It was a joy to work on. My primary job was hydraulics/pneumatics (what they called "pneudraulics"), but I got my hands dirty on nearly every system on the bird except egress. I also got to work on a lot of T-33's (we had the last squadron of them), our QF-100 drones, and pretty much anything that came in transient including quite a few A-10's, F-4's, T-38's, F-111's, F-16's, F-106's, F-14's, F/A-18's, AV-8B's CF-18's, the last CF-101 (shortly before the Canadians retired her) and one rather miserable afternoon working on an SR-71 that made an unscheduled stop courtesy of a blown fuel pump over the mid-Atlantic. We got a large variety of fighter aircraft because of our instrumented offshore range as teams competed for invitations to William Tell (a biennial international live-fire air-to-air weapons meet held at Tyndall).

 

I went into school this morning and put each of the batteries I was using on the volt meter and found that some of the batteries I had with me (11) were charged to 3.7V which is what the packaging says it should be. Four were measuring 3.81V to 3.83V which is slightly high (including the one that was connected to the negative spring, can still see some burn marks on the plastic coating). There was also a completely drained battery that read negative milivolts. Could a dead cell in a three cell serial wired set up cause this kind of thing? I know that in parallel setups that would cause major problems, but in a serial setup?

Also, what bases were stationed to? Many of the bases my dad was posted to had those same planes, but that was in the 70s and early 80s.

 

A shorted battery certainly could have contributed to the melting. The other three tried to charge you to 4.6 volts.

 

Battery source? Hopefully not cheap junk/fake 18650 cells from eBay?