I have a box of 1000 4P 21700 hex holders if someone needs some and doesn’t want to wait. Probably 900 or so left, they’re quite tight on P42A, but only just stick up over the cell edge so they work quite well with sheet nickel.
2 Likes
Kind of seems like my video upload to YT is stuck at 99% “processing”, lol. I tried uploading to here but I guess its too big. Any time now I guess it will be done. I got the results I expected, and I’m willing to make more videos for science if necessary. I’ll change things, I’ll hold it on there even longer, whatever makes people happy.
edit: soldering station was at 450c for this test.
2 Likes
The second reading is 78f, or 25c.
6 Likes
Although air is normally an excellent insulator, when stressed by a sufficiently high voltage (an electric field of about 3 x 106 V/m or 3 kV/mm), air can begin to break down, becoming partially conductive.
Nice test 
In case you decide to do the test again, I’d be really interested to see the temperature right on top of the cell where the nickel is mounted. Hope that makes sense 
1 Like
Yes, that’s a good idea. I’ll do that after I eat something. I expect it to get very hot very fast.
3 Likes
AKA “arcing”
1 Like
That was unexpected. It fared much better than I thought it would, never really exceeding about 113f. Upload incoming. I still don’t advise soldering on the terminals, heat degradation isn’t the only thing you have to worry about. You want the insulators on the positive side of the can to stay as virgin as possible, since you’ll be relying on them to protect you from vibration wear.
4 Likes
“How much does it cost?”
I always give a PC response because tbh I have let the hobby get out of hand…
“Not much factoring in selling the old board” (Which I no longer intend to sell)
“A reasonable amount considering the performance”
Quoting only the raw parts cost and sometimes even considering quoting it in scrap metal value, not counting the wrong ordered parts, tools such as a 3d printer, anodizing kit, safety gear, drill press, or the other boards in progress…
“Ahh its a penny or two but look at it!”
And many other fun quotes like this.
3 Likes
Thanks for doing those tests! A couple of thoughts…
- The field of view of the IR gun could be larger than the width of the cell. I recommend placing the gun directly against the cell.
- The temperature reading of the strip itself, or any bare metal, will be much too low due to the very low “emissivity” (ability to shed heat via IR) of bare metal. Here’s a video I did demonstrating this effect: https://youtu.be/a6AMbbf4Ol8). If you have flat black paint, actually any flat paint, you can paint over the strip on top and then get an accurate reading. Even a well-pressed-down strip of electrical tape works well.
- The top contact of the cell is thermally (and electrically) well separated from the metal can. This means that the temperature under the strip is considerably higher than the can temp. Since the top plastic crimp sealing ring is at the base of the top contact it can easily be overheated by too much heat on the top contact.
- I would recommend this for a test…unwrap the cell and remove the top insulating ring but keep a good double layer of fishpaper under the strip to prevent shorts. This will expose most of the top contact structure. Paint it flat black (if you have it) or press down some electrical tape into all the nooks and crannies at the top. Then rerun the tests using the gun directly against the top contact. Slide it around a bit to find the peak temp. I think you will find it will get considerably hotter than 113°F.
- A separate thought…soldering at the bottom will directly heat up the electrolyte which starts decomposing at around 75°C/167°F. This is a temp easily reached on the other side of the metal can where the soldering is being done. The byproducts of this decomposition then deposit onto parts of the cell, increasing its internal resistance. The plastic separator (between the positive and negative parts of the cell) starts melting at about 115°C-135°C, also easily reached at the bottom on the other side of the metal can where the soldering is being done. I am much more concerned about soldering at the bottom. I don’t know if this was part of the conversation or a possibility for testing…just mentioning it.
10 Likes
I like the idea!
I don’t think they will thermally insulate the cells anymore than using the fishpaper/kept on tape/strapping tape method, with all its air spaces. If the cells are a press-fit into the holders then there is a conduction path for heat to get to the outside of the holder, as good or perhaps even better than through fishpaper/tape.
Perhaps pull back the plastic away from the cell’s top contact and make it thinner? This will make it easier to weld the strip.
I was thinking that if you removed some of the plastic on the sides, via slots/cutouts/holes/whatever, then you could save a bit of $$ on filament and also cool the cells better but if the pack is being shrink-wrapped then those air pockets would trap heat better than having solid plastic up against the shrink wrap IMO.
3 Likes
I was able to get the cell temperature upto 130f taking your advice. I added electrical tape, and measured right up against the can, but as I think you mentioned, this doesn’t really capture internal temperatures. It’s kind of just a glimpse of what is going on inside. Granted, I really had to hold it on there a long time. I’m not measuring any difference in internal resistance with a 4 wire kelvin meter.
3 Likes
Thanks for testing that!
Yea, any damage to the electrolyte would probably only happen via soldering on the bottom and would need a cycle or two to be measurable, if at all. My much bigger concern is melting of seals and the separator. Everything would seem okay until it was very much not okay because of the cell pretending it was a model rocket engine and flying around the room. 
But since this certainly doesn’t happen often it’s not a huge concern of impending doom for everyone who solders their cells. It’s just a best practices thing to not do stuff that could possibly damage the cells…like soldering. I know you advocate not soldering cells though.
4 Likes
Yw, for whatever its worth. With regard to soldering on tabbed connections, its something I noticed when building my own packs- that spot welding introduced just as much, if not more heat into the cells than soldering on tabbed nickel. I noticed that today as well. Spot welding the nickel on took the can temp from roughly 75f, to 83f. So I had to wait for it to level out a bit.
I’ll check the ir of this cell over time. I’ll just store it properly in the meantime.
edit: Is it necessary to take wired measurements? I would hope not.
2 Likes
That’s possible. It would depend a lot of the duration, strip size/thickness, heat flux, etc. But welding would certainly add much, much less heat than soldering directly to the cells.
Sorry, I don’t understand.
2 Likes
I think he’s referring to jamming a thermocouple into places for a physical-contact measurement.
2 Likes
No need for the potentially better accuracy of thermocouples if the surface you’re reading the IR from has a decent emissivity (a battery wrap is okay).
2 Likes
Battery cables have like ten fold the mass of nickel stripping. If not more. Also, I’m not expecting soldering irons in excess of 100w. I think that’s unusual. My iron is 70w. Your 200w Weller could probably heat an 18650 much faster than my 70w iron. Also, 18650’s will heat up faster than larger cells, like 21700’s. When one quickly solders their joints (tabbed), almost no heat is transferred to the cells.
2 Likes
Unrelated, but did everyone get in on the recent P42a batches? Li-ion Wholesale still has them.
Odd times. 
