All those cells are exactly that. I also have 8x P45B starting to bulge along with P30B’s and some 30Q’s.
1 Thank
Despite these battery issues I am extremely excited for the W5050SQ5 to hopefully give us a fix for the SFT-25R’s egg yolk hotspot
4 Thanks
M21J excluded, cause this really gets far too hot, does any of the other convoys exceed the heat of a Anduril 2 light set to 70°?
Cause my Sofirns Q8+ on 70° gets so hot, that even the (convoy)handle is at the edge of being painful. Same with smaller hosts.
Of course they reduce the output after that and cool down, but the max heat doesn’t seem to be the issue.
My M21b LHP73B doesn’t feel any different.
I know about the driver temperature issues and don’t want to deny them, I just can’t believe that the temperature is the cause for the deformation of the battery part that is farthest away from the heat.
1 Thank
Mooch reckons it’s from a build up of gases that’s causing it.
He also mentioned:
“60°C on the outside of the light can mean the inside of the cell is at least 80°C, if not higher, depending on the current and thermal coupling to the interior of the light (assuming the light is metal).”
His mention of the inside of the cell being hotter is interesting, not just the outside.
2 Thanks
The P42A cells shown in your video are only rated to 60C.
But idk what’s actually causing your issue. The springs in the host could be getting extremely hot, for example.
1 Thank
I do see the datasheet has no mention of 80, only 60, but this is out of Mooch’s review.
“The 45A rating of these cells from the datasheet is not a true continuous rating IMO. It is temperature-limited to 80°C so when the cell gets that hot you must stop. Since the Molicel P42A reaches about 80°C at 30A (down to 2.5V) I will be giving it a an estimated 30A continuous and a 40A temperature-limited rating. Never let the cell go over 80°C.”
I get a bit confused with the Molicel datasheets when the P45B says 80°C cut off at 45A, but then below that it says discharging temp at only up to 60°C.
Gotta love those two things that eliminate each other in one sentence.
Since flashlights are meant to be in the hand, the outside will never reach 60°C because noone would be able to hold it.
3 Thanks
While that is true, some do wear gloves to exceed 50C and also models with handles like the 4X21A being another model that reaches unsafe temps. One person on reddit found their Convoy light on their child’s bed switched on, it was so hot it couldn’t be touched. Another good example that there should be a safety limit.
I think this issue needs more investigating because it may end up to be a Molicel issue. My old 30Q’s are ever so slightly swollen, but every Molicel is really bad. Is it possible that Molicel is more temp sensitive than their datasheet suggest?
I just had a comment saying their P42A is bulged but their 40T is perfect. Another on reddit mentioned their P45B’s were bulged just the same after runtime testing X Convoy light.
Mooch said he’s never heard of this happening before, without abuse.
- I check cells cut-off at 4.21v on the Vapcell S4+ V3
- Cells are always cool before charging, and never in temps exceeding the datasheet.
- None have ever been over discharged
- None have ever exceeded their CDR
All it leaves is high discharge temps. If not that I’m completely stumped.
I’ll create another thread for this discussion.
1 Thank
Or 17mm 20A buck would be cool. My C8+ isn’t hot enough to heat my hand on 35%, even on 100% it’s barely warm then drops power to 35% with 6v5a driver&70.3 hi r70 5000K in winter
But still overheating is a big problem and 35% of 6v5a may overheat it in warm environments
1 Thank
Noticed Simon has marked the 3x21c as sold out and placed burn warnings on all M21J lights ..
6 Thanks
I found an interesting comment under your video. Could it be a problem with the Molicel cells?
Probably unrelated. I asked him about this and he wrote that they’re out of hosts. It’s worth noticing that he’s also working on a new LED MCPCB for 3x21C.
- Besides, there was no problem if someone didn’t modify the flashlight. Correct me if I’m wrong.
M21J / L7 - I once had a borrowed L7 and I admit it got quite hot during testing. If I remember correctly, it was as high as 80-90 degrees. If there’s a limit, it could be a bit lower.
Curious what the progress is with the W5050SQ5. Many of us are extremely excited!
7 Thanks
I have constantly been thinking about how a rosy 4000k throw beam would look, I honestly think it lets you see clearer than 5000k or 6500k, at least in the LHP73B.
2 Thanks
Absolutely. A lot less backscatter, and I find warmer CCTs and rosy tint to offer much greater distinguishability between colors in the mesopic/scotopic (low intensity) regime. Especially so for greens and browns.
2 Thanks
The 3X21C was failing without any mods with high drain cells only.
Wires desoldering off the driver and springs overheating. This only affected the LHP531 version.
OT - sure yes the holy grail. A few Firefly LEDs have this if you’re interested. And not just 4000k rosy throw beam, superb 90 plus CRI to boot.
EO4 505A 4000k if you don’t want beam too rosy, and X1L 707A 4000k if you like super rosy.
3 Thanks
******fire, the BoBoFet edition?
The problem with Fireflylite for me is that I have specific shipping requirements that made ordering from them and receiving my product a nightmare, they used a random gig-worker delivery service that had my package sent back to California after arriving in my state, just so that it could be re-wrapped and sent back with USPS. Some other times I’ve gotten the gig-worker delivery companies they just didn’t deliver my package at all and destroyed it. To this date, Simon is the only one who has accommodated my needs and added a USPS shipping option to his website, and because of that about 90% of my flashlight purchases are Convoy.
3 Thanks
Is there any news on the new 10x10mm led and the 60 amp driver?
The driver is probably undergoing a tweak considering the battery issues with the 3x21c that were thankfully brought up, the emitter was in the sample phase last update from Simon, so it’s probably about as far along as the W5050SQ5, although I suspect it will be held back longer due to the driver taking longer to develop.