Ahh okay, I think I’ve just been fooled until now. It’s really nice that we have you to give us the more useful numbers than what Molicel and the distributors have given us
7 Likes
Certainly for Molicel, the 45A number isn’t there to mislead us in any way. 
While Molicel supports the sale of individual cells to end users (the only big manufacturer who does) their datasheets are still set up for use by their big industrial/commercial customers. It often takes a lot digging into the details, and understanding a company’s customers, to “decode” a datasheet properly.
13 Likes
@Battery_Mooch
I have this question as well…but fuck this isn’t the place for it. WE NEED MODS FOR SPLITS FUCKING CHRIST
5 Likes
@Battery_Mooch as long as you aren’t running them to 75-85c do you think we could run them at 45a? I know I don’t hold full throttle more than a couple seconds. I have a smart bms with dual temp probes, I’ve never seen my pack get over 45c.
1 Like
You can run them at any level you want. But…
As mentioned, Molicel rates them at 45A so I assume they’ve done plenty of testing at that level. The datasheet says running at that level reduces cycle life and I don’t think that needs the cell to get hot. There are non-heat related aging mechanisms that occur when the current level is high and 45A is a lot.
I don’t think short bursts at 45A will age them fast but it’s, of course, more abusive than bursts to a much lower level. I haven’t done cycle life testing at high current levels so I’m not able to give you any kind of numbers. All I can say is that 30A is more abusive than 25A, 35A is more abusive than 30A, etc.
For any setup, keeping the temperature as close to ambient is important for long cycle life. If you are using thermistors for temp measurement your cells are probably getting hotter than 45°C since thermistors are so slow to respond and connecting well (thermally) is difficult due to their size.
That is warm enough to speed up aging but there’s always a tradeoff. Sometimes we can’t have the performance/size/weight/cost we want and still have long cycle life. We just have to choose which are the most important and let the other things fall where they may.
20 Likes
I wonder if you/we could ever come up some definitive tests for cell longevity…
I feel esk8ers probably abuse our batteries a bit more than the vape community, but I don’t know…
I really like your high load tests so maybe it’s just me… i remember a test you ran that I read the other day, 3 18650s high drain tests, heat and resulting capacity… it was awesome!!
Hey Mooch… or Battery… what should we call you?
what do you think about this philosophy for assembling a battery pack…
I usually do a voltage check to make sure they’re the same… let them age a couple of days and compare the voltages and throw out the variances, and I check the IR of the cells, one to make sure they’re within spec and then to try and get the IR of the P groups close to similar…
am I picking flyshit out of the pepper?
5 Likes
Mooch is fine.
Checking for excessive self-discharge often take weeks. You can check after a couple of days but it takes up to a day for the cells to settle to their true testing voltages and one more day is no time at all for differences in the self-discharge rates to show up. I’d let them sit for a month if you want to match cells up but at least a week. A week just isn’t long enough IMO but it’s better than a couple of days. The longer you wait the better the matching can be.
Checking the DC IR for p-group matching is good, necessary actually for the best pack performance. But all the cells in the pack need to be matched for best performance. If you have a low IR p-group in series with a higher IR p-group then the pack’s performance will be limited by the higher IR p-group (assuming capacities are identical, which they aren’t) because this group will sag in voltage before the low IR p-group will.
Matching capacity, self-discharge rate, and DC IR needs to be done. But that’s a colossal PITA and after a certain point the ROI for all that work gets to be too small to be worth it.
Match where you can and use a really good BMS to protect the cells. Sure, another month’s work might give you a bit more Wh of usable energy but would it be worth it?
Critical to matching cells is a test unit that is accurate when measuring both capacity and DC IR. Most analyzing round cell chargers are only moderately accurate (some are terrible) and very few of these chargers can measure DC IR accurately unless you press (with consistent pressure) against the bottom contact when measuring.
12 Likes
Oops…true resting voltages…
5 Likes
Most of us have been doing that and I personally havent found any defective cells, so I’m glad you verified it as a true test
1 Like
True test? Nah, I’m just a battery geek who reads a lot and has his own preferences. 
I like to think those preferences are based on sound science though.
11 Likes
If you told me right now batteries were water resistant, I’d probably believe you 
6 Likes
13 Likes
I’ve having a tough time making a decision about going with the p42a vs the 40t. I’m looking at a 12s9p build and my calculations show on the 40t if I set the cell continuous discharge to 25amps my system demand shows a battery pack peak discharge of 225 amps and the potential for my motors to pull 178 amps, so about 50 amps of headroom.
On the p42a cell in the same 12s9p configuration with the settings set at 35 amps cell continuous discharge my pack can then do 315 amps continuous and again my motors can potentially pull 178 amps. Thats a considerable amount more headroom… but Mooch, you also just said perhaps a better safe continuous output for the p42a is around 25 amps. Would that also mean the 40t’s safe continuous output should be lower than 25 amps? Any suggestions on which way I should go. I’m satisfied with the predicted performance of both cells, the p42a offers a bit more range and headroom but I would be happy with either cell. What I’m trying to understand is would one cell or the other have a particular advantage when talking about longevity and cycle life? Current prices show I gotta spend about $755 Canadian for the required p42a cells vs $587 for the required 40t cells. Are the p42a worth the extra 170 bucks under my circumstances? For context its for a hybrid street light off road build, I’ve got a few 40% grade hills I climb but mostly its pretty flat and range is a priority for me.
1 Like
Umm…actually…round cells are completely waterproof (hermetically sealed) and can easily be used underwater. Not at all recommended but possible.
At the low voltage for each cell you get only microamps flowing through clean water. Even a salt-saturated solution only slowly discharged a li-ion cell in my testing. The cell self-destructed from electrolysis though when in salt so definitely not recommended.
15 Likes
8 Likes
The P42A beat out the 40T a little bit for most situations. When just looking at performance the P42A is the slightly better choice IMO. Staying under 25A or so improves performance, increases cycle life, and reduces risk versus running at higher levels.
They perform about the same though and a lot of people wouldn’t notice The difference.
I haven’t tested the cycle life for them so I can’t comment on the better choice for that.
8 Likes
Cool, thanks Mooch. Sounds like i have to weight the dollars vs slight performance gain. Appreciate your feedback
2 Likes
That’s easy…run the hell out of at least four of each cell and compare performance every hundred cycles.
They just need to be wrapped like a pack would (to trap the heat) and charged and discharged hard enough to bring the temps up.
They don’t need to be pulsed or to simulate the different current draws when used in a board. Any discharge over a few milliseconds long is just more of the same for a cell. Continuous discharge testing will come within a couple percent of what would happen if pulse discharged. The best cells identified using continuous discharges would be the best cells identified using a simulation of “typical” board use.
Only if the cells were performing to within a couple percent of each other could the results change for continuous vs pulse/variable discharge testing. But if they are performing that close then it’s very hard for anyone to notice the performance difference anyway.
9 Likes
Hold up, is this Mr Battery Mooch from the YouTubes?
Welcome, thanks for visiting and commenting in detail 
14 Likes
damn that soda looks good.
edit:
I just stole some of my roommates fanta that shit looked so good.
2 Likes