New builder - advice on best cell testing methods/options in 2022?


SLS BATTERY AUCTIONS are now live at www.batteryhookup.com EV, Lithium, LifePo4, Inverters & More
Joined
Jun 1, 2022
Messages
4
Hi all,

I'm a RV/caravan solar installer who's been lurking on and off here for years, wanting to do an 18650 build, but just never had the time, or the cells. As it happens, both finally presented themselves to me at once. I've packed in the business and in doing so, by chance found a regular source on bulk 18650 cells.

Time is of the essence though, and having followed a few of the long-running threads by folk like @Wolf , @OffGridInTheCity @hbpowerwall , I appreciate two things: 1) theres a lot of reading ahead before I start actually building packs and 2) I have a LOT of time also to be spent on testing cells.

I'm hoping to brush up on said reading while doing said testing, but even that seems to need a bit of planning, and I'm trying to settle on the most efficient use of money and time, to work through testing several hundred (even thousands?) of cells.

@Wolf 's setup with the automated/scripted megacell charger seems to be about at good as it gets, but clearly involved a lot of background work I may not have the time/knowledge for, and the megacell isn't readily available anyway. Which leaves me at looking like something like screening cells for high IR with my RC3563, then batch capacity testing with multiple 'cheap' diagnostic chargers like the Opus 3100, or the XTAR VC4S, or the VP4 Dragon, or the LiitoKala, or the Zanflare or ....something on lgyte-info.

All these basic chargers though seem to have their own quirks with cooling and/or accuracy, depending on what year/model/firmware/batch you have, and reviews/expereinces seem to vary greatly from one to another. Which makes me wonder if I - should just bite the bullet and get a few SkyRC MC3000s. But then I have to try and find the updated dual-fan models, they have their own learning curve which slows getting started wirth the actual testing, AND they cost HEAPS here in Australia.

And then there's the matter of what data to even record. I'm seeing some pretty impressive record-keeping as I trawl through some of these threads!

So, to boil it all down to some succinct questions, I was looking for advice on the following:
-In 2022, what would the experienced advise as a 'best use of your money with decent results' setup, for screening and cap-testing large volumes of 18650s, assuming starting from scratch?
-While doing this testing, and indeed over the life of the project, what data would be advisable to log at the commissioning of a batch of cells, and over their service life?

I realise these are questions that have been asked before, but many such threads are now years old, and it seems to be this is a hobby where the community knowledge is constanlty growing, and advice may now have changed.

Thanks very much in advance :)
 

OffGridInTheCity

Active member
Joined
Dec 15, 2018
Messages
2,218
So, to boil it all down to some succinct questions, I was looking for advice on the following:
-In 2022, what would the experienced advise as a 'best use of your money with decent results' setup, for screening and cap-testing large volumes of 18650s, assuming starting from scratch?
I use 3 x OPUSs and do 12 cells per up to 36 cells for 2 days. My current battery (260ah @ 48v) needs 1,512 cells and I've had to do 3,000 cells to get 1,512 good ones (50% keep rate) - which is the worst I've ever encountered. In other cases I've had 80% to 99% keep rates. But going with 3,000 cells at 36cells per 2 days that's 167days for cell processing. This is OK for me because I do it in the 'background'.

Question: How many cells do you plan to process over what time frame? If thousands, then you might want to do bank of OPUSs (or similar) or multiple Mega Chargers.

-While doing this testing, and indeed over the life of the project, what data would be advisable to log at the commissioning of a batch of cells, and over their service life?
I do a basic test process - check IR, charge, discharge, charge and keep 90% of original capacity or better but discard any heaters. Then let sit at least 4 weeks and discard any self-dischargers. Evenly distribute capacity / types thru the packs of a battery so all packs have same ah + makeup.

For my powerwall its designed for <200ma/cell charge & discharge! Then I track hi v, low v, daily cycles to get cycles + average DOD. I also track ah/volt with an idea to get an estimate on loss of capacity over time - but this has not been useful yet.

And above all - HAVE FUN!
 
Joined
Jun 1, 2022
Messages
4
I use 3 x OPUSs and do 12 cells per up to 36 cells for 2 days. My current battery (260ah @ 48v) needs 1,512 cells and I've had to do 3,000 cells to get 1,512 good ones (50% keep rate) - which is the worst I've ever encountered. In other cases I've had 80% to 99% keep rates. But going with 3,000 cells at 36cells per 2 days that's 167days for cell processing. This is OK for me because I do it in the 'background'.

Question: How many cells do you plan to process over what time frame? If thousands, then you might want to do bank of OPUSs (or similar) or multiple Mega Chargers.


I do a basic test process - check IR, charge, discharge, charge and keep 90% of original capacity or better but discard any heaters. Then let sit at least 4 weeks and discard any self-dischargers. Evenly distribute capacity / types thru the packs of a battery so all packs have same ah + makeup.

For my powerwall its designed for <200ma/cell charge & discharge! Then I track hi v, low v, daily cycles to get cycles + average DOD. I also track ah/volt with an idea to get an estimate on loss of capacity over time - but this has not been useful yet.

And above all - HAVE FUN!
Thanks!!

Seems like the OPUS is still something of an icon at this point. It's going to be a bank of charger/testers for sure, so keeping costs down is a concern.

Do you do your discharges at 1A during testing? The LiitoKala can be had here for 2/3rds the price of the Opus but it only does 1A charge/ 0.5A discharge, which I imagine extends the testing cycle substantially vs 1A for both

I'm a bit gun-shy on the megachargers having seen the ....difficulties people have encountered. It sounds like a bit of a project unto itself, and I need to simply get started on processing them.



As a side note - I would like to get a second IR probe to have a mate start helping with the testing. How 'fussy' does one need to be with accuracy for IR testing in this application? Are cheaper units typically good enough for 'screening'? Something like this: https://www.jaycar.com.au/universal-battery-tester/p/QP2260?
 

floydR

Member
Joined
Aug 23, 2017
Messages
1,527
Best to use two IR meters of the same type then test with a known resistor or a 18650 to make sure they both check out. You can use diferent types of IR meters (4wire ac method is best) if you calibrate the offsets on each meter with a known resistor/18650. I use an YR1030+ IR meter. As long as you and your mate test the same way, same pressure, etc consistency is what you are after it should be good. Just don't use the tester on the Opus.
Later floyd
 

Redpacket

Active member
Joined
Feb 28, 2018
Messages
1,457
This thread & others with flow charts might help:
If you start with an IR test, it's the quickest way to dump bad cells....
 
Joined
Jun 1, 2022
Messages
4
This thread & others with flow charts might help:
If you start with an IR test, it's the quickest way to dump bad cells....
I am indeed using that very thread as the basis of my method!

Just curious as to whether there's been any developments in the community wisdom since they were first posted
 
Joined
Jun 1, 2022
Messages
4
Best to use two IR meters of the same type then test with a known resistor or a 18650 to make sure they both check out. You can use diferent types of IR meters (4wire ac method is best) if you calibrate the offsets on each meter with a known resistor/18650. I use an YR1030+ IR meter. As long as you and your mate test the same way, same pressure, etc consistency is what you are after it should be good. Just don't use the tester on the Opus.
Later floyd
Noted! I'll grab one and do a comparison
 

Korishan

Moderator
Joined
Jan 7, 2017
Messages
7,071
Since you're starting out, and want to bulk charge lots of cells for testing (faster to bulk charge a lot of cells, finish top off if needed in the opus, and use the opus to capacity discharge), then go with the TP5100 modules and stay away from the TP4056 modules. Not only do they waste less energy, they are more reliable and consistent. Plus they'll operate directly off a 12V system w/o issues.

Just don't try to use them in 2S charging, even tho they are claimed to do so. Most of them have the wrong capacitor/resistor combo to do 2S charging. So a swap of these components would need to be done to properly charge 2S. But in 1S they are really great for that task.
 

harrisonpatm

Member
Joined
Jan 5, 2022
Messages
168
I use 3 x OPUSs and do 12 cells per up to 36 cells for 2 days.

This is OK for me because I do it in the 'background'.
I'll put in my two cents that this is a good way to do it, if you the spare time, and you can increase this rate if you'd like, while still having it basically as a background task. I would recommend not just having a bank of opus' but also a good workflow space in general. For example: box of cells needing to be tested, empty box for finished cells, box of charged cells waiting for self discharge check, box of cells waiting for capacity check... if that sounds like a lot, I can recommend my system:
IMG_20220602_092215.jpg
The green tape is on the inside of the drawer, so I can just use and see a whiteboard marker on the front for date charged, ect., then erase with my finger when that batch of cells is done or moving on to the next step. You can see that I only have one drawer in use currently, on first charge and waiting 10 daysto check for self discharge. But with this system I can fill it up, 30 per drawer, and not lose my progress.

@OffGridInTheCity does 36 per day using 3 Opus, @ 1-2 rotation of cells per day, so probably like only 5-10 minutes work per day. If you want that done in less than 167 days, I have 4 Opus and can process 2-3 rotations per day, depending on the cells. 4 chargers= 16 cells x 3 rotations per day = 48 per day, so 3000 cells could get processed in 62.5 days. This is not to say @OffGridInTheCity is slow; on the contrary, I am emphasizing his point that if you only have a few chargers and a few minutes per day, with a good system in place, you can easily get a few thousand cells tested in less time than youd think.
 

OffGridInTheCity

Active member
Joined
Dec 15, 2018
Messages
2,218
I'll put in my two cents that this is a good way to do it, if you the spare time, and you can increase this rate if you'd like, while still having it basically as a background task. I would recommend not just having a bank of opus' but also a good workflow space in general.
Excellent point! I have a 'permanent battery work station' area in my office. A dedicated desk for solder/assembly/supplies a battery rack, and a large bookshelf I use for tested cells before they go into the battery when needed. There's also a workbench area in the garage for busbars. Right now it's a mess as I'm 2/3's the way thru battery #8 :)
1654181916471.png

As @harrisonpatm says, a permanent setup facilitates spending time each day or hear and there to advance the process. However, there's no rush if you're patient. The pink cells in the pic above started arriving last Dec and are fine sitting a long time - its me that get's impatient!. I've gone as long as a year and 1/2 to build a battery.

One can cut down on the time significantly by spending $ on higher quality cells. For example, at one point Battery Hookup had taken cells out of the modem packs for you and you could just buy them 'ready to test' - wish they still did that. At another point Battery Clearing House was selling RING packs and those cells were 99.9% 'brand-new' - quick/easy to process and at 2950mah each it takes less cells per battery.
 
Last edited:

TeresaRogers

New member
Joined
Jul 23, 2022
Messages
1
This thread & others with flow charts might help:
If you start with an IR test, it's the quickest way to dump bad cells....


Thanks a lot. I would also like to help you by sharing the https://casinosanalyzer.com/online-casinos/advanced-cash-aka-advcash website with you where you can find real money gaming sites.
Thanks a lot :)
 
Last edited:
Top