B: "Although many of the products commercially available do not incorporate a proper BMS (battery management system), I would highly recommend including that in the mix especially if the cells are second hand.
The cells are not identical even from the factory and different operation conditions will have induced further deviation in performance and most importantly internal resistance and capacity. Over time these differences tend to create voltage differences between series cells and for this reason, the battery should be periodically balanced. The most popular way is to use dissipative balancing at the end of a charge cycle."
A: I have been thinking about integrating a BMS to the pack, but there is some contradictory information about them.
On the other hand, it would obviously be better for the cells and would be nice to know what the cells are doing, but it would add another point of failure to the pack and it would limit the usability of the pack by cutting the use too early.
Also there is EV packs, that have been running for multiple years without a BMS and the cells are "balanced" naturally.
You can check it here https://goo.gl/63CVc3 and here https://goo.gl/ZqSdmh. That is of course only one man's experiences, but nevertheless.
I'm also going to check all the cells by charging them and then discharging them and again charging them full and leaving them to storage so I will see if they lose their voltage.
I will be using this kind of charger for this: http://goo.gl/IvCAj2.
This way I can see if they are OK as if they are not, they will warm up when charged and that cell will be disposed off. Charging current will be safe at ~0,3A and therefore no problems should arise.
Again Jehus' views in this matter: https://goo.gl/yG5SMI.
Thankfully he's collaborating with EV West so maybe he knows what he is talking about...
By discharging them I will see how high the Amperage of the individual cell is and I can group them together so the series packs are as similar as they could be.
B: "To cancel out individual cell differences and to drastically reduce the channel count of the BMS, I would suggest a
Well yeah... my thinking on this one was way off and I misunderstood the comment.
As the commenter said, if I FIRST make 4 sets of 30 cells in parallel and THEN connect them in series, my "channels" for BMS for example drop to 4 and then are also more easily measurable with battery monitors.
My bad and now just I need to update my plans...
B: "Definitely an interesting idea. Are you perhaps planning on a 3D printed housing or what is the plan for mechanics?"
A: I'm going to use ready made holders as in http://goo.gl/0D4UwK and further isolate the cells with http://goo.gl/WqycEd.
For contacts I will be using http://goo.gl/1fECp9 and soldering the cells to it via individual fuses, 0.25 wire could be good, but I will need to check if it will be usable in my situation.
Soldering is not the best method, but I will clean the cell contacts thoroughly and use good flux so the heating will be minimized.
Seems to do the trick for quite a lot of people, but we will see!
I will be updating this post with further commentary and questions, so it will be up to date as best I can manage.