DIY Electric Car Forums banner

1 - 17 of 17 Posts

·
Registered
Joined
·
215 Posts
Discussion Starter · #1 ·
Hello,

I'm contemplating batteries for my 83 Rabbit and our company has a resident battery expert from China. I was discussing lithium prismatics with him and he recommended using smaller batteries like 40aH cells and placing them in parallel to reach the intended current rating requirements. According to his research at Chinese manufactures the smaller prismatics hold up better since their plates are physically smaller they are much less likely to bulge or "puff" and the resistance between the anode and cathode is much less than in a larger cell. I was wondering what others thought about these facts and if anyone else is using smaller capacity batteries and hooking them up in parallel. Seems like a lot more points of failure to me but he brought up what seem like valid points and made sense from a purely physical sense. Any suggestions or disagreements with his assertions?

Steve
 

·
Registered
Joined
·
102 Posts
Sounds very logical. Add the ability to fit batteries in smaller spots and the lower cost of a replacement cell if one should loose capacity and multiple small cells makes a lot of sense. Single cell charging systems would become more expensive but the batteries banked in separate groups and paralleled at the controller would allow a balance gauge type meter that was mentioned on this forum to be used and give a quick indication if one battery bank had a problem. It would also give redundancy capability albeit at reduced power to get home if one battery bank dropped a cell.
Weight might be an issue and a lot more cabling.

T1 Terry
 

·
Registered
Joined
·
870 Posts
All of those 40Ah cells will have a bunch of extra size and weight due to the plastic shells and since they are shorter, depending on how they are configured and the space they will fit in, they may take up too big of a footprint. Nothing that a calculator and a layout sketch wouldn't answer. I've done similar calculations with going with smaller cells but more cells higher voltage to get similar capacity. I personally wouldn't want to parallel more than 2 prismatic cells due to the cabling requirements and trying to fit other cells in. You are right though, smaller cells will sag less at the same C rate, CALB for example rates the 40Ah cells as 400amps peak(10C) for 10 seconds, 100Ah at 8C, 130Ah at 8C(1000 amps), the 180Ah cells at 1000 amps. The 1000 amps probablu involves the limits of the terminals too.
 

·
Registered
Joined
·
2,122 Posts
In my opinion, you will be trading internal losses within the large cell with external losses at all extra connections you have to make, so the bottom line will likely be a wash. However, you will waste room and weight on all the extra plastic casing. Also, smaller cells cost more per AH due to extra casing and terminals.

It is true that larger cells swell easier due to larger surface of their walls, so you should restrain them in their battery boxes to prevent swelling. But don't go nuts with restrains, any reasonably rigid box construction will work fine.

I can only see one benefit of using smaller cells in parallel, if your battery space is tight and you want cells to fill every niche and corner, but that comes at the cost of carrying more copper connections.

As for C rates, its best to plan your pack size not to go over 2C for long periods, afterall you'd want to drive your car for more than 30 minutes, right?
 

·
Registered
Joined
·
325 Posts


This is a thundersky 40Ah that I took apart.
You see that the cells are composed of very large number of thin aluminum and copper foils that are parallelled together to create 40Ah.
Having removed them, the plastic box contributes almost nothing to the weight of the cell. The big aluminum and copper connecting blocks would add extra weight compared to using larger cells.

But I believe your friend is correct, the thin foils must be coated evenly to make a good cell, and what makes the batteries premature fail is the uneven heating along the foil surface area, and as such the larger cells increase the liklihood of early capacity loss. Much like large LCD screen are more prone to failure than small ones.

It would be my approach to parallel the small cells to get a larger pack,
but frankly it probably doesn't make a huge difference relatively to the general treatment of the cells.

Another big issue is how you pack them so they all have the same temperature. If you have interior cells they can get hotter than the border cells, so avoid this type of packing, and that might be harder with smaller cells.

Hello,

I'm contemplating batteries for my 83 Rabbit and our company has a resident battery expert from China. I was discussing lithium prismatics with him and he recommended using smaller batteries like 40aH cells and placing them in parallel to reach the intended current rating requirements. According to his research at Chinese manufactures the smaller prismatics hold up better since their plates are physically smaller they are much less likely to bulge or "puff" and the resistance between the anode and cathode is much less than in a larger cell. I was wondering what others thought about these facts and if anyone else is using smaller capacity batteries and hooking them up in parallel. Seems like a lot more points of failure to me but he brought up what seem like valid points and made sense from a purely physical sense. Any suggestions or disagreements with his assertions?

Steve
 

·
Registered
Joined
·
251 Posts
There is one other big risk with putting smaller cells in parallell instead of using larger cells.
I had one of my TS LFP 160ah cells going busbar for some reason, which I don't know. There has been another one do that here in sweden also.

That failure mode would be very dangerous when putting smaller cells in parallell, the other cells in the parallelle pack would dump all the current they can into that busbar cell. Creating a perfectly good heater.......

regards
/Per
 

·
Registered
Joined
·
903 Posts
There is one other big risk with putting smaller cells in parallell instead of using larger cells.
I had one of my TS LFP 160ah cells going busbar for some reason, which I don't know. There has been another one do that here in sweden also.

That failure mode would be very dangerous when putting smaller cells in parallell, the other cells in the parallelle pack would dump all the current they can into that busbar cell. Creating a perfectly good heater.......

regards
/Per
What is the difference between one cell in a parallel set going bussbar and a portion of a single larger cell going bussbar? The pic of the 40Ah cell clearly shows two 20Ah "sets" of plates. If one side were to go bussbar the other half would dump into the first. Isn't that the same as if two smaller cells were paralleled?
 

·
Registered
Joined
·
3,218 Posts
What is the difference between one cell in a parallel set going bussbar and a portion of a single larger cell going bussbar? The pic of the 40Ah cell clearly shows two 20Ah "sets" of plates....
Good catch... I didn't notice that myself until you pointed it out, David, but it certainly does look like there are two 20Ah cells packed back-to-back inside that case.

That said, we've managed to make several TS-160Ah cells "go busbar" in series packs and though they did tend to limit power to the controller, they didn't go up in smoke or otherwise fail catastrophically, despite pulling 500A+ through them. If we still have any of those cells around I'll cut open the case to see if they are constructed in the same manner as the 40Ah cell that nimblemotors cut open.

I really worry about parallel connected cells failing like this... it almost seems we need to bring back the ultimate ozone killer, Halon, to protect against the fire potential of these cells...
 
G

·
From what I know all the cells have double packs, each half the AH rating of the total cell size. So yes, there are two bundles in parallel within each cell block. I found that with the Hi-Power 100 AH cells and saw that with Jacks larger cell that he took apart as well. So it seams all these cells use a dual pack in parallel. So you should be able to do parallel cells quite nicely. They are built that way.

Question? What do you mean by going busbar? It sounds like you mean shorting out somehow. Not quite sure. New term here to learn. Any ideas on the cause?

Pete :)

You will notice the twin packs within the 100 AH cell.
http://youtu.be/mpzjkF5XMPk Standard size
http://youtu.be/mpzjkF5XMPk?hd=1 HD size
 

Attachments

·
Registered
Joined
·
2,122 Posts
I really worry about parallel connected cells failing like this...
Above comments portrait a situation where busbar cell is paralleled with healthy charged cell, but it doesn't happen like that in reality.

I bet they don't go busbar suddenly, nor when they are full. Chances are they go busbar when abused and drained to zero, in which case the parallel cell will share the same level of abuse and likely share the same fate. In any case, once one parallel cell becomes a busbar, the one parallel to it won't take any further charge/discharge, so no risk coming from that pair anymore.
 

·
Registered
Joined
·
3,218 Posts
...Question? What do you mean by going busbar? It sounds like you mean shorting out somehow. Not quite sure. New term here to learn. Any ideas on the cause?
I had never heard the phrase "going busbar" (or "bussbar") either, but figured it meant the cells turned into a direct short.

The cause in our case has always been from overdischarging of unbalanced cell packs. I've mentioned before I'm not really in one camp or the other when it comes to top or bottom balancing, but I will say that if you have a top-balanced pack then you really need to know when the first cell is about to run out of juice or you will destroy it.


Above comments portrait a situation where busbar cell is paralleled with healthy charged cell, but it doesn't happen like that in reality.....
That's PROBABLY true, dimitri, but I still worry about internal "mechanical" failures happening at some point post-assembly because the consequences are likely to be highly unpleasant.
 

·
Registered
Joined
·
903 Posts
I had to think through the parallel or non-parallel situation and determine what I was willing to do. I wanted enough weight in my battery box for vehicle stability so I knew that going from eight 6V flooded batteries to LiFePO4 meant that I needed to fill the majority of my battery box. The only option was a partially filled box with 90-130Ah cells or go with a lower voltage of higher capacity cells so I decided to parallel 100Ah cells. Assuming that the pairs of cells were properly connected I didn't see any difference between one 100Ah cell shorting out and part of a 200Ah cell shorting out. The cell terminals would be at the same potential in either case. Furthermore, due to the nature of these cells, even if one cell had a different capacity than the other they both would reach full/empty at the same time. In fact I did a crude capacity comparison of the cells and matched the largest capacity cell with the lowest capacity cell and then moved to the next pair and so on to build my pack. It is working just great after delivering 15K Ah.

I agree with dimitri and don't worry about internal "mechanical" failures because I figure that the 200Ah cells would have had the same probability of failure. While I don't have any data to back this up I don't believe that the probability of a mechanical failure in a 100Ah cell is the same as in a 200Ah cell. I would think that the 200Ah cell has a higher probability of failure so even though I have 40 100Ah cells in a 2p20s arrangement, 20 200Ah cells wouldn't have been much better off. Naturally, there are more connection points with a buddy-paired pack so a higher chance of external connection failures.
 
G

·
Thank you. Sorta what I was thinking but you helped clear it up.

Pete :)


I had never heard the phrase "going busbar" (or "bussbar") either, but figured it meant the cells turned into a direct short.

The cause in our case has always been from overdischarging of unbalanced cell packs. I've mentioned before I'm not really in one camp or the other when it comes to top or bottom balancing, but I will say that if you have a top-balanced pack then you really need to know when the first cell is about to run out of juice or you will destroy it.




That's PROBABLY true, dimitri, but I still worry about internal "mechanical" failures happening at some point post-assembly because the consequences are likely to be highly unpleasant.
 

·
Registered
Joined
·
325 Posts
The two sets are .5ahr plates (foils) paralleled together as well.
So what I gather can happen is the cell will overheat from overdischarge boil away the solvent and melt the seperator between the foils and the foils become shorted together, which would likely cascade into melting the whole thing shorted. Who has done some test of failure modes of these cells? Someone did them with A123 cells and headway cells as I recall.
It seems to me that such disaster with a 200ah cell would have as much energy shorted as two 100ah cells or four 50ah cells.
With paralleled cells, at least you could use a fused link between them if you are concerned about this.
Another option is to parallel multiple strings, then you can monitor that status of each cell independently, but that is a lot more complicated.

What is the difference between one cell in a parallel set going bussbar and a portion of a single larger cell going bussbar? The pic of the 40Ah cell clearly shows two 20Ah "sets" of plates. If one side were to go bussbar the other half would dump into the first. Isn't that the same as if two smaller cells were paralleled?
 

·
Registered
Joined
·
903 Posts
Another option is to parallel multiple strings, then you can monitor that status of each cell independently, but that is a lot more complicated.
The problem I have with this method is that if a cell shorts in one of the parallel strings then you have the whole other string trying to charge the first one unless you have some high current diodes on the packs. With buddy pairs you have at most the energy in two cells. It has been reported that Dave Kois, of Current EV Tech, discharged a TS 100Ah cell at 25C and got 100Ah out and could still hold the cell in his hands. It really doesn't look like the safety problem with these cells is on discharge but on over charging. I'm only talking LiFePO4 here and no other battery chemistry. This is another reason I only charge to just under 3.5vpc with my pack. It is a long way from issues and preserves the life of my pack.
 

·
Registered
Joined
·
325 Posts
It would be a lot easier to have one point where you monitor the two packs, rather than 40 of them per cell, but yes it is more complicated.
I'd like to see the video of the 25c discharge. Rickard has a video of 400amp (4c) 100AH full discharge that vented a cell and they got VERY hot.

The problem I have with this method is that if a cell shorts in one of the parallel strings then you have the whole other string trying to charge the first one unless you have some high current diodes on the packs. With buddy pairs you have at most the energy in two cells. It has been reported that Dave Kois, of Current EV Tech, discharged a TS 100Ah cell at 25C and got 100Ah out and could still hold the cell in his hands. It really doesn't look like the safety problem with these cells is on discharge but on over charging. I'm only talking LiFePO4 here and no other battery chemistry. This is another reason I only charge to just under 3.5vpc with my pack. It is a long way from issues and preserves the life of my pack.
 
1 - 17 of 17 Posts
Top