Understanding Series amperage load across a pack.

RiskyJoint

100 µW
Joined
Dec 3, 2021
Messages
9
Hello, I am new here. A long time lurker and am not much of a fan of facebook groups.

I am currently building and designing a 20S7P Pack. It's using 45A Rated 2170 cells, Copper for the conductor 0.2mm in thickness. When designing this pack I designed it to have as much copper surface area as possible to handle the BMS that is 150Amps.

My question is, how does the amperage split across a pack? My current understanding is that is you had a 20S battery, with a 100 amp bms for example than each cell is only outputting 5A to achieve the 100 amp output.

How ever when I think of the voltage drop if you were to request 100 amps from a 20s1p pack it does not add up.

Series = Increased amperage same AH?
Parallel = Same amperage higher AH?

IS this the correct break down of how it works? Or would each cell need to be required to be able to output 100 amps a piece to handle this load? Thank you for any help in breaking this down and appreciate the information. I want to learn now before attempting to finish my build so I have a proper understanding.
 
The series count determines the pack voltage.

The parallel count determines the pack amps and Amp-hour capacity.

7P with a 100A BMS limit means if you truly draw 100A, then you are taking 14.5A from each cell. Of course that is only the temporary peak amps during acceleration. During any cruise-phase, the amp-draw is lower...just enough to maintain speed.
 
So what is the difference in this setup? This is where I get confused.

If I took two 48V 50AMP packs and wire them in p I get 48V but still only 50 AMP output but double the range.
If I take two 48V 50 AMP packs in hook series then I have 96V and 100AMP output. Same range.

Maybe my testing was flawed? I tried to get the most accurate data possible but With the packs having BMS I am not sure how it could would affect some of the connections.
 
RiskyJoint said:
So what is the difference in this setup? This is where I get confused.

If I took two 48V 50AMP packs and wire them in p I get 48V but still only 50 AMP output but double the range.
If I take two 48V 50 AMP packs in hook series then I have 96V and 100AMP output. Same range.

Maybe my testing was flawed? I tried to get the most accurate data possible but With the packs having BMS I am not sure how it could would affect some of the connections.


Paralleling 2 48v 50ah packs would give you a 48v 100ah pack.

Series of 2 48v 50ah packs would give you 96v 50ah.

Both packs are the same capacity of 4,800 watt hours.

Don’t rely on bms ratings to guarantee the current that a battery pack can deliver. You can put the largest 10000a bms on a tiny 12v 1ah battery but it won’t matter if the cells can’t deliver the power.
 
Right that part I get, but the total amperage out between those two difference is what?

Ignoring AH, and speaking in terms out amp output.
 
RiskyJoint said:
Right that part I get, but the total amperage out between those two difference is what?

Ignoring AH, and speaking in terms out amp output.

Divide the number of amps you want to pull by the number of parallel connections and that will tell you how many amps each cell group is going to deliver.

If you have a 48v 13s10p group and try to pull 100 amps from the battery, each group of 13s cells will see 10 amps.
If you have a 48v 13s5p group and try to pull 100 amps from the battery, each cell group of 13s cells will see 20amps.

Amps per cell group is determined by the number of parallel connections, it has nothing to do with series or amp hour capacity.
 
Got ya, so series or parallel is not going to affect the overall amperage output, only voltage.

So in my case my 7P Each cell "rated at 45A" Would be able to Deliver 315 AMP. Using a 150A BMS keeps me within the safe and happy limits.

Now the next question is, How is the amperage carried across the connections? Starting at group 1 does the next group now handle the additional load? Meaning the last series connections would be handling 150A and first would be only 7.5?
 
RiskyJoint said:
Got ya, so series or parallel is not going to affect the overall amperage output, only voltage.

So in my case my 7P Each cell "rated at 45A" Would be able to Deliver 315 AMP. Using a 150A BMS keeps me within the safe and happy limits.

Now the next question is, How is the amperage carried across the connections? Starting at group 1 does the next group now handle the additional load? Meaning the last series connections would be handling 150A and first would be only 7.5?

Glad I could clear that up! As far as how the load is shared, I’ll let a dedicated battery builder answer that question. I have seen that cable length and routing can make difference but I really don’t have any experience in that department.

Word to the wise- drawing max power from them will kill them quickly. A max 150a draw from them as a pack will keep them very happy over their lifetime.
 
RiskyJoint said:
Got ya, so series or parallel is not going to affect the overall amperage output, only voltage
No, this was answered above, series only increases Voltage, Ah and A remain. Every cell/group has the same current level pasding through it.

The LOAD determines the amps (attempted to be) drawn.

The BMS usually does not LIMIT that draw, programming your controller, or using a CAv3, or self control by the user must do that. Or a fuse.

The ampacity of the usual BMS must be sized at least 20% higher than the maximum amps you will ever inflict on it. Or you use external contactors and size them that way, current not flowing through the BMS at all.

Paralleling keeps Voltage the same, only increases the Ah and amps capacity, C-rate constant because that includes Ah.

 
If it helps:

All amps (A, current) in series are the same.

All volts (V) in parallel are the same.

Volts V add in series but stay the same in parallel.

Amps A and Amp Hours Ah add in parallel but stay the same in series.
 
RiskyJoint said:
So in my case my 7P Each cell "rated at 45A" Would be able to Deliver 315 AMP. Using a 150A BMS keeps me within the safe and happy limits.
Didn't you mean to say ... "Using a 150A Controller keeps me within the safe and happy limits." If that is your plan than using a 180A BMS would be a 20% increase over the Controller rated amperage output.

Considering the higher cost of even the best 2170 cell with a 45A rating you might want to consider it more of a burst rating than continuous output rating. If you want to get the most performance cycles out of your 20S7P DIY pack treat it NICE & RIGHT.
 
I’ll just point out that the controller can be a 1000A controller set at a limiting current of 150A (or whatever needed). Needn’t be a 150A controller..

Sizing of the drive system is a balancing game and my advice is to size everything up one size if you can afford it. Want 3kW? Get a 5kW battery, controller and bms. Want 20km driving distance? Get a battery for 40 etc.
You’ll avoid bottoming out your drive the first month and get some margin to the manufacturers exaggerations
 
I was looking at some 21700 cells rated for 45A, Wow, but most are still about 4.5AH, so you might get that for .10 hour or 6 minutes. I wonder if you could do that twice or would the cell simply be worn out. Your 20S-7P is still only around 30-35 AH. Is that enough for 100A use? That's 14A per cell. Don't know your intended use, but maybe you ought to size for less current per cell for longevity?

At 140 cells for 20S-7P, you're still up close to $1000 USD for cells. Hate to see that get burned up,
 
docw009 said:
I was looking at some 2170 0 cells rated for 45A, Wow, but most are still about 4.5AH, so you might get that for .10 hour or 6 minutes. I wonder if you could do that twice or would the cell simply be worn out. Your 20S-7P is still only around 30-35 AH. Is that enough for 100A use? That's 14A per cell. Don't know your intended use, but maybe you ought to size for less current per cell for longevity?

At 140 cells for 20S-7P, you're still up close to $1000 USD for cells. Hate to see that get burned up,
What about the Lishen 21700 4000mAh 35A Battery (LR2170LA) at $4.35/cell x 140 = $610
https://www.18650batterystore.com/products/lishen-21700-4000mah-35a-battery-lr2170la
150A Controller Setting ÷ 20 p-groups = 7.5A per ea of the p-groups 7 cells at 150A maximum 20S-7P output
300A Controller Setting ÷ 20 p-groups = 15.0A per ea of the groups 7 cells at 300A maximum 20S-7P output

Is my figuring anywhere close? Being 77 yrs young am satisfied with a top speed of 12mph on my LT Cargo Trike
 
eMark said:
docw009 said:
I was looking at some 2170 0 cells rated for 45A....
At 140 cells for 20S-7P, you're still up close to $1000 USD for cells. Hate to see that get burned up,
What about the Lishen 21700 4000mAh 35A Battery (LR2170LA) at $4.35/cell x 140 = $610
https://www.18650batterystore.com/products/lishen-21700-4000mah-35a-battery-lr2170la
150A Controller Setting ÷ 20 p-groups = 7.5A per ea of the p-groups 7 cells at 150A maximum 20S-7P output
300A Controller Setting ÷ 20 p-groups = 15.0A per ea of the groups 7 cells at 300A maximum 20S-7P output

Is my figuring anywhere close? Being 77 yrs young am satisfied with a top speed of 12mph on my LT Cargo Trike

Yes, Lishen cells are cheap, compared to Samsung, so for my first battery build, I bought some Lishen cells from that seller, a good seller by the way, I bought the LR2170SF which are 4500ma 13,5A I built a 13S-2P. I was disappointed in the range, At 50% capacity, I had to be careful with power, or the pack would shut down from voltage sag, At 20A, the sag was about 5 volts. However, if I was careful, I could draw 8AH out of the battery, which was good given its 9AH rating, I took it apart and will rebuild it as a 13S-3P. Knock the sag down to 3V.

I conclude that the series resistance is too high. Using a 1 ohm resistor, I'm measuring 50-60 milliohms at 4A on a Lishen LR2170SF, I haven't seen a DC spec for IR on this cell, but if my measurement is in the ballpark, it's high,

A better choice was the Samsung 40T's. DC IR is around .015 ohm. Might have dropped the sag to 2 volts, Then when the pack was down to 44 volts, it could still deliver 10-12 amps w/o shutting down the BMS. The Samsung cells were 2X the cost of Lishen cells and less AH, but I believe they would have been more useable.

One can build a working ebike pack with cheap cells, but you need a lot more cells.





,
 
For longevity their 50E is much better than 40T, only go with the latter if you **really need** the higher burst C-rates.

Molicel P42A, LG HG2, SONY VTC6 also highly reco'd for similar use cases

Avoid 30Q
 
eMark said:
Being 77 yrs young am satisfied with a top speed of 12mph on my LT Cargo Trike

Then what in heaven's name are you going to do with 150 to 300 amps?
 
Chalo said:
eMark said:
Being 77 yrs young am satisfied with a top speed of 12mph on my LT Cargo Trike

Then what in heaven's name are you going to do with 150 to 300 amps?

^^👍👍 yes , why ??
Age is irrelavent ( but i am also well past 60 !) .. 48 v, and 30A peak takes me places you may need a ladder to climb, and also roll at 40mph, ...if necessary !
 
spinningmagnets said:
I typically cruise along at 20-MPH, but a couple of times a 3,000-lb car tried to kill me, and I used every watt at my disposal.
I've noticed the past few years that Watt hours seems to be the more meaningful/useful measure (for some reason) than Amperage capacity for certain applications, but don't know why. Perhaps you can post a link to another ES thread explaining this conundrum :wink:

Anyway i'm thinking my previous comparison is wrong figuring ... so posted what may be the correct figuring at bottom ...

With his Controller pulling 150A from 45A Rated 2170 cells (20S-7P pack) what is theoretically the amperage draw (amp capacity / watt hours aside) from each of the 20 parallel groups when the Controller is passing 150 amps of current ...

150A Controller Setting ÷ 20 p-groups = 7.5A per ea of the p-groups 7 cells at 150A maximum 20S-7P output
300A Controller Setting ÷ 20 p-groups = 15.0A per ea of the groups 7 cells at 300A maximum 20S-7P output

... OR IS THIS CORRECT ...

150A Controller Output ÷ 7 cells in 20 p-groups = 21.5A demand of each p-group cell at 150A maximum Controller output
300A Controller Output ÷ 7 cells in 20 p-group s= 43.0A demand of each p-group cell at 300A maximum Controller output

What is the correct way to express his 150A vs 300A comparison being that there's a definite advantage of one over the other when having to evade a 3,000-lb car trying to run-over an ebiker. Do you remember what Tony Stewart said about competing in the Indy 500 against light-weight Danica Patrick ... "I'll have to start eating more salads" :D
 
This makes me curious of what motor bike and controller you'll be running with this pack.
The 48v pack can put out 100amps the 98v pack can put out 50amp.
 
Finished up spot welding the battery, at 4 welds (4 touch down's 8 contact points) Nickel/copper .2/.2 .

K-Weld really worked well calibrated at 1900, Just needed to clean the tips every 100 weld's or so ( Pain) 140J Is where it seemed to have best weld thru the copper.

Purple circles are just to keep track of when a cell is fully welded. Now time to run balance leads and finish up the securing of pack. Going to take a break and at it again soon.
 

Attachments

  • 264031697_604568387268220_1696796661267819386_n.jpg
    264031697_604568387268220_1696796661267819386_n.jpg
    342.3 KB · Views: 530
300a @ 80v = 24000watts

That is a lot of heat, maybe we can make a motor glow in the dark :shock:
 
Back
Top