Pack internal resistance

E-HP

100 GW
Joined
Nov 1, 2018
Messages
7,507
Location
USA
In playing with the Grin simulator, I've been modeling my pack, under "custom battery" using the total voltage and storage capacity Ah. I left the internal resistance value at whatever it defaulted to. Anyway, I tried backing into the internal resistance value of some of the existing packs from the drop down, and noticed they seem to be in the range of 0.1 ohm to 0.07 ohm, depending on the pack. My thought is that the simulator estimates the IR based on pack configuration (series and parallel groups), and applies some estimated/average values for IR at the cell level to determine/calculate the pack IR.

Since my "pack" is a combination of different cells and chemistries, in parallel and series, I will need to determine the IR for each component, based on the IR of the cells used in its parts, before combining everything into a single "pack" IR. Does this seem correct?
Main pack seems straightforward. Look at the cell spec sheet. Calculate the resistance of 8 cells in parallel, multiply by 14.
I have 14S of lipos in parallel with the main pack, so similar calculation for each battery, then summed up.
Then calculate the resistance of main pack and lipos in parallel.
Does this sound right?

I'm mainly trying to model my battery pack accurately, since I noticed, and was surprised by, how the simulator's performance results change with changes to the IR. Prior to that, I always thought in the simulator, the controller values dictate how the performance curves are calculated (battery and phase current limit), and only using voltage from the battery as an input for performance, and the pack Ah as an input for range. I can understand logically how the IR would affect performance, but puzzled by how the simulator incorporates it into the calculations.
 
Last edited:
Since my "pack" is a combination of different cells and chemistries, in parallel and series, I will need to determine the IR for each component, based on the IR of the cells used in its parts, before combining everything into a single "pack" IR. Does this seem correct?
Main pack seems straightforward. Look at the cell spec sheet. Calculate the resistance of 8 cells in parallel, multiply by 14.
I have 14S of lipos in parallel with the main pack, so similar calculation for each battery, then summed up.
Then calculate the resistance of main pack and lipos in parallel.
Does this sound right?
Sounds right.

FWIWI, if you have a Cycle Analyst, it can estimate the pack Ri over a charge/discharge cycle based on performance; don't know if it gets more accurate over time or not. (I don't know what it does, but the results seem realistic based on what I know about my pack; I don't have a direct way to measure my pack Ri.)



I can understand logically how the IR would affect performance, but puzzled by how the simulator incorporates it into the calculations.
That, I don't know.

But I can say that if a pack has high Ri, it will sag in voltage more and be less able to deliver power, so depending on how the controller itself responds to the voltage sag, and how the motor then responds to the lower power conversion, it could seriously affect performance--so if the simulator is realistically simulating, it's results should reflect that for controllers already modellled in it. For controllers not modelled, using the generic user input, it might either use a generic simulation profile for that or it might not simulate it's response to Ri at all.
 
That, I don't know.

But I can say that if a pack has high Ri, it will sag in voltage more and be less able to deliver power, so depending on how the controller itself responds to the voltage sag, and how the motor then responds to the lower power conversion, it could seriously affect performance--so if the simulator is realistically simulating, it's results should reflect that for controllers already modellled in it. For controllers not modelled, using the generic user input, it might either use a generic simulation profile for that or it might not simulate it's response to Ri at all.
I think you’ve got it. The battery inputs for the model can be adjusted for voltage sag. I always assumed the model just inputs the voltage and assumes no sag, but now it seems clear why they include a value for internal resistance under the custom battery parameters. I’m going to try inputting custom batteries based on a few configurations and inputs using the data sheets for a few different cells and see how they affect performance. Thanks!
 
Back
Top