Internal Resistance - How much is too much.

[dsullivan]

I have spent much time researching this topic and finding many references but no exact answers.

Is there a discussion listing the Internal Resistances of different cell types and brands when new and/or used.
I am seeing many people saying 5mohm should be new and normal and anything over 20mohm is bad and some posts are showing 150mohm for used cells.

I have some around 150 and some around 380.

Also can someone address how the C-rate is related. I believe lower resistance means higher C-rate. Is this true?

What are the Internal Resistances on new cells of li-ion/limn/lipo/nicd/nimh etc and what would be acceptable for used 1C 18650s?

Thanks everyone

 

[999zip999]

Wow. Do you mean the short answer.

 

[IBScootn]

Check your manufacturers spec sheet for what the IR and C rate should be? Too many chemistries and varying companies products to give general answers.

 

[liveforphysics]

If you have an 18650 with 380mOhm Ri, lets say it's 2.4Ah (pretty typical for LiCoO2 18650's made in the last 3 years).

Loaded at 1C (2.4A), you have 0.91v of sag. This means, if the cell was at 4v, now loaded at 1C, about 1/4 of the cells stored energy is going into just heating the cell rather than doing useful work.

Loaded at 2C (4.8A), you have 1.82v of sag, so close to half the energy stored in the cell is just going into heating itself rather than moving your vehicle.

Loaded at 3C (7.2A), you have 2.73v of sag, and once you cross the point where sag is over half the cells unloaded voltage, you actually get more power out of the cell with less current load, and in this case about 70% of the energy in the cell is just going into heating itself rather than doing useful work.


When you say 1C is your design load for the battery, keep in mind that average C-rate makes no difference on the useful energy you get from the cells. For example, if you have a 10Ah pack and a 20A controller, if you're on-throttle half the time pulling 20A and drain the pack over the hour period, your average C-rate was only 1C, however, if every time you were on throttle it was drawing 20A, then you had the same energy loss as if you had drained the pack at 2C (only you had more time to let it cool, but that energy still got wasted at heat rather than moving your bike forward).


Personally, I would cull anything over 30mOhm if I was making an 18650 pack, and yes I realize this means culling most every sort of cell ever used in a laptop.

 

[dsullivan]

Wow thanks for the explanation and the number. I appreciate it immensely.
I am finding many sells settling at 4.1v after 24 hour rest and the IR is around 119mohm.
I am also using these cells to make packs for our old rc trucks and boats so these are great.

Thanks again.
Dave

 

[chilledoutuk]

Personally, I would cull anything over 30mOhm if I was making an 18650 pack, and yes I realize this means culling most every sort of cell ever used in a laptop.

But you should also remind people that your not a typical ebike user when it comes to power output demands from battery.

I have a 4p 13s pack of sanyo 2400 18650 cells that i run on a controller limited to 27amps and if i were to constant discharge them on the bench they will get quite warm at 2c but the bursty nature of ebike use seems to generate a lot less heat and the cells seem to get only slightly warm and this is even with them in a thermally insulated bag.

One thing i have noticed when going on long rides with my 10s4p paralleled with my SubC 30cell 36v 2.5mohm per cell pack is that it seems to act like a buffer for the 18650 cells and after a long commute completely sealed in a coolbag they don't even feel warm to my cold hands.

 

[flathill]

If you have an 18650 with 380mOhm Ri, lets say it's 2.4Ah (pretty typical for LiCoO2 18650's made in the last 3 years).

Loaded at 1C (2.4A), you have 0.91v of sag. This means, if the cell was at 4v, now loaded at 1C, about 1/4 of the cells stored energy is going into just heating the cell rather than doing useful work.

Loaded at 2C (4.8A), you have 1.82v of sag, so close to half the energy stored in the cell is just going into heating itself rather than moving your vehicle.

Loaded at 3C (7.2A), you have 2.73v of sag, and once you cross the point where sag is over half the cells unloaded voltage, you actually get more power out of the cell with less current load, and in this case about 70% of the energy in the cell is just going into heating itself rather than doing useful work.


When you say 1C is your design load for the battery, keep in mind that average C-rate makes no difference on the useful energy you get from the cells. For example, if you have a 10Ah pack and a 20A controller, if you're on-throttle half the time pulling 20A and drain the pack over the hour period, your average C-rate was only 1C, however, if every time you were on throttle it was drawing 20A, then you had the same energy loss as if you had drained the pack at 2C (only you had more time to let it cool, but that energy still got wasted at heat rather than moving your bike forward).


Personally, I would cull anything over 30mOhm if I was making an 18650 pack, and yes I realize this means culling most every sort of cell ever used in a laptop.

Is it possible that the heating of the cell is not always "wasted" and you actually can have a synergistic effect in certain circumstances? Look at this discharge curve of some 26650 NMC cells:



They seem to have more capacity the harder you push em (rated 3.6Ah at 1C but deliver closer to 3.8 at 5C)...I imagine the self heating of the cells is a good thing if kept in balance

http://www.batteryspace.com/prod-specs/6459_1.pdf
http://www.batteryspace.com/prod-specs/6459_6.pdf

I just picked up 50+ of these cells so I'll let you guys know if it is a real effect or measurement error next week. So far all the cells are extremely well matched with no duds. The IR spec is <20mOhm with a PTC switch which isn't too bad

 

[Farfle]

when we were load testing some konions that were freezing fking cold, @3c they would drop to 3.6v and stay there for awhile, then in a few minutes in they would start to come back up in voltage to about 3.8v, we noticed that it was directly related to cell temp, as the warm konions just started at 3.8v.

 

[Russell]

They seem to have more capacity the harder you push em (rated 3.6Ah at 1C but deliver closer to 3.8 at 5C)...I imagine the self heating of the cells is a good thing if kept in balance

Average voltage is lower at the higher discharge rate so Watt-Hours (a better measure of capacity) is less than at a lower discharge rate.


-R

 

[flathill]

They seem to have more capacity the harder you push em (rated 3.6Ah at 1C but deliver closer to 3.8 at 5C)...I imagine the self heating of the cells is a good thing if kept in balance

Average voltage is lower at the higher discharge rate so Watt-Hours (a better measure of capacity) is less than at a lower discharge rate.


-R

Very good point! I'm just used to seeing most discharge graphs show lower Ah capacity with increasing discharge rate even with increasing sag (which is what you would intuitively expect)
example

I don't usually see it trend the opposite way like the NMC cells which makes me think it is suspect...
Does anyone have data on IR vs temp and capacity vs temp for various chems?
Here is one I found:

The point is internal resistance can be synergistic up to a point maybe
maybe more so for NMC
not sure

 

[neptronix]

Here's another wrench to throw into the works.

Cells actually have an internal resistance curve. Internal resistance climbs exponentially after they fall off their cliff. It also, depending on the battery, tends to be higher when fully charged.

A 1C cell is going to have a super huge internal resistance, a 40C cell is going to have a super tiny resistance. The question is, what is normal for the battery's chemistry at nominal voltage? That's what you need to know in the first place.