Battery capacity test? (technical)

[dangerHV]

I thought I knew batteries well. Not well enough though. My question is how do I know the true capacity of the HV battery in a used car I'm interested in buying? I wasn't too concerned until I read the Battery University article on prolonging the life of lithium batteries. A fully charged battery sitting in the hot summer sun for months degrades the capacity fairly rapidly. Some of the cars I'm looking at have sat in the hot California summer sun for two years now, presumably "idle". I doubt very much that most of the dealers know - or would bother - to keep the batteries charged to no more than 50%. (Jeffrey please tell me Georges does, especially with you there).
>Are there test points under the hood that I can test with a meter ?(1kv meter)
>Is there a way to interpret the screen data?
>How about a voltage reading before and after a test drive of "20" miles?
>Can anyone tell me what the peak charge and fully discharged voltages are? I'll compare this to A123's data sheet.
(Engineering level responses will be greatly appreciated and understood)

David

 

[kevin]

I thought I knew batteries well. Not well enough though. My question is how do I know the true capacity of the HV battery in a used car I'm interested in buying? I wasn't too concerned until I read the Battery University article on prolonging the life of lithium batteries. A fully charged battery sitting in the hot summer sun for months degrades the capacity fairly rapidly. Some of the cars I'm looking at have sat in the hot California summer sun for two years now, presumably "idle". I doubt very much that most of the dealers know - or would bother - to keep the batteries charged to no more than 50%. (Jeffrey please tell me Georges does, especially with you there).
>Are there test points under the hood that I can test with a meter ?(1kv meter)
>Is there a way to interpret the screen data?
>How about a voltage reading before and after a test drive of "20" miles?
>Can anyone tell me what the peak charge and fully discharged voltages are? I'll compare this to A123's data sheet.
(Engineering level responses will be greatly appreciated and understood)

David

There is no way of determining the battery capacity from a single measurement.

One way is to use the Energy Screen which gives a energy used and percentage used measurements and extrapolate from those.

To use this approach effectively you would have to charge the battery fully then drive at least 50 miles to get reasonable accuracy.

This is the capacity of my 2014 Spark over the last two years. It has degraded about 10%.

I only used discharges that were >50% but even so there is a fair amount of variability. I collect the data by taking a photo of the various screens just before I charge and then enter it into the spread sheet later.

kevin

 

[dangerHV]

Thank you Kevin. 10% seems like a lot for 2 years/14kmi. If you can, could you let us know how you use and charge your car?
>Deep or shallow discharges?
>Kept fully charged in hot climate w/o being plugged in? (This seems almost unavoidable for many people. Charge to full, go to work <10 miles away, car sits 8+ hours in hot sun, go back home and charge to full again. Worst conditions for these and most other batteries)
>DC Fast charge frequency, if any?
>Odd as this sounds, is your car Black or Titanium by chance?
If you averaged 50 miles between charges (I'm just picking a random number), then in 14,000 miles you would recharge 280 times, DOD approx 60%. This seems like minimal use for my understanding of A123 pouch cells. I would have expected less than 5%. There is a decent chance that the linear decline you've seen so far will slow it's rate. So maybe at the 5 year point you'll only be down say 15%. Just hoping. This assumes you keep your that long. (I keep my cars til they don't run anymore)
Thanks again, Kevin. Keep the data coming!

David
ps, Bingo on the energy screen data. Just what I needed!

 

[kevin]

Thank you Kevin. 10% seems like a lot for 2 years/14kmi. If you can, could you let us know how you use and charge your car?
>Deep or shallow discharges?
>Kept fully charged in hot climate w/o being plugged in? (This seems almost unavoidable for many people. Charge to full, go to work <10 miles away, car sits 8+ hours in hot sun, go back home and charge to full again. Worst conditions for these and most other batteries)
>DC Fast charge frequency, if any?
>Odd as this sounds, is your car Black or Titanium by chance?
If you averaged 50 miles between charges (I'm just picking a random number), then in 14,000 miles you would recharge 280 times, DOD approx 60%. This seems like minimal use for my understanding of A123 pouch cells. I would have expected less than 5%. There is a decent chance that the linear decline you've seen so far will slow it's rate. So maybe at the 5 year point you'll only be down say 15%. Just hoping. This assumes you keep your that long. (I keep my cars til they don't run anymore)
Thanks again, Kevin. Keep the data coming!

David
ps, Bingo on the energy screen data. Just what I needed!

My usage is not far off what you described, maybe a bit fewer miles before charging. I don't have the DCFC option (it wasn't available when I got mine). I normally charge level 2 at work.

I'm in the bay area so although the car does sit much the day with full charge I don't know that it gets that hot. I have only seen battery conditioning occur once and that was after being left out on a cold night.

I was surprised at the amount of degradation although I don't now how accurate this measurement is; There will be some error due to the BMS estimation algorithms, especially as I rarely discharge to a very low SOC to allow it to recalibrate.

If you notice, the original capacity value was ~19.5kWh which is more than supposed to be available, even now at ~17.3kWh it is close to the usable capacity that is normally published.

My estimated range at full charge has probably decreased a bit to ~110miles from ~116 when the car was new.

The furthest I have driven on a charge was about 2 weeks ago when I did 87 miles with 24 miles indicated remaining.

I rarely use A/C or heating and commonly get ~6.5mi/kWh. My overall average for the first year was 5.7mi/kWh (DC) 4.7mi/kWh (AC).

If anything I think the energy efficiency has improved with time - I suspect that is mainly due to wear on the tires; I noticed that on my Prius, new tires would get about ~10% worse fuel consumption and would take a few thousand miles to get back to the mpg of the previous ones.

kevin

 

[Zoomit]

FWIW, my 2015 (LG Chem) battery has shown no capacity loss after 4000 miles, using a dense dataset and a similar calculation technique as Kevin's. Kevin's sparse data looks like you could indentify the loss at 3000 mi and see it clearly at 6000 mi.

 

[dangerHV]

Zoomit, do you adhere to any extra TLC schedule to prolong the LG's battery life? And do you plan to keep it for 5 years or more?

David

 

[Zoomit]

I discharge the battery down to 10-15% SoC 5-days a week and always recharge it full. It sits outside in the summer heat as well. During the summer, it typically used a few percentage of the battery for thermal management on the way home. So I'd say I'm giving the battery quite the workout. I am leasing it, but wouldn't change how I'm treating it if I owned it.

 

[NORTON]

Wouldn't any plot like above require really consistent driving to be meaningful?

I know on my daily commute I can be very consistent.
But when I'm on weekends or after work driving through the city I'm anything but consistent.
I'm enjoying this little EV hotrod ! I like silently leaving ICE cars in my fumes at stop lights! Wait,,, I don't make fumes, so forget that expression.
My technique is to just stay ahead a few feet and when I hear their engine getting louder, meaning 'It's on!', then I really pull away!

It is not good to store a Li-ion at 100% charge at high temp. Any Li-ion, phone, laptop, flashlight, etc.
In fact the Tesla S series uses battery power to power its TMS from 100% to 80% SOC when the car is not being used and parked in high heat.

From what I read no other EV uses TMS unless the car is plugged in or powered ON.
This function could be made configurable in the Spark EV if an owner can afford up to 20% drop in range. Think leaving your Spark EV at the Tucson airport for a week.
Who knows how many days of TMS 20% SOC would provide?
But at least you'd be left with a battery at 80% SOC and, theoretically, a happier and longer life at high temps.

 

[Zoomit]

Wouldn't any plot like above require really consistent driving to be meaningful?

No, the two legs of my commute are drastically different, yet each yields very similar battery capacity numbers. What's really important is long drives, or more specifically large discharge percentages, to reduce the rounding error. With short drives, the percentage of battery used is small relative to the two digit data precision we can get from the car (or app).

 

[kevin]

FWIW, my 2015 (LG Chem) battery has shown no capacity loss after 4000 miles, using a dense dataset and a similar calculation technique as Kevin's. Kevin's sparse data looks like you could indentify the loss at 3000 mi and see it clearly at 6000 mi.

What capacity are you indicating?

I have a complete data set for the first year, I only included data for the battery capacity calculation where the battery had been discharged a sizable amount (>50%) to reduce quantization error in the measurements.

kevin

 

[Zoomit]

About 17.8 kWh.

I collect my data immediately after arriving, so any voltage recovery is not included.

 

[dangerHV]

This discussion is producing extremely valuable information, Thanks guys.
Can I request a few people put a thermometer on the back seat while car will be parked in the sun for at least an hour.(windows closed) I'd like to see the difference between inside & outside temps with the different colors offered. I wanted the Titanium but if the inside temp is significantly higher, I'll have to go with "generic" Silver. Black and white are out of the question for my tastes. I might be able to tolerate "sky blue", but would really rather not. (2014)
Norton I like your driving style. "cruel", but safe and controlled! You mentioned something that really caught my interest in above post.

(Norton)
"It is not good to store a Li-ion at 100% charge at high temp. Any Li-ion, phone, laptop, flashlight, etc.
In fact the Tesla S series uses battery power to power its TMS from 100% to 80% SOC when the car is not being used and parked in high heat.
From what I read no other EV uses TMS unless the car is plugged in or powered ON.
This function could be made configurable in the Spark EV if an owner can afford up to 20% drop in range. Think leaving your Spark EV at the Tucson airport for a week.
Who knows how many days of TMS 20% SOC would provide?
But at least you'd be left with a battery at 80% SOC and, theoretically, a happier and longer life at high temps."

I have a 10 Cu-Ft fridge/freezer that uses 1 Kwh per 24 hours. This is at an ambient temp of around 75f. It just seems like it should be "easy" to keep the battery pack between +/- 40-70f without a lot of energy. I wonder just how well insulated it is. Is the TMS "available" with the power off for those who want to see how long a battery can last? You used the term "configurable". Is this in the owner's manual? (I'll look when I'm done here)
(I couldn't find this in the online owners manual.)

David

 

[NORTON]

..... Is the TMS "available" with the power off for those who want to see how long a battery can last? You used the term "configurable". Is this in the owner's manual? (I'll look when I'm done here)
(I couldn't find this in the online owners manual.)

David

Oops, What I meant was "It could/should be configurable in a perfect world of EV's". It is not.

Park your car for months in the Tucson heat with 100% SOC and your battery is using up its 'shelf-life' much faster than if it was at 40%.

Maybe GM could send us a new SW load OTA through OnStar (when they get back in the office) in this 'perfect world of EV's' !! Just like Teslas sends out new SW updates!

 

[dangerHV]

Well, thanks for the disappointing clarification, Norton. I didn't have my hopes up since I hadn't heard of this before. Is it possible to leave the car "on", lock the doors and walk away? Aside from the pump running for the 1st few seconds after turning on, it doesn't seem to need power for anything else, as long as everything else is turned off.
It really would be nice to have control over this stuff, though. OTA software updates?, what a concept. If we could afford a Tesla we wouldn't have any more challenges in life! ok, on the driving side anyway...

David

 

[dangerHV]

On the subject of TMS & battery life, I found a 2010 thread on GM Volt forum. It's 19 pages of intense discussion mostly referencing the Volt with TMS and the leaf w/o. Glad I've still got coffee left! I don't know how to create a proper link, sorry:

gm-volt.com/forum/showthread.php?5243-Volt-thermal-management-system-temperature-band

(hope I got that right)

David