Forscan. When the ICE is stopped, record a "discharge" of the HV battery using electrical accessories (blower, headlights, rear window defroster, in "R" with parking brake set and foot off brake) monitoring SoC, current and the module deltaV. You will be able to determine X mAh extracted over Y SoC and then can extrapolate the total battery capacity, e.g.,

650mAh extracted between 55% and 40% (or 55-40 = 15%)

650mAh/0.15 = 4333mAh = 4333/5500 = 79% SoH

Abrupt drops in SoC under a constant current load and/or divergence of the deltaV number (max difference between all voltage blocks) indicate an imbalanced or ailing battery.

 

Thanks, I can certainly try that.

Why "in "R" with parking brake set and foot off brake" ?

 

Pull more current. With the foot on the brake, creep is disabled at the motor, i.e., it senses the pedal is pressed, so it doesn't even try. Parking brake only means it will still attempt to creep.

More current will yield more conservative results. Very low current draw may give optimistic results.

 

Thanks,

How much will the actual battery temperature at the time of this load test effect the mAh/SOC ratio?

 

Can't say for certain, but you're picking fly turds out of pepper at that point. Battery temperature between 60 and 90°F should produce meaningful results.

 

I haven't gotten to Forscan yet but I did get some data with the Scangauge over the past couple of weeks.

MDV (module deltaV) is stable at 0.01 - 0.02 under all driving conditions, EV mode, charging, regen, etc. I double checked that I'm using the correct X-code.

I had an opportunity today on a flat, no-traffic road to run in EV mode at 20mph from SoC of 50% down to 40% with an odometer reading of 0.4 miles. So best case it lost 10% SoC in 0.49 miles. That's about half the distance I should have gone if a healthy battery can go 1.3 miles on 53% to 40% SoC.

So the driving distance test seems to indicate a failing battery (capacity reduced by 50%) but the MDV confuses me. Seems unlikely that all modules would age together at the same rate, so why am I not seeing more of a delta in the module voltages?

As always, appreciate your comments.

 

MDV of .01-.02 is pretty unlikely particularly under any kind of load. I question the accuracy.

0.4 mile with 10% SoC loss sounds pretty good to me. Where are you getting the 1.3mi/13% ?

Did you record current over that 0.4mi drive?

MDV increase is an indication of imbalance, and one can get substantial variation in cells before it shows up in MDV especially when you're talking about cells that operation over a very narrow SoC range with a voltage curve that's relatively flat. Factor in that MDV is checking 10 cells (I think) at a time, you're seeing variation between groups of 10 cells, so a single cell is going to be masked by the others in the string.

The 250 cells have been subjected to the same conditions for X miles over Y years. Wouldn't they age together at the same or at least similar rates?

 

Thanks Steve,

The MDV is the standard X-code from the Scangauge database. Maybe Forscan will give a different number, and I will do that soon.

The 1.3mi/13% EV range has been reported multiple times in the past, there's a couple of posts below.

I probably misspoke about battery pack aging. I agree the cells should age at similar rates. I'll report back when I get a capacity check from Forscan as you suggested.

 

I'm not sure that I buy this as a valid test. There are too many variables, and temperature will play a very significant role.

A better test would be to try to hold a constant CURRENT in EV mode or during a controlled discharge. The current multiplied by the time will give you the capacity extracted. Then you can extrapolate the battery capacity.

0.4mi @ 20mph takes .02 hr
Let's say you averaged 20A over that time (if you can't log the data, you can record the display on video and make a fairly accurate estimate of the current).
That would be 20A * .02hr = 0.4Ah = 400mAh
If you extracted that over 10% SoC, then your battery capacity is 400mAh / 10% = 4000mAh, which is 4000/5500 = 72.7% of rated capacity.

Here's some battery test data:
https://avt.inl.gov/sites/default/fi...escape8237.pdf

That's where I got the 5500mAh, which is consistent with my understanding of the capacity of the Sanyo cells (also used in the HCH2).

 

I setup Forscan and collected some data for battery capacity.

Used your methodology except I had it in "L", parking brake set, no accessories, which gave me a relatively stable 3.6A of discharge (I hope TRAC_M_AMP was the correct parameter).
BTM = 79°F
DVM = 0.12 - 0.19v (mostly near 0.12v) This differs greatly from the Scangauge.
SoC 53% to 44% (when uncommanded restart occurred)
Elasped time = 5m:48s

By my math that's 70.3% battery capacity relative to the new battery of 5.5Ah. That seems pretty good for a 12 year old battery.
I'm still puzzled by why the regenerative braking system is so weak compared to a few years ago, that's my big complaint, but the battery condition would not seem to explain that.