Complete Discharge with log

[magudaman]

I did a complete battery discharge driving my FFE and recorded the event using Forscan. I'm just over 12,400 miles and have never discharged the car past 2%. The battery depleted came up and and continued driving.

Interesting finds from the complete discharge:

- The actual percent (not displayed percentage) of charge at shutdown was 4.4%!
- The lowest cell voltage was 3.09v! (Generally 3.0v minimum, but closer than I thought it would get)
- My car had around .7kw/h after the battery depleted came up
- I couldn't get my car to restart and had to coast home
- It appears it started to shut KW limit when the lowest cell hit 3.14V, but I was still accelerating which gave the low voltage above.
- It seems my power was limited by something else beside the power limit because I wasn't able to get over 40kw. Then again I didn't have the room to really pick up speed in the hood, but it felt like it was limping.

Below is overview starting at around 2% on the displayed percentage (~11% actual), the data shows around .5kw/h remaining.

Overview:

Detailed:

Personally I will not do this again. I don't think it is exactly good for the pack despite keeping it above 3v per cell.

 

[michael]

I really doubt you can do any harm using this process. 3 V/cell is far above the damage threshold.

However, I'm puzzled by your pack and cell Voltage. They don't seem right.

My pack Voltage runs between approximately 360 and 300. Yours was pretty steady at around 290.

It also doesn't seem right that with the battery fully charged, the cell Voltage is about 3.48. I would have expected it to be in excess of 4 V.

What am I missing?

 

[magudaman]

It not super clear but what you see in the overview is actually around 11% to 4% true pack state of charge. I only started recording when my car was showing 2% on the my touch screen. The cell and pack voltages are low because the state of charge is so low.

 

[michael]

What is the x-axis? I assumed Wh from 0 to 18000. Not so?

 

[WattsUp]

What is the x-axis? I assumed Wh from 0 to 18000. Not so?

I would assume it is some sort of measure of elapsed time? I can't figure out the units easily, though.

For example, if we assume milliseconds, it would mean the whole graph only spans only 18 seconds, which seems too short. If seconds, then the span would be much too long. So, I don't think it is any traditional unit of time.

My next conclusion is that the X-axis represents the "sample number", captured at some interval. For example, using a 20 millisecond sampling interval, 18000 samples would account for 6 minutes of elapsed time, which starts to seem more realistic in this case.