1. Attachments are working again! Check out this thread for more details and to report any other bugs.

Predicting HV health by delta V and discharge curve

Discussion in 'Gen 2 Prius Care, Maintenance and Troubleshooting' started by borrowedcarbon, May 4, 2017.

  1. borrowedcarbon

    borrowedcarbon Junior Member

    Joined:
    Aug 13, 2015
    16
    23
    0
    Location:
    Canada
    Vehicle:
    2017 Prius
    Model:
    Base
    Hi all,

    I'm trying to estimate my HV battery health using advice on PC and other sites. I've looked for large deltaV especially when amps are high. I hope someone can look at the graphs and tell me if I've interpreted this correctly.

    My major concern with this data is the non-linear discharge from 58% SOC to 40% SOC. I think that means my battery is worn out and I shouldn't expect it to last long. Delta V looks small enough except a few values above 0.3 ... so its worn evenly.

    Its a 2008 Prius with 170,000 km. No codes or problems, just noticed the battery is charging to bar 7 & 8 a lot more frequently than usual so I started logging.

    Thank you for your help,
    David
     

    Attached Files:

    SFO likes this.
  2. RobH

    RobH Senior Member

    Joined:
    Sep 18, 2006
    2,369
    980
    70
    Location:
    Sunnyvale, California
    Vehicle:
    2006 Prius
    Now if you could calculate block voltage times amperage, integrated from 80% SOC to 40% SOC, you'd have a measure of pack capacity. Adjust that 40% range to 100%, divide by 2, and you'll have a number you can compare to the new module design value of 6500 mAh.

    Be interesting to see how that computation comes out with various age/usage vehicles. In particular, when a pack gets the first bad module.

    I suspect that the module with the highest deltaV is the weakest, and thus the most likely failure point. But that's complicated by the fact that it is cells that fail, and the measurement available is in units of two modules which is 12 cells.
     
    SFO likes this.
  3. borrowedcarbon

    borrowedcarbon Junior Member

    Joined:
    Aug 13, 2015
    16
    23
    0
    Location:
    Canada
    Vehicle:
    2017 Prius
    Model:
    Base
    Thanks, Rob. I'll see if I can figure that out and post back.

    Could I get a rough estimate using something like:

    delta time = 8.85 minutes
    delta SOC = 38% (from 78% to 40% SOC)
    average I = 8.78A

    capacity = current x time
    = 8.78 Amp x 0.1475h
    = 1.3 Ah
     
    #3 borrowedcarbon, May 5, 2017
    Last edited: May 5, 2017
  4. Fred_H

    Fred_H Misoversimplifier

    Joined:
    Apr 20, 2007
    555
    476
    28
    Location:
    Germany
    Vehicle:
    2007 Prius
    Model:
    Touring
    Yes, no signs of excessive imbalance or bad individual modules. The few delta V data points above 0.3 V are mostly just signal noise. Block Voltages that are consistently slightly higher or lower (typically the ones at both ends) is normal. A block that conspicuously swings from being highest to being lowest Voltage depending on SOC or current would indicate a weak or defective cell.

    Your battery appears to have just a normal slight imbalance. If your pack ever becomes more imbalanced, you might consider balancing it with a grid powered charger.


    That is not a nonlinear discharge. That is a nonlinear estimation of SOC by the battery ECU. Look at the actual Voltage and current. They are close to linear. The ECU was looking at those and estimating the SOC. Then it noticed that voltage was dropping too quickly to be consistent with its estimate, and decided to change the calibration of its estimated SOC algorithm.

    There does appear to be some loss of capacity, but it could still be years before it has a significant effect on gas mileage or a cell failure.
     
    borrowedcarbon likes this.
  5. borrowedcarbon

    borrowedcarbon Junior Member

    Joined:
    Aug 13, 2015
    16
    23
    0
    Location:
    Canada
    Vehicle:
    2017 Prius
    Model:
    Base
    Thank you - that makes sense - and a big difference in interpreting the graph.

    I'm assuming the hybrid system regulates based on block voltages, and not SOC. ... so a miscalculated SOC would affect my MDF view but not the actual operation of the car?

    I'll see what I can find about recalibrating SOC.
     
  6. Fred_H

    Fred_H Misoversimplifier

    Joined:
    Apr 20, 2007
    555
    476
    28
    Location:
    Germany
    Vehicle:
    2007 Prius
    Model:
    Touring
    Only Toyota knows exactly, but all my data logs and everything I've seen from others indicates that basic operation is mostly based on the SOC estimate of the battery ECU. But the battery ECU keeps a close eye on many parameters, and when Voltage or anything else gets too far out of line, it will signal for a change in operations to bring it back in line. So the miscalculated SOC would affect the operation of the car, but probably only insomuch as recharging or EV mode might begin or end a little sooner or later.


    As far as I know, it is not possible to directly recalibrate the SOC with readily available means. But it does occasionally recalibrate itself. I could speculate that force charging to the full 80% and then running it down to the 40% might encourage it to recalibrate itself, but I don't know. However, I do know that it sometimes does this all by itself while driving at a steady, moderate speed on a long flat road.
     
    borrowedcarbon likes this.