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

DIY Prius HV battery rebuild

Discussion in 'Gen 2 Prius Care, Maintenance and Troubleshooting' started by MikeGreen, Jul 29, 2015.

  1. MikeGreen

    MikeGreen Junior Member

    Joined:
    Jul 29, 2015
    23
    5
    0
    Location:
    Troy, NY
    Vehicle:
    2006 Prius
    Model:
    N/A
    Hi guys, first post here but I've been lurking since I was considering buying a Prius.
    I was hoping that some of you guys could offer some advice for my situation.

    I bought an 06 with 216k miles on it and what seemed to be a bad battery pack, the warning lights were on and the battery read at 1 bar. We left it idling to charge for an hour before I attempted to drive it home, an hour down the park way. On the way back the battery entirely cut out and I only had the motor to power me, according to the info screen the battery was not being used at all, and the car could barely make it up to speed with my foot to the floor perpetually.

    Once I got it home took the battery out of the car and disassembled it and saw that a cell was reading at 6.8 volts so I replaced it with another I bought off ebay then reassembled the pack and put it into the car.
    The car drove fine for about 50 minutes until the warning light went on and it quit using the battery entirely again. During the 50 minutes that it did work the mileage seemed to be lower then it should be, around 36 mpg.

    I've noticed that the car perpetually thinks it had a full tank since I've filled it up when I got it, even though I've burned at least two gallons. The 12v battery also appears to be a Toyota one, is it even possible that it's the stock one?

    I limped the car home and pulled the pack out of the car, read the voltage of every cell and then did a load test with a 55w automotive headlight bulb. All of the numbers seem normal so I'm not really sure where to go from here.
    here are the cells and their results
    cell #, unloaded V, loaded V
    1, 7.99, 7.91
    2, 8.03, 7.80
    3, 8.03, 7.85
    4, 8.03, 7.89
    5, 8.03, 7.95
    6, 8.05, 7.91
    7, 8.03, 7.94
    8, 8.10, 7.93
    9, 8.03, 7.87
    10, 8.04, 7.82
    11, 8.03, 7.85
    12, 8.03, 7.92
    13, 8.04, 7.90
    14, 8.03, 7.92
    15, 8.03, 7.85
    16, 8.03, 7.82
    17, 8.04, 7.92
    18, 8.03, 7.82
    19, 8.04, 7.92
    20, 8.03, 7.89
    21, 8.03, 7.80
    22, 8.04, 7.87
    23, 8.03, 7.90
    24, 8.03, 7.85
    25, 8.03, 7.87
    26, 8.03, 7.90
    27, 8.03, 7.86
    28, 8.02, 7.87

    After searching and reading whatever I could find online I think the next step is to balance charge them? So I bough 6 DC6 chargers on amazon. here's their title since I can't post a link
    "SKYRC 100% Original iMAX B6 OEM Balance Charger for LiPo, 1s-6s Li-ion,1~15cell NiMH Akku 50W Balancer Ladegerät Netzteil"
    The car has been throwing a P3000 code but that's all my code reader will say.
    Thanks for the help in advance!
     
  2. bisco

    bisco cookie crumbler

    Joined:
    May 11, 2005
    110,156
    50,059
    0
    Location:
    boston
    Vehicle:
    2012 Prius Plug-in
    Model:
    Plug-in Base
    welcome! there are some great threads here on battery rebuilding, all the best!(y)
     
    Jimmie Ray likes this.
  3. MikeGreen

    MikeGreen Junior Member

    Joined:
    Jul 29, 2015
    23
    5
    0
    Location:
    Troy, NY
    Vehicle:
    2006 Prius
    Model:
    N/A
  4. gdanner

    gdanner Member

    Joined:
    Jul 29, 2007
    86
    39
    0
    Location:
    Central Illinois
    Vehicle:
    2005 Prius
    Model:
    N/A
    It appears that all 28 of your HV battery modules are good!

    But there are several other things than can cause DTC codes:
    • Is HV battery safety lock pushed DOWN all the way?
    • Are the thin wires from HV battery busbars to HV ECU (orange plug and socket) in good shape? The HV battery ECU uses these thin wires to measure the voltage of each block. I've heard stories of them getting corroded until they break.
    • Are all temperature sensors inside the HV pack (1 on top of modules, 3 underneath the modules) properly positioned and in good condition?
    • Are all nuts which connect the busbars to the modules tight? (But don't overtighten).
    • Do you have a device that can scan the DTC codes?

    Finally, I do recommend reading and re-reading all threads on Priuschat that deal with DIY HV battery repair.

    I replaced 2 modules in my HV battery pack 4 weeks ago. So far it is working perfectly. I started on my 3rd tank of fuel today. I haven't had any DTC codes. Mpg is up to 43.5.

    -EB
     
  5. MikeGreen

    MikeGreen Junior Member

    Joined:
    Jul 29, 2015
    23
    5
    0
    Location:
    Troy, NY
    Vehicle:
    2006 Prius
    Model:
    N/A
    The battery is out of the car now so I can work on it.
    The safety lock was certainly pushed down because it wasn't originally, and it took me four hours to figure out it was causing the car not to start lol.

    The thin wires running to the bus bars all appear to be in good shape, and I cleaned all of the buss bar's copper plates and nuts the first time I had the battery out of the car to clear off any corrosion. Nice and shiny now :D
    The nuts were all torqued to whatever "26" is on my cordless drill's torque setting, which was snug but not over tightened when I checked by hand.

    I have the scanner that they lend out at pepboys, it says it reads DTC codes but it would only show the generic p3000 high voltage code
    As for the temperature sensors I've only removed one, it was the one on the bottom farthest from the ecu end of the battery on the last cell, but I put it back in place before reinstalling the battery the first time. I haven't touched the other temperature sensors so I assume they're still installed the way they should be.

    I forgot to mention that once the car went into safe mode when I was driving it, the battery fan stayed on high. Not sure if this is standard or a sign of overheating problems?

    I've been suspicious of the battery's ecu but if that was bad the car wouldn't run at all right?

    Thanks for the quick response and all the help!
     
  6. ericbecky

    ericbecky Hybrid Battery Hero

    Joined:
    Mar 12, 2004
    4,379
    3,238
    1
    Location:
    Madison, Wisconsin
    Vehicle:
    2005 Prius
    Model:
    Two
    I would get a scanner so you can read each vblock voltage while car is on.

    Or take time to load test each vblock and record/compare results.

    If you knew the code, you might be better informed to solve the problem.
     
  7. MikeGreen

    MikeGreen Junior Member

    Joined:
    Jul 29, 2015
    23
    5
    0
    Location:
    Troy, NY
    Vehicle:
    2006 Prius
    Model:
    N/A
    I thought I did load test them? I hooked a 55 watt automotive headlight to each cell, one at a time and measured their voltage drop, that is the data that I posted in my first post. Unless there's another test I could do?
    I don't want to take it to a stealership because even after knowing the mechanic they still want 114$ just to read the codes.
     
  8. gdanner

    gdanner Member

    Joined:
    Jul 29, 2007
    86
    39
    0
    Location:
    Central Illinois
    Vehicle:
    2005 Prius
    Model:
    N/A
    It's worth getting either a mini-VCI (works with a Windows XP laptop) or a OBD2-Bluetooth adaptor (works with Android devices) so that you can view the details of the DTC codes. And also clear them in case you get more codes when you drive the car. The mini-VCI does a more complete analysis, since it is basically running Toyota techscan. That's the same app that Toyota dealers use.

    Is your HV battery fan working?
     
  9. usnavystgc

    usnavystgc Die Hard DIYer and Ebike enthusiast.

    Joined:
    Jul 29, 2011
    3,159
    989
    0
    Location:
    Tucson, AZ
    Vehicle:
    2005 Prius
    Model:
    Four
    The thing that jumps out at me is the balancing of your cells. The Prius will throw a code if the batteries are more than .3V apart. Although you don't show any in your test, it is likely that the variance is greater as its running. The quick fix is to hook them all in parallel for at least eight hours (24 is better) and that will likely fix your problem.
     
  10. wjtracy

    wjtracy Senior Member

    Joined:
    Sep 19, 2006
    11,341
    3,596
    1
    Location:
    Northern VA (NoVA)
    Vehicle:
    Other Hybrid
    Model:
    N/A
    As far as your gas tank, the first little pip can last up to 150 miles or so. After that they average 25-30 miles per pip. Just approx. You are the proud new owner of flexible bladder "guess" tank
     
  11. MikeGreen

    MikeGreen Junior Member

    Joined:
    Jul 29, 2015
    23
    5
    0
    Location:
    Troy, NY
    Vehicle:
    2006 Prius
    Model:
    N/A
    Hey guys, so I left the batteries in parallel over night then put them back into the car and took it on a test drive.
    I averaged 43 mpg which gives me some confidence, but during the beginning of the ride the battery fluctuated a lot going from purple to green in the span of five minutes. So I'm a little worried about that, the fan never came on loud enough for me to hear from the driver's seat though. Towards the end of the ride it stabilized around 2/3 full on the display.
    My balance chargers are supposed to be delivered tomorrow so I'm debating just returning them or going through what appears to be an arduous process of balance charging all of the 28 cells
     
  12. gdanner

    gdanner Member

    Joined:
    Jul 29, 2007
    86
    39
    0
    Location:
    Central Illinois
    Vehicle:
    2005 Prius
    Model:
    N/A
    Here's my recommendation: Leave the HV battery in the car until you've driven it far enough to ID which blocks still have issues.

    With Torque/Android (or mini-VCI and laptop) you can monitor the 14 individual block voltages while you drive. This data will point out the weakest block under actual operating conditions. This may work better than most of the bench testing techniques that I've read about.

    Although I haven't done enough testing of my own to prove this yet, I have a theory that we don't need to spend hours and hours and hours "cycling" and bench testing individual modules. Those who are in the business of rebuilding Prius HV battery packs DO need sophisticated bench testing gear. But I propose that us DIY folks (who just wish to fix the HV battery in our own car) can use a more streamlined procedure that doesn't take a lot of time or require a heap of special test gear.

    I've been driving my 2005 Prius for >3 weeks after changing out 2 modules in my HV pack. Recent mpg is 43.5. No DTC codes.

    I monitor individual block voltages, HV battery temps, and several other items using Torque/Android every time I drive my car. I'm "logging" this data so that I can download it to my PC and analyze it later. So far all 14 blocks are quite similar to each other in performance.

    FYI, my Prius MFD SOC display almost always stays at either 6 blue bars or 7 green bars. On 1 or 2 occasions where I was in a drive-up lane with A/C running for >10 minutes, SOC gradually dropped to 4 blue bars. That's about the lowest I've seen it since I fixed my HV battery.

    -EB
     
    Neural Net likes this.
  13. MikeGreen

    MikeGreen Junior Member

    Joined:
    Jul 29, 2015
    23
    5
    0
    Location:
    Troy, NY
    Vehicle:
    2006 Prius
    Model:
    N/A
    The car threw the red triangle of death again, but thankfully this time my code reader said that block 6 became weak.
    I have an extra good cell I bought off eBay so all I have to do now if figure out which of the two in the block is back then swap it out and put the cells in parallel for the night. Any tips for identifying which cell is bad?
     
  14. MikeGreen

    MikeGreen Junior Member

    Joined:
    Jul 29, 2015
    23
    5
    0
    Location:
    Troy, NY
    Vehicle:
    2006 Prius
    Model:
    N/A
    Holy crap! The connection between cell 17 and 18 melted the orange bus bar plastic to the terminals. Which is odly not block 6
     
  15. MikeGreen

    MikeGreen Junior Member

    Joined:
    Jul 29, 2015
    23
    5
    0
    Location:
    Troy, NY
    Vehicle:
    2006 Prius
    Model:
    N/A
    Now it just so happens that the melted terminal ends would be block six if I counted from the non ecu end, maybe it's a coincidence but I'm suspicious. The copper plate that formed the connect where the melting occurred in now heavily corroded. Which is super weird considering I cleaned every copper terminal bridge before reassembling the pack the first time.
    The entire pack is now measuring between 8.05 and 8.25 volts
    There seems to still be liquid around the melted terminals, battery acid?
     
  16. gdanner

    gdanner Member

    Joined:
    Jul 29, 2007
    86
    39
    0
    Location:
    Central Illinois
    Vehicle:
    2005 Prius
    Model:
    N/A
    From which direction are you counting blocks?

    Do you count block 1 at ECU end of HV pack? Or do you count block 1 from battery fan end?

    For gen2 Prius (2004-2009) I believe "blocks" are counted with block 1 (modules 1 and 2) on the battery fan end, and with block 14 (modules 27 and 28) on the ECU end.

    Counting blocks from battery fan end puts your overheated bus bar on block 6 (modules 11 and 12). This corresponds to your DTC code, right?

    But if you count backwards (block 1 at ECU end), these same 2 modules are 17 and 18 (block 9)

    Can you post photos of the overheated bus bar?

    The most likely cause of overheated bus bar is loose hardware, stripped or broken stud on module, or corrosion that didn't get removed.

    Up to +/- 100 amps of peak current can flow through these bus bars.

    Check all connection bolts inside the ECU section too: I've seen photos of those overheating due to loose hardware.

    Due the modules themselves show any signs of internal overheating, bulging, burned/melted areas, or leakage of liquid?

    -EB
     
  17. gdanner

    gdanner Member

    Joined:
    Jul 29, 2007
    86
    39
    0
    Location:
    Central Illinois
    Vehicle:
    2005 Prius
    Model:
    N/A
    Yes, block 1 is on the battery fan end of the pack. So your DTC code "block 6 failed" is correct.

    If one of the modules in block 6 is reading 8+ volts and the other module is reading 0, this confirms which of the 2 failed. But if you have enough spare modules it might be a good idea to change both.

    The fluid that leaks out of Prius HV modules is a "base," not an acid. It is a solution of KOH (potassium hydroxide). Similar to lye (NaOH, sodium hydroxide). Avoid getting it on your body! Clean it up with plenty of clean water, then make sure to dry everything off thoroughly. Even a tiny amount of the battery fluid can cause the DTC code for "ground fault" (P3009). There's a discussion of this in the Priuschat "rehydrating the cells" thread.

    -EB
     
  18. MikeGreen

    MikeGreen Junior Member

    Joined:
    Jul 29, 2015
    23
    5
    0
    Location:
    Troy, NY
    Vehicle:
    2006 Prius
    Model:
    N/A
    If I count from the fan end then the suspect cells would be block six, and yes the plastic on the cells are melted and there seems to be battery acid running down them, here's a pic.
    I see liquid on the the ends of the modules under where the terminals melted.
    I'm about to go out and take the pack apart to see the sides of the cells
     

    Attached Files:

  19. gdanner

    gdanner Member

    Joined:
    Jul 29, 2007
    86
    39
    0
    Location:
    Central Illinois
    Vehicle:
    2005 Prius
    Model:
    N/A
    which of the 2 modules in block 6 is reading 0V now?
     
  20. MikeGreen

    MikeGreen Junior Member

    Joined:
    Jul 29, 2015
    23
    5
    0
    Location:
    Troy, NY
    Vehicle:
    2006 Prius
    Model:
    N/A
    That's the thing, they both read 8.1x and both do fine under a headlight load test.