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

HV Battery Balancing Questions

Discussion in 'Gen 2 Prius Technical Discussion' started by Ed Carmack, Oct 14, 2013.

  1. Ed Carmack

    Ed Carmack New Member

    Joined:
    Aug 28, 2013
    25
    4
    0
    Vehicle:
    2005 Prius
    Model:
    Two
    I am in the process of balancing the pack out of a 05. It failed at 157 K miles. Took it to the dealer. Code indicated failure of module 9. Did a minimal amount of research, bought a couple of cells off ebay, watched a you tube on how to remove the pack, put a new battery in my multimeter and went at it. Found clearly low voltage on cell number 11 vs.9.. Installed a new/old cell and it worked. Drove it for 9500 miles including a 6500 mile trip through the western high mountain country and its is a minor miracle but had nary an issue. The moral of this story is that foolishness sometimes pays off. Did a little more research and realized how lucky I was and decided on doing a more elegant repair vs. just slapping in a new/used module. Based upon what I learned, figured that the next failure was probably not far off. Also as there are three middle aged priis in the family I am thinking that I need to know how to deal with the failures. All live in the hot steamy south and spend summers baking in the sun.
    Bought a pack from a 09 low mileage donor from the cool northeast and installed in my car. Car behaves differently than with old battery but no change in MPG that I can see. Much more regeneration going on with the 09 battery. Maybe by a factor of three according to the little green cars showing up on the consumption display. I could probably drive it farther in reverse but haven’t tried.

    My plan is to balance the 05 pack and keep it as a spare.

    I am not an electrical engineer but did manage A’s in circuit design 101 and 102. However, that was so far back that Dr. Faraday was the prof. Not his fault but most of what he taught me has evaporated. Not to say that I won’t do it but I don’t need advice on how not to electrocute myself etc.

    On to the questions.

    1. From what I have gathered from other posts it is important to have modules that are of similar capacity. Makes sense as the weaker modules could be driven into overcharge/undercharge. What is an acceptable variance?

    2. Have seen much discussion on having to use multiple packs to get enough similar capacity modules. What if I did not balance the better modules until optimum capacity is reached – stopped short and then worked the weaker ones to match the better ones? Obviously my pack would not have as high an overall capacity but other than that is there a downside to this approach?

    3. Bought a Thunder T6 4 channel charger/discharger which I suspect to be the same as the Hitech charger only with an even less decipherable manual. I ended up downloading the Hitech manual and only then was I able to get things moving. The 1st charge for module 9 (which was maybe not coincidentally the one identified by the code) was halted at 3822 MaHrs on high delta V. I restarted the charge and got another 1370 before it again halted on delta V. It’s initial d/c value was low. About half of the other 3 that I have done. Subsequently it is improving nicely. Is restarting the charge after a high delta V a bad idea? I realize that some overcharging is needed but what is too much?

    4. I am thinking that my relatively early battery failure was helped by heat along with a clogged cooling fan. Driving environment was mostly open road with some stop and go and the occasional mountain. Would it be beneficial to put a layer of insulation on the area above the battery to help keep the summer cabin heat away from the battery? Any downside to this? OEM is basically a piece of plastic with carpet attached. Not much insulation but some.

    5. My plan is to do the balancing on the 05 HV pack, remove the modules and store them tightly wrapped in a freezer. When needed warm them, check the voltages and charge as needed to get voltages close, rearrange them in the pack based upon capacity and do the final balance on the pack. Is this a bad idea? Mainly the freezer part.

    Will appreciate help from the battery gurus.
     
  2. jdenenberg

    jdenenberg EE Professor

    Joined:
    Nov 21, 2005
    3,852
    1,845
    1
    Location:
    Trumbull, CT
    Vehicle:
    2020 Prius
    Model:
    LE AWD-e
    You want all of the modules as close in capacity, SOC and series resistance. If module pair voltages differ by about 0.3 volts the ECU will set a code. Note that there are two things that can kill a module. One is heat as you have noted. The other is if a cell has a lower SOC that the others in the long series string, it is at risk of being totally discharged and reversed when the battery is being discharged. That will permanently destroy the reversed cell.
    This is not a wise approach as it leaves the cells in the module out of balance and the weaker ones are subject to a reversal as noted above.
    I never restarted charging, choosing to discharge and then recharge to the first delta V again. The danger in going further is that you may overheat the module which is a bad idea.
    Insulation is the wrong approach. You should try to increase cooling air flow and/or try to lower the temperature of that cooling air.
    Store the modules at about 80% SOC at cool temperature but not frozen as either temperature extreme can hurt module life. Recharging them after a few months of storage and leveling the charge before installation in the HV assembly is a good idea.

    JeffD
     
    Cyon Kim likes this.
  3. uart

    uart Senior Member

    Joined:
    Sep 7, 2009
    4,215
    1,200
    0
    Location:
    Australia
    Vehicle:
    2005 Prius
    Hi Jeff. I don't own one of these chargers but I'm interested in their operation. What exactly is a "high delta V" on these things. Only one voltage (that of the module) is being monitored isn't it, so I assume it's not referring to a voltage difference between cells.

    Is the delta V referring to the increase or the decrease in module voltage?
     
  4. jdenenberg

    jdenenberg EE Professor

    Joined:
    Nov 21, 2005
    3,852
    1,845
    1
    Location:
    Trumbull, CT
    Vehicle:
    2020 Prius
    Model:
    LE AWD-e
    Uart,

    This discussion of Delta V refers to one module being charged. As an NiMh (or NiCad) cell approaches full charge the voltage peaks and then drops a bit. These embedded processor controlled hobbyist battery chargers track the slowly increasing voltage of the series string of 6 cells and tries to detect the negative Delta V of each one to determine when to shut down the charging cycle. If the charging continues beyond this point, cell voltage will then increase as the cell tries to dissipate the excess energy as heat. While the highest SOC cells are heating the lower SOC cell are still charging thereby balancing the module a bit. This has to be done very carefully for as Bob Wilson says, "Heat is the enemy".

    see: How to charge Nickel
    Metal Hydride Batteries.


    My SuperMate DC6 also uses a "Pulse Cycle" algorithm (described in the link above) in addition to Delta V that shuts down charging as the cell approaches full charge, measures the voltage at zero current and then reads the voltage increase with a reduced charging rate. This cycle continues for a while before the charging is stopped. It also supports temperature measurement, but this is mainly to handle LiOn batteries safely.

    JeffD
     
    uart likes this.
  5. uart

    uart Senior Member

    Joined:
    Sep 7, 2009
    4,215
    1,200
    0
    Location:
    Australia
    Vehicle:
    2005 Prius
    Thanks Jeff. Yes I already knew what a "negative delta V" was in regard to Ni-Cd/MH charging. I just didn't know if this was what the "high delta V" that the OP was referring to or if it was some other delta V.

    Anyway thanks for the link, it was interesting reading. :)
     
  6. Ed Carmack

    Ed Carmack New Member

    Joined:
    Aug 28, 2013
    25
    4
    0
    Vehicle:
    2005 Prius
    Model:
    Two
    Hey Jeff,
    Thanks for the input. Yes I aware that overheating is bad. What I am doing to minimize heating is to rebalance staggered modules with fans running while running the charge cycle. The adjacent modules should somewhat act as heat sinks. Am doing 1,5,9 and 11 first and then will do 2,6 etc. Have not experienced much heat at least on the surface. Number 9 did get up to around 90F the others around 80 . Ambient temp around 72. I went ahead and restarted 9 as it was not going to be usable with its poor capacity. So not much to lose even if I did toast it. After three cycles it has improved but is still only about 4.5 amp-hr. the others are in the high 5s. One of which did so after only two cycles. #9 has yet to take the full 7.25 amp-hr charge but is getting close. I doubt I will use it even if it does improve to match the others capacity. Sort of using it as an experiment.

    Back to the insulation question. My intent was to not insulate the battery case but to put a layer on top of the cargo deck in plain view. I do see that this would impair HV case cooling as air flow is not impacted and neither is heat transfer through the bottom of the case to the car body. I doubt that is much transfer upward into the cabin as blower air flow is to the top of the modules and then down through/between them and out the bottom duct. Point is to try to keep cabin heat from getting to the battery as the car bakes in sun.

    I understand the importance of having reasonably matched capacities but do not understand the significance of matching resistance other than a high resistance module is indicative of a problem with the module? What I intend to do is to measure the module voltage open circuit and then under a 1 amp discharge and see if any look anomalous. Is this a good approach?
    Thanks.
     
  7. jdenenberg

    jdenenberg EE Professor

    Joined:
    Nov 21, 2005
    3,852
    1,845
    1
    Location:
    Trumbull, CT
    Vehicle:
    2020 Prius
    Model:
    LE AWD-e
    A 1 amp discharge would not show much as a good module has only a few milliohms of resistance. so pull about 10 amps to get a measurable change in voltage (not for very long as the resistor, or better, a light bulb will get hot). The Battery ECU will report 14 module pair resistances if asked by an OBD2 code if you have a Prius aware scanner.

    The reason the module pair resistances should match is:
    1. It indicates that the modules are of the same age and will stay "balanced" longer
    2. Your Prius will pull/charge at up to 100 amps which can cause module pairs with different resistances to momentarily have a voltage difference that will cause the ECU to throw a code.
    Remember to measure the voltage at the module terminals as the wires pulling the load current also have resistance.

    JeffD
     
  8. Ed Carmack

    Ed Carmack New Member

    Joined:
    Aug 28, 2013
    25
    4
    0
    Vehicle:
    2005 Prius
    Model:
    Two
    Thanks Jeff,
    Understand importance of resistance now.

    Looked at the link you provided for charging nimh batteries. Gave me an idea which is generally the first step in my process of running amok. There is a statement in the guide to the effect that nimh cells are ok with a 0.1 C continuous overcharge. So if the charger stops on neg. delta V would reducing the charge rate to 0.5 amps (vs original 5 amp) balance the weak cell while not causing damage to the good ones or is that rate to low to do any good? Your thoughts are appreciated.
     
  9. jdenenberg

    jdenenberg EE Professor

    Joined:
    Nov 21, 2005
    3,852
    1,845
    1
    Location:
    Trumbull, CT
    Vehicle:
    2020 Prius
    Model:
    LE AWD-e
    Since that is what my SuperMate DC6 does (continued charging at a low rate for a while after seeing the peak), it worked for me.

    JeffD
     
  10. Ed Carmack

    Ed Carmack New Member

    Joined:
    Aug 28, 2013
    25
    4
    0
    Vehicle:
    2005 Prius
    Model:
    Two
    thanks Jeff

    You have been a great help. Will try not to ask you any more questions. At least for a day or two.