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

Prius Battery rebalance thread

Discussion in 'Gen 2 Prius Technical Discussion' started by Oreynid, Dec 12, 2013.

  1. JoeX

    JoeX Junior Member

    Joined:
    Aug 26, 2015
    10
    6
    0
    Location:
    PA
    Vehicle:
    2006 Prius
    Model:
    Two
    New here reading all the post, lots of good inf. I'm setting up My HiTechx4 and on the screen SAVE there is a default 3.5V on the left side of the led and the the 7500mAh on the right. What should the V be set to? I see no reference to it in any of the post, or I just don't understand the literature. Thanks for your time.Also I cannot find how to get the charger into auto mode.
     
    #101 JoeX, Aug 26, 2015
    Last edited: Aug 26, 2015
  2. terramir

    terramir Member

    Joined:
    Aug 3, 2015
    305
    47
    2
    Location:
    Los Angeles
    Vehicle:
    2004 Prius
    Model:
    II
    assuming your charger is looking for you to input the rated voltage it would be 1.2V per cell (actually 1.24V but that is getting down to the nitty gritty). so for a 6cell NiMH module like the ones in the prius I prius II prius III etc, you be looking for a voltage of 7.2 volts. this should tell your charger that the V max should be 1.5V or 1.55V per cell depending on the programming and design of the charger.
    As for the 7500mAh chances are that will not fully charge the cell at a coulomb metric efficiency of about 66% overall (the efficiency is much higher between 0 and 70% SOC it,s when it get's toward +80% where it gets really lousy from what I read. so to fully charge a module from a complete useable discharge (to 6V or 5.4V per cell) you would need about 6500/66x 100= 9848mAh. but here is the thing you should start with like 7500 because if your cells are not that good of your capacity might be even lower than 66% of 7500. The hiTech charger is a DeltaV charger so it needs a charge rate of at least 1/2 of the rated C to be able to detect that.
    but yes a minimum setting of 7.2v as for the cut-off capcity you gotta experiment a little
    My $0.02
    terramir
     
  3. JoeX

    JoeX Junior Member

    Joined:
    Aug 26, 2015
    10
    6
    0
    Location:
    PA
    Vehicle:
    2006 Prius
    Model:
    Two
    I just don't want to burn up a few good modules as a learning curve. Thanks
     
  4. terramir

    terramir Member

    Joined:
    Aug 3, 2015
    305
    47
    2
    Location:
    Los Angeles
    Vehicle:
    2004 Prius
    Model:
    II
    as long as you discharge 1st to about 5.4 V first at a high drain rate then a slow one again you could technically pump in 150% of capacity catch is to determine the capacity.
    I just pumped a total of 9200 mAh in a module, in three stages because I had to go somewhere inbetween first 3900 @5A and that took an hour because my charger drops down to 0 often to do some measurement then 2300mAh @ 6A and then 3000mAh @ 6A about 14 hours later, only in the last 1000 mAh did the temp slowly rise from 94F to about 100F since the temp of all the other cells was 91.5F I'd say while it was towards the end of charge it was not full yet.
    if you have an infrared gun or a temperature sensor you should be fairly safe. with 7500 to 8000 but it's doubtful you'll get anywhere nesr full.
    terramir
     
  5. MTL_hihy

    MTL_hihy Active Member

    Joined:
    Sep 2, 2013
    536
    261
    0
    Location:
    Maritimes, Canada
    Vehicle:
    Other Hybrid
    Model:
    N/A

    I think your calculations are a bit off above because using an arbitrary number like 66% can get you into alot of trouble when checking the condition of modules (some may be ALOT less than that in warmer locations). This is not like RC cars where the modules are maxed out for a few cycles and then thrown out. The principle is to use hyper-cycling which extends the life exponentially but for the most part they are only used within the 40-80% SOC window. I would never want to pump 9200 mAh into any of these modules because you could end up damaging even one in perfect condition that way. These modules are very slow to build up heat so relying on temp sensors is also not a good idea because the pressures can also distort the modules significantly (vents don't really release except in a catastrophic failure). My suggestion to someone just getting started is to start low and then increment up on the capacity target (ie if you live in a warm area start lower, colder places start higher but in all cases you never should need to exceed 7500 mAh when testing modules for service).
     
    terramir likes this.
  6. terramir

    terramir Member

    Joined:
    Aug 3, 2015
    305
    47
    2
    Location:
    Los Angeles
    Vehicle:
    2004 Prius
    Model:
    II
    66% is the charging efficiency in general for all NiMH charging types, i.e you need to put 150% mAh in them to reach 100% SOC.
    But yes you never know what the SOC is your right. But you just gave me a genius idea TY TY on how to determine actual capacity of the cells without ever needing to charge them to 100%
    the one thing odd about the NiMH chemistry is that to about 70% SOC the coulomb metric efficiency is very close to 100% and then drops rapidly the closer you get towards 100% SOC the worse the efficiency is.
    so in order to determine a cells SOC mind you I said near 100% I should drain a module to 5.4V and then charge it let's say with 4000mAh if at the next drain cycle I get close to 4000 out and in the next module I do 4500 (assuming most modules will be close together) I will be able to estimate 70% soc i.e also estimate an increase or decrease in capacity without frying the modules. I hope I didn't permanently fry module 1, well if I did replacement will be the cost of learning.
    Thanks for your input
    You might have just saved me from a lot more costly mistakes
    terramir
    PS: Because I discovered that the modules do not get hot on top, they get hot where they are compressed together, so the infrared thermometer seems to be pretty useless.
     
    Doug T and gdanner like this.
  7. MTL_hihy

    MTL_hihy Active Member

    Joined:
    Sep 2, 2013
    536
    261
    0
    Location:
    Maritimes, Canada
    Vehicle:
    Other Hybrid
    Model:
    N/A
    Yes, you are spot on because there are 2 issues at play. First is coulomb metric efficiency (ie how much mAh you put in vs what you get back out) and the second is the current capacity of the battery (since it degrades over time/use). Due to this it's much better to ramp up capacity slowly and then you can see what is the actual capacity of your modules at 100% SOC (really you don't need to find maximum SOC, just relative SOC to match modules) without overshooting too much and thereby cause damage from excessive heat/pressure. The Hitec X4AC charger I used was pretty good at finding this point so I didn't need to go much further in cycling since you don't want to cycle too much at 100% SOC because it shortens module life (ie opposite of hyper-cycling).

    Also the not sure about the 70% number in practice, seems to me I was inputting 7200 and getting back over 6000 (may have been below 100% SOC but that's fine, again it's relative characteristics that are important) but this is why it's important to take it slow and experiment with modules outside the pack before you jump in with both feet on the real deal. Like you I also tried an IR gun and temp sensors but gave up early because it's just not useful in this application. Don't worry about frying a module here or there (I did one in myself in the beginning), the key is to get the methodology down so that when you go through the pack you are able to complete testing without doing any damage to your modules you want to use in the rebuild. Also don't forget load testing your modules, since that is arguably just as (maybe more) important as capacity testing when doing the rebuilds.
     
    #107 MTL_hihy, Aug 28, 2015
    Last edited: Aug 28, 2015
  8. kiwi

    kiwi Member

    Joined:
    Jul 12, 2012
    245
    85
    0
    Location:
    New Zealand
    Vehicle:
    2012 Prius c
    Model:
    N/A
    Folks: majority of RC chargers you are using in your experiments are designed to be used with the Linear Temperature Sensor - which is easy to source and attach to your devices.
    I can go into more details if the need be - please do not hesitate to ask.
    Please support me here with your Votes - competition for me to win Tesla Powerwall with Solar closes midnight Sunday 30 Aug New Zealand time (UTC+12Hours)
    Future of Energy | Yury Botkov
    Vote only counts when you click the link in the email coming from Vector energy - our local electricity company.
     
  9. terramir

    terramir Member

    Joined:
    Aug 3, 2015
    305
    47
    2
    Location:
    Los Angeles
    Vehicle:
    2004 Prius
    Model:
    II
    you olks discharge the cells to about 6V from what I hear in this thread, now reality is that does not do by far as much good than a deep discharge will do discharging to 6V is just called cycling really, it supposedly takes a discharge to 0.4V per cell which would be 2.4V in our case to really restore the surface, and from what I read that's Mil-spec, Anyway besides self discharge to sort of get there safely, like maybe a 50mA discharge with a resistor or something?
    terramir
     
  10. kiwi

    kiwi Member

    Joined:
    Jul 12, 2012
    245
    85
    0
    Location:
    New Zealand
    Vehicle:
    2012 Prius c
    Model:
    N/A
    Point to the source please. What Mil specs? Where else apart from Toyota / Lexus hybrids have you seen the EV NP1/2 Modules being used so that any "Mil specs" would apply? Please advise.
     
  11. MTL_hihy

    MTL_hihy Active Member

    Joined:
    Sep 2, 2013
    536
    261
    0
    Location:
    Maritimes, Canada
    Vehicle:
    Other Hybrid
    Model:
    N/A
    Are you maybe getting Nicd and NiMH cells confused with that Mil spec?

    Use the charger to both charge and discharge your batteries so you can at least measure relative input/output of the battery. There's a difference between rebuilding battery packs and experimenting so the numbers we typically quote for cell discharge are what you can repeatably use in order to ensure you don't damage anything during your rebuild. If you want to experiment on how far you can draw them down then by all means try it and let us know how you make out.
     
  12. gdanner

    gdanner Member

    Joined:
    Jul 29, 2007
    86
    39
    0
    Location:
    Central Illinois
    Vehicle:
    2005 Prius
    Model:
    N/A
    For many many years the US military extensively used NiCd cells in aircraft. One of the military standards called for putting a shorting wire across individual NiCd cells prior to storing them for long periods. Of course the cell needed to be deeply discharged first before putting on the shorting jumper.

    I've read reports that "deep discharging" of NiCd cells actually did reduce the size of the individual metallic crystals in the electrodes (cathode, anode, or maybe both?). Reducing the size of the metallic crystals increased the mAh capacity of the cell and reduced its series resistance. Deep discharging was also said to reduce the tendency for large metallic crystals to penetrate the insulating separator layer between anode and cathode. If a metallic crystal penetrates the separator on a NiCd cell it permanently destroys the cell: The cell becomes a 0.00 ohm short circuit. So it's a good thing to avoid letting the metallic crystals get too big.

    But I haven't been able to find any credible research papers or other documents stating that NiMH cells actually develop "large metallic crystals." Perhaps the differences in physical and chemical structure between NiCd and NiMH are large enough to prevent this phenomena?

    To me it is significant that most of the cells in our Prius HV battery packs are surviving >10 years. How much does limiting the minimum SOC to 40% and the maximum SOC to 70% actually lengthen the life of the cells? In contrast small consumer type NiMH cells rarely last longer than 3-4 years. Consumer NiMH packs are frequently subjected to polarity reversal and also to overcharging. Maybe that's what prevents them from lasting as long as Prius HV modules?

    I think it's time for us to learn more about the chemistry of NiMH cells.
     
    strawbrad and kiwi like this.
  13. jdenenberg

    jdenenberg EE Professor

    Joined:
    Nov 21, 2005
    3,831
    1,824
    1
    Location:
    Trumbull, CT
    Vehicle:
    2020 Prius
    Model:
    LE AWD-e
    To me it is obvious that limiting the minimum SOC to 40% is good insurance against reversing a cell. limiting the max SOC to 70% (? I thought it was 80%) avoids the increased heat generated due to a lower charging efficiency as you approach 100% SOC and heat is the enemy of battery life.

    JeffD
     
    usnavystgc likes this.
  14. kiwi

    kiwi Member

    Joined:
    Jul 12, 2012
    245
    85
    0
    Location:
    New Zealand
    Vehicle:
    2012 Prius c
    Model:
    N/A
    Claims from manufacturer state that NP1 & NP2 modules do not have memory effect, hence full discharge which was practiced in NiCd is not required.
    The answer to that "how" question is quite simple:
    If one have a look at the real charge / discharge graphs of all modules captured simultaneously under load - they will notice that BMS is making the battery operating in the shallow range where the performance of aged modules is similar.
    Even if there are some degraded modules - they will still operate "in cahoots". The reason being: the capacity used (or in other terms - the energy drained) during those bursts is much less than design capacity.
    e.g. to start your Prius - less than 1AH is required. And when that 1AH is drained - the discharge graphs I've observed were the same with Delta V <0.5V on pairs. Voltage did not drop below 6.87V/module
    With time when the URC becomes less and less - the pack will still perform up till the point when URC drops below what Prius will demand.
    Don't be confused to assume that if individual module has remaining capacity of 4AH - it is good enough. The pack assembled from modules with different RC would have URC much less than individual RC.
    e.g. pack with 4.0 AH URC was operating with no prob. Another pack with 2.7AH URC is still performing for 6 month from being installed (do not know the ODO though).
    I have right now another rebuild pack with 2.7AH URC. Would be interesting to see how long that would last.
     
  15. usnavystgc

    usnavystgc Die Hard DIYer and Ebike enthusiast.

    Joined:
    Jul 29, 2011
    3,159
    988
    0
    Location:
    Tucson, AZ
    Vehicle:
    2005 Prius
    Model:
    Four
    Future of Energy | Yury Botkov Vote only counts when you click the link in the email coming from Vector energy - our local electricity company.

    Please vote for me. :)
     
  16. MTL_hihy

    MTL_hihy Active Member

    Joined:
    Sep 2, 2013
    536
    261
    0
    Location:
    Maritimes, Canada
    Vehicle:
    Other Hybrid
    Model:
    N/A
    X2

    The enemy of these modules is usually dehydration not crystal growth as is the primary issue with NiCd batteries.
    I wouldn't go deep cycling them myself except if it was an experiment designed to test module limits (ie won't help life).
     
  17. kiwi

    kiwi Member

    Joined:
    Jul 12, 2012
    245
    85
    0
    Location:
    New Zealand
    Vehicle:
    2012 Prius c
    Model:
    N/A
    That is in line with the manufactuer patented technology of refilling. They also suggested in another patent discharge and idling after discharge but were not specific about level of discharge.
     
  18. MTL_hihy

    MTL_hihy Active Member

    Joined:
    Sep 2, 2013
    536
    261
    0
    Location:
    Maritimes, Canada
    Vehicle:
    Other Hybrid
    Model:
    N/A

    Can you link to those patents kiwi?

    I would be very interested in reading that material to see if we could find anything to apply in a DIY setting.
     
    #118 MTL_hihy, Sep 4, 2015
    Last edited: Sep 6, 2015
  19. terramir

    terramir Member

    Joined:
    Aug 3, 2015
    305
    47
    2
    Location:
    Los Angeles
    Vehicle:
    2004 Prius
    Model:
    II
    dehydration can be a serious issue, but that nimh cells have no memory effect G_D I hate that term, it's is a voltage depression in load based on reduction in surface area. Also all nickel based chemistries suffer from this load voltage depression, which also reduces capacity.
    As for mil-spec (US military specification) I read this on one of the battery sites I will look for it when I am not at work.
    Also limiting the charge cycle from 82-38% will increase the life of any rechargeable, Li chemistries included, but on occasion a deep discharge in nickel chemistries can help tremendously. The example with regular AA NiMH cells is mostly based on dumb chargers and frequent deep discharge, they would last a lot longer as well if the charge level was kept between 82-38% but regrettably they get discharged fully too frequently and charged over full too frequently.
    terramir
    PS: Ni-CD cells of course suffer from it way worse.
     
    nicholas_k likes this.
  20. Warmonger

    Warmonger New Member

    Joined:
    Aug 20, 2015
    2
    0
    0
    Location:
    Atlanta
    Vehicle:
    Other Non-Hybrid
    Model:
    N/A
    All

    This may not be the correct place for this post, if so i'll repost elsewhere. Does anyone have a used Hitec X4 eighty they want to sell?
    Russ
    russnagel1@yahoo