YAPiP - recreating pEEf's approach

Discussion in 'Prius PHEV Plug-In Modifications' started by jdh2550, May 23, 2012.

  1. Flaninacupboard

    Flaninacupboard Senior Member

    Joined:
    Apr 25, 2010
    1,297
    212
    0
    Location:
    Midlands - UK
    Vehicle:
    2010 Prius
    Model:
    N/A
    Maybe it just means upping the cell count a bit, but personally I believe this indicates the cells are struggling, and would definitely not push them to 2C, never mind 2C+
     
  2. ericbecky

    ericbecky Hybrid Battery Hero

    Joined:
    Mar 12, 2004
    4,357
    3,161
    1
    Location:
    Madison, Wisconsin
    Vehicle:
    2005 Prius
    Model:
    Two
    Agreed. Up the cell count a bit and things should be fine.
     
  3. ericbecky

    ericbecky Hybrid Battery Hero

    Joined:
    Mar 12, 2004
    4,357
    3,161
    1
    Location:
    Madison, Wisconsin
    Vehicle:
    2005 Prius
    Model:
    Two
    Wait, are you buying the 80ah realforce prebuilts packs from Enginer? Or are you making your own?
     
  4. jdh2550

    jdh2550 Co-Founder, Current Motor Company

    Joined:
    Nov 16, 2010
    317
    246
    0
    Location:
    Ann Arbor
    Vehicle:
    2005 Prius
    Model:
    N/A
    I'm buying re-configured packs from Enginer. Price of $900 for 24V80AH plus $100 for reconfiguration to 48V40Ah. I'm buying 4.5 packs. The half pack will be in the normal size container but only containg 24V40Ah cells and extra space filled with a suitable inert filler.

    Bear in mind that this is for an assembled, tested pack with cell voltage leads and thermal probe housed in a battery box. How much would you charge for a similar assembly with the cells that you recommend?

    I'm committed to this course (the check is in the mail) - but also I still think it's the right course for my particular needs and wants. I'll just have to wait and see with regard to voltage sag - but, as you say, there's a relatively straightforward fix for that (more cells).

    BTW - just to make sure we're comparing apples-to-apples the pack is made up of these cells: F20-08180225. Is that the same cell as in your packs?
     
  5. ericbecky

    ericbecky Hybrid Battery Hero

    Joined:
    Mar 12, 2004
    4,357
    3,161
    1
    Location:
    Madison, Wisconsin
    Vehicle:
    2005 Prius
    Model:
    Two
    I currently only sell loose cells and not assembled packs. I have 24v80ah packs from Enginer that have the 3-bolt one through hole internal connection design. A couple years old, so I don't recall the exact cells.

    Since you were choosing RealForce cells I was going to suggest that you have Enginer obtain/create what you need, but I see that is the route you have taken.

    I am certain that Enginer will get what you need. Jack and his crew have been good business partners with my company for some time.
     
  6. Flaninacupboard

    Flaninacupboard Senior Member

    Joined:
    Apr 25, 2010
    1,297
    212
    0
    Location:
    Midlands - UK
    Vehicle:
    2010 Prius
    Model:
    N/A
    Yes

    IMG-20120705-00117.jpg

    This cell was overcharged and as you can see has failed.

    72*3.35V will give you a resting voltage of 241.2 which looks good, but if you load them hard and they drop to 2.9V you'll get 208.8V which is a little lower than ideal for high speed EV.

    Having the taps for cell voltages already in place is useful, i'd reccomend 4x16 channel slave boards and 1x8 channel master board of miniBMS from Cleanpowerauto. OK it's $912 for that, but the Enginer solution (the old BMS16D) is what caused the cell you see above.
     
    jaso likes this.
  7. jdh2550

    jdh2550 Co-Founder, Current Motor Company

    Joined:
    Nov 16, 2010
    317
    246
    0
    Location:
    Ann Arbor
    Vehicle:
    2005 Prius
    Model:
    N/A
    It might be an interesting process to figure out what you'd need to charge for an equivalent, assembled pack. Or not ;) I'm not trying to be a PITA - just trying to get a feel for the value of assembly, packaging and testing for the different cell solutions.

    Yes, I'm enjoying working with Jack - he's been very responsive to all my questions.

    I take full responsibility for how I decide to use the cells and how well they perform. If I abuse them it won't be Enginer's fault :)
     
  8. Flying White Dutchman

    Flying White Dutchman Senior Member

    Joined:
    Dec 29, 2007
    4,374
    305
    0
    Vehicle:
    Other Non-Hybrid
    Model:
    N/A
    Thats not a a123 to the looks of it
    I use 70 cells just like plainitaire lopez and mr plugin
    This seems to work best


    -Htc Tapatalk ( sorry for auto spell correct )
     
  9. Flaninacupboard

    Flaninacupboard Senior Member

    Joined:
    Apr 25, 2010
    1,297
    212
    0
    Location:
    Midlands - UK
    Vehicle:
    2010 Prius
    Model:
    N/A
    A123 have a higher resting and working voltage though, and jdh2550 is replacing the BECU so can be more flexible :)
     
  10. jdh2550

    jdh2550 Co-Founder, Current Motor Company

    Joined:
    Nov 16, 2010
    317
    246
    0
    Location:
    Ann Arbor
    Vehicle:
    2005 Prius
    Model:
    N/A
    What a horrible picture!! :eek:

    Hmm, that is worse performance than I'm expecting - so I guess I'll need to budget for some more cells (and more BMS, and more charger). I've attached the discharge curve from the RF spec sheet. It shows a 1C nominal of ~3.15V and a 3C nominal of ~3.0V (the big ugly lines are ones I added to determine the approximate nominal).

    rfe-discharge.PNG

    Oh well, it won't be the first time a spec sheet has been "optimistic"...

    (but on the flip side Jack says the real capacity is more like 26Ah rather than the advertised 20Ah).

    I did look at the miniBMS - but I'm planning on using Current Motor's BMS.
     
  11. ericbecky

    ericbecky Hybrid Battery Hero

    Joined:
    Mar 12, 2004
    4,357
    3,161
    1
    Location:
    Madison, Wisconsin
    Vehicle:
    2005 Prius
    Model:
    Two
    I would be very interested to see some quantifiable tests of individual realfoce cells. I don't know of anyone doing that.
    Perhaps I should send a few to my colleague who has been doing some testing on our A123 cells. Then we'd have the same tests being performed on the RF as the A123's.
     
    jdh2550 likes this.
  12. jdh2550

    jdh2550 Co-Founder, Current Motor Company

    Joined:
    Nov 16, 2010
    317
    246
    0
    Location:
    Ann Arbor
    Vehicle:
    2005 Prius
    Model:
    N/A
    I think that would be an excellent idea. More data is always good. And cells tested by the same people using the exact same equipment and test methodology would be fantastic.

    Edit: BTW, I don't doubt that the A123 cells are better in a lot of aspects than the RFE cells (the only non-contended attribute is that the A123 cells have less capacity than the RFE - although usable capacity may be open for debate). It's a question of cost/value analysis - do I need the better cells or are the cells I've chosen adequate for my usage? Data will help to quantify the old adage "you get what you pay for".
     
  13. ericbecky

    ericbecky Hybrid Battery Hero

    Joined:
    Mar 12, 2004
    4,357
    3,161
    1
    Location:
    Madison, Wisconsin
    Vehicle:
    2005 Prius
    Model:
    Two
    Actually I realize it won't be possible to do it exactly the same.
    Why? The A123 cells have a much higher C rating (10+), and that is how we tested them.

    The first test we did was at 68 amp load. ~3.5C
    Second test we was a 103 amp load. ~5c (Cells lasted for ~15 minutes)

    In both cases the cells performed amazingly with no droop in voltage.

    If the RealForce cells are only rated for 2c, we can even begin to do the basic tests without exceeding their C rating.

    Dang.
     
  14. jdh2550

    jdh2550 Co-Founder, Current Motor Company

    Joined:
    Nov 16, 2010
    317
    246
    0
    Location:
    Ann Arbor
    Vehicle:
    2005 Prius
    Model:
    N/A
    Yeah, that's a shame. Is the 10+C rating a continuous rating? I couldn't find a clearly spelled out continuous vs. peak rating for the A123s (but I didn't look very hard!).

    However, your results highlight exactly the sort of thing that I'm talking about. I don't need 5C, I don't even need 3.5C continuous. In an ideal world I think I'd need a 3C continuous rating for sustained high speeds and only a 3C peak for suburban speeds. So, why spend money to achieve something I don't need? That's not a facetious question because there's certainly truth in the fact that if I use a 10C capable cell at only 30% of it's capability I'll get more life out of it. But that amount of life isn't quantifiable from the data you give and, with the greatest of respect, unless you're buying A123 cells from A123 then all bets are off in regards to them matching the spec sheet so even if it says it in A123 literature it needs to be seen with the cells you're actually sourcing.

    BTW, I'm trying to separate the two questions: (1) Are A123's more than I need? & (2) Are RFE's going to be adequate or will they turn out to be a false economy? The questions are certainly related because it might be that if RFE's are a false economy then A123s and other more expensive options are back in contention.

    The question I'm not trying to answer is "Are A123's better than RFE's?" That answer is simple: "Yes".

    So, I think a useful test would be a 2C and a 3C test of both cells measuring voltage sag and temperature rise for those tests. These rates are useful for me because I'm planning a 40Ah pack. If you're planning a 20Ah pack then the higher C rates that you use are much more appropriate. For my pack 2C = 80A draw and 3C = 120A draw and my guesstimate (and supported by "the flan man" and pEEf) is needing slightly north of 100A for sustained high speed - of course the problem is compounded by sag because at a lower voltage I'll need higher current to give the same power.

    If you'll sell me a single A123 cell and I can find a convenient source for single RFE cells (Jack only wants to sell packs - which makes sense) then I'll do that testing...

    This is interesting conversation and I'm enjoying it. But, the check is in the mail, yaPIP will be starting life with a 72 cell RFE setup. What it evolves into is anyone's guess! :)

    Peace, love & plug-ins (tm). ;)
     
  15. ericbecky

    ericbecky Hybrid Battery Hero

    Joined:
    Mar 12, 2004
    4,357
    3,161
    1
    Location:
    Madison, Wisconsin
    Vehicle:
    2005 Prius
    Model:
    Two
    I may be able to link you up with both cells.
    Or if I get time I may simply do the tests myself. I have a Prolab8 battery tester at my shop. 40amp max load.
    I'll be seeing Jack at Green Drive Expo in a couple weeks.

    I look forward to watching your progress. You are right about not buying more than you need. And having an open source project will help get this out to as many people as possible.
     
  16. jdh2550

    jdh2550 Co-Founder, Current Motor Company

    Joined:
    Nov 16, 2010
    317
    246
    0
    Location:
    Ann Arbor
    Vehicle:
    2005 Prius
    Model:
    N/A
    Thanks and good luck with Green Drive Expo.
     
  17. lopezjm2001

    lopezjm2001 Senior Member

    Joined:
    Apr 14, 2009
    1,146
    403
    5
    Location:
    Sydney Australia
    Vehicle:
    Other Electric Vehicle
    Model:
    N/A
    I guess the real world results do not reflect the RFE data sheet. According to the data sheet the RFE cell will sag to about 3.0v at 5C or 100 amps. So with 70 cells it will sag to 210v. To make up for the sag you will need to add more cells in series. Say 76 cells will give you 228v which is a lot better. The RFE packs have a history of having been overcharged due to a dodgy BMS16D which would have effectively increased the internal resistance and hence the extra sag.
     
  18. lopezjm2001

    lopezjm2001 Senior Member

    Joined:
    Apr 14, 2009
    1,146
    403
    5
    Location:
    Sydney Australia
    Vehicle:
    Other Electric Vehicle
    Model:
    N/A
    According to the EBay seller Xueming08 who is selling the A123 cells on eBay he quotes them on having a continuous rating of 3C and a 10C peak for 10 seconds.
     
  19. Flaninacupboard

    Flaninacupboard Senior Member

    Joined:
    Apr 25, 2010
    1,297
    212
    0
    Location:
    Midlands - UK
    Vehicle:
    2010 Prius
    Model:
    N/A
    Yeah it's interesting weighing up all the pros and cons, and with the packaging taken care of I can see the attraction of the RFE cells - I would probably find a headway pack assembly more fun than the coding you're doing, but to each their own! :)

    At least you're going in with some information and field experience, forewarned is forearmed. Something I'd reccomend you do is start your software with a 40A CDL, and not increase this without being able to monitor the pack closely (i.e. have someone else driving, increase the CDL and observe the sag, and back it off again if it's excessive). We also haven't touched on CCL too much, but from memory the RFE cells only permit charging at 0.2C, so your CCL must be limited to 8A. Again, this is something an A123 system would be a lot more forgiving with.
     
  20. jdh2550

    jdh2550 Co-Founder, Current Motor Company

    Joined:
    Nov 16, 2010
    317
    246
    0
    Location:
    Ann Arbor
    Vehicle:
    2005 Prius
    Model:
    N/A
    Flaninacupboard : You don't like coding? Oh dear, poor misdirected man ... ;) (yup to each their own!)

    Good point with incrementing CDL. I do plan on logging and real-time visualization - both CAN data and raw data from the sensors. So I should be well setup for seeing the effects of my changes. The RFE data sheet lists charging rates of 0.5C (Standard) and 1C (quick). But I guess if we can't believe the discharge info then we should be leery of the charging info as well.

    BTW, my BMS solution can easily (and cost effectively) add another 6 cells (we sell a 24 cell and a 30 cell bike - the 30 cell bike uses the 24 cell BMS plus a 6 cell expander). I'll also make sure that I buy a charger with enough headroom - so adding cells shouldn't be a problem. Although I will need to source them - but I'm sure Jack or Eric can help me out there.

    BTW - a very important question: What's with the name flan-in-a-cupboard? There's got to be a story there!