A123 LiIon Battery Pseudo Extra

Every so often I have to slap myself in the head. I’m surprised that we haven’t covered these things by now. DeWalt’s been selling a LiIon 36 volt battery pack that’s full of the latest A123 cells. These are the same ones that were used in the Killacycle. (I think they’ve got a new batch of cells now).

A while back, [Jeff] sent in a circuit for using multiple packs, leaving the internal BMS in place. [The link is fixed now]

[Robert] sent in a scooter that’s been designed to run these same cells. The custom fabrication and machine work looks fantastic.

18 thoughts on “A123 LiIon Battery Pseudo Extra

  1. So… Okay. They are Li-Ion cells… Still, the chargers for those are redonkulously expensive.

    Anybody got a circuit for charging them outside of the one that comes in the battery box?

  2. Yea if you actually read the stuff that was linked, you would see that there infact is.

    Actually you didn’t even need to read the page that was linked, it says RIGHT IN THE POST!

    “A while back, [Jeff] sent in a circuit for using multiple packs, leaving the internal BMS in place.”

    And it even links to it!

    Imagine that, read a bit, and find answers, good lord!

  3. Okay, the A123 cells are some very nice engineering. I’ve looked over the linked sites, but either I missed the physical size of the cells or it’s not on there.

    I’m curious if I can rebuild my T42 ThinkPad battery pack with these cells. Are they the same (or a compatible) size as “standard” lithium-ion cells? Would the charge circuit in the ThinkPad be able to properly handle the difference? Is there anyone else out there doing this?

    (As an aside… http://www.philpem.me.uk/elec/welder/ has a much better solution for connecting the batteries. Soldering might get a bit too much heat into the cell; a quick, concentrated spot weld is what is apparently preferred by the manufacturers — and you don’t risk plugging the vent holes.)

  4. The nominal voltage of the LiFePO4 chemistry is lower than other Li-ion chemistries, so it is not a good direct replacement for li-ion cells in laptop batteries. they also have a slightly lower energy density than the li-ion cells.

    the lifepo4 cells have some wonderful benefits however: they are much safer (won’t burst into flame on overvolting, undervolting, or puncturing); can have a ~2,000 cycle life; they don’t degrade over 2 years (a lithium battery with shelf life?); and the dewalt cells have a wonderful power density (100A out of a 2.4Ah cell!).

    A number of folks have taken to tearing these packs apart for use in e-bike and e-scooter applications. check out the endless sphere forums if you are interested in light ev’s.

    The remote-controlled aircraft hobbyists have also taken to the cells. They appreciate the fast charging capability, and especially like the fact these batteries will not burst into flame on a hard landing. I guess they like salvaging their wrecks and rebuilding them rather than watching the fire that would often follow a real aircraft crash.

  5. The schematic describes how to combine these packs in series or parallel, without opening them up. Charger is not modified either. The speed control of the Dewalt drill was reverse-engineered to use the pack’s switched negative output lead. Battery maintenance system(BMS) is used during discharge to safely prevent over discharge and cell reverse bias. This was designed to drive a bike to moped upgrade described in detail at http://www.neodymics.com

  6. Thanks jeff, i have seen the “rural roads” review of your neodymics prototype and it is quite promising.

    I noticed you were providing the bms with a voltage on the speed sensors to signal its maximum output. What is the allowable output current through each bms based on the setup you diagrammed? It looks like you are limiting the overall current with a 25A fuse.

    Who is the hub-motor and controller provider for your prototype? Or is the controller custom designed by you?

  7. Hi Mike,

    Thanks for your feedback.

    I don’t know what the current limit of the series parallel setup is, except that it is more than the Crystalyte 406/409 hub motor requires. Motor controller also came from Crystalyte.

    I routinely exceed 22 A total current for tens of seconds, and 16 A continuously. I believe that each pack is capable of providing at least 15 A, since that is the size fuse in the unswitched lead. There is a nice description of the DC9360 dissected by ruddman430 on this thread: http://www.rcgroups.com/forums/showthread.php?t=587041&highlight=a123+dewalt+circuit+fuse

  8. I can’t agree with the opinion that those spot welds are any less electrically and mechanically secure, than using solder as conductive “glue” to attach conductors the the cell cases. Whatever happened to; an electrical connection should be both electrically AND mechanically secure BEFORE solder is applied?

  9. Ok, I was wondering if you were able to pull more from the pack than the 15a unswitched lead. looks like you are keeping things down in the 10-15a range anyway.

    Does the bms shut off a pack as it runs down to low voltage or an individual cell voltage drops below 2.0v? other e-bike users have been tapping the unswitched / 15a fused negative terminal, which occasionally leads to an overdrawn pack if they get careless and run too long. If it is a true BMS that would be great. I haven’t seen too much indicating the functions the dewalts circuitry is performing. so your experience with the system is a great resource.

  10. The BMS has leads to every cell, so i am assuming it monitors each cell during discharge. That would give greatest capacity. The more cells that are monitored, the lower the cutoff voltage. Presumably the A123 cells do not like to see a reverse bias any more than the other chemistries. There is a nice summary of the BMS issues involved in building a high voltage rechargeable pack here:


    (Click on “more engineering resources,” then “special design rules for large battery packs.”)

  11. yup, i believe the a123’s short closed when discharged to beyond 0v, and the precipitous cell voltage dropoff at end of discharge means you don’t want to bring a cell below 2.0v. at least with these lifepo4 cells you don’t have to worry about an explosion if you do, just a dead cell.

    i was wondering if you had tried a full discharge untill the bms stepped in and cut off the pack, and then follow up with a measurement of the unswitched pack voltage and maybe cell voltages to verify the bms was doing its job. with the info from your schematic i’ll forward on to other e-bike enthusiasts (many of whom are also using a crystalite motors) and see if they can verify the bms will monitor cell voltages and cutoff at the appropriate time.

    when do you expect to start selling your kits commercially? i saw on your website you are still in a preproduction / market research phase. Is the prototype demonstrated to the “rural roads” guy pretty close to what you are going to go into production with?

  12. Weakest pack terminal voltage was 30.4. I don’t have a way to measure individual cell voltage without opening a pack up.

    We are seeking investment and marketing help to bring the Cyclemotor to the masses. We will probably make some cosmetic changes to the prototype tested by Roger and release it as soon as possible.

  13. some users over at endless sphere took the schematic jeff produced for a single pack, and played with it to see how many amps they could pull from a pack without the bms intervening, and whether the bms would cut power when the pack reached a low voltage condition. By adjusting the resistor values in the neodymics schematic doctorbass was able to pull 19a from a single dewalt 36v battery pack. the bms stepped in and cut power as the pack dipped below 26v. so they were able to confirm the bms provides a low voltage cutoff function. the dewalt 36v charger will not charge packs at less than 26v, so the bms keeps the packs above that value.

    discussion and test results are here:


    thanks to jeff ratke for the great information. the dewalt 36v packs are becoming an interesting choice of power supply for electric bicycles.

  14. I am using 2 x dewalt 36v without the bms as unfortunately i blew the fuse in them. In series on a 408 Crystalyte . The performance is amazing. 45kph for 10 kms drops voltage from 67v to 62v but am not sure how long they will last without the BMS.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.