0
\$\begingroup\$

I've been using a 36V motor that draws ~13A current at peak. I initally used a set of three lead-acid (12V 12A) batteries in series to power the motor. They had sufficient power to easily take a payload of more than 60kg and output ~35km/h at full speed. I have now replaced them with a set of Li-ion batteries (10 x 12.8Ah + 10x 12.8Ah = ~26Ah joined using short copper wire strips ~1cm) however, I am worse off than the lead-acid batteries. They can't even provide enough power. Without the load, battery voltage stays constant at ~4.2 but when I put load on it, battery instantly drops below 3.8 and is unable to provide neither speed nor power.

Can anybody tell what is going on and how can I get the power from the batteries?

Thanks,

enter image description here enter image description here

\$\endgroup\$
6
  • \$\begingroup\$ What are the new batteries rated at, for discharge? Might you have bought cheap, useless, batteries? part number or photo would help. \$\endgroup\$
    – Neil_UK
    Nov 30, 2019 at 8:20
  • \$\begingroup\$ How are you connecting the Li-ion cells in order to get about the same voltage as the two lead-acid batteries in series? \$\endgroup\$
    – joribama
    Nov 30, 2019 at 8:20
  • \$\begingroup\$ I am connecting 10 batteries in series with each battery facing opposite to each other while being side by side with the other. I use copper wire strands cut from a 1.5mm2 silver coated copper wire. Then I charge 2 sets of 10 batteries separately and combine them in parallel when I feed them to motor controller. \$\endgroup\$
    – azad.wolf
    Nov 30, 2019 at 8:30
  • 1
    \$\begingroup\$ 13 Ah is a measure of capacity, not current which is measured in amperes (A). \$\endgroup\$
    – Transistor
    Nov 30, 2019 at 8:39
  • 1
    \$\begingroup\$ When you have 10 cells in series you may have problems if any of the cells is not properly charged or is defective. Have you measured the voltage across each of the cells to make sure this is not the case? \$\endgroup\$
    – joribama
    Nov 30, 2019 at 9:04

1 Answer 1

7
\$\begingroup\$

I can see a few issues:

  1. The UltraFire batteries shown in your photo (and in your calculations) are fake. It is impossible to have 12,800 mAh capacity 26650 batteries. This is similar to my answer here about 18650 batteries. The largest capacity 26650 batteries currently available are less than 6000 mAh.

    For example:

    UltraFire BRC26650 7800mAh (Red) - test by lygte-info.dk
    Claimed capacity 7600 mAh, measured capacity around 2000 mAh, but unable to sustain fast discharge and mismatched capacity between batteries.

    UltraFire BRC26650 7300mAh (Gray) - test by lygte-info.dk
    Claimed capacity 7300 mAh, measured capacity around 2500 mAh, and mismatched capacity between batteries.

    Various larger Li-ion battery capacity test results, including 26650 - lygte-info.dk
    Note that not all batteries on this page are 26650, some are 21700 etc. Search for 26650 on the page, if that is the size you are interested in.

    Batteries with ridiculous ratings (actually comparisons of good & bad batteries) - lygte-info.dk
    You can see some batteries with clearly fake ratings, as well as capacity vs. discharge rate available from different genuine models.

  2. Not all Li-ion / LiPo batteries are equal in their ability to deliver all the energy inside. Currently, with 18650 / 26650 batteries, there are effectively "medium capacity, high discharge rate" battery models, and "high capacity, lower discharge rate" battery models.

    If you get the highest capacity batteries, you are unlikely to be able to discharge them as quickly as some lower-capacity batteries. This is not my specialist area, but from the research I did a while ago, there appears to be a trade-off between the battery manufacturing processes for these two features - capacity and power delivery (effectively the internal resistance). This is a topic frequently discussed in flashlight and RC model forums.

    You need to investigate the "C" discharge rating of whichever Li-ion / LiPo batteries you are considering, and read their datasheets very carefully, to understand how quickly you are allowed to discharge them (safely). Just looking at their capacity (in this case, the fake 12,800 mAh) does not tell you how quickly you can discharge them, nor does it tell you their internal resistance (which is related), even if the capacity value was true (which it isn't).

  3. Building your own battery packs, without any form of BMS or other protection (as shown in your photo) is high-risk, due to the lack of balancing or voltage control for the individual cells. That is a whole topic on its own.

(P.S. Soldering directly to 18650 / 26650 batteries does them no good at all, and can easily damage the internal plastic separator sheets, leading to reduced capacity or even overheating, fire and venting. Can people get away with doing that sometimes? Yes. Is there a risk in doing that? Yes, you will see warnings on 18650 / 26650 battery datasheets not to do it. There are good reasons why professional battery pack manufacturers don't do that.)

\$\endgroup\$
5
  • \$\begingroup\$ Thanks! If it is about discharge rate then it seems it is quite high as in just about few minutes of use, the voltage has dropped from 42 to 39 for both sides. So, how can I know the "true" current carrying capacity of these batteries? (I can't seem to find exact batteries listed in those lists that you mentioned) \$\endgroup\$
    – azad.wolf
    Nov 30, 2019 at 10:20
  • \$\begingroup\$ @azad.parinda - "how can I know the "true" current carrying capacity of these batteries?" Ideally, you should read their datasheet. But I bet you don't have one, or even if you do, then it's garbage if it claims the capacity is 12800 mAh. So you have a problem - you don't really know the answer. If you have a constant-current load which you can control, and which can simulate your load (13A?) then there are tests you can do - but those batteries may be missing safety features (e.g. vents) too! Also, there is no reason to think the individual batteries have matched capacities either. \$\endgroup\$
    – SamGibson
    Nov 30, 2019 at 10:30
  • 1
    \$\begingroup\$ @azad.parinda - "I can't seem to find exact batteries listed in those lists that you mentioned" Correct, your specific fake batteries are not on the web pages I quoted. No-one has the time to test every fake battery being sold. I linked to those pages to show you that the claimed 12,800 mAh capacity for 26650 batteries is clearly fake, and to show the results for other fake "UltraFire" branded 26650 batteries. \$\endgroup\$
    – SamGibson
    Nov 30, 2019 at 10:32
  • \$\begingroup\$ @azad.parinda - If you want to risk testing your batteries, to try to find what discharge rate they are "happy" with, then this answer from Russell McMahon shows one approach. As I mentioned in an earlier comment, you need a constant-current load which you can control. Also as I mentioned, since there is no guarantee that the batteries will behave the same as each other (the test results on the links in my answer show that fake batteries often behave differently, even when they appear to be the same) you need to test every battery! :-( \$\endgroup\$
    – SamGibson
    Nov 30, 2019 at 10:59
  • 1
    \$\begingroup\$ But beware, these batteries could fail in unpredictable ways (fire, explosion etc.) since we can have no confidence in these clear fakes. If you are going to use these batteries, then I would charge and discharge them far away from people, and take advice from whoever is responsible for fire precautions where you are. || Also, I just realised that you have already soldered the batteries together. That is going to make it impossible to test the batteries individually without desoldering them again. Repeated soldering & desoldering directly to the batteries is high risk (see my answer). \$\endgroup\$
    – SamGibson
    Nov 30, 2019 at 11:06

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.