0
\$\begingroup\$

I have got some li-ion batteries which I don't know anything about and there is nothing written on them. Can you advise some testing methods to find out about their specifications and charging data so I can safely charge and use them for my own circuits?

Some of them are 18650 size and others are flat square shaped.

You are right to advise me to be on the safe side and get rid of them. But right now I am more curious to solve the puzzle than caring about the economic side. I have the device for two of them and can put the batteries in and charge. So I think I can find about the max voltage, and may be other data.

\$\endgroup\$
  • 2
    \$\begingroup\$ They might be batteries that got sorted out because they were unsafe to operate. In fact, you having them, probably from a cheap source or an unmarked pile, does make that relatively likely. If that's the case, there's no safe way of operating them... \$\endgroup\$ – Marcus Müller Jul 25 '18 at 8:15
  • 1
    \$\begingroup\$ Li-ion battery cells can be dangerous when you know exactly what you've got. My advice is to put those cells in a suitable waste electrical bin. \$\endgroup\$ – Puffafish Jul 25 '18 at 8:23
  • \$\begingroup\$ also, any lithium cell does need relatively elaborate charging logic. The amount of money you save by not going out and buying new batteries that come with a datasheet, built-in protection and guarantees might be negative, because you'll have to implement protective circuitry yourself. So, I think from an economic point of view, I'd agree with @Puffafish. \$\endgroup\$ – Marcus Müller Jul 25 '18 at 8:26
  • \$\begingroup\$ If they are unmarked how do you know that they are li-ion? \$\endgroup\$ – Andy aka Jul 25 '18 at 9:14
  • \$\begingroup\$ @Andyaka On the 18650 ones it is written Li-ion. For the others it is a guess according to the device date. Not sure about all of them though. \$\endgroup\$ – Sohail Jul 25 '18 at 9:39
0
\$\begingroup\$

Most Li-ion cells have a nominal voltage of 3.6 or 3.7 V, and a maximum voltage of 4.2 or 4.22 V. LiFePo cells have lower voltages, don't mix them up.

So here is a rough guideline to get you started with testing them. Note that this is probably not the best way to do it, not even necessarily a safe way to test them, but if you must test them, this method is probably better than nothing:

  • To test max voltage, charge to 4.2 and see if the cell puffs up or pops. If it does, the max V is lower than this, and the battery is now destroyed. Dispose of it safely. If not, let the battery rest for a while. Confirm that the voltage doesn't drop below 4.15 V or so during the next hour. If it does, the battery either has a lower max voltage than 4.2V and should be disposed of, or it has an internal short and should be disposed of. If all is good, you can continue with the next step.

  • To test capacity and losses, discharge from max V to a set V (e.g. 3.0 V) at a constant safe-ish current (e.g. 1 C). See how much charge you get out of it. Then recharge to max V using a proper charging algorithm, see how much charge goes back into it. This will tell you the capacity. The difference between the charge and discharge will tell you about efficiency.

  • To test max discharge current, discharge at 1 C, measure the temperature all the way. Then recharge, and discharge at 2 C. Then at 3 C, and so on. Once the temperature reaches a certain threshold that you define, e.g. 60 or 70 °C, you have exceeded the max continuous discharge current. Remember to take the casing into consideration, as you're measuring the temperature externally, but if the internal temperature reaches 120 °C or so, things will get ugly fast.

  • To test max charge current, do something similar while charging, but start lower and increase more slowly. Generally, Lithium cells cannot be charged as fast as they can be discharged. Also be advised that fast charging can degrade the battery even if the temperature doesn't skyrocket, so go gentle.

| improve this answer | |
\$\endgroup\$
0
\$\begingroup\$

Here is some method I found in eevblog but the capacity were known. I don't know if it is applicable to my case.

You need a multimeter and a laboratory power supply with voltage and current regulation.

Measure each cell and check the voltage. It should be between 3.6 V and 4.2 V. Put to one side any cells with a voltage below about 3.0 V. These are probably weak or failed cells and should be recycled.

If you have six cells that were wired to give 10.8 V and 5.2 Ah, then the pack would have been a 3s2p arrangement and the capacity of an individual cell should be about 2.6 Ah.

Take one of the cells with a voltage above 3.6 V and try charging it with the power supply. Set the voltage limit to 4.2 V and initially set the current limit to 500 mA. Put the cell on charge and check that it accepts charge. The power supply should be in CC mode, the cell voltage should slowly rise and it should not get warm. If this is OK, you can increase the current limit to 1 A or even 1.5 A and continue charging.

Eventually the voltage should reach 4.2 V and the power supply should switch from CC mode to CV mode and the charging current should start tailing off. When the charging current decreases to 200 mA you can stop charging.

If you want to check the capacity of the cell you will need to conduct a controlled discharge and measure the current over time. A data logging meter is helpful for this. Otherwise you can just use a load like a 6 V 6 W lamp and record the current every 5 minutes or so by hand. Discharge the cell down to about 3.3 V under load and then figure out the capacity as average current times discharge time.

| improve this answer | |
\$\endgroup\$
  • \$\begingroup\$ I think this is decent advice. But I don't see anything about the capacity being known. On the contrary, the last paragraph starts with "If you want to check the capacity of the cell" ... Did you mean that the max voltage is known? Yeah, you'll probably have to assume that the max voltage is 4.2 V, provided that the chemistry is Li-ion. I don't know of a way to test this other than to go for it, and see if it goes bang or not. \$\endgroup\$ – Dampmaskin Jul 25 '18 at 10:57

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