As a student group project, we are going to use Li-Ion batteries to test our charging and balancing circuit. These batteries at risk of explosion. How should we avoid that danger? Is there any intelligent battery cells to use? or any other suggestion?


1) Do the first test with a battery simulator. Don't connect it to a real battery until you're sure it's working as expected.


simulate this circuit – Schematic created using CircuitLab

Set R1 to a value according to the expected charge current and make V1 an adjustable lab power supply. You can then gradually increase V1 and check that charge stops at the appropriate point, and that the current flowing in R1 does not exceed expected values (don't forget to use a suitable wattage resistor).

2) If you're using the type of battery that comes with a thermistor, make sure you incorporate that correctly.

  • \$\begingroup\$ Do you mean I should charge the battery through a reversed diode? Also do you think using lead acid for the first time can be another option? \$\endgroup\$ – barej Mar 2 '15 at 11:59
  • \$\begingroup\$ No, I mean don't use a battery at all for the first time, use a circuit that simulates the properties of a battery. Lead-acid is different from Li-ion and requires a different charger design anyway. \$\endgroup\$ – pjc50 Mar 2 '15 at 12:06
  • \$\begingroup\$ Thank you, could you tell me how the reversed diode is expected to charge the battery? \$\endgroup\$ – barej Mar 3 '15 at 6:29
  • \$\begingroup\$ It doesn't. There is no battery. V1 should be an adjustable bench power supply. \$\endgroup\$ – pjc50 Mar 3 '15 at 9:32

There are mostly two situations that can get critical with lithium ion batteries: extreme physical abuse and overvoltage. Extreme physical abuse can be avoided by not performing the experiments with your charging and balancing circuits inside crashing R/C planes, and overvoltage can be prevented by using a voltage clamp in parallel that is able to sink the full charger current.

The most simple clamp circuit would be a Zener diode, put those diodes are not sharp enough. You want no current at 4.2V, but full charging current at 4.3V. The simplest design with enough precision and steepness is the TL431 shunt circuit, as shown in http://www.ti.com/lit/ds/symlink/tl431.pdf, figure 31 ("High current shunt regulator"). You don't have Vin and the resistor on the left in your application, tough.

Actually, such a voltage clamp is identical to one kind of balancing circuit, just with its trigger voltage set slightly (typically 50mV) higher. You can test the this kind of clamp circuit with just a lab supply. No need to have a battery cell in parallel or a charger. The supply should switch quite steeply from constant voltage/nearly no current to current limiting when the output voltage passes the protection circuit voltage. Make sure the clamp circuit is able to dissipate the power (e.g. charge current * 4.3V).

If the battery chemistry is still open for suggestions, take a look at LiFePO4 batteries, as they need only to be charged to around 3.7V for (nearly) full capacity, but are claimed to be safe up to at least 4.2V. There is a youtube video showing a single LiFePO4 cell explode, which has been connected to a 12V source for 13 minutes, though.

  • \$\begingroup\$ How to make a clamp circuit? should I use Zener diode? \$\endgroup\$ – barej Mar 2 '15 at 12:00

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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