0
\$\begingroup\$

UPDATED:

I am developing a wireless device composed of :

  • 2 Varta CP 1654 A3 Li-Ion 3.7V 120mAh (supporting 2C charge) with protection circuit module
  • 2 MCUs that can handle Li-ion CC/CV charging

My wireless device is composed of 2 earbuds connected with a cable, I can't use any other type of battery. There is a Protection Circuit Module (PCM) on the board where each cell is mounted in each earbuds (PCM recommended by Varta).

Correct me if I'm wrong but fast charging at 2*2C of parallel cells seems to be problematic for this reason:

If during charge cell_A reaches 4.2V before cell_B, cell_B could either receive a dangerous current > 2C if PCM_A triggers overcharge or ultimately both PCM_A and PCM_B would prevent the charging.

My GOAL is to have the fastest charging device possible while being safe and the fact that I have two ICs able to manage CC/CV charging makes me wonder if I could trigger switches between the batteries when the device is connected (Vcharge available) so then I could manage a 2C fast charge independently for each cell :

Switch between parallel batteries

I am trying to find a circuit that could act as a switch between the two cells. I thought at first about two MOSFETs in parallel but the body diodes will prevent the use that I want.

  • LOGIC_A represents if Vcharge is available to the device

  • The switch should be closed when no charge and open during charge.

I know this is unusual and most designs have single cell battery while some others have parallel batteries but with slower charging current than 2*2C. Please my question is not about modifying the design of my product, I just want to know if there is the possibility of such a "switch" circuit that could allow me to use the two chargers at fast charge instead of just one at slower charge.

\$\endgroup\$
  • 1
    \$\begingroup\$ The body diodes of the MOSFETs are surely going to defeat the purpose of the MOSFETs? \$\endgroup\$ – Dampmaskin Aug 14 '18 at 8:34
  • 1
    \$\begingroup\$ Indeed, this circuit isn't going to behave as you want it. The presence of the body diodes in the MOSFETs means that the difference between Vbat1 and Vbat2 can never be larger than 1 diode voltage as then one battery starts charging the other through the body diode. Also: why this dual charging? What would this achieve? Why would one battery not suffice? \$\endgroup\$ – Bimpelrekkie Aug 14 '18 at 8:47
  • 3
    \$\begingroup\$ I heard that parallel fast charging (2*2C) can be troublesome if one of the two batteries get damaged. If a battery is damaged that always means trouble, not only during fast charging. Also watch out with "I heard..." statements, when you hear/read this immediately ask WHY because understanding the issue is essential. If you don't (understand the issue) you could be over complicating your design, doing things that are simply not needed. Like dual charging. The fact that I do not know of any device using "dual charging" leads me to believe that there's no advantage to this scheme. \$\endgroup\$ – Bimpelrekkie Aug 14 '18 at 8:51
  • 1
    \$\begingroup\$ I don't understand from the question why you don't just keep the cells in parallel all the time. That seems like the easiest solution. Just pretend it is really only one cell with 240 mAh capacity. \$\endgroup\$ – mkeith Aug 16 '18 at 5:21
  • 1
    \$\begingroup\$ @mkeith it's all about applying a fast charge 2C (current equal to 2 times the capacity of the cell). If you 2C charge a 240mAh stack that means you send a current equal to 480mA (which is 4 times the capacity of one individual cell). In case of the failure of one cell the other could receive a current superior to 2C which is dangerous so it is blocked by the PCM. The charging would stop there. Keeping them parallel is maybe the easiest but means a slower charge which is not the best for consumers. I am trying to figure out a hack \$\endgroup\$ – chillyjee Aug 16 '18 at 6:43
0
\$\begingroup\$

Is there any warning about this configuration ?

Yes, two:

1) It does not work as you expecting it to, and
2) It is absolutely, utterly pointless.

There are two primary dangers of parallel connection:

1) If one cell is shorted inside it will short the other cells too.

This is easily avoided by adding a fuse in series with each cell. The reason it is not done all the time, I think, is that if lithium cell is shorted inside you are already in enough trouble to worry much about the other.

2) The charger does not see two cells, it is all one cell for it.

Therefore if one cell is fully charged earlier it might not be detected by the charger and it will continue charging the battery, overcharging that one cell. There are several things you can do to minimize this danger:

  • use only cells with same capacity, same manufacturer, preferably from same batch;
  • verify cell condition and capacity by individually charging and discharging them first;
  • before connecting cells together fully charge them and let them rest; then measure voltage and use only cells with very close voltage;
  • use cells with built-in protection circuit.

Having said all of the above, here is one thing you didn't ask about: why?

Connecting cells in parallel can usually be avoided by using single bigger cell.
If you need bigger capacity - use bigger cell;
If you need higher current - use bigger cell;

In short - save yourself a lot of troubles and use a bigger cell.

UPDATE

Batteries are Varta CoinPower cells

OK, first of all, this information (actual battery model would be better) belongs to the question. People wasted a lot of time trying to help you while making wrong assumptions based on incomplete data.

Second, these cells do not have built-in protection circuit, as you implied in the question ("Li-Ion 3.74V w/ protection circuit module"). You did not specify the external PCM module you use, nor provided circuit for it.

Nevertheless, I've reviewed the datasheet of those cells and I must say: my answer still stands, use single bigger cell.

Assuming you use CP 1654 A3 cells (120 mAh, 16.1 x 5.4 mm) with battery holders you need about 35 x 18 x 6 mm space or 3780 mm3 volume. In addition you'd need one extra protection circuit + one extra charger + whatever switching circuit you can come up with to split batteries for charging. These alone can take as much space as yet another cell.

Just 30 seconds search returned these:

LP602025 250 mAh, 25 x 20 x 6 mm, 3000 mm3
LP402035 250 mAh, 35 x 20 x 4 mm, 2800 mm3
LP501540 280 mAh, 40 x 15 x 5 mm, 3000 mm3

They all have higher capacity than your two cells while taking about the same space and less volume. Furthermore, they already include protection circuit, so you only need space for tiny charger or BMS chip. I am sure deep research will return plenty of suitable cells by current rating, capacity and size.

\$\endgroup\$
  • \$\begingroup\$ Thanks for your answer! At first I thought of using just one cell because I need a large capacity but I realized that the size of one cell would be too big to fit in my device whereas two batteries (with a smaller capacity) could fit in the design. What you refer to in your point 2) is also what I heard from someone working for a battery manufacturer. Parallel fast charge at 4C could damage one cell if the other is charged before. Hence I thought that with two chargers I could handle normal CC/CV 2C-fast charge for the two batteries independently and have a safe circuit. \$\endgroup\$ – chillyjee Aug 15 '18 at 5:45
  • \$\begingroup\$ Just what batteries are that? 2C is not a fast charge for anything but tiniest cells. And if your cells are so small then any "circuit for charging independently" will probably be bigger than battery itself. Also note, that with the precautions listed above the chances of failure could be same as your mobile phone's battery. The protection circuit in each cell will prevent overcharging and matching them will ensure longer service without losing capacity. \$\endgroup\$ – Maple Aug 15 '18 at 7:24
  • \$\begingroup\$ Batteries are Varta CoinPower cells, they can handle a 2C charge. The PCM won't trigger overcharging at 2C. However at 4C it would be troublesome to be trapped in an overcharging scenario where the charger IC would not be able to charge one of the two cells (or ultimately both). \$\endgroup\$ – chillyjee Aug 15 '18 at 8:13
  • \$\begingroup\$ See an update, then fix the question and maybe somebody will be able to help you. \$\endgroup\$ – Maple Aug 15 '18 at 11:43
  • \$\begingroup\$ I am sorry to hear that you wasted your time trying to help me... Thank you to suggest design modifications but as I mentioned, I have design restrictions. My question is not about what kind of battery I should use but about the theoretical possibility of a switch. I guess it wasn't formulated clearly enough so I updated it. Of course I did a lot of research before posting here... Over the last months I collected a lot of similar chinese ebay batteries that you found in 30sec and 100% of them had very different capacity than advertised! Anyway sorry for your time. \$\endgroup\$ – chillyjee Aug 16 '18 at 4:09

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.