0
\$\begingroup\$

I am trying to build a lithium battery charger (Li-Ion) for charging banks of 6s batteries of varying capacities (2-30Ah) from a varied DC input.

Given that I need to supply CC/CV to charge these batteries I plan on using a buck/boost converter such as this one: https://www.ebay.com/itm/1500W-30A-Voltage-Step-Up-Converter-Boost-CC-CV-Power-Supply-Module/292853831480?epid=13026837672&hash=item442f72d738:g:EfsAAOSwtYZcCiBH

Connected directly to a BMS (or several in parallel) which offers charge balancing: https://www.ebay.com/itm/Waterproof-BMS-6S-25A-35A-45A-60A-24V-25-2V-Li-ion-Battery-PCB-PCM-with-Balance/153021323145?hash=item23a0c81789:m:md0CttkjID5Rn60hEU8c1BA

My questions are as follows:

1) Will the above parts safely charge a 6s li-ion battery, assuming current supply is set appropriately for their cells' capacity?

2) Is it necessary to add a separate system for charge termination at low current (the above BMS claims to offer overcharge protection but I am dubious, not to mention 4.25v seems high)? If this is necessary what would be the easiest way to integrate such a function?

3) Final question - what would I need to adapt about this system for charging LiFePo batteries in series?

Thanks in advance

\$\endgroup\$
  • 3
    \$\begingroup\$ Note: Li-Ion batteries are very dangerous. Improper manipulation can lead to overheating, fires and explosions. If you are not 100% sure of what you are doing, don't! \$\endgroup\$ – jcaron Jun 19 '19 at 11:58
  • 1
    \$\begingroup\$ @jcaron Unfortunately, that misconception is widely spread. Yes, failure modes of Lithiums often include fire and chain reaction. But treating good Lithiums 'right' is easy: Limit current and voltage and you're pretty safe. - If a cell should be already damaged, 'safe' charging, not causing a fire failure, is indeed more challenging. \$\endgroup\$ – JimmyB Jun 19 '19 at 12:24
2
\$\begingroup\$

Reading the comments I feel the need to point out:

Warning

"Good", i.e. healthy, i.e. non-defective, cells are safe and easy to charge, see below.

If a Lithium cell is damaged, through previous mechanical, thermal, or electrical abuse (over-discharge, over-charge, over-current), which may not be visible at all, charging that cell (as well as using it, sometimes even just storing them) bears significant risk of thermal runaway with fire, likely causing a chain reaction of the other cells in a battery pack.

A good/sophisticated charger will employ different mechanisms to try and detect defective cells and abort charging. Some symptoms of some defects can be detected by the charger before it's too late. Some may not be so pronounced and go undetected. That's why even with the most expensive/sophisticated chargers there's still a chance of disaster.

Answer to OP

1) Will the above parts safely charge a 6s li-ion battery, assuming current supply is set appropriately for their cells' capacity?

Yes.

2) Is it necessary to add a separate system for charge termination at low current (the above BMS claims to offer overcharge protection but I am dubious, not to mention 4.25v seems high)? If this is necessary what would be the easiest way to integrate such a function?

No, not necessary. As long as you make sure to never apply more than 4.2V to a cell, you're safe. This means that the PSU in your case must be set to never output more than 6*4.2V = 25.2V, besides limiting the current to appropriate value for the batteries to charge. The BMS will take care of balancing to ensure that you don't end up with one cell at e.g. 4.0V and another one at 4.4V (i.e. apparently "safe" 8.4V for a 2s).

Consider the overcharge protection of the BMS (@4.25V) as a safety feature to prevent fire, not as part of the charging algorithm.

Don't rely on this to normally terminate your charge. It's like a circuit breaker in your home: You don't stick a screwdriver into your wall outlet to turn off the light, although it will seem to work as desired most of the time.

You may want to play it extra safe or to be gentle to your batteries to prolong their life and set V[max] to 4.15V or even 4.10V per cell. This significantly reduces stress and may double the cell's life by limiting the SoC to about 90% instead of 100% (=4.2V per cell).

Note that while terminating charging is not required as long as you stay at or below 4.2V per cell, holding a cell at 4.2V for extended periods will reduce the cell's useful cycle life. Hence the cells will suffer if you keep the charger running e.g. over night or even over a weekend. Apart from that, it's not too dangerous to apply 4.2V for some time even after the cell is already full. Although a safety timer does make sense as one means to detect a cell defect: If the battery is expected to be fully charged after e.g. 1 hour, and the charger detects that it does not seem to be full even after e.g. 2 hours, it should assume a defect and abort charging.

3) Final question - what would I need to adapt about this system for charging LiFePo batteries in series?

Because LiFePo's operate at a lower maximum voltage (about 3.6V instead of 4.2V for Li-Ion), you'll need a different BMS, and you need to set the PSU's maximum output voltage to 3.6V per cell, i.e. 6*3.6V = 21.6V max.

| improve this answer | |
\$\endgroup\$
  • 2
    \$\begingroup\$ Most of your comments are OK, but the suggestion that LiIon can be safely held at 4.2V on charge is bad advice and will lead to life reduction for moderate extended periods and rapid destruction if left allowed to CV tail down to low current levels. I realise that you have warned about not leaving them overnight or for a weekend BUT you need to emphasis charge termination at some Imin || Terminating at eg C/10 is pushing cells quite hard and shortens life. Even terminating at C/4 is is 'enthusiastic' and C/2 termination meets most needs. || Your recommenation re Vmax of 4.15 or 4.1V is good. \$\endgroup\$ – Russell McMahon Jun 19 '19 at 13:03
  • 1
    \$\begingroup\$ +1 .................... \$\endgroup\$ – Russell McMahon Jun 19 '19 at 13:04
  • 1
    \$\begingroup\$ @JimmyB You are correct - permissible charge rate is cell model specific. Usually either C/1 or C/2 are specified as Imax in CC mode. I'll open a few data sheets in the order I find them and see what they say. || Panasonic NCR18650B 3250 mAh -> C/2 max | Samsung INR18650-30Q usual C/2, max 4C/3 (full charge period). | LG HD2 2 Ah - Nominal : 1A 4.2V, 50mA End-current (CC-CV) / Fast : 4A 4.2V, 100mA End-current (CC-CV) <- ie C/2 and 2C ! - I_terminate = C/40 & C/20 = Rpad Warrior range :-). \$\endgroup\$ – Russell McMahon Jun 19 '19 at 13:29
  • 1
    \$\begingroup\$ @Thom Stop the charger - if at output a series FET is easy and cheap. | I would lean towards removing voltage from LiFePO4s BUT they have a "strange" end of charge mode - they will charge all the way to 4.2V with very very little capacity above 3.6V. I have found nothing in formal specs on this and minimal mention in literature. I SUSPECT that there is perhaps a very small amount of active material outside the LiFePO4 Olivine matrix which effectively adds a low capacity LiIon front end, | ... \$\endgroup\$ – Russell McMahon Jun 19 '19 at 13:36
  • 1
    \$\begingroup\$ That said - I have had LiFePO4 fail rapidly and unexpectedly in a situation where I left 3.6V on it. I do not know why and I hope to investigate that further. For now I'd remove Vchg and it's easy t o do. \$\endgroup\$ – Russell McMahon Jun 19 '19 at 13:36
-1
\$\begingroup\$

I would not recommend to jury rig a charger for Li-based chemistries. Li-ion and LiFePo batteries are very sensitive to being properly charged and, due to the large energy they pack, can be disastrous if they even feel like being unkindly handled. While what you propose is safe for NiCd and NiMH chemistries, it is not for Li chemistries. If you want to build a charger, first read up how such chargers are supposed to work (there are good articles and appnotes from the charger chip manufacturers, Ti and Linear/Analog are good places to start). If you cannot find a board with a suitable charger chip somewhere, you can build your own PCB (not too difficult, just requires some care).

For the first experiments, I urge you to use battery packs where each cell has its individual protection circuit. Or use them in an environment where a metal fire wouldn't cause damage or safety issues.

BTW: there is a reason why chargers for Li batteries are custom builds for a specific type of battery. Understand why this is, before attempting to build your own charger.

| improve this answer | |
\$\endgroup\$
  • \$\begingroup\$ I disagree: "are very sensitive to being properly charged" - Lithiums are much easier to "properly charge" than e.g. NiMH. And "due to the large energy they pack, can be disastrous" is also not the right conclusion. It's not the (electrical) energy in them that causes disasters, it's the fact that they contain flammable material (electrolyte) and bring their own oxidizer which makes it very hard to extinguish them. \$\endgroup\$ – JimmyB Jun 19 '19 at 12:32

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.