Author Topic: 1S LiFePO4 BMS  (Read 446 times)

0 Members and 2 Guests are viewing this topic.

Offline ThermallyFrigidTopic starter

  • Frequent Contributor
  • **
  • Posts: 266
  • Country: us
1S LiFePO4 BMS
« on: June 19, 2026, 03:18:45 pm »
I have a project that uses single cell AA size LiFePO4 cells to power a blue LED
The circuit also consists of a QX5252, axial inductor and small solar cell.

I want to use a small 1S 3.2v BMS to protect the cell from overdischarge.
However, when I look for such a BMS, they are all set to 2.1v for their overdischarge cutoff limit.

I want the cut-off higher, at 2.6v to 3.0v for the LiFePO4 cell.

I am wondering why all manufacturers for 1S, 3.2v LiFePO4 BMSs' would have the over-discharge limit set so low that damage will occur to the cell after repeatedly dropping to that voltage?
Why not set it at a safer (for the cells lifespan) voltage?

 

Offline Peabody

  • Super Contributor
  • ***
  • Posts: 2719
  • Country: us
Re: 1S LiFePO4 BMS
« Reply #1 on: June 19, 2026, 03:51:41 pm »
The same complaint is often heard about 3.7V protection circuits.  Many suggest that 2.54V is too low for overdischarge, and the battery may be damaged.  However, there are probably hundreds of millions of lithium ion and LIPO cells out in the world protected by the DW01, which has the same shutoff point, and apparently most do fine.  One suggestion is that these protection circuits are only emergency shutoffs, and your circuit should provide a separate "normal" shutoff at a higher voltage.

There is an alternate protection IC for 3.7V lithium with overdischarge at 3.0V, the part number of which I can't remember at the moment, and maybe something similar is available for your LiFePO4 cell.  But if not, in theory you could use something like the DW01 in combination with your normal BMS, which would give you at least 2.5V instead of 2.1V.

Actually, they now make protection ICs that move the mosfets into the chip, so the entire circuit is now a very small three-pin IC, a capacitor and a resistor.  XB7608A and FM5056 would be examples.
 
The following users thanked this post: Whales

Online tunk

  • Super Contributor
  • ***
  • Posts: 1389
  • Country: no
Re: 1S LiFePO4 BMS
« Reply #2 on: June 19, 2026, 04:37:04 pm »
Would it be possible to use two BMSs?
- Li-ion for discharge
- LiFePO4 for charge
 

Online mariush

  • Super Contributor
  • ***
  • Posts: 5307
  • Country: ro
  • .
Re: 1S LiFePO4 BMS
« Reply #3 on: June 19, 2026, 05:56:12 pm »
Monolithic Power Systems have some charger ICs which have configurable undervoltage levels through i2c

For example, MP2664 and MP2667 are lithium chargers that have the battery undervoltage configurable between 2.4v and 3.1v (defaults to 2.8v) . Technically they're lithium-ion charger chips, but the charge voltage can be configured between 3.6v and 4.545v so technically if you set the charge voltage to the minimum of 3.6v you could charge the LiFePO4 cells.

MP2664 https://www.digikey.co.uk/en/products/detail/monolithic-power-systems-inc/MP2664GG-0000-Z/18089944

MP2667 https://www.digikey.co.uk/en/products/detail/monolithic-power-systems-inc/MP2667GG-0000-P/11610986

Also MP2710 is worth the mention, though it's less hobbyist friendly with the bga balls : https://www.digikey.co.uk/en/products/detail/monolithic-power-systems-inc/MP2710GC-0000-P/25543918


 

Offline thm_w

  • Super Contributor
  • ***
  • Posts: 9542
  • Country: ca
  • Non-expert
Re: 1S LiFePO4 BMS
« Reply #4 on: June 19, 2026, 10:16:31 pm »
HY2112-HB has overdischarge voltage of 2.5V
FH2120-NB has overdischarge voltage of 2.8V

Look around on LCSC and you'll find a few, some are up to 3V. https://www.lcsc.com/category/1005.html
Profile -> Modify profile -> Look and Layout ->  Don't show users' signatures
 

Offline ThermallyFrigidTopic starter

  • Frequent Contributor
  • **
  • Posts: 266
  • Country: us
Re: 1S LiFePO4 BMS
« Reply #5 on: Yesterday at 02:45:38 pm »
HY2112-HB has overdischarge voltage of 2.5V
FH2120-NB has overdischarge voltage of 2.8V

Look around on LCSC and you'll find a few, some are up to 3V. https://www.lcsc.com/category/1005.html

These would be great...except they are for 2S applications according to the datasheet
 

Offline ThermallyFrigidTopic starter

  • Frequent Contributor
  • **
  • Posts: 266
  • Country: us
Re: 1S LiFePO4 BMS
« Reply #6 on: Yesterday at 02:46:24 pm »
Monolithic Power Systems have some charger ICs which have configurable undervoltage levels through i2c

For example, MP2664 and MP2667 are lithium chargers that have the battery undervoltage configurable between 2.4v and 3.1v (defaults to 2.8v) . Technically they're lithium-ion charger chips, but the charge voltage can be configured between 3.6v and 4.545v so technically if you set the charge voltage to the minimum of 3.6v you could charge the LiFePO4 cells.

MP2664 https://www.digikey.co.uk/en/products/detail/monolithic-power-systems-inc/MP2664GG-0000-Z/18089944

MP2667 https://www.digikey.co.uk/en/products/detail/monolithic-power-systems-inc/MP2667GG-0000-P/11610986

Also MP2710 is worth the mention, though it's less hobbyist friendly with the bga balls : https://www.digikey.co.uk/en/products/detail/monolithic-power-systems-inc/MP2710GC-0000-P/25543918

These look promising for charging but I don't see over-discharge protections ?
 

Offline ThermallyFrigidTopic starter

  • Frequent Contributor
  • **
  • Posts: 266
  • Country: us
Re: 1S LiFePO4 BMS
« Reply #7 on: Yesterday at 02:47:01 pm »
Would it be possible to use two BMSs?
- Li-ion for discharge
- LiFePO4 for charge


Interesting thought
 

Offline Peabody

  • Super Contributor
  • ***
  • Posts: 2719
  • Country: us
Re: 1S LiFePO4 BMS
« Reply #8 on: Yesterday at 09:45:58 pm »
Would it be possible to use two BMSs?
- Li-ion for discharge
- LiFePO4 for charge


Interesting thought

And it should be pretty simple.  You just install both of them, and  the one that is the most limiting would automatically be controlling in each function.  If you end up with two sets of mosfets that the current goes through, it's probably not a major problem.  I've used protected batteries in circuits that also had protection built in, and didn't have any problems.
 

Offline thm_w

  • Super Contributor
  • ***
  • Posts: 9542
  • Country: ca
  • Non-expert
Re: 1S LiFePO4 BMS
« Reply #9 on: Yesterday at 10:39:38 pm »
HY2112-HB has overdischarge voltage of 2.5V
FH2120-NB has overdischarge voltage of 2.8V

Look around on LCSC and you'll find a few, some are up to 3V. https://www.lcsc.com/category/1005.html

These would be great...except they are for 2S applications according to the datasheet

"HY2112 ICs is best created for single-cell LiFePO4 rechargeable battery protection"
Profile -> Modify profile -> Look and Layout ->  Don't show users' signatures
 

Online mariush

  • Super Contributor
  • ***
  • Posts: 5307
  • Country: ro
  • .
Re: 1S LiFePO4 BMS
« Reply #10 on: Today at 09:36:37 am »
Monolithic Power Systems have some charger ICs which have configurable undervoltage levels through i2c

For example, MP2664 and MP2667 are lithium chargers that have the battery undervoltage configurable between 2.4v and 3.1v (defaults to 2.8v) . Technically they're lithium-ion charger chips, but the charge voltage can be configured between 3.6v and 4.545v so technically if you set the charge voltage to the minimum of 3.6v you could charge the LiFePO4 cells.

MP2664 https://www.digikey.co.uk/en/products/detail/monolithic-power-systems-inc/MP2664GG-0000-Z/18089944

MP2667 https://www.digikey.co.uk/en/products/detail/monolithic-power-systems-inc/MP2667GG-0000-P/11610986

Also MP2710 is worth the mention, though it's less hobbyist friendly with the bga balls : https://www.digikey.co.uk/en/products/detail/monolithic-power-systems-inc/MP2710GC-0000-P/25543918

These look promising for charging but I don't see over-discharge protections ?

Go to page 6  in the datasheet: https://www.monolithicpower.com/en/documentview/productdocument/index/version/2/document_type/Datasheet/lang/en/sku/MP2664/document_id/4382/

Battery under-voltage lockout threshold  :

i2C programmable range 2.4 3.1 V
Falling, programmable, REG01h, bits[2:0] = 100 2.6 2.8 3 V 

Also page 19

Battery Discharge Function
If battery is connected and the input source is missing, the battery FET is fully on once VBATT is above the VBATT_UVLO threshold. The 100mΩ battery FET minimizes conduction loss during discharge. The quiescent current of the IC is as low as 11μA in this mode. The low on resistance and low quiescent current help extend the running time of the battery.

Over-Discharge Current Protection
The IC has an over-discharge current protection in discharge mode and supplement mode.
Once IBATT exceeds the programmable discharge current limit (default 2A), the battery FET turns off after a 60µs delay, and the MP2664 enters hiccup mode in over-current protection (OCP). The discharge current can be programmed high to 3.2A via the I2C. If the discharge current goes high to reach the internal fixed current limit (about 3.7A), the battery FET is turned off and starts hiccup mode immediately.
Similarly, when VBATT falls below the programmable VBATT_UVLO threshold (default 2.8V), the battery FET is turned off to prevent over-discharge.


So like I said, the only risk is having the charge voltage default to 4.2v and forgetting to override it and set the charge voltage to 3.6v through i2c.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf