Isolating Li-poly battery during deep charging

Thread Starter

DJ_AA

Joined Aug 6, 2021
296
Hi

Is it possible to have some type of isolation circuit to isolate the battery from the circuit that is powering when the battery is charging, when the voltage drops less then 3V?

There was an incident when the battery voltage drop to about 2.5V, the Li-Ion charger IC provides about 10mA of charge to recondition the battery, but after a while the microcontroller or remaining part of the circuit power ups and uses the little provided charge in the battery. This cause the battery not progress past the deep charging cycle. This continued till I had remove the battery and use a standalone charger.

Is it possible that we can place a MOSFET or switch between battery and remaining circuit, so that when the battery voltage reaches 3.5V it then powers the reaming system?

I have tried a simple toggle switch, but notice some unusual performance issues on my RF circuit. I presume the switch must have added some resistance or interference.

I am looking if there is a better option, so that the MOSFET or discrete switch can help recover a battery during charging?
 

Audioguru again

Joined Oct 21, 2019
6,647
The 10mA low current charging when a Lithium cell is at a low voltage is not for "reconditioning". Instead it is for a test of how badly the cell is damaged by discharging too low and prevents an explosion and fire. If the cell's voltage does not rise then the charger determines a severely damaged cell and shuts off with an alarm.

You must not allow the microcontroller to power up during this test. Instead, sense that the charging current or the voltage becomes normally high then power up the microcontroller.
 

Thread Starter

DJ_AA

Joined Aug 6, 2021
296
The 10mA low current charging when a Lithium cell is at a low voltage is not for "reconditioning". Instead it is for a test of how badly the cell is damaged by discharging too low and prevents an explosion and fire. If the cell's voltage does not rise then the charger determines a severely damaged cell and shuts off with an alarm.

You must not allow the microcontroller to power up during this test. Instead, sense that the charging current or the voltage becomes normally high then power up the microcontroller.

Yes that is what I am to do but need some type of switch
 

sghioto

Joined Dec 31, 2017
5,370
What else besides the micro is powered by the battery, you mentioned a RF circuit?
What part # charging IC are you using?
 
Last edited:

Thread Starter

DJ_AA

Joined Aug 6, 2021
296
I have the GSM module BG95 also on the board, and small Microchip ZigBee module.

We have are using the following Li-poly chargers
LTC4054
and
LTC4162

Different boards are using different Li-Poly Chargers
 

Ya’akov

Joined Jan 27, 2019
9,044
Hi

Is it possible to have some type of isolation circuit to isolate the battery from the circuit that is powering when the battery is charging, when the voltage drops less then 3V?

There was an incident when the battery voltage drop to about 2.5V, the Li-Ion charger IC provides about 10mA of charge to recondition the battery, but after a while the microcontroller or remaining part of the circuit power ups and uses the little provided charge in the battery. This cause the battery not progress past the deep charging cycle. This continued till I had remove the battery and use a standalone charger.

Is it possible that we can place a MOSFET or switch between battery and remaining circuit, so that when the battery voltage reaches 3.5V it then powers the reaming system?

I have tried a simple toggle switch, but notice some unusual performance issues on my RF circuit. I presume the switch must have added some resistance or interference.

I am looking if there is a better option, so that the MOSFET or discrete switch can help recover a battery during charging?
Why was the battery allowed to discharge so far?
 

sghioto

Joined Dec 31, 2017
5,370
Different boards are using different Li-Poly Chargers
Are you using different battery types? The LTC 4054 is much less complicated and has a status pin that the micro can monitor. Would need to see the schematic of how you have this configured.
 

Thread Starter

DJ_AA

Joined Aug 6, 2021
296
Are you using different battery types? The LTC 4054 is much less complicated and has a status pin that the micro can monitor. Would need to see the schematic of how you have this configured.
The RF Zigbee devices use the LTC4054

The main control unit uses the LTC4162 as its a MPPT charger so it has solar charging option.
 

Thread Starter

DJ_AA

Joined Aug 6, 2021
296
Yes our batteries come with this protection circuit, but the issue is even when battery drop below the charge initially is very low, but seems to get stuck to about 10-20mA charge current till I disconnect the battery from the system to charge the battery independently for short period (few mins). It seems once the battery get to about 3V its back to the normal charge current, instead of 20mA.
 

Ya’akov

Joined Jan 27, 2019
9,044
Yes our batteries come with this protection circuit, but the issue is even when battery drop below the charge initially is very low, but seems to get stuck to about 10-20mA charge current till I disconnect the battery from the system to charge the battery independently for short period (few mins). It seems once the battery get to about 3V its back to the normal charge current, instead of 20mA.
If the protection circuit is working properly, it shouldn't be possible to discharge the battery below the threshold where you are having this problem. I can't quite tell what is actually happening with these devices so I will leave it to others to work out.

I hope you find a good solution.
 

sghioto

Joined Dec 31, 2017
5,370
I still need to see how all this is configured.
The battery charge ICs all have auto shut off when fully charged and recharge when the voltage is below the programmed threshold as long as the input supply is connected.
 

Audioguru again

Joined Oct 21, 2019
6,647
The Protection Circuit" allows a deadly discharge to only 2.5V then the battery charger IC knows that is way too low and "test charges" with a low current to see if the battery is not damaged. The voltage of a surviving battery will rise.

Your circuit has a current that fools the charger IC that the battery voltage is not rising normally then it shuts down.

Most Chinese BMS's allow a discharge voltage too low maybe so the battery fails soon then you buy another and another.
 

Thread Starter

DJ_AA

Joined Aug 6, 2021
296
I connected my battery to multiple PCB to drain the battery, and saw voltage drop even below 2.5V. So I am thinking this protection circuit is not quite what it says it is.

It seems that when the IC is proving minimum voltage about 10mA, something in my PCB is using that power never progresses till I use an independent charger(LTC4162 with not other circuit )

I am now thinking in future design where Li-Poly battery is required , I should keep a Mosfet in Series between the battery and circuit with the gate controlled by a toggle switch.

This way I can isolate the remaining circuit, if I need to just charge the battery to recover it.
 

Audioguru again

Joined Oct 21, 2019
6,647
My RC model airplanes pulse the main motor to alert me that the battery voltage is low at 3.2V per cell. Then the main motor cuts off but there is enough voltage for a low current steered glide when the battery is at 3.0V per cell. The manufacturer says, "Repeated flying to low voltage cutoff will damage the battery".

Li-PO batteries are sold and stored at the recommended half-charge of 3.7V per cell.
 

Thread Starter

DJ_AA

Joined Aug 6, 2021
296
My devices report the battery voltage via a message on a app , but sometimes this can be ignored by some users, which causes this loop.

Is there a reason why this does not happen on mobile phone or tablet, what do they use ?
 
Top