Circuit not charging when the battery is below a certain voltage

Thread Starter

Mywk

Joined May 10, 2018
16
Hello everyone,

I have a simple battery charging circuit that seemed to be working perfectly, that is, until the battery was completely discharged, afterwards I'm unable to re-charge the battery.

Charging controller is a MCP73831/2

The battery is a tiny cheap 55mAh 3.7 V battery and the circuit is as follows:
upload_2018-9-25_16-18-47.png

I set everything according to the datasheet, being the 20k resistor on R3 for a 50mAh battery.

When the battery is fully charged it outputs 4.1V, after discharging it stays at 1.950V

Here's the deal, if I take out the battery and charge it somewhere else (let's say up to 3V) then put it back in my circuit, it keeps charging, but not on the circuit.

I'm not the most experienced in electronics and I'm out of ideas so I'm asking here as it seems the most appropriate place to do it:
What could the problem be?

Note: I think the problem may be related to the Undervoltage Lockout (UVLO), page 13 of the controller pdf.

///// EDIT

Unlike the battery specs say so, it does NOT have over-discharging protection, so the microcontroller will just keep trying to suck the battery dry.

I'm basically searching for a charging ic with over-discharging protection.
 
Last edited:

dl324

Joined Mar 30, 2015
8,918
Note: I think the problem may be related to the Undervoltage Lockout (UVLO), page 13 of the controller pdf.
That's not the problem:
upload_2018-9-25_7-28-54.png
It comes out of shutdown when the supply voltage is 150mV above the battery.
What could the problem be?
Battery could be bad.

Need to read more of the datasheet...
 

Thread Starter

Mywk

Joined May 10, 2018
16
@dl324 Thanks, I'm currently trying to debug the circuit but have not yet found a solution.

I'l add some more info:
The VBUS is basically coming out of a regulator that always outputs around 4.8V (500 to 2000 mAh), the battery basically powers a microcontroller which always discharges the battery to 1.95V, if I connect a 3.3V power supply to the microcontroller for just a few seconds (as the battery seems to charge a bit) then the controller seems to start charging the battery.

Also feel free to correct me if any of my explanations seems incorrect!

Edit: This is a 55mAh 3.7V battery, can't find the datasheet (manufacturer says it has over-charging and over-discharging protection, if this is true how is it that I'm reading 1.95V coming out of it?)
 
Last edited:

Thread Starter

Mywk

Joined May 10, 2018
16
Sorry for the double-post, just wanted to update the situation.

It seems that the controller is acting as it should, since the battery is way below what it should be it simply doesn't charge it.. And, unlike the battery specs say so, it does NOT have over-discharging protection, so the microcontroller will just keep trying to suck the battery dry.

Could anyone suggest a good controller that offers over-discharge and over-charge protection?

Thanks once again!
 
Last edited:

Dodgydave

Joined Jun 22, 2012
8,389
Sorry for the double-post, just wanted to update the situation.

It seems that the controller is acting as it should, since the battery is way below what it should be it simply doesn't charge it.. And, unlike the battery specs say so, it does NOT have over-discharging protection, so the microcontroller will just keep trying to suck the battery dry.

Could anyone suggest a good controller that offers over-discharge and over-charge protection?

Thanks once again!

I can recommend the Tp4056 chip.

https://www.homemade-circuits.com/li-ion-battery-charger-circuit-using-ic/
 

Thread Starter

Mywk

Joined May 10, 2018
16
@Dodgydave Thank you for the suggestion. Just had a look at the datasheet, interesting IC but it doesn't seem to have over-discharging protection which is what I can't seem to find an IC with.

Edit: Just had a look at the STBC02 which is way too overkill for what I need (basically over-discharge protection and if possible automatic power path management)
 

ebp

Joined Feb 8, 2018
2,332
If the microcontroller is always connected to the battery it may be that the preconditioning function never gets the battery to a voltage sufficiently high to allow the main charging to start. Preconditioning circuits are implemented to prevent chargers from attempting to charge bad batteries, with potentially disastrous consequences. Preventing over-discharge may or may not solve this.
 

Thread Starter

Mywk

Joined May 10, 2018
16
@ebp I had absolutely no idea, thanks for the insight! What would be your suggestion for solving this issue?
 

Dodgydave

Joined Jun 22, 2012
8,389
@Dodgydave Thank you for the suggestion. Just had a look at the datasheet, interesting IC but it doesn't seem to have over-discharging protection which is what I can't seem to find an IC with.

Edit: Just had a look at the STBC02 which is way too overkill for what I need (basically over-discharge protection and if possible automatic power path management)
Tp4056 has a Trickle Charge threshold of 2.8V and is 120mA, so if the battery gets to this voltage it will initiate charging .
 

Thread Starter

Mywk

Joined May 10, 2018
16
@Dodgydave Yes but the device will not be always connected to power, that means if it's running on battery then the battery will keep discharging until it's way below the recommended minimum charge.
 

Dodgydave

Joined Jun 22, 2012
8,389
@Dodgydave Yes but the device will not be always connected to power, that means if it's running on battery then the battery will keep discharging until it's way below the recommended minimum charge.
Then you sould be using a bms pcb to run the load on.
 

Thread Starter

Mywk

Joined May 10, 2018
16
@Dodgydave So I'l have to add an extra IC for the over-discharging protection? I searched the whole day for charging IC's with that also included a over-discharging protection but couldn't find anything else than the super expensive STBC02.
 

crutschow

Joined Mar 14, 2008
23,349
It's hard on a lithium-ion battery to discharge it to a low voltage.
You should add a low-voltage cutoff circuit to the system to limit the minimum battery voltage to about 3V.
That should also solve your charging problem.

Below is the LTspice simulation of a circuit using a cheap TLV431 programmable reference as a comparator, to cut off the power at <3V.
The circuit changes state at a Ref voltage of 1.24V as determined by the voltage divider consisting of U2 and R3.
Pot U3 adjusts the cutoff voltage.
The P-MOSFET must be a logic-level type device (Vgs max threshold <2V).

upload_2018-9-25_19-41-39.png
 
Last edited:

ebp

Joined Feb 8, 2018
2,332
On rethnking, I realized that low voltage protection should prevent issues with preconditioning since the LV protection circuit disconnects the load, leaving only a tiny leakage current through the FET that switches the load and another tiny current for the LV protection circuit itself. The sum of these will be far below the preconditioning current.

The preconditioning circuit may actually be bringing the cell voltage up but at a rate so low as to be useless. You might try accurately measuring the cell voltage every half hour or so after connecting the charging power supply when the battery is deeply discharged. If it isn't too difficult, you could also try just measuring the net current into/out of the cell.
 

Thread Starter

Mywk

Joined May 10, 2018
16
@crutschow Thanks, that seems like a very simple solution.

I'm considering a BQ296xxx as it seems to offer everything I need in one single IC (even a 3.3v regulated output!!)

Quoting from an article about batteries here:
TI offers the BQ296xxx line of overvoltage protection ICs for 2-4 cells. These devices include a regulated 3.3V output supply that is disabled when any of the cells fall below the undervoltage threshold, thus avoiding over discharging the batteries.
So my problem is, the battery is a 1 single cell battery, would something like this work?

upload_2018-9-27_15-29-40.png

If I can use the above with a single cell battery, I have two questions regarding the circuit:
1- The Load would be where R4 is, is this correct?
2- What would be the recommended MOSFET's for Q1-3 ?
 
Last edited:

crutschow

Joined Mar 14, 2008
23,349
So my problem is, the battery is a 1 single cell battery, would something like this work?
Don't know if it will work with only one cell.
If yes, in that circuit, the Load would be where R4 is right?
No.
The OUT is the overvoltage signal.
The R4 load must go to ground. As shown there's no path to ground for the load current.

If you want to boost the REG 3.3V output current, a MOSFET follower would reduce the voltage by its Vgs threshold voltage.
You need to add an opamp with a transistor follower at the output to boost the current without significantly affecting the voltage.
 
Top