BQ24014 STAT1 periodic blinking/reset or sleep loop

I don't think I understand the situation.

You have two chargers on the PCB and plan to connect to two separate batteries? When you connect a battery to one of the charger, the BAT pin of the other charger goes up to 5V and STAT1 starts blinking. Once you connect a second battery to the second charger, the blinking stops. Is this correct? Can you please further explain what was happening with charger1, charger2, battery1, battery2?

Do you have a schematic that can be shared with us?

Thanks,

Wenjia

Hi,

yes you got it right. but here is the flow for clarity:

1. plug in 5V supply (2A charger, or desk psu)

2.connect battery 1 to charger 1

3. with high probability at this point the charger 2 will go into a reboot loop led blinks like it does on power on, but this time repeatedly every 570ms

4. disconnect battery 1 from charger 1 and with a 1 in 20 chance the 2nd charger will go into the same loop

5. reconnect either batteries and they will change like nothing happened and everything seems normal

but again on disconnecting battery 1 and 2 can cause this random lockup state with the blinks.

And this state of the chip doesn't seem to be described in the reference manual, because BAT goes to VCC when blinking happens.

(as a test we tried a quick short on the output and that removes the chips from this blinking state, but for example raising the CE# will only suspend the blinking until it is high, but on release they continue.

I attach schema.

thank you for the prompt answer

Thanks for the clarification. I have some followup questions about the flow you described. Please see below:

1. plug in 5V supply (2A charger, or desk psu)

ok

2.connect battery 1 to charger 1

ok

3. with high probability at this point the charger 2 will go into a reboot loop led blinks like it does on power on, but this time repeatedly every 570ms

at this point, only battery 1 is connected to charger 1 and no battery is connected to battery 2, is it correct？Charger 1 is charging fine, charger 2 is the one that has this blinking issue.

4. disconnect battery 1 from charger 1 and with a 1 in 20 chance the 2nd charger will go into the same loop

do you mean that after battery 1 is disconnected, 1 out of 20 times charger 2 will remain in the led blinking state? What about the other 19 times? Will it be back to normal?

at this point, there should be no battery connected to either charger.

5. reconnect either batteries and they will change like nothing happened and everything seems normal

In this case, are you connecting both batteries or just one of them? What is the difference between this condition and the condition in step2?

but again on disconnecting battery 1 and 2 can cause this random lockup state with the blinks.

Do you mean even without a battery connected to either of the charger, the led will still blink? One of them or both?

And this state of the chip doesn't seem to be described in the reference manual, because BAT goes to VCC when blinking happens.

When battery 1 is connected to charger 1, is the BAT pin of charger 1 or charger 2 going to VCC?

Is the 5V input power the only thing that is shared by the two charger circuits?

Thanks,

Wenjia

Wenjia Liu said:

Thanks for the clarification. I have some followup questions about the flow you described. Please see below:

1. plug in 5V supply (2A charger, or desk psu)

ok

2.connect battery 1 to charger 1

ok

3. with high probability at this point the charger 2 will go into a reboot loop led blinks like it does on power on, but this time repeatedly every 570ms

at this point, only battery 1 is connected to charger 1 and no battery is connected to battery 2, is it correct？Charger 1 is charging fine, charger 2 is the one that has this blinking issue.

That is correct. since when they have batteries connected they are charging them. This blinking can only happen when there is no battery on the charger.

4. disconnect battery 1 from charger 1 and with a 1 in 20 chance the 2nd charger will go into the same loop

do you mean that after battery 1 is disconnected, 1 out of 20 times charger 2 will remain in the led blinking state? What about the other 19 times? Will it be back to normal?

What I mean is once a charger gets into blinking only 3 things we found can make it stop.

-full power down

-short the out pin

-connect a battery

the 1:20 ratio is just that the charger1 can get into the exact same blinking state independently from the charger2s state as well on battery removal.

at this point, there should be no battery connected to either charger.

that's correct, and I am either sitting with a none blinking, either or both blinking. :)

5. reconnect either batteries and they will change like nothing happened and everything seems normal

In this case, are you connecting both batteries or just one of them? What is the difference between this condition and the condition in step2?

What I meant is that the charger will start regardless of it's state if you reconnect batteries to the outs.

but again on disconnecting battery 1 and 2 can cause this random lockup state with the blinks.

Do you mean even without a battery connected to either of the charger, the led will still blink? One of them or both?

yes, 1-2 1+2 or none

And this state of the chip doesn't seem to be described in the reference manual, because BAT goes to VCC when blinking happens.

When battery 1 is connected to charger 1, is the BAT pin of charger 1 or charger 2 going to VCC?

charger2 if it goes blinking, otherwise it is doing the slow sweep battery detect if I remember correctly.

Is the 5V input power the only thing that is shared by the two charger circuits?

Yes that's the only thing shared. I don't know if it is relevant but this blinking never happened if we had the scope probe hanging on the output of charger2 while connecting-disconnecting charger1+batt1 combo. Also using higher VCC it doesn't happen either. (but these might be coincidental, not much tested)

Thank you so much bearing with me, it is hard to explain, but tomorrow I can upload a video if you want showing what is happening.

Levente

This is good information. Can you also measure the signal at the IN pin of both chargers? A video will definitely help.

Thanks

Wenjia

So we spent today debugging, and here is a new information:

-the blinking happens if the VCC drops for a short burst under 4.2V, if VCC always stay above it no blinking happens

-IN pins are on VCC both connected together and they have short drops on battery1 connections, depending on PSUs we have seen drops to between 4.1V and 4.7V. This is how we concluded the previous point.

Here is the video:

We think that it has to be som glitch in the internal state machine in cases where the battery detect signal gets higher than the VCC for a short time. (and then of course as it happens and blinking state starts, in which case it will always be happening, since the battery detect sweep seize to work and the out goes to VCC)

Any ideas?

Thank you 

Levente

Hey,

no the drop is not independent, it happens right after the battery is connected, but only for a very short time, like a transient, and it happens with all the power supplies, just the amount is different (I suspect maybe some sort of inrush current?)

Well the 5V supplies limits are their rated currents 1.5A, 2A, and tested even with a desktop PSU @ 8A, all suffered from that short drop. We haven't tested the charging current if it is at 500mA, but when we conenct both batteries we see 1A flow from the PSU so I assume it is charging both at 500mA as set by the ISET res.

No load, except the battery being connected.

That happens when it starts blinking, and it stays there (no longer does the sweep for battery connect detection), and it is exactly VCC 5V.

Levente

Wenjia,

We have added two big capacitors on the Vin, but it didn't really make any difference unfortunately. (we have some size restrictions, so no big elcos could fit)

Yeah, we figured that the Out should never go there, that's why I contacted you :) And it only goes there when the chip gets stuck in an invalid state in its internal state machine. And it only can be triggered in this kind of dual chip setups, because the state machine resets correctly on battery insertion.

I will have to check if we have two development kits from TI on this charger and I can try to connect those two together to reproduce the problem. 

For now I am out of ideas how to make it go away, but I think for now we are going to give specific PSU recommendation to the customer, and then this will not surface.

Thanks

Levente