BQ27621-G1: BQ27621YZFT-G1A

mark greenwood

Part Number: BQ27621-G1

hi,

we are seeing a high number of BQ27621YZFT-G1A devices that appear to be failing in the same way.

on the units that fail vdd and vss measure 3-5ohms whereas a unit that works measures 35ohms plus.(device fitted to board)

if the chip is removed the 3-5ohms can be measured between the vdd and vss pins on the device itself.

on the units that fail the unit will power up with a battery connected but there is no vdd output.

I have tried toggling a vdd voltage as suggested on another post but it made no difference.

the boards are assembled in an antistaic environment so I do not believe that it is an esd problem.(which another post suggested can lock up devices).

is it possible that the devices are shipped from the manufacturer with different "states" of the vdd output?

if so is there a command that needs to be issued to remove the device from that state which is giving the low impedance output?

any help or suggestions would be appreciated

over 6 years ago

0 Onyx Ahiakwo over 6 years ago

TI__Guru* 76230 points

Hi Mark,
On the units that are failing, what voltage does Vdd pin measure? Do you have the gpout pin connected to your host on your board? Can you toggle that pin high then low, or low then high depending on what state it is on?

thanks
Onyx

0 mark greenwood over 6 years ago in reply to Onyx Ahiakwo

Prodigy 80 points

hi onyx,

on a good board vdd line to vss measures greater than 30ohms and when battery is applied then vdd is 1.8v.

on a bad board vdd line to vss is around 3-5 ohms (but not dead short). when battery is connected vdd is zero.

by toggling gpout with battery connected by applying 470ohm resistor between bat and gpout vdd rises top about 0,5v and then drops again as if it tries to wake up and then shuts down again.

gpout is not used on the board and is tied high to vdd with a 10k resistor.

I have enclosed a copy of circuit.

0 mark greenwood over 6 years ago in reply to Onyx Ahiakwo

Prodigy 80 points

0 Onyx Ahiakwo over 6 years ago in reply to mark greenwood

TI__Guru* 76230 points

Hi Mark,

You need to toggle the pin from high to low(ground) and not to a tri-state mode. Connecting the gpout to a gpio pin of your host is a requirement as stated in the data sheet for shutdown events such as this. To exit shutdown which is what your chips are experiencing, you have to need the gpout pin to be toggled.

thanks

Onyx

0 mark greenwood over 6 years ago in reply to Onyx Ahiakwo

Prodigy 80 points

hi onyx,

we have tried toggling the pin, it appears to try and start i.e about 0.5v then shuts back down.

my understanding is that the chip should start up on a battery insertion event, with gpout tied to vdd this should control this event by pulling the pin high when the battery is connected.

is there a reason why the chips appear to be shipped in different states?

is there any other way to force the device to wake up?

or do we have a large number of defective devices?

what is interesting is that we have had similar problems on previous builds but as the qty was smaller i.e 3 or 5 we had put it down to defective parts (although potentially from different batches).

or is there something in the design we are missing that is preventing the device from wake up correctly?

0 Onyx Ahiakwo over 6 years ago in reply to mark greenwood

TI__Guru* 76230 points

Hi Mark,
Can you pls send a scope shot of your toggling action? With our other custoers who have experience such failures, Toggpling the gpout pin hi, then low(ground) allows the device exit shut down mode.
thanks
Onyx

0 mark greenwood over 6 years ago in reply to Onyx Ahiakwo

Prodigy 80 points

hi onyx,

please find attached scope pic.

2018.tek00001.tif

0 Onyx Ahiakwo over 6 years ago in reply to mark greenwood

TI__Guru* 76230 points

Hi Mark
Which is the GPOUT pin? I do not see any one of the signals going high to 3.3V

thanks
Onyx

0 mark greenwood over 6 years ago in reply to Onyx Ahiakwo

Prodigy 80 points

hi onyx8814.tek00002.tif tek00003.tif

Input 1 is GPOUT pin.

Input 2 is VDD pin.

Tek00002.tif is when I tried to apply signal (510 Ohm resistor between GPOUT and BAT+) for length less than 100ms. You can see some debouncing.

Tek00003.tif is when I connected resistor and held for a while to have only positive transition on GPOUT pin.

0 Onyx Ahiakwo over 6 years ago in reply to mark greenwood

TI__Guru* 76230 points

Hi Mark
How many units are seeing this behavior? Does unplugging the battery and relpugging cause the device to recover? Toggling the gpout pin high then low worked for our other customers that experienced similar issues. I am not certain why it isn't working for you.

thanks
Onyx

0 mark greenwood over 6 years ago in reply to Onyx Ahiakwo

Prodigy 80 points

hi,

on the current batch of 70 boards there are approx. 40 with the problem.

the datecode of the batch is 1523

from philippines

lot numbers 5582050cl5,558205665,5582049cl5.

is there anything else we can try?

we do not want to rework these boards if there is no need.

we have similar failiure % on previous builds although smaller qty.

could it be design related?

0 Onyx Ahiakwo over 6 years ago in reply to mark greenwood

TI__Guru* 76230 points

Hi Mark

Can you try holding the clock and data lines low for 2s and then releasing to see if that helps? I am not sure how else to help you here.

thanks
Onyx

0 mark greenwood over 6 years ago in reply to Onyx Ahiakwo

Prodigy 80 points

hi onyx,

Holding both I2C lines for at least 2 seconds doesn’t make any difference. I tried to do it before triggering GPOUT and after. Still the same behaviour on VDD line and no answer from the chip over I2C..

0 Onyx Ahiakwo over 6 years ago in reply to mark greenwood

TI__Guru* 76230 points

Hi Mark,
I am sorry, I don't know what else to try. you will want to submit some ics to TI failure analys8s
thinks
Onyx

Power management

Power management forum

BQ27621-G1: BQ27621YZFT-G1A