BQ2057C: BQ2057SN not charging

Hello Leon,

We will review your questions and schematic and respond by the end of the day.

Sincerely,

Wyatt Keller

Hi Leon

The problem could be Q1 and heatsink, the device could dissipate a lot of power. The IRF7410 looks like an SO8 package with no thermal pad.  Charge current is set for 600mA by 0.20 ohm R-SNS resistor. Depending on input voltage and battery voltage the power dissipation on Q1 could be high.

Check Q1 if it has been damaged.   A device with better thermal dissipation is recommended.

Hi Bill,

I want to charge a 1-cell LiPo with 4.2 V and 2500mAh. A USB-C port with the default setting of 5V 500mA is used as the power source.

Could the problem also be that the IRF7410 is a P-channel mosfet and actually needs a pull-up from source to gate?

What is the best way to test whether the mosfet is faulty when installed?
The only thing I can think of at the moment is to use a multimeter to test the resistors when switched off.

Best regards
Leon

I don't think the problem lies in the heatsink of the IRF7410.

I found the following formula in the data sheet of the BQ2057:
PD = (Vi-Vd-Vcs-Vbat)*I.

Vi: is limited to 5V
Vd: According to the IRF7410 data sheet, the diode forward voltage is -1.2V, i.e. 1.2VVcs 0.5A*0.2Ohm = 0.1V
Vbat not below 3.3V

This gives PD 0.2W, which results in a theta in the operating range of 40-150°C of eta 550 °C/W.
The IRF7410 has a specified theta of 50 °C/W, which is 11 times lower than the calculated theta.

Or did I miss something in the calculation?

But yes, I have already thought about damage to the BQ2057 or IRF7410, but I thought it was more likely to be due to the missing pull-up, as the BQ2057 pulls the CC pin to GND via a transistor and the mosfet cannot be switched off and theoretically cannot be switched on at all.

HI Leon

Yes will need a pull up on the Gate, see below figure from datasheet for reference.

Take another look at the power dissipation on Q1.  The VD term will apply to D1 in below figure, it is about 0.4V .  Your schematic does not show this diode.

PD = (Vi-Vd-Vcs-Vbat) *I= (5V - 0V - 0.1V - 3.3V)*0.5A=0.8W   This may be a problem.  

The resistor in the pull-up is only there to limit the current, isn't it?

Can I use the existing resistor as shown in the circuit diagram as a pull-up resistor or do I need an additional resistor in the connection?

I would recommend the pull up resistor configuration from data sheet figure 10-4.

Thank you, I will try that.

and about your idea: do you have an idea how I can test if the mosfet is broken?

Most common mosfet fail is gate damage, shot to drain or source.  Use ohm meter to check, should be open or very hi-Z to drain and source.

The connection from gate to source and to drain each has about 10MOhm. Does this mean that the mosfet is defective or is this normal behaviour?

We have now installed the pull-up and measured the voltages at the gate, source and drain. The pull-up is a 1k resistor between pin 1 of RSNS and pin 2 of R13.
Due to the pull-up, we have between 3.3 and 3.6V everywhere.

However, the LED on the status pin of the BQ2057 is still not lit. I then also measured the voltage at the LED and it was around 0.5V, whereby the LED started to light up slightly when the multimeter was applied. So I wouldn't completely trust the measured value.

I now assume that the BQ2057 is either in sleep/temperature error mode or defective. Is there a way to measure the BQ2057 and check whether it is defective?

Is it possible that the BQ2057 was damaged when I switched on the load during charging?

I read in an old forum post of yours that someone should measure the pins of the BQ:
I did that once with the board where it no longer works and a board that has never been loaded:
non-functioning board: e2e.ti.com/.../4763776
Pin 1: 16k
Pin 2: 60k
Pin 3: 16k
Pin 4: 7.3k
Pin 5: 10M
Pin 6: 0.3
Pin 7: 18k
Pin 8: 16k
never used board:
Pin 1: 16k
Pin 2: 83k
Pin 3: 16k
Pin 4: 7.3k
Pin 5: 10M
Pin 6: 0.3
Pin 7: 5.5k-11.6M (I measured several times and came up with different values)
Pin 8: 16k
Do you think the BQ is broken or is the problem somewhere else?

I have also recalculated the setting of the NTC via RT1 and RT2 and this also fits. We should be able to charge between 10 and 45 °C.

So I have continued testing.
Since the STAT pin measured with an oscilloscope showed 0V, I assumed that the BQ2057 was broken and replaced it.

However, the behaviour is the same as before. I therefore measured all the voltages on the BQ2057 pins:
SNS 3.37V, BAT 3.68V, Vcc 3.37V, TS 1.52V, COMP 3.37V, CC 3.37V, Vss 0V, STAT 0V.

I think the problem is that the battery voltage is above the Vcc voltage and therefore the BQ2057 goes into sleep mode /off.

Unfortunately, I don't quite understand why the voltage of USB-C drops from 5V to 3.37V. When the battery is not connected, the Vcc voltage is 5.21V. Could it be that something has broken in the mosfet or the freewheeling diode could be the problem?

Hi Leon

Looking at the voltage drop from 5V to 3.37V with no battery voltage drop would point to a current overload.

1.) Voltage drop across the SNS resistor is 0V so current is not go to the mosfet

2.) Voltage on MOSFET gate is high, it is turned off.

The ohm check does not point to problem.

Have you tested with a power supply that would indicate current?

Hi Leon

The BQ2057 family of devices is very old device we have newer and better devices.  Have you looked at BQ2404X family or BQ25170 family?

Yes, I've just done that and since then the loading has worked. I am now continuing to test whether loading is also completed and whether I can reproduce this error.

I'll have a look at the suggested components for future boards, but the board has already been ordered and is here, so I'll try to use it.