bq24250 VIN_DPM in SA mode

You have to write to one of the charger's I2C registers in order to enter HOST mode.

Is the voltage at the charger IC's IN pin 4.7V or is that the voltage at the power source?  There could be a voltage drop from the power source to the IC.

Yes, that was my first instinct. But, as I lower the input voltage from 5V, the current draw from the source goes to zero around 4.7V.

I plan to query the register to see where the VDPM is set, and to see if changing it has an effect. I just need some firmware time...

I found an Easter Egg! After setting up the I2C link I discovered the problem. The chip operation is flawed. There is a discrepancy between the datasheet and the actual operation of the VDPM level.

Reading register #5 on cold restart shows xxxx x010 for the VDPM bits. This is the expected default value. However, testing the VDPM reveals the actual set point to be 4.68V. Suggesting a difference between working register and data register, or something like that.

Writing xxxx x010 to reg #5 changes the actual VDPM to 4.36V (where it should have been all along), and reading the register is still xxxx x010.

Writing xxxx x110 to reg #5 changes the VDPM to 4.68V (where it was at power-up), and reading the register confirms it is still xxxx x110. Likewise the VDPM now correctly follows and reports all other VDPM settings.

This undocumented operation happens after a cold reset, OR after sending the "all register reset" command to resister #2.

Are there any other register values that don't operate as described in the datasheet?

Regards, Bob

Hello, Bob,

I checked this on bench, and here is what I saw.

- I shorted VDPM pin to ground. With cold start, the charge current dropped to 20mA around 4.36V, which is the same as what DS described.

- I sent reset command to reg01, the charge current dropped to 20mA at 4.36V again.

- I programmed VDPM to 4.68 through I2C and the current dropped to 20mA around 4.69V.

Can you share your schematic so that we can take a further look?

Thanks,

Wenjia

Hi Wenjia,

Thanks for looking into this. Here is the chip I'm using along with the circuit. Is this the latest rev of the chip?

I need to get back to the lab to re-run the tests and give you a step by step of what I've found. I'll post back at that time.

I don't see the .jpg photo or schematic in my last post, did these get filtered out? How do I post a schematic?

Hi, Bob,

You can send me your schematic by email.

Thanks,

Wenjia

Not sure exactly how your setup differs from mine, I trust you received my schematic?

I just tried this simple approach in the lab:

Variable supply at Vin (VBUS2 on my sch.), 100 ohm resistor from Vsys to GND, battery and thermistor open circuit, VDPM pin grounded, ILim resistor 453 ohm.

Measure voltage at Vsys while increasing/decreasing voltage at VBUS2.

With my circuit Vsys measures near zero with Vin between 0V and 4.70V. At Vin at 4.71V to 5V Vsys jumps to 4.3V. It drops back to zero when Vin is reduced to 4.7V and below.

I'm not able read or alter the registers as this bare bone circuit does not have I2C connected.

I'm curious to hear what you find.

Bob,

I received your schematic and will reply through email.

Thanks,

Wenjia

In our implementation the BQ24250 is in SA mode with EN1=1, EN2=0. This allows both input current limit and VDPM to be externally programmable with resistors. I selected a resistor for ICL but chose to save two resistors by grounding the VDPM pin expecting the a 4.36V threshold. I haven't tested it yet but I suspect that setting VDPM with resistors to 4.36V would give the desired operation.

This seems reasonable doesn't it?

Bob,

I checked this on bench using two resistors to program the VDPM to 4.36V. With the two resistors I used, which are not very accurate, I was able to program the VDPM to 4.4V. So I think you will be able to use the same EN1/EN2 setting and the external resistors to get the 4.36V DPM threshold.

Thanks,

Wenjia