BQ25792: Help with initiating charge

Hi Daniel,

I don't see any obvious issues on the schematic.  I don't understand " 0.2ohm resistor before the BQ chip".  Can you elaborate?  What is the voltage that I IC sees at BATP pin? 

Did you follow the PCB layout example in datasheet and user's guide, especially for PMID and SYS 0.1uF caps?  

Lastly, when you move to the 4S battery, you will need a TVS zener diode on BAT/BATP to prevent the 20V internal BATFET from damage at battery hotplug.

Regards,

Jeff

The positive battery connection first passes through a 0.2 ohm current sensing resistor so that a seperate IC can keep track of battery capacity by measuring the tiny voltage drop across the resistor. My multimeter isn't accurate enough to detect a voltage change from one side of the resistor to the other without any load on the system so I'm still seeing the same voltage that's at the battery terminals (7.7v). Yes I've copied the example layout for most of the design so hopefully that should be fine.

I didn't realise I needed a zener diode on BAT/BATP. What about on the input side of things? I'm supplying 20v to Vin.
I have already had a 4s battery connected before I realised that I had the incorrect resistor value choosen. I guess that means I could've just fried the IC already, explaining why nothing is working. I'll get the i2c hooked up as you suggested and try to work out how to read some registers. Which ones would you recommend reading for debugging?

Cheers,

- Daniel

Hi Daniel,

The 0.2 ohm is okay but will cause to measure V(PACK+) +ICHG*0.2ohm at BATP.  This means the charger enters CV mode early and has a long taper time down to ITERM.  

On the input, you have the 2ohm, 2.2uF snubber which is sufficient.

If fried, the unpowered IC usually shows either SYS or BAT with a low resistance to ground or SYS = BAT. For the registers, you can start with the system and fault registers, 0x1B to 0x27.

Regards,

Jeff

So the final stages of charging may just take a little longer? Assuming that its not a dramatic change that should be fine for my purposes I think.

I will test the resistance from SYS and BAT to ground this afternoon to confirm, but I suspect they are still okay as I have been periodically checking those two (as well as VBUS) against ground to ensure I haven't created a short the few times I've re-soldered the component.
I'll read those registers and let you know the results.

Cheers,

- Daniel

I've had no luck getting anything out of the I2C connection so far. I have an arduino board hooked up to the I2C pins with external 10k pullup resistors but I'm getting no response at the 0x6B address listed in the datasheet (or any others, for that matter). SYS and BAT don't have low resistance to ground however I do get the full pack voltage at SYS as soon as I connect a battery. I think  that would be expected behaviour as long as the pack was within the correct range of voltages? I've tested for continuity between BAT and SYS with no power connected and I get nothing. The stat pin is oscillating from roughly 0 to 0.3v when compared to ground at roughly 1 pulse per second. I've only got a multimeter so I can't connect this up to a scope to get accurate readings at the moment unfortunately although I might be able to borrow a scope in the coming days if its neccessary. Any other ideas?

Appreciate the help so far!

- Daniel

Hi Daniel,

There is nothing special about the I2C so the arduino should be able to read the I2C registers.  I can't help further without register settings or oscope plots.

Regards,

Jeff

No worries, I'll have to see if I can get my hands on an oscope. What should I be checking with it?

Cheers,

- Daniel

Hi Daniel,

Let's start with VBUS, SW1, SW2 and SYS.

Regards,

Jeff

Sorry about the lengthy delay in response. I've found a handful of issues with my design in the past week and a bit. I've been able to resolve most of the issues however I've just discovered that the i2c is actually not routed, which obviously explains the lack of communications. I've also replaced the IC with a brand new one just in case the initial power on via 4s battery without the TVS diode killed the original. I did manage to get my hands on an oscilloscope although its very old and I'm still trying to learn how to how to interpret the results properly but I do have a photo of the waveform I got from it. This same waveform was present when I checked all of the pins you suggested. This reading was taken with 7.7v on the battery input and 20v from the charger on VBUS. The waveform wasn't present when the charger was disconnected.

Obviously I'm at the stage now where this requires a re-design and a new PCB but I'd really like to try and get it at least charging a battery before I commit to another set of PCB's so I know my mistake is sorted. The only other thing I've been able to find is that I don't have D+ and D- connected to anything because I didn't realise this IC could handle the USB PD handshake so I've got another IC in use for that currently and I'm just supplying 20v to the BQ. Would this lack of a connection on the usb data lines be causing any issues?

Could I use the INT pin to determine what is going on? I do have that one broken out so I can easily connect up to it.

Thanks for the help so far, greatly appreciate it!

- Daniel

Hi Daniel,

I do not understand the oscope plot. Is that AC coupled? The datasheet has example plots that you can try to emulate.

Regarding INT, it reports when the status or fault registers change.  If you want to use it to know when to read those registers, it might help.

Regards,

Jeff

Hi Jeff, sorry I think that might've been a bad contact on the probe connection as I've been unable to replicate this reading. I may just not be using the oscope properly but I haven't been able to get any sort of waveform from any of the sections you've asked me to test. I may see if I can track down another scope just to prove its not at fault but in the mean-time I'm looking over my schematic and PCB layout to see if I can find anything that might be causing my issues.

I did have the 4.7uF REGN capacitor a fair distance from the BQ chip so I don't know if that would've caused an issue like this? I'm also going to swap out the Inductor for the one that's listed on page 130 in case the one I chose isn't up to the job. The inductor I originally picked has 32mohms of resistance as opposed to the 7.8 of the suggested one and its also unshielded which may not be helping matters either. This is the inductor I'm currently using: https://www.digikey.com.au/en/products/detail/tai-tech-advanced-electronics-co-ltd/TMIM-322512A-1R0MG/17835491
The description says shielded but I just checked the datasheet and realised its actually unshielded so I guess the description is incorrect. I did initially have the CE pin floating too because I haven't fitted the micro controller to the PCB yet but I've wired it to ground for now so it should be fine.

I did notice the thermistor I picked is PTC not NTC but at room temperature it has the correct resistance so it shouldn't be causing this issue either.

Besides these problems and the unrouted I2C lines, I haven't yet been able to find anything that seems like it should cause the problems I'm having, unless the inductor is a bigger problem than I understand?

Cheers,

- Daniel

Hi Daniel,

The inductor with higher DCR will likely only cause bad efficiency.  I can review your layout if you like.  Your ILIM_HIZ resistors are setting input current a bit low.  Is there enough input current to power both your system and charge the battery?  If you force a TS fault by removing the thermistor to disable charging, does SYS regulate as expected?

Regards,

Jeff

I will test both of these later today and let you know the results, thanks!

Hi Jeff, I tried removing the ILIM_HIZ resistors (R3 and R4) and connecting that pin directly to REGN as the datasheet says that's an acceptable way to set the input current to maximum. I'm now getting about 8.5v on SYS which is pretty close to what I'm expecting for a fully charged 2s pack so I assume that side of things is now working correctly. The battery still doesn't appear to be getting charged at all though. I left it all connected for around 10 minutes and checked the battery voltage again and it hadn't changed at all so it's not even trickle charging. I'll organise some pictures of the layout for you tomorrow and send them over, but please let me know if you think of anything else that could be preventing it from initiating a charge. I've double checked that the CE pin is pulled to ground and I've had yet another go at resoldering the IC. I've also checked to ensure the thermistor is providing a resistance within the appropriate range. It's sitting at around 7kohms at ambient temperature so that shouldn't prevent it from initiating a charge.

Huge thanks for all the help so far and for putting up with my stupid questions. I'm determined to get this working!

 Cheers,

-Daniel