BQ33100EVM-001: Communcation Error in BQ33100EVM-001

Hello Alex,

Thank you for such a quick response. Can you let me know where exactly do I have to input the "REG27". In the SBS window write 2 word command, or in the Pro Window. Yes I have connected power to the system_PWR to power the IC externally, still no response.

Regards

Yeshwendra Singh

Hello Yeshwendra,,

By REG27 I was referring to the pin on the device. It is the LDO internal power supply which is exposed on an external capacitor. On the EVM it is available on TP4. If you see 2.7V, it indicates that the device is not in shutdown mode. 

Otherwise, I would suggest checking the SMBUS lines (SCL/SDA) to check if there is communication occurring. 

Also, it is worth checking that the EV2300/EV2400 is all connected properly and verifying their functionality if you can. Since you applied system_PWR already, I suspect the issue lies with communication rather than the gauge being shutdown. 

Thanks,

Alex M.

Hello Alex,

Thank you for the much needed guidance. I checked the voltage between the TP4 and system GND, it shows ~ 2.7 V. I also checked the continuity between the SCL/SDA (between the pin 14 & 16 with the SMBD & SMBC at J1). There is continuity. I tried changing the USB cable which connects my laptop with the EV2400. It is now displaying the error "No Acknowledge: VB_T2H_NACK" & "SMB BAD PEC: VB_BAD_PEC". I am attaching the screeshots for the same.

Currently, I am supplying 8.5 V from my power supply to the system PWR for the 5 supercapacitors connected in series. Due to some reason, the voltage between TB3 1N & TB2 5P is ~ 0 V (5 mV). The same is at TB4 between CAPOUT & VSS (5mV).

Requesting you to please assist me on this issue. Much thanks for such quick response.

Best Regards

Yeshwendra Singh

Hello Yeshwendra,

It looks like maybe some of the connections are bad based on those voltage measurements. One thing that can often be helpful to check, is to use a power supply and a voltage divider instead of supercaps to rule out any issues with the caps. It may be something as simple as a loose GND connection on 1N that causes the issues.

Also, checking continuity on SDA and SCL is good, but if you have a logic analyzer or scope you can see if there is actually communication occurring. On your EV2400, are you using the SMbus port connection to connect to the board?

Thanks,

Alex M.

Hello Alexander,

Apologies for the delayed response. We tried using other commercially available supercapacitors instead of our experimental coin cells. But, the result remained the same. The EVM module displayed the same error as I mentioned in the earlier reply. Unfortunately, we do not have a logic analyzer or scope with which we can rule out the communication error.

By any chance, do you think their is any issue with the BQ33100 EVM itself? If so then we can proceed the replacement, if possible. Kindly suggest.

Best Regards

Yeshwendra Singh

Hello Yeshwendra,

Thanks for the update. All of our EVMs are tested by the manufacturer prior to being listed/sold. It is still possible for something like ESD to damage the board though. In my experience, it is more common that something bad gets shorted while connecting everything that damages the board. 

I still think it is best to start with a voltage divider and a power supply before moving onto cells/caps. The fact the REG27 is showing 2.7V is a good sign the board is alive. I assume you have done it like this, but can you confirm you did this procedure? 

The reason I like to use the resistor divider, is that you can easily tell what state the device is in by reading the current draw from the power supply.

There are a lot of connections to make with 5s, if even one connection to the terminal block is bad, it will cause problems. You mentioned previously that the voltage differentials between the pins were wrong, this could be a reason why. 

As a last suggestion, are you running BQevsw as an administrator? I have seen it cause issues in the past (but not this exact error).

thanks,

Alex M.

Hello Alexander,

Thank you for the guidance. As soon as I test the PCB with the voltage divider, I shall write you back with the results. Hope that this method lets you identify what is exactly wrong with the BQ33100 EVM.

Best Regards

Yeshwendra Singh

Hello Alexander,

Yes I did proceed in the similar manner as mentioned in the User Manual of the EVM.

Hello Alexander,

Do you suggest any particular resistor divider or any rating would work fine. I am attaching a link below for resistor / voltage divider by CADDOCK which has a rating of 1K to 9M ohm. Or else, if you have some suggestion at your end regarding which works best, please suggest.

1776-C681 Caddock, RESISTOR, VOLTAGE DIVIDER, 1K TO 9M | Farnell EN

Best Regards

Yeshwendra Singh

Hello Yeshwendra,

The value of the resistors doesn't matter much, anything from 1k to 100k would be fine. The issue with going higher is that the variation between the voltages will increase. By the way, you don't need a special resistor divider product, you could just use through hole resistors. I don't know the spacing of the pins on the product you listed, but if they don't match the spacing of the connectors on the board it could be really difficult to use here.

Thanks,

Alex M.

Hello Alexander,

Thank you for your recommendation. I do have spare through hole 30 k ohm resistors. I shall use them. Should I check the current draw for each resistor by connecting the five resistors in the same manner we connect the five supercapacitors or should it be one by one? The TB2 is closely stacked but I shall try connecting them.

Best Regards

Yeshwendra Singh

Hello Yeshwendra,

I would connect all of the resistors in the 5s configuration, then apply the nominal voltage you want to use multiplied by five on the whole stack. So if you wanted 2.7V, you can apply 13.5V across the entire stack and each resistor will be at roughly 2.7V and mimic the capacitor voltage for the gauge. The only way to this is simultaneously, which is fine. 

As for current, you will have 30K*5 which is 150K. With 13.5V, the current draw from the divider is only roughly 90μA, so the gauge's current draw should be easy to see in comparison. If your power supply's current resolution is not good, you can attach an Ammeter in series also. 

Thanks,

Alex M.

Hello Alexander,

Thank you very much for providing me this valuable information. Surely, I shall check the current draw by the gauge in the same manner as you have suggested. I shall get back to you with the results by friday after I perform this test on the EVM.

Thanks

Yeshwendra Singh

Hello Alexander,

I did proceeded as per your instructions. The power we use has a resolution of upto two decimal places. Niether the power supply nor the ammeter in series did gave a current draw reading, it only showed 0. I checked the voltage across the five 30 K ohm resistors. The voltage is same like the supercapacitor, in mV (around 5mV). The REG27 still shows 2.7 volts.

I am not sure as to what this means. Should we opt for a new BQ33100EVM or can we send it to Texas Instruments to diagnose what is wrong.

Awaiting your further suggestions.

Best Regards

Yeshwendra Singh

Hello Yeshwendra,

It sounds like your DMM may not have enough resolution to measure the current, what is the exact resolution for the measurement equipment you are using? Do you mean the voltage is at the same voltage of the super capacitors +/-5mV?

The EVM may have been damaged during testing or from other causes and it would be very hard to determine this.

Sincerely,

Wyatt Keller

Hello Wyatt,

Yes the resistors are showing the same voltage as in the super capacitors, i.e. 5 mV. Kindly let me know what should be done in case the EVM is damaged. Will Texas Instrument provide any assistance in regards to this repair. How should we proceed ?

Best Regards

Yeshwendra Singh

Hello Wyatt,

Thank you for the guidance. I shall contact the customer service then.

Best Regards

Yeshwendra Singh