DRV8842 nFault Status

Hi Kevin,

What is the VM Voltage you're supplying to the device? Are you seeing the nFAULT pin driven low during your testing?

Hi Phil,

Thank you for your prompt response. I am using 12V rail. The nFAULT pin is low during the test. The temperature of the chip is around 55°C, so I think it's not a thermal problem.

Kevin

Hi Kevin,

Thanks for the information. Are you able to measure the output current going to the TEC with a current probe?

The OCP on the device has a tblank time associated with it that the driver will ignore the current sense resister voltage. This is intended to allow the current due to the inductance of the motor to settle before the driver begins limiting it. I suspect you maybe be hitting the OCP threshold in that tblank time on some of the devices.

Hi Phil,

Thank you for your response. For current, I can directly see the current output from the power supply. That's why I can read 0.45A and 2.4A for each driver bank. 

If I understand OCP correctly, the tblank time of OCP is in nano second level, but what I experienced is after 10s, the output current ramps up from 0.45A to 2.4A.

Sorry I didn't capture the voltage scope. In this case, I found the output voltage on the driver has 21V spikes (0.45A output current at the same time), and then go to 12V and Low (2.4A output current which is normal). I applied 100% duty-cycle PWM as my input. In other tests, there is no current through the TECs if I use other duty-cycle (50%).

I found out the bulk cap(220uF) is a key factor for motor driver, what is the purpose for the bulk cap? Is it critical for TEC circuit?

I'll appreciate your suggestions.

Thank you,

Kevin

Hi Kevin,

Thanks for the additional information.

When the driver begins switching the TEC it will require a large amount of instantaneous current. Any stray inductance between the power supply and motor driver will limit how quickly the current can change in the system. If the local bulk capacitance is to small, the VM voltage will change. This can result in the driver flagging an Under Voltage Lockout (UVLO) fault until the voltage rises above approx. 8.2 V.

Page 16 of the DRV8842 datasheet outlines bulk capacitance design considerations.

I would suggest adding the 220uF Bulk to your design and redoing these tests. You may need more bulk depending on the amount of inductance between your driver and power supply, as well as how much instantaneous current the TEC requires.

Hi Phil,

Thank you for your valuable information. I have already put a 470uF Bulk capacitance, it helps smooth the circuit. Now I find out if I increase the VREF value from 3.3V to 7V, the output current raises up to 2.4A and the captured scope is perfect! But I know 7V is over the max rating for Vref(4V), so I changed the Isense resistor from 0.1ohm to 0.03ohm(3*0.1ohm in parallel), I4:0 are all high. The result is also great.

Now I am confident that the circuit is working fine. But I am still curious why the previous case happened? Equation for current calculation is Vref/(5*Rsen)? I also attach the captured scope on OUT1 & OUT2 in error status (0.45A case)for your review. Any suggestion would be appreciated!

Thank you,

Kevin

Hi,,,i am a new user. As per my experience i am agree with this thing that the OCP on the device has a tblank time associated with it that the driver will ignore the current sense resister voltage. This is intended to allow the current due to the inductance of the motor to settle before the driver begins limiting it. I suspect you maybe be hitting the OCP threshold in that tblank time on some of the devices.

board assembly

Thank you for your information! That makes a lot of sense for the differences between TEC module and Motor.