UCC27201A malfunctioning in motor controller application

Hello Tim,

I'm sorry that you're experiencing this trouble. I've alerted the Drivers group and someone there should be able to address your issue.
Since this is received on a late Friday afternoon, I'm not sure how soon you will receive a response. Maybe not until Monday.
In that possibility, please examine the inputs and outputs voltages of the driver and ensure that they are not being driven outside of their min/max ratings per the UCC27201A datasheet. Look for large voltage spikes, for instance, which could result from stray inductances in a high-power system. Use low-noise tip & ring probing to avoid GND-noise pickup such as when using a long GND "pigtail" on the scope probe.

Regards,
Ulrich

Hi Tim,

To help us diagnose this issue, please provide oscilloscope screenshots for the input (HI/LI) and output (HO/HS/ and LO) pins? Negative voltage transients that exceed the datasheet limits on HS or either input pin could lead to device failure. A small R-C (50-100ohm, 10-100pF) circuit can be used at the input pins to avoid noise or other transients.

If you still have the devices that have a VDD to VSS short, please confirm which (if any) other pins are shorted to VDD and/or VSS. This will help narrow down the possible source of malfunction.

With regards to layout, hopefully the board has ceramic bypass caps across VDD-VSS and HB-HS. Feel free to post the layout portion near the UCC27201A and FETs.

Please let us know the items above so we can continue working on this.

Regards,

- Daniel

Hi Tim,

Has this problem been resolved? I’m seeing the same type of failure with the UCC27201A. The power section design is similar to figure 22 in the data sheet, but with three phases for a motor driver. When I increase the supply voltage above 24V with about 25% PWM duty cycle, a random driver fails. Typically, the driver shorts the 12V supply and the low side output shorts to ground. Sometimes the driver doesn’t short the 12V and the low side output is open. Replacing the driver fixes the circuit. This is a hand-wired prototype, so I suspect poor grounding and transients are the cause. The 12V supply is a TPS61093 and the FETs are STL140N6F7s.

Regards,
Alan

Hey Alan,

No we have not resolved our problem. We had to ship the controller to competition and it is now on its way back. We found that keeping the voltage lower (below 40V) reduced the chances of the problem occurring but did not notice a correlation with duty cycle. After examining the chips we found that the HB pin and the supply pin were shorted to ground. We did have the same result that unless the FETs were destroyed replacing the chip fixes the problem and everything goes back to running.

We are waiting to get our equipment back from shipping so we can send debugging images to TI. We are working on a theory that we are getting voltage doubling for some reason (maybe reflected waves?) which would put the HB pin above the absolute rating when the high side is active.

Tim

Tim,

Thanks for the reply. It sounds like a different failure mode, but maybe related. I didn’t find any shorted HB pins. There is a lot of ringing on the motor phases, so that could be causing trouble. The FETs have failed a few times, though none since I added an 18V zener to the 12V supply. I’ve seen the 12V supply rise quickly when a driver fails. That would explain the shorted FET gates.

Alan

Tim, I regret hearing that your issue has not yet been resolved. Please do provide as much info as possible, such as layout and scope screenshots, once you have your board available. 

Alan, for better tracking, would you mind starting a thread and provide details, such as requested in my post above, from your application? 

- Daniel