- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
HI,
I have been experiencing a startup issue when using the EVAL-M1-101TFTOBO1 to control a motor. I am using this eval board with a custom driver board. I am using MCEDesigner to control the motor. I am seeing that the motor starts up correctly if I load the parameters just before operation, then select the Green Traffic Light button on the app. However, if I restart the hardware and try to start up again, then the motor stalls. The parameters should be correct and setup from memory of the device, so why would I not get the same behavior?
The startup should be the same in these two cases, no? Is there some command sequence that occurs when programming the parameters, and one that I can replicate to get the right behavior at startup? I want to be able to start the hardware and tell the motor to run without a parameter load each time.
Thanks,
Aaron
Solved! Go to Solution.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi @aneedles,
This is related to the initial ADC calibration process. The work-around solution can be either one of the following.
- The 3.3V power-up slew rate need to be fast enough.
- Generate a fault intentionally and clear it, and this will trigger the current offset calibration process again which will update the affected 2 parameter values.
A thorough fix can be expected in the newer FW release.
Thanks,
Krupashankar
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi @aneedles,
Thanks! for posting in the Infineon Community.
Yes, the startup must be the same in both cases.
Could you please share your .mc2 file and let us know if you are configuring any other register in MCE designer before starting the motor.
After powering it ON are you immediately trying to run the motor? and Please let us know what is the error which you are getting.
Thanks,
Krupashankar
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
The platform does not accept the .mc2 file directly, so I put it in a zip folder.
For this version, the motor starts either way, but then runs with significantly more vibration when the controller is restarted and the motor run without re-programming parameters. It seems that one of the parameters is not getting updated except in the case of programming through MCEDesigner just before running.
Yes, after powering on, immediately running the motor using the motor start button on MCEDesigner. There is no error, just a significant difference in motor operation.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
And here are two plots of activity at 200rpm showing the difference in behavior in the two cases.
After programming parameters and start:
After controller bootup and start:
Motor operation is visibly rougher, more vibration, more noise.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
I have verified this also occurs when set to Analog VSP.
We really need a fix or work-around since this makes certain testing impossible since we cannot always load the parameters using MCE Designer right before operation.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
For those who are still reading, one temporary work-around seems to be to force a fault and then restart. At least with VSP control mode, a fault and automatic restart appears to fix this behavior once the controller starts up again after the fault.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi @aneedles,
This is related to the initial ADC calibration process. The work-around solution can be either one of the following.
- The 3.3V power-up slew rate need to be fast enough.
- Generate a fault intentionally and clear it, and this will trigger the current offset calibration process again which will update the affected 2 parameter values.
A thorough fix can be expected in the newer FW release.
Thanks,
Krupashankar