DRV8353RH-EVM: Failed to connect to TMS320F28027 device

Thank you Anthony for your Help !

Unfortunately, this is not a current value issue (my Current limit is set to 5A).

I tried with 24V => Same result.

1.With the desktop application, I do not succed to connect to the board at all.

2.With the online version, I successfully upload the software into TMS320F2027F (At 24V, the current vary from 0.042A to 0.07A).

3. After uploading the softwatre : impossible to connect to the hardware (online or desktop version). If I push the RST Button, consumption is 0.042A.

About your note : I have not the S Version, but the H. So I can not change de SPI register. How could I limite the IDRIVE value to 50mA/100mA ?

One more precision : I'm by default in "split rail" configuration (R19 connected). I see in datasheet that VM has an operating voltage. Do I need to switch to "single rail" configuration ? In "split rail" configuration, the TMS320F2027F is powered through USB ?

Thank you !

Sebastien

Hi Sebastien,

Would you be able to let me know for R45 or R47 which one is populated with a 33k ohm resistor and which one is not populated? I believe that these resistors (R45 and R47) are used to indicate to the controller whether or not the device is the SPI variant or the hardware variant. If the particular resistor is populated to indicate to the controller that the device is a SPI device, but it is in fact a hardware device, this might cause some issues. From what I can tell, R47 should be populated for the hardware variant and R45 should be depopulated.

Regarding IDRIVE, since you are using the hardware variant and I can see that there is a pulldown resistor to ground on the IDRIVE pin, If this is a 0 ohm resistor, then that will set the IDRIVE setting to the lowest setting which will be safe for the MOSFETs to handle. If this is not populated with a 0 ohm resistor, switching it to a 0 ohm will allow for a safe IDRIVE setting for the MOSFETs. 

If you don't see any issue with the population of R45 and R47, then do you have another computer that you could try to see if it seems to be more of a hardware problem or more of a software/computer problem?

Using a split rail shouldn't present any issues with connecting to the device, the split rail simply means that the VM pin of the DRV is powered by the 12V from the buck instead of being powered directly from the VM supply that is used for the MOSFET/VDRAIN power supply.

Regards,

Anthony Lodi

Hi Anthony ! I just checked the R45 and R47 : R45 is depopulated which is coherent with th H version I have (R47 is populated).

I verified with another Windows computer : same issue.

=> I will check tomorrow with a Linux version. I will keep you in touche.

Sébastien

Hi Sébastien,

Sounds good, if the Linux version doesn't work then we will keep digging further. 

Regards,

Anthony Lodi

Hi Anthony,

I tried the online version on a UBUNTU Computer.

Step1 : connection to TI Cloud Agent => OK

Step2 : Downloading Program => KO

My Computer stay in the "Downloading program" status and nothing happens (I waited 30 minutes in this state).

I don't know what could I do now

For information : I just buy a S Version in parallel (We don't want to be blocked with this card)

Let me know i  you have another test in mind !

Hi Sébastien,

Could you provide a waveform showing nFault during the flashing of the device as well as the 3V3 voltage? I want to see if there is any fault that occurs during flashing, or if 3V3 supply dips while the F28027 is flashing, possibly causing it to fail to fully connect. Also, could you double check to ensure that the F28027 launchpad is mated correctly onto the EVM? Have you tried switching out the USB cable? 

Regards,

Anthony Lodi

Hi Sébastien,

Thanks for the additional information regarding the fact that you were at first successful in connecting to the GUI on your first try. 

1. Do you have another F28027 or another EVM to try to test to see if perhaps some damage occurred either to the F28027 or the DRV EVM during the initial operation that is causing it to now fail to connect? 

2. You could try pushing the reset CPU button on the ISO-F28027 to see if that helps with the connection issue. 

3. I noticed in your most recent picture that S4 is in the On position. Based on the Users Guide, S4 should be in the Off position.

Regards,

Anthony Lodi

Hi Anthony ! 

1. I bought a drv8353rs-evm => will receive it tomorrow : I will tell you if this new version works.

2. I Already tried the reboot button, but no result. 

3. I took the picture with de S4 to ON, but I always test it with S4 to OFF. I tried everything just in case of :)

I will keep you in touch when my other card received.

Thank you !

Sebastien

Hi Sebastien,

Sounds good, hopefully will have good results As a reminder, for the SPI variant make sure you adjust the IDRIVE settings in the SPI register settings to the lowest IDRIVE setting before operating the motor to avoid damaging the MOSFETs on the EVM due to too high of an IDRIVE. The IDRIVE register values default to 1A/2A source/sink current, which is too high for the MOSFETs that are used on the EVM. I recommend using an IDRIVE setting of 50mA/100mA source/sink current.

Regards,

Anthony Lodi

Hi Sebastien,

Glad to hear that this board works! Please let me know if there is anything else I can help out with.

For sensorless FOC control, if the motor identification parameters are too inaccurate that can result in the motor not spinning well or not being able to spin at all. You could try adjusting parameters such as the Resistance Estimation Current, Induction Estimation Current, Motor Flux Estimation Freq, and Max Current based on the instructions given when you hover over the  symbol next to it. 

It may also be worth trying the old DRV8353RHEVM with the new ISO-F28027F to see if the issue was with the EVM itself or with the F28027F microcontroller.

Regards,

Anthony Lodi

Thank you anthony ! I have more questions now, do I need to open a new post ?

1. Resistance measured by estimation seems to be twice the resistance value indicated in motor datasheet, is it normal ? Is ut resistance beetwen A and B phase of the motor ?

2. Inductance (Ls-d and Ls-q) has not been measured. I measured it with a 10kHz Sine waveform and i found 4 microHenri. Do you think a normal value ? (It is a small brushless motor). Is this method ok to measure inductance ?

3.Flux : estimation is 0.021V/Hz. How could I measure it with instruments to check if the value is correct ?

I attached a screenshot of software for information. 

Thank you in advance !

Hi Sebastien,

If you could repost this in a new thread that would be helpful since the original issue of this thread was resolved. That will be more beneficial for people who may have similar problems, so that there is an individual thread that is focused on on a particular issue.

Thanks!

Anthony Lodi