DRV10975: My driver doesn't work properly.

Hello Mitja,

Thanks for posting to the MD  forum the videos links the  were also helpful. The schematic to your device looks mostly correct. The only issue is that I dont see the I2C lines going back to the ESP32. It also makes this issue hard to diagnose because we cant read the registers for any possible faults.

The mentioned behavior is typical of this device. The device has an IPD setting which detects the state of the motor, since this is a sensorless motor then it must know its initial position to commutate the motor properly. The amount of time provided for IPD is set via the internal registers. This device is typically used for fans such as computer fans or HVAC, applications that typically dont need a quick startup time and spin continuously. I am not sure if this is what your system needs are. 

Now the reason that the motor is stopping leads me to my next question. Were you able to configure the DRV10975 EEPROM registers with your motors specifications? If you did could you please provide the register settings?

Best,

Isaac

Hello Lara, 

I did not configure the registers. I can add some wires to connect I2C lines to ESP32 to configure them. But how do I connect ESP32 with GUI. 

Thanks for help, 

Mitja

Like how do I configure DRV10975 with ESP32. 

Hello Mitja,

Yes you will need to add some wires to connect the I2C to the ESP32. Unfortunately the GUI will not work with the ESP32 so you would have the to write the EEPROM registers using the ESP32 using the following instructions:

1. Set SIdata = 1.

2. Write the desired motor parameters into the corresponding registers (address 0x20:0x2B)

3. Write 1011 0110 (0xB6) to enProgKey in the DevCtrl register.

4. Ensure that VCC is at or above 22 V.

5. Write eeWrite = 1 in EECtrl register to start the EEPROM programming. The programming time is about 24 ms, and eeWrite bit is reset to 0 when programming is done.

After writing the registers you can power cycle the device and read back the registers in order to make sure they were written correctly. And every time you power cycle if done correctly these will not have to be written again. I hope this helps let me know if you have further questions.

Best,

Isaac

Where do I need 22V, won't this burn chip. 

Hello Mitja,

Device is rated for max of 23V at VCC, please use this only to program EEPROM, and not to run a motor at this voltage. The recommended voltage for regular operation is typically 12V.

Best,

Isaac

Hello, 
I don't have clue how to write in registers, can you maybe help me, maybe if you know any videos that could help me. 

Hi, 

I managed to read and write registers. I changed 2 registers R (phase-CT) and Kt, but motor acts the same as before (like in that video), but this time it stops earlier (it acelerates for only 3 seconds). Should I change any other registers as well or do I just keep changing these two until it works. 

Thanks for help.

Hello Mitja,

Since you now have access to the registers I recommend setting up all the registers. We have a tuning guide that you can follow in order to help you set up the device to run your motor: https://www.ti.com/lit/ug/slou395d/slou395d.pdf 

Feel free to let us know if you have any questions!

Best,

Isaac

I have everything except inertia, beause I can not measure it, what do I put in then?

Mitja,

This can typically be found in a motors datasheet, but if you do not have the value it is not a big deal. This parameters is only used to calculate what the ideal align time setting that should be used. 

Best,

Isaac

In which registers do I write other parameters. I've only found resistance and Kt.

Hello Mitja,

No all values collected in the tuning guide have registers to enter these values. These values are just collected to help you calculate values for other settings.

Best,

Isaac

My motor spins in open loop. I get fault code 4(hex) when stopped and 0 when running

Hello Mitja,

So you are seeing a Lock2: Kt Abnormal fault?

This typically indicates that your programmed Kt value may not be correct or that your lock threshold might set too high. My recommendation would be to verify the value of Kt is correct and verify your threshold values, what setting are you using for register 0x2B? You can also disable the abnormal Kt lock using register 0x27 if you do not wish to use it. 

Best,

Isaac

Dissabling Kt lock solves the problem.

How would I set up chip that it wouldn't kick in the start and so it would be easier and acurate to program camera stabelizer. 

Hello Mitja,

I didn't quite understand your question. Are you asking how to setup the device at startup for your application?

If you could clarify that would be helpful. Thank you!

Best,

Isaac

So motor kicks in start, how do I get rid of it? Video

Hey Mitja,

Thanks for the video that was very helpful in understanding the problem. Can you provide me with all your register from 0x20 to 0x2B?

Best,

Isaac

0x20: 5B

0x21: 1F

0x22: 2A

0x23: 0

0x24: 98

0x25: E4

0x26: 7A

0x27: B4

0x28: 69

0x29: B8

0x2A: AD

0x2B: 0

Hello Mitja,

Thank you for the settings, please allow me to review them and I will get back to you later on today.

Best,

Isaac

Mitja,

Looking at the video again it looks like you might need to adjust the OpenLCurr[1:0], this is the current that is used during the align state so I would suggest perhaps using a low value for this setting and seeing if that fixes the issue. Also make sure that you clamp the motor down in some way. The movement of the motor could be making the kick at the beginning seem worse than it might be. Thanks!

Best,

Isaac

Unfortunately none of that seem to do any changes. I also tried holding motor and it was vibrating a lot in open loop. 

Hello Mitja,

Instead of using align did you try to implement IPD? This could cause the slight kick at startup. When you reduced the current did you notice any slight improvements or was performance essentially the same?

Open loop commutates the motor regardless of its state until enough BEMF is generated to enable the closed loop. In this case it jumps into closed loop when the Op2ClsThr value is reached. Did the device stop vibrating when it reached closed loop or did it stop vibrating as it got close to closed loop?

For you camera stabilizer application did you need the motor to run at a high speed during operation?

Best,

Isaac

When I reduced the current, its performance was essentially the same. Motor roughly spins in open loop and when it reach closed loop it just stops vibrating. 

I don't need to run my motor at high speeds, opposite I need it to do quick actions and slow speeds. Usualy just few degress. I will use gyroscope to read position of motor in all 3 axes. I have 3 motors (3 drv10975), each for one axis (YAW, ROLL, PITCH). Motors will then align with help of gyroscope and program.

I know that would be the best option, but I have deadline for this school project in a little bit over one week. I think if I manage to remove that kick on start, I will be able to program it, so it would kind of work. 

I got rid of kick with turning IPD off. It was on by deafult. Start is now smooth, I will tune it a little bit more and when I get the best I will head back to you. 

Hello Mitja,

This makes sense because the value you had for the IPDCurrThr was pretty high ~2.2A. There was too much current being delivered during the IPD state. If just starting the device using the align state is also giving you problem you can always try reducing the current during IPD to see if that helps. Looking forward to any updates!

Best,

Isaac

I have set the fastest accelaration posible and lowest align time. These are just current settings, I might change them in future, but here it is video of it working. I will send some updates when I do a some more progress on project. https://youtu.be/FkL_G-ctJNY 

Thanks for help so far. 

Not yet. 

I tried to make one axis gimbal, but chips reaction time was too slow. Always around 1 second. 

Hello Mitja,

Once the device receives the command the start-up timing is mainly dependent on the align time or IPD (which is disabled in this case). Here is a chart from the datasheet for reference.

Also make sure your algorithm is not delaying sending the information from the controller to the DRV10975. This could also potentially be a huge source of the delay.

Best,

Isaac

Hi, why speed doesn't work. I have potentiometer setup, and until I turn it to around half motor doesn't even start, when it starts it goes to full speed and if I try to change the speed it is always full, or if I turn potentiometer down it will stop. 

Even DIR is random. If I connect it to GND direction will be random. 

I disabled Mechanical AVS and now it works. 

except DIR still doesn't work. 

Never mind, the motor was broke.