DRV8870: Analysis of periodic fluctuation of motor current

Dong Shen1

Part Number: DRV8870

Hi all，

I'm a FAE of TI. When I tested with drv8870evm, I encountered a problem and wanted to get your answer.

When the motor is unloaded, we find that the motor current will fluctuate periodically. What is the reason? By providing 12V, 15V, 18V, 21V and 24V input voltages, we found that the current fluctuation period basically showed a positive proportional change. The test is as follows (there is an error in the interval between the two spikes)

12V：The frequency is approximately 362Hz.

15V：The frequency is approximately 423Hz.

18V：The frequency is approximately 510Hz.

21V：The frequency is approximately 510Hz.（There are errors here）

24V：The frequency is approximately 862Hz.

What causes this? How to verify?

over 2 years ago

+1 Pedro Arango Ramirez over 2 years ago

TI__Expert 4955 points

Hello Dong,

I am happy to be able to help. I have confirmed with one of our systems engineers that specializes in ripple counting.

The ripples you presented are very similar to those we see on the bench. The reason it increases proportionally with supply voltage is because motor voltage is proportional to steady state speed, and steady state speed is proportional to ripple frequency.

If the ripple frequency trend does not match voltage exactly, we can blame it on unknown loading conditions that change with speed, such as viscous friction in the bearings and magnetic effects.

Best,

Pedro Arango Ramirez

0 Dong Shen1 over 2 years ago in reply to Pedro Arango Ramirez

TI__Prodigy 110 points

Hi Ramirez，

In fact, what I want to know is the reason for the ripple appear, the problem of uneven motor structure, or something else? How to verify?

BRs,

Sanders.

+1 Ryan Kehr over 2 years ago in reply to Dong Shen1

TI__Guru 54725 points

Hello,

Please see section 2.3 in the following app note for the theory behind this ripple:

https://www.ti.com/lit/pdf/tidud30

Regards,

Ryan

0 Dong Shen1 over 2 years ago in reply to Ryan Kehr

TI__Prodigy 110 points

Hi Kehr，

The ripple displayed in the file you provided is different from the ripple I mentioned here. The current ripple in the file is caused by the switching behavior of H bridge. In my experiment, green is the motor current waveform and blue is the out waveform. You can see that my motor current is low-frequency ripple current.

How does this low frequency ripple current come into being?

Fig. 1 Current waveform in file

Fig. 2 Waveform of my experiment

BRs,

Sanders.

0 Dong Shen1 over 2 years ago in reply to Dong Shen1

TI__Prodigy 110 points

It suddenly occurred to me that if it is caused by BEMF, it is essentially caused by uneven magnetic field;

I can directly connect the motor output to the oscilloscope, manually rotate the motor and observe whether the voltage waveform has low-frequency ripple.

+1 Ryan Kehr over 2 years ago in reply to Dong Shen1

TI__Guru 54725 points

Sanders,

Please take another look and note the 1/deltaX in the scope capture and vertical cursor lines. Looks like ~208Hz. The "purple" trace is the output of the comparator used for ripple counting.

Regards,

Ryan

0 Dong Shen1 over 2 years ago in reply to Ryan Kehr

TI__Prodigy 110 points

Hi Kehr，

Thanks. I have fully understood this result and completed the verification in lab.

BRs,

Sanders.

Motor drivers

Motor drivers forum

DRV8870: Analysis of periodic fluctuation of motor current