DRV8800: Failure cause investigation

Hello Kensuki,

K. Suzuki said:

Q1: Should I avoid using the DRV8800's ITRIP for current limiting?
・The starting current of the motor in use exceeded ITRIP.
・When the motor is locked, the state continues to exceed ITRIP for several seconds until the software stops

Unless I am reading the waveform incorrectly, I don't see the current exceeding ITRIP. Can you clarify how it is exceeding ITRIP.

K. Suzuki said:

Q2: Will the IC be damaged if I continue to operate it as is?

As long as the current is not exceeding the motor current rating or IC abs max, there should not be a concern for damage to the IC. 

K. Suzuki said:

Q3: I'm thinking of further lowering ITRIP to keep the current down when the motor is locked, but should I avoid this?

You can lower the ITRIP current if the current you are measuring is higher than what you need it to be. However, ITRIP should not be too close to the operating current (0.2A in your case) or the motor may not have enough current to spin.

Regards,

Pablo Armet

thank you for your answer.
Add waveforms as suggested.

If ITRIP is set to 2A or higher, the motor starting current will rise to about 1.5A.
(The waveform shows that the direction of the current is reversed.)

In my circuit this time, Itrip=1.163A, so the motor start current has a PWM-like waveform. (See previous image)
Similarly, 1.5A continues to be consumed even when the motor is locked, so it will operate intermittently beyond Itrip.

Waveform at motor lock (Itrip=2A)　　　　　　　　　　　　Waveform at motor lock (Itrip=1.163A)

If it is okay to use Itrip as a current limiter, the cause of the IC failure is considered to be different.
We suspect the possibility that common mode noise, etc., was directly applied to the SENSE pin, and we will investigate ways to strengthen noise countermeasures.

Pablo Armet said:

You can lower the ITRIP current if the current you are measuring is higher than what you need it to be. However, ITRIP should not be too close to the operating current (0.2A in your case) or the motor may not have enough current to spin.

You're right, when I increased Rsense above 0.43 ohms, the motor couldn't fully rotate.
However, we are considering 0.6Ω because we want to suppress the torque when the motor is locked as much as possible.

Regards,
Kensuke Suzuki

K. Suzuki said:

You're right, when I increased Rsense above 0.43 ohms, the motor couldn't fully rotate.

to correct

You're right, when I increased Rsense above 0.86 ohms, the motor couldn't fully rotate.

HI Suzuki-san,

K. Suzuki said:

In my circuit this time, Itrip=1.163A, so the motor start current has a PWM-like waveform. (See previous image)
Similarly, 1.5A continues to be consumed even when the motor is locked, so it will operate intermittently beyond Itrip.

I'm not seeing the current higher than 1.163A in the right waveform with ITRIP=1.163A. I only see the current = 1.5A for the other two waveforms with ITRIP=2A. I see proper driver operation. Did I misinterpret the waveforms. Please correct me if I did.

K. Suzuki said:

If it is okay to use Itrip as a current limiter, the cause of the IC failure is considered to be different.
We suspect the possibility that common mode noise, etc., was directly applied to the SENSE pin, and we will investigate ways to strengthen noise countermeasures

The purpose of ITRIP is to limit the current. Can you describe what exactly is the IC failure. I'm still not fully understanding what the issue is.

K. Suzuki said:

You're right, when I increased Rsense above 0.43 ohms, the motor couldn't fully rotate.
However, we are considering 0.6Ω because we want to suppress the torque when the motor is locked as much as possible.

As long as ITRIP is enough to provide the required torque and current for your motor, it should be okay to use 0.6Ω.

Regards,

Pablo Armet

Pablo Armet said:

I'm not seeing the current higher than 1.163A in the right waveform with ITRIP=1.163A. I only see the current = 1.5A for the other two waveforms with ITRIP=2A. I see proper driver operation. Did I misinterpret the waveforms. Please correct me if I did.

I understand that the original current value (1.5A) is higher than the Itrip set value, but Itrip correctly limits it to 1.163A.
I think that Itrip is usually used for current control, but since an IC failure occurred this time, I suspected that it was actually a fail-safe function.
For example, when the motor is locked, Itrip operates and continues to operate intermittently, so I thought that it might cause the IC to break down.

Pablo Armet said:

The purpose of ITRIP is to limit the current. Can you describe what exactly is the IC failure. I'm still not fully understanding what the issue is.

There are two main failure modes.
It can be restored by replacing DRV8800.
・Itrip operates at about half the current consumption (less than 0.6A), and the motor does not rotate due to insufficient torque.
・OUT+/- of the IC is not output and the motor does not rotate.
　(I measured the pin-to-pin resistance value, but there was no clear sign of dielectric breakdown.)

As shown in the first image, it rotates in one direction for 400ms and stops repeatedly.
(In rare cases, the motor locks due to the transported object getting caught, but it is not possible to check whether the motor locks when the IC fails.)

If there is no problem in using the IC, I would like to proceed with noise countermeasures on the board first.
(Another motor is driven on the board, and although it is not large, noise is generated by the motor.)

Regards,
Kensuke Suzuki

Hi Suzuki,

Thank you for the clarification. I understand now. Is it possible to share the driver schematic or layout? I can check if something in the design is possibly causing the problem. 

Can you provide a waveform of VREF during the failure mode where ITRIP operates at half the value. 

Regards,

Pablo Armet

I am attaching the circuit diagram and layout though it is partial.
・VCC is 5V.
・FL5 and FL6 are not mounted and not used.
・F4 is for cable shorts and will not trip under motor load.
・The MODE and ENABLE pins are controlled by the CPU for unidirectional rotation.
    (The SN74HC buffer IC is used between them.)
・0.1uF of C122 is based on the circuit diagram (CPG002_Schematic.pdf) of "DRV8801 Evaluation Module".

             


Regards,
Kensuke Suzuki

Pablo Armet said:

Can you provide a waveform of VREF during the failure mode where ITRIP operates at half the value. 

Sorry, I can't measure waveforms.
The board needs to be repaired and is no longer on hand.

I'm sorry, I pressed the solve button by mistake.

Thank you for your interesting tip.
However, the faulty board so far needs to be repaired and has been replaced with a new DRV8800, so I can't confirm it.
Since failures do not occur frequently, the next opportunity to check is undecided.
If I get any follow-up information, I would like to add or reply.
This time, I would like to solve the problem once by obtaining the following knowledge, but please let me know your opinion.
・C122 (100nF) may have an adverse effect.
・There is no problem in using the Rsense terminal.

Hello,

Today is a US holiday.  Please expect a response tomorrow.

Regards,

ryan

Alright, I'll add it to this thread.
(Even if I do something, I won't know how effective it is until I release it, so my answer may be delayed.)

Thank you for your kind attention.
Regards,
Kensuke Suzuki