DRV8711: BOOST board

Hey Jerry,

Have you tried different iDRIVE settings?  Here's a page about choosing it for your system:  https://e2e.ti.com/support/motor-drivers-group/motor-drivers/f/motor-drivers-forum/796378/faq-selecting-the-best-idrive-setting-and-why-this-is-essential

What we found the most effective way to avoid this is to setting the predriver output current to its lowest level. Please try following settings.

• IDRIVEN = 100mA
• IDRIVEP = 50mA
• OCPTH = 750us
• OCPDEG = 8us

Here are some other threads on this issue:  Link1  Link2 

The DRV8711 is a fairly old device, if this is a new design you might want to consider a newer device - the DRV8452 is an integrated stepper motor driver so you wouldn't need external FETS, it has a peak of 4A or the 8462 has a peak of 5A.  https://www.ti.com/motor-drivers/stepper-driver/products.html#p248=3.5;8&p2716max=40;70

Regards,

Jacob

Thanks for the suggestion and also for the links to threads. I used the settings you suggested and also put 100ohm resistors in series of both A and B LS drive and it worked for about 30 minutes before I started getting Channel A overcurrent shutdown. Since then I have not been able to run it successfully without the same overcurrent fault re-occurring every single time. I believe the channel A MOSFETs is likely damaged as I can smell something bad coming from the board though there is no obvious signs of burnt component. None of the MOSFETs are warm either.

What I found is that in our application, the back emf is really high. Is there a recommended way to address this?

I will need to repair the board before testing again.

Hey Jerry,

That's progress, but obviously the board failing after just 30 minutes still isn't what we want.  

What microstep and decay modes are you using?  I would recommend using a high microstep (1/128 or 1/256) as it should make a smoother signal and more controlled BEMF.  I also recommend trying the auto-mixed decay mode, as that should result in a smooth signal.  Try all the decay modes and compare the signals on an oscilloscope to find the cleanest-looking mode.  

Given the likely driver failure you've experienced I would also probably set the OCP level lower next time at the beginning of your testing, simply to try to protect the device while you tune settings.  

Here's an application note relating to BEMF:  https://e2e.ti.com/blogs_/b/industrial_strength/posts/art-of-stopping-the-motor-vm-pumping  and the different decay modes:  https://www.ti.com/lit/an/slva321a/slva321a.pdf#page=1

Regards,

Jacob

Hi Jerry,

If current is around 3.5A or more I would monitor mosfets temperature and check if they are not getting too hot.

Regards,

Grzegorz

Hi Jacob,

I am using 1/4 microstep. I was not getting the speed that I need at high microstep although higher microstep seemed to work well and is a lot quieter. I am using mixed decay and I have tried auto mixed decay too. Mixed decay gives me a slightly smoother signal.

I managed to repair the board and tried several different settings but it has failed again with the same failure.

Meantime I will shelve this since I don't have any more parts to repair the board. Thanks again.

Regards,

Jerry

Hi Gregorz,

Strangely the MOSFETs are just warm. They never got hot. I was watching them with a FLIR camera.

Regards,

Jerry

Hi Jacob,

To provide a little more insight into the failed MOSFETs. The first time it failed, Q4 which belongs to channel B2 LS and HS failed. Both MOSFETs are non-functional.

Regards,

Jerry

Hey Jerry,

Can you give me more details on your system requirements (voltages, current, special features like stall detection or current monitoring, etc) so that I can maybe recommend a different part and EVM for it?  

Currently I see motor specs:

• Phase resistance: 1.0mohms
• Phase inductance : 3.5mH
• Rated current: 3.5A

Regards,

Jacob

Hi Jacob,

We are operating from a 48VDC power supply with up to 10A current source. Features like current monitoring are optional but stall detection is a requirement. We also need to be able to switch microstepping easily.

Thanks.

Jerry

Hi Jerry,

More important to choose the right driver would be max. Irms or peak current in motor phase that will be needed.

Regards,

Grzegorz