• State
  • Locked Locked
  • Replies 6 replies
  • Subscribers 30 subscribers
  • Views 353 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Original question:

DRV8825: Vref values higher than calculations and squeaking issue

DRV8434: DRV8434EVM shutdown issues

TI User
Prodigy 75 points
Part Number: DRV8434
Other Parts Discussed in Thread: , DRV8424

Hi,

I designed a stepper driver board using a DRV8434 and was having problems so I got a DRV8434EVM ("EVM") to compare.  The good news is that the EVM board performs exactly like mine (shutting down after running some time).  The bad news is that I don't know what the problem is.

I suspect that the issue is an overheating problem.  That said, the EVM board says that it can handle 2.5A (peak) without external cooling, however, I can only get 1A from the EVM before it shuts down.  This is very similar to what my board does.  Vref gets about 1.5V before I get shutdown problems. 

I hooked up the "external inputs" from the EVM to my microcontroller board and drove the stepper motor in that way and get the same result as if I run the EVM directly from the TI's DRVxxEVM application. 

The best results I can get are with Toff at Hi-Z, putting Toff at 0 causes shutdown at about 0.5A (actually, I haven't tried Toff at 330k or 1).  The main difference between my board and the EVM is that I guess the EVM sets Vref via a DAC but I set Vref via a resistor divider.  Both the EVM and my resistor divider produce stable Vref and the corresponding output matches the drawn current: Ifs (A) =Vref (V) / 1.32 (V/A).  I just can't get higher currents from the EVM or my board without them shutting down. 

I guess the EVM is a 4-layer board and follows the design recommendations with respect to heat sinking, etc.  Is there any reason why it can't get close to 2.5A? 

A couple of notes, due to availability, I could only get DRV8434 in QFN packages (as well as DRV8424) chips so I had to design my board using them.  Furthermore, my board is smaller than the EVM so it is to be expected that the thermal characteristics would differ and the DRV8434 on my board would overheat faster than the EVM.  That said, the datasheet has a reference design for the QFN chip so I expect that given enough copper and/or external cooling, the chip should be able to handle the rated power.  In any case, at this point, the EVM (which uses a HTSSOP package), shuts down almost as quickly as my board does and does not come close to 2.5A peak.  

Can you provide rough calculations of the temperature of the DRV8434 running on the EVM board?  Or, if you can suggest any reasons the board would shut down other than over temperature?  (The EVM runs fine at lower currents). 

Thanks in advance.

  • 0 in reply to Grzegorz Pelikan
    Prodigy 75 points

    Hi Grzegorz,

    Thank you for the quick reply.  I did review that thread before posting. 

    Because nFAULT can have many different conditions causing the fault, I cannot be sure the problem is a temperature issue.  That's why I would like to rule it out. If the fault is that the EVM chip is overheating, then I'm sure it's the same problem as with my board.  But I don't know the specifications of the EVM with regard to copper amount/area, number of vias, etc, to estimate the temperature rise of the DRV8434.

    The EVM board says it can handle 2.5A peak without external cooling.  What is the maximum continuous current that the EVM can run at without external cooling?

    Also, if it is a temperature issue, then why does setting Toff to 0 makes the problem happen faster than when Toff is Hi-Z. 

  • 0 in reply to TI User
    Genius 3590 points

    Hi,

    From  https://www.ti.com/tool/DRV8434EVM  you can download design files including PCB layers for DRV8434EVM.

    The shorter Toff the higher switching frequency and switching losses what can overheat IC faster.

    Regards,

    Grzegorz 

  • 0 in reply to Grzegorz Pelikan
    TI__Mastermind 23140 points

    Hello,

    Thanks for your assistance Gregorz, I think you're on the right track.

    TI User, could you monitor the switching frequency of the outputs?  I'm guessing you're switching very quickly with a low-inductance motor.  

    You can find the thermal calculations in the datasheet:  8.2.2.5 Thermal Application, specifically check the switching losses section.

    Please let us know when you have that info - this board should easily be able to handle 2.5A, and the behavior you're seeing isn't desired.

    Regards,

    Jacob  

  • 0 in reply to Jacob Thompson
    Prodigy 75 points

    Hi Jacob,
    I will check the switching frequency when I get back but I know that the driving PWM signal is at 48Hz.  I am using "full step" mode so I can't imagine that the problem will be with a high switching frequency on the outputs.  Maybe the low input frequency might cause large current spikes causing OCP errors rather than over temperature errors?  I have only been watching the current draw on my power supply which wouldn't show quick spikes even at 48Hz.  I'll look at the switching frequency of the outputs later and report back then.

  • 0 in reply to TI User
    TI__Mastermind 23140 points

    Hey TI User,

    For a PWM signal as low as 48Hz I would agree with you that it's unlikely to be the switching frequency. 

    Let me know when you get back and can check on an oscilloscope.  If you could provide some zoomed in scope shots of the behavior that could help us diagnose this issue, so we can see if it might be bouncing between Itrip or having some odd ripples.  

    Regards,

    Jacob