• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Subscribers 30 subscribers
  • Views 2070 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

DRV8825 Overcurrent Protection Issues

Jordan Meyer
Prodigy 160 points
Other Parts Discussed in Thread: DRV8825, DRV8825EVM, MSP430F5528, DRV8824

I have been evaluating the DRV8825 for a single axis stepper motor drive.  We purchased a DRV8825EVM board and I was able to develop most of my code on that platform.  I also bench marked the current in the motor windings using a current probe on an oscilloscope while driving a 6 port Valco Valve with Nema 17 stepper motor.  On the Evaluation board the DRV8825 worked flawlessly and I was able to generate 6 - 8 A peak to peak and 2A RMS.  However, when using our own PCB with the latest pre-production 8825 (EUROB1P) the overcurrent protection trips prematurely and I cannot move even a simple stepper with no load.  I know for a fact that the PCB is working correctly because we also are using the board with an 8824 and it works as intended.  Nothing changes on our pcb when the 8825 is used other than the i sense resistors which are currently at 0.2 Ohms. 

What is the difference between the current RTM version of the 8825 and the EVM version of the 8825?  I would expect it to get better and not worse.  When measuring the current on the newest pre production 8825 I cannot even get 200ma RMS before the current protection trips.  Also, could you offer any explanation as to why the current protection would be tripping early.  Like I said the same pcb is used with an 8824 and there are no issues.  Our designed uses an MSP430F5528 and either an 8824 with 0.4 Ohm Isense resistors or the 8825 with 0.2 ohm resistors.  Most of the board layout is also based on the EVM layout.

Thanks,

Jordan Meyer

  • 0
    TI__Expert 7455 points

    Hi Jordan,

    This is very odd indeed. Can you send me a copy of the board Gerbers to my personal address at jquinones@ti.com. I am positive we can figure this one out. Thanks!

    Best regards,

    JIQ

     

  • 0 in reply to Jose Quinones
    Prodigy 10 points

    Hi Jose,

    I'm also designing in the DRV8825 to replace the DRV8824. Are there issues, like the overcurrent issue above, that I should be aware of?

    Thanks,

    Dave

     

  • 0 in reply to David Lyons78424
    TI__Expert 7455 points

    Hi Dave,

    The DRV8825 does not necessarily has an issue with false OCP. The failure mechanism we have observed has to do with the system. As it turns out, and with like any other Power Pad device, it is very improtant to have that Power Pad very well soldered to the board with a nice copper structure to work as a heat sink.

    If you can remove the heat quickly from the device, the die will be maintained cool and your RDSon will be maintained within spec. System will operate as expected, without fear of thermal runaway which eventually ends up in TSD. We have been able to operate our EVM boards non stop for weeks (way much more than a month), while running at 2.5A peak.

    However, if the thermal impedance is too high, then not only the die starts getting hot, you also increase the chance of localized heating. This is where problems start to arise. If localized heating occurs, parasitic effects increase as well. Anything can happen at this point. Do note that if this is the case, the device will be basically operating outside of the Abs Max ratings, or out of spec.

    What can you do to ensure success? If you have a four layer board, make sure the top and bottom layer have a nice Power Pad and the most copper flooding you can afford, forming some sort of a bat wing extruding from the Power Pad and out into the board, in order to work as a heat sink. Top and Bottom layers should be 2 oz copper. Then, from the Power Pad have a few 13 mils or smaller vias connecting top and bottom layers to a GND dedicated plane in a mid layer. Mid layers can be 1 oz copper.

    If you only have 2 layers, the top and bottom layer copper structure become even more important. We recommend taking a look at SLMA002 which has some examples as to how to design the top and bottom layer copper structures, as well as to how many vias to use.

    One last note. The board size will also play a role and we realize this is something you may not have a lot of head room to play with, if at all. Whatever you reduce on board size, will need to be overcome with any other technique to improve thermal impedance.

    Hope the info helps. Best regards,

    Jose Quinones

  • 0 in reply to Jose Quinones
    Prodigy 160 points

    Jose,

    Thanks for your help with our design.  As it turns out, the over current and over temp was tripping because I had the decay pin set to fast decay.  Once that was changed to mixed, our design could match that of the evaluation board.  To anyone with concerns, see Jose's previous comment and also try different settings for the decay pin.  For our application and board layout, mixed decay gives the best performance.

    Jordan