• State TI Thinks Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 2 answers
  • Subscribers 31 subscribers
  • Views 1000 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.

BOOSTXL-DRV8323RS: How to use BOOSTXL-DRV8323RX with LAUNCHXL-F280049C

Kustaa Nyholm
Expert 2005 points
Part Number: BOOSTXL-DRV8323RS
Other Parts Discussed in Thread: LAUNCHXL-F280049C, BOOSTXL-DRV8320RS, MOTORWARE, DRV8323, LAUNCHXL-F28069M, , DRV8320

Hi,

we have been developing our system with LAUNCHXL-F280049C and BOOSTXL-DRV8320RS.

In preparation for the next step in development before we have our own PCB we purchased 5 pc BOOSTXL-DRV8323RX

boards with the presumption that it would 'just work' because the chips are identical apart from the on registers which seems to have sensible default values.

Alas, it turns out we made a mistake.

When we tried to run our  code we consistently get over current error flag in Instaspin FOC labs.

On investigation we noticed that the nFAULT line is in a different pin on the BOOSTXL-DRV8320RS and BOOSTXL-DRV8323RX.

Is that the only difference?

What do we need to change in the hardware AND software to be able continue with the BOOSTXL-DRV8323RX kits?

We are working based on the instaspin labs for motorControlSKD 3 specifically lab is_07_speed_control.

Having read through and traced the signals in the BOOSTXL interface for number of different LAUNCH and BOOSTXL boards,

not to mention the reading the soucecode and struggling with the always changing pin numbering and conflicting comments and file names

and other inconsistencies  in related  MotorWare and MotorControlSDKs to say that I'm frustrated would be an understatement...

I noticed in this thread that the boards are compatible but the thread does not say in which way they are incompatible:

e2e.ti.com/.../798912

  • 0
    TI__Mastermind 18900 points

    Hello Kustaa,

    Yes, I can definitely see how this is frustrating experience. I think I have to talk a bit about business strategy and the intent of enabling DRV83x EVM + Instaspin. For the shortest answer, you either take on the task of converting the LAUNCHXL-F280049C to work with the DRV8323 EVMs or using the streamlined path mentioned in the DRV8323 user guides.

    Business Strategy:

    From a DRV83x perspective, we acknowledge that the DRV83x is a gate driver at heart and we are looking show the performance of the gate drive. As a result, any DRV83x EVM is made with a launchpad format so any user can interface with the MCU or processor of their choice and start making their motor commutation algorithm.

    From experience, trying to learn the processor or MCU, and the DRV83x can be tough so we have streamlined ways to get users spinning their motors faster. This might be a sensored or senorless trap but we also acknowledge that instaspin is a huge brand for TI. As such we have tried our best to streamline a DRV83x with a processor that will enable instaspin quickly.

    The problem is, you ran into the case where the processors and DRV83x don't have the same streamlined path in both hardware and software. I'll get into the device specific details in a bit but we try to plan that users go through 3 stages of "playing with" a DRV83x device:

    • The driver looks good on paper and I just want to see if it can spin a motor (I don't care about the MCU or processor just yet)
    • The driver performed well but now I want to make the motor algorithm more realistic to the use case I would use (and I care more about to a style of commutation than a particular processor or MCU right now)
    • The user is ready to start making my own motor algorithm (usually with a specific processor or MCU)

    To reiterate, if you have started to develop your own algorithm and absolutely have to use the LAUNCHXL-F280049C, then you will have taken on the task of making sure that inputs between the DRV83x and C2k make sense (which the C2k team can help you with and I can assign this thread to that team).

    But if you're more interested in spinning the motor with Instaspin and not necessarily the LAUNCHXL-F280049C, then the BOOSTXL-DRV8323RS has instructions how to get instaspin up with the LAUNCHXL-F28069M/F using the quick start guide, user guide, and software guide provided on the EVM product page.

    Specific problems with DRV8320 vs. DRV8323:

    The big one here is integrated vs. no current sense amplifiers. Instaspin requires information about the current in order to commutate properly. As such, the most important information to algorithm needs to dramatically change in architecture between the two devices. With more signals routed to processor, I'm not surprised some other signals are changed around, like nFault.

    Closing Thoughts

    Hopefully you can understand how and why this happened. I completely acknowledge that having these 2 separate flows for the same device part number (or at least family) isn't very intuitive. As such, we do have initiative to make things similar.

    If you decide to continue to use the LAUNCHXL-F280049C, I'll notify the C2k team to this thread and together we can offer some guidance.

    Best,

    -Cole

  • 0 in reply to Cole Macias
    Expert 2005 points

    Hi Cole,

    thanks for the speedy answer. I appreciate what you write about your business strategy and I had actually sort of deciphered that from TI's offering. ;)

    I think we are on the stage that is the  third  of you bullet points.

    I understand the current sensing issue (I think) but since both BOOSTXL board (8323 and 8320) have those current measurement circuitry the only difference being if they are integrated or not I would expect that it should no be a big task to convert the BOOSTXL-DRV8323RX to work with LAUNCHPAD F280049C and especially the software.

    I think I will need just some hand holding. So please alert the C2k team.

    My own thinking is that we need to rewire the nFAULT and just initialise the integerated amps (one register) and we should be pretty close to finish line already, All the other signals (at least in hardware) seemed to go to the right inputs and ouputs on the F28004C, did no yet check software.

    wbr Kusti

  • 0 in reply to Kustaa Nyholm
    Expert 2005 points

    On closer inspection it seems that ISENA, ISENB, ISENC, ENABLE and VSENVM need to be rerouted => too much trouble, we have ditched the DRV8323RX kits we purchased and order more DRV8320RS kits.

    BTW The BOOSTXL DRV8323RX schematics and BOOSTER PACK interface signal naming explicitly and implicitly refer to DRV8323RH chip but the actual chip on the PCB seems to be DRV8323RS.


    wbr Kusti

  • 0 in reply to Kustaa Nyholm
    TI__Guru** 115470 points

    Good, you might use the DRV8320RS kits and there are the related examples supporting this kit in motorcontrolSDK.

    C2000WARE-MOTORCONTROL-SDK: http://www.ti.com/tool/C2000WARE-MOTORCONTROL-SDK