• State
  • Locked Locked
  • Replies 13 replies
  • Subscribers 33 subscribers
  • Views 1659 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.

BOOST-DRV8305EVM Current sense

Robert Rice1
Prodigy 150 points
Other Parts Discussed in Thread: EK-TM4C1294XL, DRV8301, DRV8305, MOTORWARE

Hi,

When mounted piggy-back on an EK-TM4C1294XL BoosterPack, two of the BOOST-DRV8305EVM current sense outputs do not align with analog input capable pins. Is there an easy way to combine the outputs to one pin, or do I need to mount the BOOST-DRV8305EVM on a separate carrier and use three analog inputs?

Thanks,

Bob Rice

  • 0
    Guru 47900 points

    Robert Rice1 said:
    Is there an easy way to combine the outputs to one pin

    Unless yours operates under very special conditions - that lack of "ease" may prove your salvation.

    Normally/customarily individual outputs are "not" to be "joined/connected" to one (common) pin as your writing suggests.   (there are exceptions - such as open collector/drain outputs - which may be made common - yet these seem outside your issue.)

    Such joining may prove destructive to one or both outputs.   Any such "joining" should be avoided.

    Your idea of employing a "signal re-organizing board" which routes those current sense signals to ADC inputs - makes sense.   It proves wise to minimize signal length and to provide a small capacitor to ground and "match" the sense output impedances to the input expectations of the MCU's ADC.

  • 0
    TI__Guru**** 244440 points
    Hello Robert,

    Which pins on the BOOST-DRV8305EVM are not compatible. Can you please mention that?

    Regards
    Amit
  • 0
    Prodigy 150 points
    Hi Amit,

    The standard BoosterPack pinout has analog inputs on pins A2, A6, B3, B4, B5, B6, B7 and B8. The BOOST-DRV8305EVM has ISENC analog output on pin B9. On the EX-TM4C1294XL pin B9 cannot be muxed to an ADC converter.  To read ISENC with the board piggy-backed I could probably repair the board by jumping pin B9 to pin A6 so that ISENC could be read from pin A6. If you update the board, you should move ISENC or VSENPVDD to either A2 or A6.

    At least two VSEN outputs are needed for FOC, but is there a need for independent current sense outputs? I combine the conversions in my firmware but I’m new to 3-phase motor control so someone could tell me if there is a use for independent current sense outputs.

    Thanks,
    Bob Rice

  • 0 in reply to Robert Rice1
    TI__Guru**** 244440 points
    Hello Bob,

    Yes, you may need to re wire the board to bring the pun B9 to another available and spare ADC Input.

    Regards
    Amit
  • 0 in reply to Robert Rice1
    Guru 47900 points

    Robert Rice1 said:
    ...is there a use for independent current sense outputs?

    Should two - or even three - "current sense devices" be present - someone must have thought so.

    The more advanced BLDC Commutation schemes all benefit from - and several require - knowledge of each motor phase's current draw.   A less obvious benefit is the reduction in power dissipation which such current sense devices experience - when all three phase currents are routed through a single sense device.  Why might this prove of interest/importance - we have "melted" 60 watt, TO-220 style power resistors - mounted to (proper) heat-sinks!   We now employ - almost exclusively - hall effect current sensors with far lower internal (burden) resistance.   And wide, short pcb traces along w/selectively plated - 4 ounce copper.

    If you're new - and "starting" w/FOC - that's a clear violation of KISS!   My belief - you'd learn so much more by employing hall sensored BLDC commutation - at least during your initial test/verification.

  • 0
    Prodigy 150 points
    I have obtained a hall sensored motor to start with but I intend to eventually support a several motor types and feedback devices with my system. Comments like “someone must have thought so” and “KISS” are not useful or productive for this forum.

    Does the BOOST-DRV8305EVM use linear hall effect current sensors? If so, you should add that to the feature list.

    Bob Rice

  • 0 in reply to Robert Rice1
    Guru 47900 points
    I thank you for expressing your (long & well considered) beliefs. I stand by those earlier comments - completely...

    Would not your (overwhelmingly) motor specific questions (better) meet your demanding response @ vendor's more, "Motor-Centric forum?"

    I have no interest - nor obligation - to "add anything" to your (undescribed) feature list.
  • 0
    Prodigy 150 points
    What? Now I don’t even understand your reply.

    Bob Rice

  • 0 in reply to Robert Rice1
    TI__Guru**** 244440 points
    Hello Bob

    Specifically to the following part of the post "Does the BOOST-DRV8305EVM use linear hall effect current sensors?", it is the MOTOR E2E forum that should be able to help you and not TM4C E2E forum.

    Regards
    Amit
  • 0 in reply to Robert Rice1
    Expert 2415 points

    Bob,

    Robert Rice1 said:
    At least two VSEN outputs are needed for FOC, but is there a need for independent current sense outputs? I combine the conversions in my firmware but I’m new to 3-phase motor control so someone could tell me if there is a use for independent current sense outputs.

    As cb1 pointed out, you're not adhering to the "Keeping It Simple is Smart" principle here - you state you're new in 3-phase motor control yet you start out at the very top - and showing your lack of knowledge while doing so. I must suggest you "remain humble".

    I've been looking into the subject more or less closely for the last two years and only now am I realizing how much there is still to learn. But I do know that FOC does not require any phase voltage requirements, yet requires knowledge of each and every phase current. Now you can "cheat" and get all three phase currents with less than three sensors, but each different cheating method introduces some limitations to what you can do. Regarding the phase voltages, perhaps it is a sensorless observer that you are thinking about - but that is separate from the "FOC core" - you can do FOC without an observer, without any phase voltage measurements, if you have a rotor angle sensor.

    Regarding the BOOST-DRV8305EVM current sensors, I suggest you check out the documentation available online. I've seen the info you're looking for and I only skimmed through the document. But I must suggest you to consider the fact that (as far as I undestand from all public TI communication), both the DRV8301 and DRV8305 boosterpacks are primarily designed to be used with the C2000 Piccolo LaunchPads and are not supported on other platforms. As Amit and cb1 have suggested, check out the motor forums for better info on that.

    As a general intro to FOC, I suggest you check out for example TI's Dave Wilson's online videos on the subject, "Teaching old motors new tricks" is a series I enjoyed and learnt tons of information from.

    Best regards,

    Veikko

  • 0
    Prodigy 150 points
    Hi Veikko,

    Thanks for your reply.

    "Keeping It Simple is Smart” is a nicer variation of the original KISS principle acronym "Keeping It Simple, Stupid” coined by Kelly Johnson, lead engineer at the Lockheed Skunk Works, promoted by the U.S. Navy in 1960 and in popular use by 1970. Other variations are: "Keep it Simple, Silly", "keep it Short and Simple", “Keep it Simple and Straightforward" and “Keep it Small and Simple”.

    I don’t intend for my CNC system to remain humble. I intend to eventually support several motor types and feedback devices, so I need to know a little about a lot to get my software architecture correct. Then I can fill in the details.

    But I think I need to respond to any E2E Community response having a generally condescendant tone such as this one. And if you are going to be condescending you had better get your information correct because I will be very critical of your responses.

    TI advertises the BOOST-DRV8305EVM and Motorware to work with the C2000 Piccolo LaunchPads. The Piccolo LaunchPad documentation shows BoosterPack pin B9 as analog capable. However, TI recently advertised a release of Motorware to work on other LaunchPad boards so I don’t think TI intends to limit the technology to the C2000 business unit. Since pin B9 is not analog capable on the TM4C LaunchPads, TI should probably avoid using that pin for analog output in future BoosterPack board layouts. But I simply wire-wrapped pin B9 to A6 to get around that problem.

    I suppose that you could use FOC motor control with a rotor angle sensor but you would probably do a simple commutation using the sensed angle. Why do vector calculations?

    I think I recall a note in the BOOST-DRV8305EVM documentation that the voltage sense outputs are intended for BEMF measurement. Is that not correct?

    I did viewed a TI online videos on FOC and learned a lot. I need to view it again to “sink in”.

    Keep up the good work, but try not be be so condescending with your responses,
    Bob Rice


  • 0 in reply to Robert Rice1
    Expert 2415 points

    Robert Rice1 said:
    "Keeping It Simple is Smart” is a nicer variation of the original KISS principle acronym "Keeping It Simple, Stupid” coined by Kelly Johnson, lead engineer at the Lockheed Skunk Works, promoted by the U.S. Navy in 1960 and in popular use by 1970. Other variations are: "Keep it Simple, Silly", "keep it Short and Simple", “Keep it Simple and Straightforward" and “Keep it Small and Simple”.

    My subtle point (backed up by your "lecture" there) was that the last S in KISS is offensive only if you choose to interpret it as such.

    I'm going to leave this at that. Best of luck with your CNC project.

    Regards,

    Veikko

  • 0
    Prodigy 150 points
    Hi Veikko,

    Now that I have fixed my SPI timing all control registers set and verify. I didn’t notice that the clock timing is different from the BOOST-DRV711 - falling edge rather than rising edge.

    Thanks,
    Bob Rice