• State
  • Locked Locked
  • Replies 1 reply
  • Subscribers 29 subscribers
  • Views 168 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.

DRV8912-Q1: Parallel mode questions

Kyson Tan
Prodigy 110 points
Part Number: DRV8912-Q1

Tool/software:

Hi,

We are tying 2x channels in parallel to our load. In our 1st use case, we have both channels turning ON and OFF at the same time by utilizing the built in 200Hz PWM. For our 2nd use, we are planning to staggering the 2 channels to effectively create a 400Hz PWM that is needed for our load.

I saw the following in the datasheet - "For PWM operation in parallel mode, all half-bridges is to be mapped to a single PWM channel selected from any of the 4 PWM channels to avoid any delay in the PWM durations which can lead to undesired OCP condition." This talks about (1) the SPI to output prop delay and (2) the OCP.

(1) For the prop delay, since each cycle is 5000us (for 200Hz), and the delay per datasheet is 25us max. This means the 400Hz output could have unequal OFF time between adjacent ON time, but the overall output waveform will still have 400 pulses in one second. Seems like the prop delay is not an issue in this case. Any other way the prop delay could make this idea not work?

(2) For the OCP, if our load will only draw up to 1A, i.e. well under the OCP threshold even if only 1 channel is turn ON, this means the OCP will not be tripped under normal 400Hz operation and seems like we are OK. Any other way the OCP could make this idea not work?

Any other limitations that would prevent us from using the DRV8912 in this configuration to generate a 400Hz PWM?

Thank you for the help!

  • 0
    TI__Mastermind 23361 points

    Hey Kyson,

    Hmm interesting. It seems like you would basically be manually implementing the parallel of two channels.  In that case, you'll need to manually make sure to avoid shoot-through conditions.  I would probably add in 2x the prop delay to guarantee avoiding shoot-through / shorting of VM to GND.  

    For 2, since you're manually implementing the parallel then the OCP threshold will be for the 1 channel OCP.  So agreed it should work, assuming you avoid shorting conditions. 

    All that said, since this is a pretty unusual use case I highly recommend testing it out on an EVM before you build prototype boards.  You might want to remove the 0Ω resistors and use your own microcontroller to control the device, as I don't think our EVM GUI supports doing what you want.  

    Best,

    Jacob