• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Subscribers 29 subscribers
  • Views 143 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.

DRV8145-Q1: Running as H-bridge, do I need deadtime on complimentary input?

Johan Moberg
Prodigy 10 points
Part Number: DRV8145-Q1

Tool/software:

Hi!

I wish to use two DRV8145 as an H-bridge, driving a single phase BLDC at 36V 35A up to 150K RPM. I intend to set them up as per the datasheet (image below). 

I intend to feed both half bridges with the same 125 kHz PWM signal to their "IN" ports. This signal will be at 125 KHz constantly and only vary the duty cycle.

For the "DRVOFF" pin I intend to feed them a 100-20 000 Hz signal at constant 50% duty, where one half bridge receives a complimentary signal of the other.

With this setup I imagine I can control the current with the 125 KHz varying duty cycle and by varying the frequency of the "DRVOFF" signal, the rotational speed.

So, my main question is if it would be acceptable to not have any deadtime on the "DRVOFF" signal? Wouldn't just the "worst case" be that the motor is tied to say VIN on both sides for a micro second?

Is this overall a good setup or am I mistaken about something? Thanks for any suggestion!

  • +1
    TI__Mastermind 32765 points

    Hi Johan,

    Thank you for your post. 

    Johan Moberg said:
    I wish to use two DRV8145 as an H-bridge, driving a single phase BLDC at 36V 35A up to 150K RPM. I intend to set them up as per the datasheet (image below). 

    The recommended maximum operating VM is 35 V. Even at this voltage it must be ensured that under no circumstances the VM voltage sees peaks greater than the absolute maximum of 40 V to avoid output FET breakdown. 

    In addition please refer to the transient current capability for the device with PCB design referenced in the datasheet - see below. With PWM at 125 kHz the switching losses would be significantly higher than at 20 kHz, I expect further degradation of the current that can be supported. In addition the dissipation in the table referenced was calculated with 12 V VM. You can plugin your applications parameter in the calculation tool "SLVRBN2 — Half Bridge Driver Junction Temperature Estimator" that can be downloaded from the product webpage, https://www.ti.com/product/DRV8145-Q1.  

    Johan Moberg said:
    For the "DRVOFF" pin I intend to feed them a 100-20 000 Hz signal at constant 50% duty, where one half bridge receives a complimentary signal of the other.

    Have you characterized this operation using our EVM for this device? The DRVOFF pin control is not intended for PWM input control especially at 10's of kHz. It is meant for disabling Half-Bridge power FETs for Hi-Z or motor coasting. 

    Johan Moberg said:
    So, my main question is if it would be acceptable to not have any deadtime on the "DRVOFF" signal? Wouldn't just the "worst case" be that the motor is tied to say VIN on both sides for a micro second?

    Assuming you characterized and have everything functional like you intended, between the two outputs because you'd always have the motor coil connected there will be no risk of shoot through current between the two Half-Bridge devices. I am sure how the timing between the DRVOFF control and the IN PWM will work. They are asynchronous control signals so when you disable a driver would it matter where in the PWM edge this needs to happen, likewise a enable driver control - would it be necessary to align on the PWM edge? 

    "Wouldn't just the "worst case" be that the motor is tied to say VIN on both sides for a micro second?". Yes and that'd mean a recirculation of energy via the FETs due to back EMF. Assuming that energy is well within the capabilities of the FETs to support without an OCP or TSD (thermal).

    Regards, Murugavel   

  • 0 in reply to Murugavel4637
    Prodigy 10 points

    Thank you very much for you extensive answer! I will run the calculator tool as advised and lower the voltage. Perhaps it is better leave the DRVOFF and focus on just using the IN pin like you suggest. If I drive the IN pin with a PWM (and complementary signal on the other half bridge and no dead time) between 100-20 000 Hz with variable duty I should also be able to achieve both current and rpm control (?). 

    This is just for a prototype so if I have to go down in performance due to thermals then its ok. But I prefer if the first prototype at least can start the motor without smoking Slight smile

    Thanks again for the support!

  • 0 in reply to Johan Moberg
    TI__Mastermind 32765 points

    Hi Johan,

    Sure! 

    Johan Moberg said:
    If I drive the IN pin with a PWM (and complementary signal on the other half bridge and no dead time) between 100-20 000 Hz with variable duty I should also be able to achieve both current and rpm control (?). 

    For one direction you should have one device HS-FET on and the other device IN should have the PWM signal and for the opposite direction the other way around. I'm not sure I understand the complementary signal on the other half-bridge. Please clarify if possible with input and output waveforms. Thank you.

    Regards, Murugavel

  • 0 in reply to Murugavel4637
    Prodigy 10 points

    Hi Murugavel! Sorry for the late reply. I will run a single phase BLDC motor in only one direction. So I need to change polarity of the coil 8 times per full rotation. At 150k rpm this means switching frequency of 20 kHz. But as you pointed out, the current/heat transfer capabilities of this chip is too low for my requirements. I would require 30-35 A RMS continuously, even though I will use the fan I'm driving as a very capable cooler I think I will look for a H bridge gate driver IC with 4 external power mosfets instead.

    Thank you for your advise, saved me a lot of time.

    Best regards

    Johan