• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Subscribers 29 subscribers
  • Views 1235 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.

DRV8307: DRV8307 is not working at Duty cycle below 10%

Adesh Jagtap
Intellectual 320 points
Part Number: DRV8307
Other Parts Discussed in Thread: , CSD18542KCS, CSD88537ND, CSD19538Q3A, CSD19538Q2

Hello,

I am testing the BLDC motor part no. B773DA3C428D using DRV8307. Its working fine at all duty cycles above around 10%. But at Lower than 10%, it is failed to work. I am not able to understand the issue. PWM frequency was 20KHz. I also tried to increase the PWM frequency so as to check whether there is any minimum Ton required for the IC, but it still shows the same behaviour of not being able to work below 10%. Please guide me to solve this issue. We need this driver to get operated from 0% to 100% with 12-bit resolution i.e. 4096 steps(which is maximum bits available in IC). 

  • 0
    TI__Genius 14485 points
    Hello Adesh,

    Can you please clarify how the DRV8307 fails to work below 10% duty cycle? Do the gate outputs show a 10% duty cycle even though your PWM input is 5%? Or, do you see 10% duty cycle on the 1/2 H-bridge outputs of the FETs.

    If you could please send a scope shot showing the failure, it can help me debug your issue. I would like to your PWM signal, the voltage of one of the gate drive outputs of the DRV8307, and the voltage of the corresponding FET output.
  • 0 in reply to James Lockridge19
    Intellectual 320 points

    Hello James,

    There is no 10% duty cycle voltage at the output.The output is as below. Please check the waveforms.

    In waveforms,

    Yellow = WHSG

    Pink =WLSG

    Green = PWM

    Blue = W

    In between, the output is changing continuously to the second waveform as shown below. In this second waveform, you can see that WLSG is becoming high for a while. 

  • 0 in reply to Adesh Jagtap
    TI__Genius 14485 points
    Hi Adesh,

    Thank you for the scope shots.

    I used the DRV8307EVM on my bench, and I was able to see duty cycles down to around 4%. This was controlled by a PWM generator on the board which stopped at that duty cycle. The frequency was 30 kHz. The DRV8307 should be able to work below 10% duty cycle.

    Here are a couple more suggestions to help debug:

    1) Check your design against the DRV8307EVM schematics to see if you've configured the device correctly.
    2) Slowly decrease the duty cycle while looking at the oscilloscope to check the nature of how the signal disappears. One possibility is that you are using FETs with too large of a gate capacitance, and the FET doesn't turn on for small duty cycles because there isn't enough time for the gate to charge.
  • 0 in reply to James Lockridge19
    Intellectual 320 points

    Hello James,

    I check the EVM schematic. We have used MOSFET CSD18542KCS whereas EVM has used CSD88537ND. Comparing the gate charge of both, CSD18542KCS has Qg=44nC whereas CSD88537ND has Qg=14nC. I observed the waveform of gate voltages and output when operating @10% duty cycle. When PWM is started Gate outputs are given from the IC initially, but an output of half bridge is not PWM, which in turn indicates that MOSFETs are unable to turn ON in that duty cycle. One more issue observed is that the relation of speed and PWM is not linear for the constant load, which shouldn't the case. I think, this can also be happen because of delay in switching in MOSFET. Please correct me if I am wrong. Thus, Issue is with MOSFET.

    You were also having the issue when the duty cycle is below 4%, right?

    But even if we used new MOSFET we will be having this issue of the minimun duty cycle. Can you suggest any MOSFET better than this CSD88537ND?

  • 0 in reply to Adesh Jagtap
    TI__Genius 14485 points
    Adesh,

    Your reasoning sounds correct.

    The CSD19538Q3A and the CSD19538Q2 have gate charges at 4.3 nC. Would those work for you? If not, submit an E2E post under one of those part numbers and ask for a recommendation from our MOSFET team.
  • 0 in reply to James Lockridge19
    Intellectual 320 points
    Yes, these will work for us. Thanks you James.