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.

DRV110

Other Parts Discussed in Thread: DRV110

I'm currently using the DRV110 to drive a 12V Relay (OMRON G5LE). The same 12V power supply powers both the Relay and the DRV110. The MOSFET I'm using is the BSS138BKS.

Here are my external setting: 

Rsense= 2 ohm

Rpeak= 1M ohm (so the peak current should be 30mA)

Rhold= 333K ohm (so the hold current should be15mA)

Rosc= 47K ohm   (so the PWM frequency should be 60KHz)

Ckeep = 1uF (Thold 100ms)

Attached is the voltage signal measured at the coil of the relay. The PWM signal frequency is around 7 KHz instead of 60KHz and the signal doesn't look like a clean square wave. Am I doing something wrong?

Thank you ,

Salvo 

  • Salvatore,

    Your settings seem correct. One issue we've seen is that in some cases having 33K ohm < Rosc < 66.67K ohm has caused some instability with particular relays. 

    I would recommend increasing Rosc to 100K ohm and see if that solves the problem.

    Could you provide details on your relay? P/N if possible, otherwise inductance and resistance at least. I'm trying to track down a particular pattern.

    Thanks,
    Kevin LaRosa

  • Hi Kevin,

    I'm using the OMRON G5LE-1-E-36 DC12. I don't know the inductance, the resistance is 400 Ohm. Here is the data sheet:

    Thank you,

    Salvo

  • Kevin,

    I increased Rosc to 100K, the PWM frequency is now around 13KHz instead of the expected 40KHz. The duty cycle is fine but still not a "normal" PWM signal.   

    I wonder if the signal is actually fine and the Relay driver reliable.  

    My final goal is to drive efficiently a 15A, 360mW Omron relay, in order to use ~60% of the power.

    See the signal attached. 

    Thank you, 

    Salvo

  • Salvo,

    There is definitely something unstable in that output. My initial thought was that it is operating at >13 KHz but that some of the cycles are reaching 100% duty cycle and thus don't really appear as a separate pulse, but that's still not good. The PWM should be distributing evenly across cycles. 

    Are you testing on the EVM or on your own board? If they are on your own board, could you post (or send me) a SCH? You may also want to take a scope shot of the sense line. If it's picking up noise it could cause the system to respond incorrectly.

    Additionally, are these taken during peak or hold? I'm guessing it is present in both cases but if not let me know. 

    I've ordered the relay you mentioned and will do some tests over here on my end once it arrives. 

    Thanks,
    Kevin LaRosa

  • I'm testing my own board.  I took the previous signal during hold. See the schematic attached. The sense line signal is attached as well.

    Thanks,

    Salvo

      DRV110.pdf

  • Hey Salvo,

    I've been doing some testing on my end and am able to replicate this behavior on the EVM with your given conditions and relay, so you don't appear to be doing anything wrong. We've begun an investigation to try to determine why the PWM is not being divided evenly between the pulses in this case. This will take some time.

    In the meantime, is the irregular PWM causing any negative symptoms in your system? It looks to me that the primary symptom you are seeing is that the effective frequency is decreased to 13 kHz? Are you still seeing the proper PEAK/HOLD currents and is the relay working as expected?

    Thanks,

    Kevin LaRosa

  • Hi Kevin,

    No, the irregular PWM is not causing any negative symptoms in my system. It's just noisy (13KHz is audibile). Also, I'm seeing proper PEAK/HOLD currents and the relay is working as expected.

    Thank you very much!
    Salvo
  • Hi Kevin,

    How is the investigation going? We are almost ready with the final design, it would be great to understand if the DRV110 will do the job.

    Thank you,
    Salvo
  • Hi Salvo and Kevin,

    I know this is an old thread, but did you guys solve the problem with DRV110?

    whether you were facing sub-harmonic oscillation since the PWM duty cycle is >50%?

    I am interested in knowing the solution to this problem.

    Thanks,

    Shihab.

  • Hi Shihab,

    Yes, ultimately we found that the DRV110 operates using a hysteretic voltage sense, and so in cases where large inductances are used and small sense resistors, the voltage may take several pulses to go above/below hysteresis. In these cases, the "PWM frequency" setting as described in the datasheet is really varying the time between when it checks to see if the voltage is above/below the voltage threshold and ends up having no effect on the macroscopic switching frequency.

    For example: if it takes 100 us for the current through the solenoid to decrease such that the voltage on the sense resistor is <Vset - Vhysteresis, then the effective frequency will be 1 / 100 us, or 10 kHz, regardless of whether it checks the voltage at 40 kHz or 100 kHz.