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.

DRV2624: Question of DRV2624 Diagnostic and Characteristics

Part Number: DRV2624

Hi Sirs,

We have several DRV2624's questions which need product line's recommendations: 

1.After finishing diag routine, customer found the result shows "DIAG_RESULT: z=31 k=48 DIAG_RESULT: R=162761886" What is R units? How to we translate 162761886 to ohm?
2.What its max and min driving voltage?
3.What is DRV2624 close loop tracking accuracy?

Thank you and Best regards,

Wayne Chen
07/03/2018


  • Hi Wayne,

    1. The equation is in section 8.3.2.6. R = 478.43*(31/(719+4*48)) = 16 ohms.
    2. Max driving voltage is determined by the supply voltage. Output voltage is equal to about VDD-0.4V. Min voltage can be adjusted with NG_THRESH to either 4% or 8% of VDD. It is 4% of VDD by default.
    3. Our tests show that the accuracy is within ±1.5Hz. The resonance frequency can change based on the input voltage.
  • Hi Kelly,

    If our VDD is 3.8V, what is the max and min output Vrms?
  • Hi Eddie,

    As I mentioned above, the max output should be about VDD-0.4V, so in your case 3.4Vp. Converting to RMS, ~2.4Vrms.
    The min voltage will be 0.04*3.8 or ~0.152Vp --> 0.1Vrms.
  • Hi Kelly,

    I set DRV2624 to open loop with 170 Hz as below setting, but the frequency I measure is 173 Hz,I have to set 167 Hz to get 170Hz result.
    It seems there is a 3 Hz gap on the setting and measurement result.
    Please help to confirm why there is a 3 Hz gap? Is there something problem with the setting?
    2.1Vrms/170Hz open loop setting:

    Reg:[0x0] = 0x2 DeviceID
    Reg:[0x1] = 0x0 Status
    Reg:[0x2] = 0x1f InterruptMask
    Reg:[0x3] = 0x23 DiagZResult
    Reg:[0x4] = 0xe5 VBAT
    Reg:[0x5] = 0x0 LRARealTimeFreqMSB
    Reg:[0x6] = 0xfd LRARealTimeFreqLSB
    Reg:[0x7] = 0x48 Control1
    Reg:[0x8] = 0xd8 Control2
    Reg:[0x9] = 0x0 Control3
    Reg:[0xa] = 0x92 BatLifeExtLevel1
    Reg:[0xb] = 0x8d BatLifeExtLevel2
    Reg:[0xc] = 0x0 GO
    Reg:[0xd] = 0x20 Config4
    Reg:[0xe] = 0x7f RTPStrength
    Reg:[0xf] = 0x1 WaveformSeq1
    Reg:[0x10] = 0x0 WaveformSeq2
    Reg:[0x11] = 0x0 WaveformSeq3
    Reg:[0x12] = 0x0 WaveformSeq4
    Reg:[0x13] = 0x0 WaveformSeq5
    Reg:[0x14] = 0x0 WaveformSeq6
    Reg:[0x15] = 0x0 WaveformSeq7
    Reg:[0x16] = 0x0 WaveformSeq8
    Reg:[0x17] = 0x0 SeqLoopLSB
    Reg:[0x18] = 0x0 SeqLoopMSB
    Reg:[0x19] = 0x0 WaveSeqMainLoop
    Reg:[0x1a] = 0x0 OverDriveTime
    Reg:[0x1b] = 0x0 PositiveSustainTime
    Reg:[0x1c] = 0x0 NegativeSustainTime
    Reg:[0x1d] = 0x0 BrakingTime
    Reg:[0x1f] = 0x4b RatedVoltage
    Reg:[0x20] = 0x6a OverDriveVoltage
    Reg:[0x21] = 0x7 CalCompensation
    Reg:[0x22] = 0xaf CalBackEMF
    Reg:[0x23] = 0x36 Control5
    Reg:[0x24] = 0x64 RatedVoltageClamp
    Reg:[0x25] = 0x80 OverDriveClampLevel1
    Reg:[0x26] = 0x0 OverDriveClampLevel2
    Reg:[0x27] = 0x18 LRAControl
    Reg:[0x28] = 0x22 DriveControl
    Reg:[0x29] = 0xc DriveControl2
    Reg:[0x2a] = 0x2 AutoCalTime
    Reg:[0x2b] = 0x0 LRAControl3
    Reg:[0x2c] = 0x0 LRAControl4
    Reg:[0x2e] = 0x0 OpenloopLRAFreqMSB
    Reg:[0x2f] = 0xee OpenloopLRAFreqLSB
    Reg:[0x30] = 0x59 ImpedanceCoef
    Reg:[0xfd] = 0x1 RAMAddrMSB
    Reg:[0xfe] = 0x48 RAMAddrLSB
    Reg:[0xff] = 0xdf RAMData

  • Hello Kelly,

    Addition, customer set DRV2624 to open loop with 170 Hz, but the frequency they measured is 173 Hz, customer has to set 167 Hz to get 170Hz result. It seems there is a 3 Hz gap on the setting and measurement result. Please help to check why there is a 3 Hz gap? Is there something wrong with the setting? I will send register setting table and waveform by Email.

    Thanks a lot

    Wayne Chen
    07/21/2018
  • Hi Eddie,

    For new questions, it is best to start a new thread.
    I've measured this on an EVM, and I see similar results. It appears that the accuracy of the open-loop frequency is around ±3Hz. I am checking with design to get the exact specification since this is not in the datasheet.
  • Eddie, I confirmed that this is the expected accuracy of the open-loop frequency.
  • Hi Kelly,

    So is DRV2624 open loop accuracy is ±3Hz?
    And for another question, if we use open loop for operation, is there still necessary to do calibration for the vibrator?
  • Hi Eddie,

    Yes, correct. When using open-loop, you do not need to calibrate the actuator.
  • Hi Kelly,

    Is there any effect to open loop auto break or over drive factor if we don't do calibration?
  • Eddie,

    If you use any part of the closed-loop algorithm, which includes enabling AUTO_BRK_OL, then you need to run calibration. Automatic overdrive can only be implemented in closed-loop mode.
    FYI, the output rated voltage in open-loop mode = overdrive voltage.
  • Hi Kelly,

    Why AUTO_BRK_OL need f0 for break? Doesn't it a reverse 170hz sine-wave for the break?
    And in closed loop, which point did you detect for the back EMF circuit?the peak of waveform or the zero cross point?
  • Hi Eddie,

    We detect the BEMF voltage magnitude after the zero-crossings (approximately 150us-300us after). We detect the zero-crossings for auto resonance tracking. For auto-braking, yes we need to drive the signal with reverse phase. The setting "AUTO_BRK_OL" changes our IC mode from open-loop to closed-loop when the braking bit is detected. When in closed-loop mode, the algorithm setting need to be correct for the Ic to properly detect the f0. This is why we need to run auto-calibration.

    This image shows the transition into closed loop at the end of a waveform.

  • Hi Kelly,

    Appreciate for you reply.
    Can we adjust calibration and diagnostic time ? What is the minimum time for calibration and diagnostic?
    There is three options for calibration: 250us/500us/1000us, is there a option for diagnostic too?
    What the difference if we using 250us or 1000us?
  • Hi Eddie,

    Yes, these can be adjusted as described in the datasheet. The minimum time is 250ms, but this time is not exact. There may be some variation depending on how quickly the algorithm converges. Longer calibration times may be more accurate. Also, for certain motors, the characteristics will change as the motor is driven for a longer period of time. Therefore, the longer calibration may be more accurate.
    This register, auto_cal_time, is also the diagnostics time. Sorry that the datasheet does not specify this.
  • Hi Kelly,

    What is the impedance tolerance generate by diagnostic mode ?Because I found there is about 20% diff on the result.
  • Hi Kelly,

    I found there are two mode in your SPEC for different power consumption, but I don't see the definition about stand-by and idle mode.

    How do I find if driver is in idle or stand-by mode? Is there any register or signal I can check with? 

  • Hi Eddie,

    The DRV2624 automatically goes into standby mode after playing a waveform as long as the AUTO_BRK_INTO_STBY register is set (it is set by default). So the device is in standby-mode most all the time.
    You can put it in idle mode if you set the "GO" bit and play like a 0x00 RTP Strength waveform.
  • Hi Kelly,

    We have met a problem,in our current setting, NRST is pull high(1.8V),TRIG pin pull high(1.8), VDD is connect to 3.8V,and no I2C signal activity.We find the power consumption in this condition is about 0.5mA. Is this a normal power consumption in this condition?
    What is DRV2624 status in this condition?Standby or idle?
    Does TRIG pin always pull high will cause DRV2624 go to idle mode?
  • Hi Kelly,

    Is there an internal pull down resistance in NRST and TRIG pin? And what is it value?
  • By default, the DRV2624 is in external level TRIG mode - RTP mode - for LRA drive. So if you power the device and pull TRIG high, then the device is in active mode and trying to drive an output signal.

    There is an internal pull-down resistor on the TRIG pin = ~2MΩ.

  • Hi Kelly.
    If we using TRIG/INTZ as a output INT pin, and always pull high, we found the power consumption on VDD is about 0.5mA, it's much higher than the SEPC 1.5uA.
    Is it a normal power consumption for this chip? Is there any power consumption SPEC for VDD when we use TRIG/INTZ as a TRIGGER pin or a INT and always pull high?
  • Hi Eddie,

    This does not happen when using the TRIG pin as a trigger because you are driving the actuator when the TRIG pin is high in this mode. Therefore, you are already driving 50-20mA of current depending on your load. However, I am seeing the same results when using the INTZ function with the pin pulled high. I am looking into this.
  • Hi Kelly,

    Yes, this problem happened when using TRIG/INTZ as INT function, please help to check this issue, thank you!
  • Hi Eddie,

    We are looking into this internally. I see the same results when using the INTZ pin this way. You can also get all of the interrupt information over I2C (register 0x01) that can be provided by the INTZ pin. Please use the software method if the extra power consumption is not acceptable.

    Also, since this E2E post is now several pages long and covers many topics, I will close this thread. Feel free to ask a new questions and create a new E2E thread. We like to keep the message threads to one topic per E2E question.

  • Hi Kelly,

    I've create a new thread for power consumption of INT pin on e2e.ti.com/.../733299.
    Let us update information here, thank you!