DRV2605: Questions in DRV2605

Hi, Scott,

Welcome to E2E and thank you for your interest in our products!

Please take a look at my responses below and let me know if you have additional questions or comments on this.

1. It looks like library is using 3.3V rated voltage. If use VDD as 4.2V  and  also want to play the library's waveform ,will it be dangerous?

There's no risk to damage the device if VDD is higher than the rated voltage. Actually, this is the correct operation. VDD must be always higher than the rated voltage and the overdrive clamp voltage. Rated voltage and overdrive clamp voltage are both configured by the user. See 7.5.2 Automatic-Level Calibration Programming section for details ( http://www.ti.com/lit/ds/symlink/drv2605.pdf#page=22 )

2. If want to change the amplitude of output in close loop , should I change RATED_VOLTAGE or RTP_INPUT?

You would need to modify the RATED_VOLTAGE parameter. This parameter is defined as the driving voltage that the driver will output during steady state. In closed-loop, in addition to this parameter, you mus also consider the OD_CLAMP parameter, which allows the output voltage go above the rated voltage during the automatic overdrive and automatic braking periods. This allows to reach the rated voltage and brake the actuator in a lower time.

In case of the RTP_input, this parameter is specifically for the RTP mode operation. This mode can be used to adjust the output from 0% to 100%, being 100% the rated voltage. For more details, please consult the section 7.5.8.1 Data Formats for Waveform Playback ( http://www.ti.com/lit/ds/symlink/drv2605.pdf#page=28 ).

3. According to "Calibration failures are typically fixed by adjusting the registers associated with the automatic level-calibration routine or, for LRA actuators, the registers associated with the automatic-resonance detection engine."What action will be taken after auto-resonance fails in the Datasheet?Because there's no definition of it in datasheet.

The auto-calibration may fail due to several reasons.So, it is important to always monitor the DIAG_RESULT flag after performing an auto-calibration procedure.

- The failure could be associated to the LRA position on the board/device. It is important to have the LRA always fixed to a solid mass >100gr. Otherwise, the DRV2605 algorithm won't be capable to measure the back-EMF.

- If the rated voltage or overclamp voltage are not set according to the actuator specs, there could be risk to have a calibration failure. So, during the design stage, it is necessary to test the different actuators with these parameters. We have seen cases where some actuators may fail even if they come from the same manufacturer. Some of them have tolerance values so they could need a higher rated voltage. It would need to be adjusted and try again with the auto-calibration procedure.

- The overdrive and brake times are also needed to be adjusted depending of the actuator capabilities to overdrive or brake the device. We don't have a specific formula to select which times are better since they depend of the actuator. So, the best practice to determine these parameters is through testing.

I also recommend to take a look at our configuration file for more details about the device configuration. This will be useful to have the device fully working and avoid auto-calibration failures.

https://www.ti.com/lit/zip/slor114

4. According to"The DRV2605 device allows the user to control how fast the driver attempts to match the back-EMF (and thus motor velocity) and the input signal level." Could explain the relationship loop-gain between Back-EMF and input signal level in more detail?

I think this refers to the actuator input signal level (or DRV2605 output). The back-EMF and the actuator voltage are finally proportional. When the actuator voltage is higher (and also speed), the back-EMF should be higher since it will oppose to the driving voltage.

The loop-gain adjust how fast you would need to reach the back-EMF. I personally suggest to select very slow for a stable operation.

5.According to Adress 0x19 "Auto-calibration back-EMF (V) = (A_CAL_BEMF [7: 0] / 255) × 1.22 V / BEMF_GAIN [1: 0]" Why the BEMF_GAIN becomes larger, the smaller value of back-EMF we get?

The Auto-calibration back-EMF formula in address 0x19 is a way to calculate the real back-EMF. Since there's a "gain" block defined by the BEMF_GAIN, this gain must be considered in the calculation since this formula throws the real actuator back-EMF removing the loop gain. This gain is normally configured for cases where the actuator back-EMF is small, so the DRV2605 can read it better.

6. If I choose a different actuator today, what parameters are critical, needed to be set on Auto-Calibration?Is there anything else that we should pay special attention to ?

As mentioned in the question 4, the rated voltage and overclamp voltage are the most critical values. However, the overdrive and brake times can be also fundamental for a good auto-calibration. I would also suggest to try with different auto-calibration times (register 0x1E), in some cases more time is required to measure the back-EMF.

I hope this is clear. Please let me know if you have further questions and comments.

Best regards,
Luis Fernando Rodríguez S.