Post Go back to editing

LTC7810 - DRVcc not stable until the DRVcc UVLO is reached. Leading to various startup times that can be as long as a few seconds.

Category: Hardware
Product Number: LTC7810

When the LTC7810 is powered up (voltage applied to Vin and RUN pin passing 1.2V) DRVcc starts to rise. But at a voltage between 3 and 5 it suddenly gets unstable. Up to the point where it gets above the DRVcc UVLO after which the LTC7810 starts switching and DRVcc is correctly at the set 10V. This behavior causes a lot of variation in time before the 3.3 V becomes available.

DRVcc is only connected to a 4.7uF 25V ceramic capacitor and the NDRV pin. And of course, for both channels, via a bootstrap diode and 10 Ohm bootstrap resistor to the boost pin.

The RUN pin of channel 1 is connected to Vin via a resistor divider and capacitor for a ~10ms delayed start. The RUN pin of channel 2 is kept low to disable this channel until its voltage is required.

Some images of the logged signals:

  • Green: Voltage on Vin, 27V
  • Blue: DRVcc voltage
  • Pink: Run pin voltage (resistor divider and 1uF on run pin for a delayed start)
  • Red: 3.3V output voltage on channel 1

Another startup, clearly a different start up time:

We like to inquire about any possible cause and how to solve this.



Deleted carriage returns to improve readability
[edited by: MGES at 6:48 AM (GMT -4) on 14 Aug 2024]
  • Since it's been over a month now with no response, I do not expect too much from this forum. But to add to the issue for future reference:

    The LTC7810 was replaced but the behavior remains about the same.

    DRVcc capacitor, bootstrap capacitors, bootstrap diodes and bootstrap resistors were replaced, but the behavior remains about the same.


    Other PCBA's with the same circuit are functioning fine.

  • It looked like the cause and solution were found. Because of clearance rules, there was only limited connection between PGND and GND on the PCBA. DRVcc was referenced and connected to PGND via the recommended 1uF capacitor. What should have been a good low-impedance connection between PGND and GND, wasn't. Shorting PGND to GND seemed to solve the issue for a moment.

    However, it later returned and the previous observation seemed a lucky shot. Replacing the LTC7810 again, just to be sure, did not solve the issue and DRVcc (blue line) does never reaches its required value:

    • Green: Voltage on Vin, 20V
    • Blue: DRVcc voltage
    • Yellow: Run pin voltage (resistor divider and 1uF on run pin for a delayed start)
    • Red: 3.3V output voltage on channel 1

    Zooming in on DRVcc, a saw tooth profile is observed:

    Is this erroneous behavior of board components that are related to the internal LDO?

  • Finally the cause and solution were found: A small 2.2 Ohm resistor was connected in series with Vin of the LTC7810.

    Though visually fine and the correct value, it measured a high resistance instead, causing the voltage on Vin pin to drop as soon as DRVcc was started. Replacing the faulty resistor solved the issue and resulted in a stable DRVcc and operation of the LTC7810 as intended.