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.

TPS56637: Is R6 adjustable with a potentiometer?

Part Number: TPS56637
Other Parts Discussed in Thread: TPL0501-100

I am inclined to make R6 (in Fig 17 of the TPS56637 datasheet) adjustable (slowly), with a potentiometer, to make a 4.8-12V (approximately) adjustable power supply.  Looking at eqn 5 (and Table 4) of the datasheet, it seems I can, e.g., select L1=5.6uH, and make R6 adjustable from about 63.4 to 173.4 kOhms (using a 0-100kOhm TC42X-2-104E pot in series with.a 0-10kOhm TC42X-2-104E trim pot and a 63.4 kOhm resistor, taking R7=9.1 kOhm) in order to achieve this.

However, I also noticed the following sentence in the datasheet:  "Please note that dynamically adjusting output voltage is not recommended."  What is the underlying reason for that recommendation?  If R6 is only adjustable slowly (via a pair of pots, not electronically), and/or the load that V_out drives equal nearly zero when such adjustments are made, is the possible reason for this recommendation mitigated?

  • I will check this and give feedback next Monday.

    Yuchang

  • Hi Bewley, Generally lower vout to high vout change is ok, but some special setting inner IC cause when Vout change from high to low, big risk will cause OVP happen and then IC will restart, that's why finally we exactly don't recommend to use under adjustable Vout application.

    Yuchang

  • Dear Yuchang -

    Thank you, I understand from your answer that increasing R6 (to increase Vout) is generally ok, but decreasing R6 (to decrease Vout), if done too quickly, can trigger the TPS56637 IC to automatically detect an over-voltage on Vout at some point, and thus trigger the IC's OVP, and initiate a soft restart of the IC.  That answer is clear to me.

    Based on this guidance, if I persist with this idea, I would notify the user of the recommended procedure to avoid such a restart: that is, turn the TPS56637 IC off, turn the main pot all the way down, center the trim pot, then turn the IC on, dial the main pot (in one direction only) to increase Vout to the neighborhood of the desired value, then adjust the trim pot SLOWLY to obtain the desired value of Vout.

    I want to understand the possible consequences of a user not following this recommended procedure.  Can the soft restart triggered by decreasing R6 too quickly somehow damage the IC?

  • PS - Alternatively (better), instead of using a manual potentiometer, I can use a software-controlled Digital Potentiometer like the TPL0501-100.  Then the driver software could enforce the "recommended procedure" I described before, starting with turning the  TPS56637 IC off and leveraging a calibrated look-up table to quick dial in to the neighborhood of the necessary resistance without ever decreasing R6 while the device is on, thereby never risking triggering the IC's OVP to begin with.  Problem solved.  :)

  • Bewley,

    It is a good idea to prevent OVP happen, but sorry forgot to tell another challenge for the wide Vout application is the loop stable, in order to make loop stable for different output, there is internal compensation setting corresponding for different high/low Vout, but bad news is the compensation coefficient is locked during soft-start, which may cause dynamically adjust Vout unstable after power on, so dynamically adjusting the Vout is not recommended, in another word, it is okay if you pre-set the FB resistor ready for the target Vout before power on, but also need to concern the output LC configuration.

    Above all it is hard to use this part for dynamically adjusting Vout and not recommended.

    Yuchang

  • Understood.  Related to the first challenge mentioned previously: I will prevent OVP from kicking in at all by using a software-controlled Digital Potentiometer, as identified earlier.

    In addition, related to the second challenge mentioned above: whenever changing to a different Vout, I will conservatively (in the software driver) cycle the power on this device off completely for a timeout period of T (thereby resetting the compensation coefficient), adjust the Digital Potentiometer based on a pre-calibrated look-up table, then turn the power back on.  This avoids any hint of "dynamic adjustment" on the device, and should keep things in the clear.

    Question then: what is a sufficient value of T for the compensation coefficient to reset?

    Thanks again for the help.

    - Tom

  • Recommend waiting until Vin power off exactly and then power on again.

    And better do some test based on your configuration with EVM.

    Yuchang