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.

LM27761: Feedback resistor effect of making lower value?

Part Number: LM27761
Other Parts Discussed in Thread: OPA192

Hi all

We are testing prototypes with LM27761. Input 5V, Output -3.5V with about 1-10mA to supply negative rail for opamp OPA192. Minimum required is -3.4V, maximum is not critical and even -5V are ok.

During thermal and humidity tests we found that output supply voltage drops and does not give the required output of -3.5V. It drops even below -2.5V @ -40degree Celsius.

We use Feedback Resistors R1=120k / R2=56k and found that these high values are sensitive to humidity. We would like to use 10x smaller resistors but Datasheet requires minimum 50k for R2.

Plan is to use R1=10k / R2=5.6k to make the system less sensitive to humidity, but that is violating the Datasheet value of minimum 50k.

Any technical reason for the minimum 50k requirement? 

  • Hi Gregor,

    My colleague Febin will look into this and come back to you by Wednesday.

    Best regards,

    Varun 

  • Thanks Varun for your support.

    Hi Gregor,

    LM27761 is a very old device and the best practice would be to adhere to the d/s recommendations.

    The device is characterized and tested with R2>50K. Using lower feedback resistances could reduce the efficiency.

    There is also a possibility that the LDO stability would be impacted. Since VFB is fed to the LDO. And VFB is dependent on the choice of resistors.

    So, I suggest you should try this on an EVM first and confirm that the device meets all the system requirements.

    Additionally, I would make sure the layout is properly designed as well. 

    I hope this helps.

    Regards,

    Febin

  • Hi Febin

    Thanks for the answer.

    We did additional tests and found that the VFB line is very sensitive to be disturbed by even small humidity and dirt impact. Even 10M parasitic pickup from the switching capacitor will affect the LDO and it gives anything between 1 and 5V. Of course we use PCBA washing with ionic test and coating but that does not help to solve the problem. We realize that Layout is not that good as it could be and FB resistors are too close to CPout and this is the path for pickup garbagage.

    Short term solution for using the alreay produced boards is to disable LDO and use CPout directly by hardwire short and putting FB to GND as suggested in another thread  "LM27761: LM27761 using only charge pump function" from 7 yers ago.

    Long term solution is to design out LM27761 with new layout.

    --->>>What type of newer DC/DC converter do you recommend? Product will be in production for minimum 15 to 30 years from now on and therefore we need good sourcable device. Requirement is +5V in (4.8 to 5.3) / Out typicaly -3.4V (-3.4 to -5V is working as well) / about typically 10mA for supplying 3x OPA192 plus some resistors (total less than 15mA). We need solution for very limitted board space, hight does not matter. Product is for harsh industrial environment, -40 to +85degree celsius and high EMC environment like several 1000A and several 1000V switching nearby.

    Thanks very much for your help.

    BR, Gregor

  • Hi Gregor,

    As you might have noticed, E2E was under maintenance last week and I could not reply earlier.

    However, our FAE had reached out to me and I hope you received an email already with my recommendations.

    Regards,

    Febin

  • Hi Gregor,

    We did not hear back from you. Do you need further support on this thread?

    Regards,

    Febin

  • we now short Cpout and Vout together and disable LDO. FB pin shorted to GND. This resolves the problem.

  • Thanks for confirming.