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.

LM5085-Q1: Output is unstable when input is equal to output for 100% duty cycle

Part Number: LM5085-Q1
Other Parts Discussed in Thread: LM5085

Hello

I am using LM5085 in one of my boards: Design is for Input: 20V to 60V; Output 24V (for Vin >=24V); Output = Vin (for Vin <24V); Max power = 75W.  I am seeing two issues:

1. When input is 22V to 28V- output fluctuates a lot and does not deliver the needed power. This has to be resolved.

2. When input is lesser than 22V, output follows input and can deliver full 75W. 

  • Hi:

    can you try to connect to output capacitor as below , and increase the Rs to 200m see any improvement

  • Thanks. I will try with 200m and with the modified connection. How did you come up with 200mohms? I tried using 480m and it did not work. 

    What about the spikes seen for higher voltage? I have ensured that the loop area is as small as possible. Not sure why there is a spike. 

  • HI 

    higher ESR will make the ripple injection larger, so that it will easy stable. 

    the spike is dv/dt coupling to Vo, normally we use 20Mhz test the ripple. you can test the ripple with SW, see it is turn on or turn off. 

    and you can add some snubber for reduce the dv/dt

    Thanks

  • I tried with 250mohms resistor and the modified connection as suggested. I did not have 200mohms in hand. The output voltage for Vin =24V is dropping down to 17V. And the voltage sags a lot for Vin <24V to 20V also. 

  • I also went back to the original connection but with 250mohms resistor. The regulator works very well for Vin = 20 and for Vin > 26V. Only between 22 to 26V, the output fluctuates a lot.

  • Hi

    let me do a simulation , see any clue and reply to you later.

    Thanks

  • ok thanks. For 100% duty cycle do we have to use type 3 ripple configuration?

  • I tried the type 3 ripple configuration and I am able to see consistent output with Vin = 24V for full load and it works for other voltages too. I used C1 = 3300pF, R3 = 191Kohms for 25mV ripple

    Now the snubber circuit has to be added to reduce the spike. Any suggestions? 

  • HI

    From simulation result, I didn't find the issue based on your schematic, but type3 is better than type2 for sure. you can also adding a resistor sereies with cboot to reduce the turn on slew rate, for snubber, pls make sure Csnubber*Vin^2*fsw<Rate Power dissipation of Rsnubber

    Thanks

  • Thanks. What simulation did you do? Did you use pspice or Simplis setup?  can you share the setup? 

    I am noticing that with type 3 config, the output is higher than 24V when no load.

    meaning when Vin = 24V, no load Vout = 24V. With load Vout = 23.35V. Ripple is very less. 

    When Vin = 48V, No load Vout = 25.4V and With load Vout = 24.9V. 

  • HI 

    The injected ripple impact the output voltage, Vout=K*(0.5*Vripple_inj+Vref), K is divid ratio.

    the ripple injection voltage will be affected by duty which means for Vin=48V, ripple injection at no load to full load change about  25mV at FB side.

    I used Pspice for simulation ,download from product folder. can you leave your email, I can send out by email.

    Thanks