Low frequency ripple on output too high
Both 24V outputs of the power supply have line ripple of around 120hz (see images). The amplitude of the ripple increases and decreases with the size of the load applied. Image 1-the amplitude is 200mV. Image 2-the load has been decreased and the amplitude is 300mV with a max of about 400mV. This is seen at line voltage of 120VAC.
Our supply also has 5V and 15V outputs. The 5V and 15V outputs have low ripple around 10mV. If you reference the schematic, you can see that the 5V/15V outputs are fed from one transformer and a TOP Switch HX and the two 24V outputs are fed from another transformer and a TOP Switch JX. Also attached are the transformer design drawings and the layout drawing.
A couple things have been tried to reduce the ripple. Output capacitance was increased (not shown on the schematic) without any reduction in the ripple. In the 24V feedback circuit, R16 was originally a 10.7K resistor. Using that resistor value, the ripple was 800mV. This resistor was reduced to 6.04K, which helped to reduce the amplitude of the ripple. Using the PI software, 6K is about the lowest acceptable value for this resistor.
The goal is to reduce the 24V ripple to around 10mV. Are there component changes that could be made to our circuit that will reduce this low frequency ripple? Or is there something that could be changed in our layout to reduce this ripple amplitude?
Thanks,
Mark
Comments
Hi PI-Cochrane,
I should've mentioned in my first post that I have already tried that. I gave it another shot just to be sure. I increased the input capacitor (C17) to 3000uF and it didn't change the ripple at all. Also, as I mentioned before, the ripple on the 5V & 15V outputs is very low = 10mV.
Something also note worthy that I didn't notice before, if I vary the load enough, I can change the frequency of the ripple anywhere from 80hz to 230hz. Low frequency oscillating? A feedback loop issue? Other suggestions?
Thank you.
Either the loop is not stable at low frequency or the loop gain is too low. If possible PLS check the loop with loop analyzer, and compensate the control loop based on loop test result.
PLS try a low value resistor for R16, and use phase boost circuit parallel with R16. I donot think R16 must be higher than 6.04k. Actually 1k resistor is used for LED drive in DER299. PLS check DER299 for the phase boost circuit and LED driving resistor (R16).
http://www.powerint.com/sites/default/files/PDFFiles/der299.pdf
Thank you. I don't have a loop analyzer, but I will try your suggestions tomorrow morning and let you know how it goes.
Hello Mark,
did you solve the problem? I have this problem with all PI ICs. I tried so many different things.. changing of all feedback loop components, transformer, PI devices. There is always low frequency ripple on the output that is NOT dependent on the line frequency and varies with different loads
Hi Andrej,
Thank you for considering Power Integrations for your solution.
Have you been able to check the loop with loop analyzer, and compensate the control loop based on loop test result?
Have you also had a chance to try out the suggestion mentioned earlier? i.e "try a low value resistor for R16, and use phase boost circuit parallel with R16"?
Best Regards,
PI-Kryten
The 120Hz line frequecy is the root cause of the 120Hz ripple.it should be solved with higher value input capacitor. PLS let me know if higher value input capacitor not help.