LNK306 - Experiencing a lot of audible noise (like other people have mentioned). I could use some advice in figuring this out.
Hello,
I have the PCB's up and running, everything is workin great, but I am experiencing a lot of audible noise. Before I designed the board, I had read through some other topics on here regarding this issue. I was hoping that it was not going to be very loud or noticeable so I went ahead and designed our boards' power supply around the LNK306.... sure enough this is the first thing the client noticed and its currently considered an unacceptable design due to the loud noise this thing is producing.
So I could really use your expertise in helping me figure out what to do here. I attached my circuit board schematic as well as the PCB layout. I will also list out the components I am using.
Here is what I have figured out so far, and what I have tried:
- The audible noise tone seems to be directly related to the current draw on the power supply.
- The noise actually sounds more similar to static (or like background noise), its not a perfect tone. This could be due to the nature of my circuit and varying digital current loads though? Not sure.
- Originally I had a 3.3K resistor as my load resistor. I figured out this was not enough of a load so I changed it to a 1K resistor (this should provide 5mA constant load). This reduced the loudness of the noise by half, but it is still present.
- If I push down on the inductor (L2) the noise quiets down a lot - pretty much to an acceptable noise level. Because of this I am assuming this is where the majority of the noise is coming from. Pushing on other components do not change the noise, only L2 changes the noise.
- Both C4 and C8 are electrolytic capacitors, which I read is supposed to help a lot with the noise.
- Pi Expert recommended using a 470uH inductor. I read that using a larger inductor helps with noise and performance, so I am currently using a 1mH (1000uH) inductor here in this design. This is a surface mount inductor and is shielded.
And here are the components we are using for reference:
D1, D3: DFLU1400-7
D2: US1JHE3/61T
D4: FM4007W-W
L1, L2: 46105C
C4: GRM219R61E106KA12D
C6, C7: 475CKE450MLN
C8: 35ZLH100MEFC6.3X11
R3: WP2S-7R5JA25
Any help is greatly appreciated, thanks!
Comments
Hello Alex,
Thank you for the information about the discussion details. I know more about this problem now.
For Mahonroy's design, i think there are two possible ways we can try. The inductor is too much over size to me. i don't recommend the big value inductor here because with big inductor, the swithing frequency will be lower than expected, which will be more close to the audible frequecy. i recommend you to change to the value PIxls or PIexpert suggested. In the forum discussin, the audible noise performance will be better when connected another inductor in parallel, which means lower inductor value.
As the output is connected to a 3.3V regulator, this mean the output load should be very low, you can try with a samller size device LNK304 or LNK305 based on the output current(or power) you need. Or you can adjust the pre-load, maybe use a lower resistor which means more pre-load to change the swicthing frequecy in the indcutor. there will be a point you can get the accpectbale noise level.
Best Regards
Hello Mahonroy,
I think the main cause is from the inductor. You can try some vanished indutors or you can vanish your inductor if possible.
Another way is to reduce your inductor. I don't know how much power you need or how much load when you can hear the audible nosie. Using a big inductor will reduce your swicthing frequecy, at some point, it will give you audible nosie.
Best Regards