EMI Test
Dir Sir/Madam
During the EMI conducted test we faced some problems in 3MHz. In this frequency band (from ~450kHz to 3.5 MHz) the we had high level conducted noise which did not change while we changed the input section filters.
Finally we changed the Y1 capacitor from 1nF to 3nF (while as we do not have the earth connection I guess the leakage current level can be acceptable) and the noise level became acceptable, you can find the results in the attached file.
It is ok while we still have some peak in those frequencies. Changing the ferrite core transformer1 with iron powder transforme2 changed the results which can found in the attached file.
It would be your kind to let me know your idea about the noise source and your recommendations to suppress it.
Comments
Dear Sir/Madam.
Thank you for your support.
We have already shield winding which is based on PI Expert offer. One side of the shield winding is connected to MOSFET and the other side is float.
In the attached file you can find the transformer build.
Regards
You dont want to connect the shield winding to the MOSFET. Try connecting it to the primary winding pin that connects to the Bulk capacitor CIN.
I have made a mistake. The shield winding is already connected to the primary winding pin that connects to the Bulk capacitor CIN.
This region in the spectrum is indicative of common mode noise - noise originating mainly from the switching MOSFET and coupling onto the secondary side (which is why the Y cap works)
Try using shield windings (see PI Expert design software for this)
Also attach a copy of your transformer build.