

I visited Scotland on holiday, a beautiful country! There is so much history in Edinburgh, and I also enjoyed visiting Glasgow. It is difficult for a consultant engineer to stop working entirely, as you can imagine. I received a call from an IT company that already has a product on the market, and they made some changes due to chip shortages, and guess what? They had problems with EMC. As a result, I asked them what the problem was, and I tried to point them in the right direction.Now since this issue is quite difficult to solve, if you don't know the reason, I decided to write a small post to direct you to solve this kind of issue.Imagine designing your own insulated DCDC converter. As a standard configuration, you have a transformer driver, an insulation transformer, and a rectifier. Here is my drawing.Can such a small circuit cause any problems?You may encounter the following issues:Excessive conductive emissionsRadiated emissions can sometimes occur
Whenever you observe harmonic multiples of a transformer driver's oscillation frequency during a CE test, note them down and determine if they are odd or even.Adding a Y capacitor between the "input" GND and the "output" GND usually cures even harmonics. I suggest leaving some physical space for a Y capacitor, maybe start the test without it and then fit it if you fail the CE test. This capacitor's value is difficult to predict until you perform the test, so I suggest leaving some physical room for it.In a previous post, I explained how this capacitor works.How about the odd harmonics?It is likely that you are emitting above 30 MHz if you see odd harmonics.We have odd harmonics because the rising edge is too fast, so don't panic.The solution is simple, reduce the rise time of your transformer driver. Adding a small resistor in series between the transformer driver and the transformer or using a transformer driver with slew rate control can accomplish this. This is what I would suggest (this is what I would suggest).Is there anything else we can do?There are still two issues to resolve.I1. SRF is the resonance frequency of every transformer. It is possible to get higher voltage and higher emission if you drive the transformer at a frequency very close to this SRF. Therefore, be careful.I2. Overdrive and saturation have occurred in the transformer. In this case, a small resistor may be able to solve any problems that may occur; a small resistor between the driver and the transformer may help solve the problem.
Written by Francesco Poderico
About the author
These articles are Francesco Poderico's work. He built EMC Design Solutions around a simple idea: EMC is not black magic, and every failure has an explanation. When he closed the company, he passed his writing to us so it could keep helping engineers. We publish it unchanged.
Facing a problem like this one?
We help hardware teams find the root cause and pass. Start with a call, or map the rules that apply to your product first.
More on Electronics

Electronics
How Parasitic Capacitance Breaks Optocoupler Isolation in Switching Power Supplies
Galvanic isolation is not high-frequency isolation. How switching noise crosses the optocoupler in isolated supplies, why it upsets the feedback loop, and the layout rules that prevent it.
Francesco Poderico · · 3 min read

Electronics
Estimating the Lifetime of a Varistor on the Mains – What an Electronic consultant should know!
Every surge ages a MOV a little. How varistors degrade on the mains, how to estimate their useful life, and what that means for products that must stay protected for years.
Francesco Poderico · · 2 min read

Electronics
Conductive Emissions due to lack of Bulk Capacitors
When common-mode filtering is not enough and low-frequency ferrites do not help, the problem is often too little bulk capacitance. How to spot it and size it before going to the lab.
Francesco Poderico · · 3 min read