Not long ago in the CE-V (conducted voltage method) experiment, the product at 530KHZ and its multiplier test value exceeded the limit.
You can see that the limit value of the base frequency is just in the limit change area, then we put this radiation out of the interference frequency from 530K lowered to 480KHZ or so, then whether the problem can be solved?
From the figure of the limit value area can be seen, 480KHZ limit is relatively wide, then this program is feasible.
The next step is to find out the location of the product radiating that frequency and make a frequency change. Conducted voltage method and failure in about 500K, this frequency is not a high-speed signal, generally related to the power supply.
Examine the power supply on the PCBA, found that there are two power supply oscillation frequency is set in this interval. One is a power supply chip, one is the backlight drive power chip.
In turn, the oscillation frequency of the two power supplies to change the test, found FAIL project is related to the backlight power supply, a power supply is not affected by this test, because its input string of an inductor, so its interference is not transmitted to the front. And to the backlight of that way there is no string inductor, resulting in radiation out.
Finally, by modifying the oscillation frequency can still let the experiment pass, but this is only a clever “to avoid the sharp end” approach, not the fundamental solution. To completely eliminate can only change the PCB, add inductors and capacitors to form an LC filter to isolate interference.
Of course, in this process, you can use the spectrum meter to detect in order to rectify.
