Radiated Emission from EUT

[bhavanisurabhi]

Hi,

We have tested our equipment for CISPR 25 radiated emission limits. The emission result is attached. I have trouble shooted the problem element. There are 5 switching components in my design - microcontroller, LED driver IC, external crystal, DC-DC module for RS485, DC-DC module for LEDs. When verified in Anechoic chamber, I have localized to DC-DC module for LEDs as the component responisble for emission. remaining component are fine. I have verified for harmonics in emission graph and found 300KHz harmonics which matches its switching frequency.

I also verified by shielding the DC-DC module with Aluminium box for it. Only I have, in and out slots for cable. I have utilized shielded cable for wiring and also grounded the shield to 24V -ve (My system works on 24V DC battery in vehicle).
Emission reduced slightly but still it is crossing limits.

Any one can suggest me how to go further. 87_1302512925.jpg
is before shielding and 48_1302512925.png is after shielding dc-dc with aluminium box.

- Regards,
Bhavani

https://obrazki.elektroda.pl/87_1302512925.jpg
https://obrazki.elektroda.pl/48_1302512925.png

 

[albbg]

Can you check if the DC-DC module alone, shielded with the aluminium box, still radiates ? I mean removing the output cables, just putting a resistive load (roughly equivalent to the actual load) inside the box. Is the box completely encircling the DC-DC ?
However I think you could also have a radiation back to the input wires (or PCB tracks).

 

[grittinjames]

Hi,
Are you using common mode filter in your driver?
your al cabinet is grounded?
if it is legal post your circuit
Thanks

 

[bhavanisurabhi]

Thanks for your reply.

Not tested the DC-DC alone. And also we have not used common mode filter.

The Aluminimum Box is totally encircling DC-DC module, however there are openings for input and output cable. The input cable is connected to WAGO connector on PCB and output cable is connected to another WAGO connector on PCB.

The cables I have used is Shielded cable and one side it is connected to Chassis. And chassis is already grounded.

Use of aluminium box decreased emission but need to decrease further. The thickness of al box is 1.6mm which is quite greater than the depth of penetration at 80MHz frequency.

What I thought was for 80MHz peak emission, the aperture length should be minimum 0.1875 meters (λ/20). Openings for cables for aluminum box is around 10mm. My assumption is it is OK.

Is it possible for emission to use shielded cable and get outside the box?

I will post the circuit schematic.

 

[klystron]

Measure the conducted emissivity on the output of the DC - DC converter. This may transform to radiated emissivity

 

[bhavanisurabhi]

I want to clarify that my EUT is LED display sign.

Front side it is only transparent polycarbonate sheet & other sides there is metallic enclosure. Inside sign housing, I have provided shield box for DC-DC modules.

Klystron - I have verified with line filter at input but not at output. Since, switching frequency of dc-dc is 300KHz (harmonics found spaced at 300KHz in emission graph), do you think line filter at output will have any impact?

If yes, please clarify.

Thanks and Regards,
Bhavani Shankar