Designer’s Guide Community

‹ Previous topic | Next topic ›

Pages: 1

SSCG's spec for EMI reduction (Read 3314 times)

Monkeybad

Junior Member

Offline

Posts: 31

SSCG's spec for EMI reduction
Apr 23^rd, 2008, 7:36pm

Hello, everybody!
Has anyone designed the spread spectrum clock?
My company's product is suffering the EMI problem. I need to redesign a PLL which has a spread spectrum function. I use a fractional-N PLL to implement it. A 40kHz triangular wave is applied to the VCO's controlling voltage. A MASH1-1-1 sigma-delta modulator is used.
My question is how much attenuation in the spectrum for the EMI reduction is enough?
Can anyone give me some suggestions? Where can I find the detailed specifications?

IP Logged

taofeng Junior Member Offline Posts: 22 Belgium	Re: SSCG's spec for EMI reduction Reply #1 - Apr 27^th, 2008, 4:30am You normally should get this from your customer. As far as I know, 5-7 dB reduction is more than enough !
Back to top	IP Logged

Monkeybad Junior Member Offline Posts: 31	Re: SSCG's spec for EMI reduction Reply #2 - Apr 27^th, 2008, 7:59pm Thanks for your reply! taofeng! I'll try to contact the manufacturer to get the information. In my result of simulation, I can get about 20dbm attenuation at maximum modulation. So it seems quite enough!
Back to top	IP Logged

taofeng

Junior Member

Offline

Posts: 22
Belgium

Re: SSCG's spec for EMI reduction
Reply #3 - Apr 28^th, 2008, 1:57am

It also depends on the interesting range of the frequency. Although you could get 20dB attenuation, it not necessary to say your product will
pass the test if the frequency with maximum attenuation is not the critical one or beyond the range of interest. Also ssc technique is only effective
for radiated emission , for conducted emission it is an another story.

Watch out the timing margin of your system under the maximum spread ratio. Normally it is within +-2%.

IP Logged

Pages: 1

‹ Previous topic | Next topic ›