header
 

 

 

Home
About
Articles
Contact Me
Podcasts
Webmail
Search
Upload
Links
Home arrow Articles on toningenieur.info arrow Common Mode Madness
Common Mode Madness Print E-mail
May 19, 2013 at 10:57 PM

Too much common mode noise on your AES3 output to pass EN55022? Almost no space left for breaking the pcb traces without destroying everything around them?
Things like these happen always after you had the first spinout of prototype boards, so that you're forced to become creative. This is a modification of a digital output that was designed to have perfect transitions and minimal rise and fall times to be as good and robust as possible. Unfortunately Monsieur Fourier cought me, and I had to change the output circuit a bit.

Usually I really enjoy fine pitch rework while prototyping, but to be honest, these 0603 TDK common mode chokes are just a mechanical mess. While it may be ok if they're really simultaneously reflowed on pads, I broke one instantly when soldering the fine wire (0.15mm diameter) to one of the pins and then bent the copper wire over to the pin of the pulse transformer. The solderable area at the edges of the choke decided to separate and live the rest of it's life on the wire instead of creating a decent connection to the choke.
The other way around it worked, but still was a bit fiddly, especially since the choke is hold in place by instant glue making it virtually impossible to remove in a clean way. By the way, the "large" resistor in the upper right corner is a 0805 1/8 Watt 110R resistor doing the termination of the AES3 output. The choke itself is smaller (0603) and has four pads ... so if you think 0805 is small, no, it's not. There is so much more that can be done by hand, I'd say down to 0603 not even a microscope is necessary for people with common vision capabilities. Though it might come handy during inspection.

But the good thing about prototypes is - you can actually test the behaviour of the whole system under real life conditions and of course, the change of parameters when applying changes to the circuits. I hope this EMC problem is fixed now, my measurements tell me it should be enough.

<Previous   Next>
 
Page generated in 0.003086 seconds