You need to keep well above the audio band, and well below the RF you are suffering from. In addition, for VHF and above, you need to not go too far below the RF as component parasitics become important. For example, to block 100MHz keep the filter corner frequency no lower than about 1MHz.
So I added two 1nF caps in parallel and the RFI is 90% gone. I'm guessing the remaining might be sneaking in through my preamp. It's a huge improvement.
One question I still have is whether the cap should be connected to chassis or signal ground. My Dynaco uses several chassis connections so for ease/quickness of assembly, I just connected the cap to signal ground.
However, with my Aikido all-in-one I'm not sure what to do. Broskie's doc mentions combatting RFI with a 100pF to chassis ground. Currently my Aikido is connected to chassis ground but I'm thinking of playing with this to see if I can lower the little bit of buzz/hum I'm getting. So if I connect the Aikido PCB through a resistor to chassis, should RFI filter be connected through this same resistor or directly to the chassis? I guess it's not that hard to try both.
Thanks for everyone's help.
One question I still have is whether the cap should be connected to chassis or signal ground. My Dynaco uses several chassis connections so for ease/quickness of assembly, I just connected the cap to signal ground.
However, with my Aikido all-in-one I'm not sure what to do. Broskie's doc mentions combatting RFI with a 100pF to chassis ground. Currently my Aikido is connected to chassis ground but I'm thinking of playing with this to see if I can lower the little bit of buzz/hum I'm getting. So if I connect the Aikido PCB through a resistor to chassis, should RFI filter be connected through this same resistor or directly to the chassis? I guess it's not that hard to try both.
Thanks for everyone's help.
You might also try the trick I mentioned earlier of putting a cap between the incoming signal ground and the chassis. Normally there is enough inductance between the signal wiring internally and chassis (earth) to make the external interconnect cable screen look like a reasonable antenna at RF. At RF, your whole PCB is bouncing up and down, and this of course capacitively couples to the chassis and thus a lot of garbage to enters sensitive signal nodes. The other option is to use some comon mode ferrites, but a lot of people say they affect the sound (I cannot comment on this).
If you do this cap trick, then I'd use 100pf to 1nf (experiment) to chassis (mount as close as possible to your incoming connector) and then leave your 1nF cap you already fitted to signal ground, as you have already done.
Anyway, looks like you have made progress, so happy listening!
If you do this cap trick, then I'd use 100pf to 1nf (experiment) to chassis (mount as close as possible to your incoming connector) and then leave your 1nF cap you already fitted to signal ground, as you have already done.
Anyway, looks like you have made progress, so happy listening!
Last edited:
Yes, 1nF to signal ground, then another 1nF from signal ground to chassis at the point where the signal enters the chassis. The idea is to stop RF from getting inside the box.
You could try Canare Star Quad wire instead.
Another possibility, though daunting, is to convert your entire system to use balanced interconnects.
I added 100pF from signal to ground at my output RCAs of my Aikido all-in-one and that killed all traces of RFI. I opted not to use chassis ground because bypassing to signal ground worked. I didn't see the point of adding more wiring.
I'll now have to try the same trick on my grounded grid. I think I used 47pF on it.
The interconnect type now seems to be moot.
I might also remove the RC filter from the ST70 to see if it was the preamp all along. I'll see if mounting the cage has any effect. But everything is sounding good so I may save that for another day.
I'll now have to try the same trick on my grounded grid. I think I used 47pF on it.
The interconnect type now seems to be moot.
I might also remove the RC filter from the ST70 to see if it was the preamp all along. I'll see if mounting the cage has any effect. But everything is sounding good so I may save that for another day.
- Status
- Not open for further replies.