"Ticking" interference [resolved]

[mcrushing]

Hi, all.

I have a 100% stock crack that normally runs quiet. Recently we started renovations to the living room/listening space, so I set the amp up on the desk in my home office, where it started picking up a LOT of noise: It sounds like RFI, but it cuts in and out a few times a second. Imagine a cross between an ancient dial-up modem and an automatic watch.

It's not the tubes, volume has no effect, and I can disconnect the source right at the RCA jacks and still hear the noise. 

On the desk, the amp was surrounded by a laptop, LCD monitor and dongle dac. That kinda troubleshoots itself, but moving to another room didn't solve it. (I should note, in its current, temporary spot, the amp needs to be in some proximity to a Raspberry Pi (via a couple dongles) – but this is also the case in my permanent setup, and I never had issues.)

But here's where it gets weird: Tonight, with 'phones still on, I picked up the amp by to move it a few inches further from the Pi – and the noise STOPPED.  I didn't actually move it...I just touched it. Waving my palm around the amp changes the volume of the noise. But placing my hand on the wooden side panel – the WOOD, not the metal, not anything in the room that's metal – and, notably, on the side of the amp that's FURTHEST from the Pi, cures it. Instinctively I think the wiring from the jacks to the volume pot must be the issue.

Hopefully, this will all disappear as easily as it appeared when I put my main rack back together. But I still want to better shield the amp. One no brainer is to replace the stock braid with a length of shielded, twisted pair (and connect shield to ground at the RCA jack.)

What else can I try to keep out the RFI/EMI nasties?

[Paul Birkeland]

I'd start by tightening down the safety ground and power transformer screws.  You can add a 0.1uF ceramic capacitor between the RCA jack ground connection and terminal 22 (provided you get a good 0 ohm reading between terminal 22 and the chassis safety ground screw).

[mcrushing]

Cool, thanks Paul.

I'll do screws first, and also check the joints while I'm under the hood.
Just for my own electronics enrichment, I have a few questions when you get a minute.

1. It occurred to me to blame the input wiring because it's the closest thing in proximity to my hand when I placed it on the side of the wooden case. But if those wires were picking up noise and I'm hearing it on the output, it seems to me the volume would affect it...which it doesn't. Did you not mention it because you'd already ruled it out? 

2. What's the cap actually doing in this application? My assumption is decoupling AC to break a possible ground loop. Why that location, though?

3. Why do body parts effectively shield electronics from RF even when they don't have an obvious path to ground? If putting my hand against the wooden side panel stops the amp from picking up interference, why wouldn't putting a hunk of permeable metal in that same spot work?

[Paul Birkeland]

The cap provides a path from your cable shielding and the input audio ground to earth that's very short and keeps it out of the input of your amp.

Problems like these are almost always just proximity to a Wi-Fi/cellular equipped device that's overly noisy.

You'd have to do more experiments to see if metal is doing a better job at eliminating whatever issue you're having compared to your watery appendages.  These situations can be very difficult to predict with much certainty unless you kinda go all the way (a full aluminum box with RF tape.

[mcrushing]

Resolved.

The screws were already mighty tight and all the joints and wiring looked fine. But you were right on the money...the culprit was a mesh router, about 4-5 feet away.  Like I said, this is a temporary configuration. The Raspberry Pi feeding the amp was hard wired into that, so I dug up the longest Cat5 I could find and moved the amp an additional 10 feet from that router.

Tomb-like silence. Thanks for your help.