Getting rid of console hum via the PSU?

[JohnRoberts]

All my gear is connected to these patchbays, and there is no significant hum with the rest of the balanced equipment. Also no hum with the console inputs and returns coming from the same patchbay. Could the problem originate somewhere between the chassis and audio grounds within the console?

Rather than hypothesize a single organic problem, lets consider the task we are trying to accomplish.

Let's ASSume as given, that the feed from the console is clean by itself (perhaps not true but for now lets pretend), when listened to with headphones, "and" the input of the product we are sending signal to is quiet when it's input is shorted (+ to -).  If both of these conditions are true then the task is simple to connect the two together without corrupting the signal in transit.

It should be simple enough to feed the external box from a different source, and/or send the console output to a different receiver to determine which interface is corrupting the signal quality.

One brute force fix, which i do not advocate or suggest, is to add a line level transformer. It is literally the easiest part of console design to provide proper differential sends and inputs (simple not cheap), while that doesn't mean they are all done correctly. Even some of my early work was not great. At least with an isolation transformer you can determine how much of your noise floor is the interface, and how much is inside either box.

It is normal to be suspicious of grounds, but kind of like weather we can't eliminate them, so we need to live with them. For proper studio audio interfaces there should be 3 wires. Audio +, audio -, and common or ground. The audio should be completely defined by (+) - (-) and independent of ground noise at either end.

The third interface lead, common or ground should be bonded to chassis ground at both ends, and this ground connection should not cause hum inside either unit. If it does, this is the well reported "pin 1" problem, and evidence that any gear with this problem has faulty internal circuit design, not some external grounding issue.

For rack gear it may be relatively simple to rewire a differential input to correct the pin 1 issue, for a console with tens of I/O more work.

Perhaps pick up a transformer for troubleshooting and determine which gear is the source of your problems.  Floating grounds is a bandaid to cover up some basic design fault. 

JR

 

[living sounds]

I was just playing around with the grounding cables from the mod (there seems to be some room for optimziations there) and realized the hum was changing with the position of the master module. Turns out the power amp sitting next to the console was introducing quite a lot of hum via the transformer... With the amp at a safe distance the problem is conciderably lessened and really good enough for all practical purposes. I guess the recent changes to the PSU were effective after all, I just didn't realize it because of the amp put there a couple months ago...  :-[

But thanks nevertheless!

 

[SSLtech]

This is probably little consolation to read now, but I'm only just catching back up, having spent some time away from the fora. (NOT in the 'jail' sense... before anyone asks!)

My standard approach with these 'console hum' problems is to ask people if it's 50Hz (60 in the US/Japan) or 100Hz (120 US/J) or harmonically-dominated 'buzz'. The causes (and cures) are significantly different.

But ALWAYS the beginning step is the same...

Unplug everything. -EVERYTHING- (both audio AND power) and power on only the console, and use ONLY passive headphones from an internal amplifier or a floating, battery-powered amplifier. Do not connect or power up ANY other equipment until you have either silenced the console or abandoned attempts to improve matters.

Also, reading the thread now, it looks as if John described the 'fix' which was retroactively applied to early SSL consoles when the frame sizes being commissioned grew significantly beyond their originally-envisioned or anticipated maximum, and the (unbalanced) mix buses grew to be sensitive to hum. -In SSL's case the solution was to add a second 'null' buss to every signal bus, and to refer these null buses to local ground at the module using the same impedance as the one with which the module fed the signal bus. Combine, invert, cancel.

When adjusting for maximum cancellation/noise rejection at the summing amplifiers, my standard test was to place a Weller soldering station transformer directly up against the buss card. -The transformer made things nice and noisy, and made things VERY easy to adjust for minimum hum. -Sounds like now you might begin to appreciate why.

Keef