Input Transformer Advice for Common Impedance Coupling (Ground Loop)

[thecheapseats]

not questioning your troubleshooting skills - but initially your issue was stated as ground loops? - yes?... is there other hash/noise as well?... to confirm (as a tone) ground loops are a slightly flat, low B-natural)... in reading your stuff again, it sounds like you have other noise issues besides a simple vanilla ground loop...

assumptions to start (just checking here)... there isn't any older balanced i/o anywhere in the system that might be pin3 hot instead of pin2 hot?- a goofy external patch-bay with miswired points? - ground-lifted a/c connections? - or an ancient console with transformers on the i/o?... like I said - just checking...

as is always the case, going from +4/bal to -10/unbal is always the problematic hookup - as there can be different wiring combinations (xlr to rca) that work - might sorta' work - or don't work at all - as @KSTR outlined above - depending on what kind of 'balanced' outputs your console really has as well as the impedance loading the outboard device presents to the console (too complicated to go into here)...

to start (yeah it's a hassle - but it's the only way to do this right) unplug each and every input and output cable from your console (obviously power-off your computer as well) - so that all you have connected is console->amp->speakers... power-up those three devices - slowly push you monitor-buss fader to normal listening levels (maybe a little higher) - and all you should hear (now) is thermal noise.. .no loops. etc...

now being methodical - attach each device one at a time that you know to be a problem (console bal-out->device unbal in - and device unbal out-> to console bal in)... listen... make notes re: the behavior (loops/noise, etc.), as well as the wiring of the send and return cables... got a loop?...if so, now try a differently wired xlr-f to rca-male (lift the cold pin from the rca shield - let it float, taped off so it doesn't short)... and llisten again... note differences (if any)...

ok that ought to keep you busy for a few days... questions about the process? - post here...

you had other questions that take a long time to answer... one step at a time for now... pm me if necessary...

 

[antcollinet]

not questioning your troubleshooting skills - but initially your issue was stated as ground loops? - yes?... is there other hash/noise as well?... to confirm (as a tone) ground loops are a slightly flat, low B-natural)... in reading your stuff again, it sounds like you have other noise issues besides a simple vanilla ground loop...

assumptions to start (just checking here)... there isn't any older balanced i/o anywhere in the system that might be pin3 hot instead of pin2 hot?- a goofy external patch-bay with miswired points? - ground-lifted a/c connections? - or an ancient console with transformers on the i/o?... like I said - just checking...

as is always the case, going from +4/bal to -10/unbal is always the problematic hookup - as there can be different wiring combinations (xlr to rca) that work - might sorta' work - or don't work at all - as @KSTR outlined above depending on what kind of 'balanced' outputs your console really has as well as the impedance loading the outboard device presents to the console (too complicated to go into here)...

to start (yeah it's a hassle - but it's the only way to do this right) unplug each and every input and output cable from your console (obviously power-off your computer as well) - so that all you have connected is console->amp->speakers... power-up those three devices - slowly push you monitor-buss fader to normal listening levels (maybe a little higher) - and all you should hear (now) is thermal noise.. .no loops. etc...

now being methodical - attach each device one at a time that you know to be a problem (console bal-out->device unbal in - and device unbal out-> to console bal in)... listen... make notes re: the behavior (loops/noise, etc.), as well as the wiring of the send and return cables... got a loop?...if so, now try a differently wired xlr-f to rca-male (lift the cold pin from the rca shield - let it float, taped off so it doesn't short)... and llisten again... note differences (if any)...

ok that ought to keep you busy for a few days... questions about the process = post here?...

you had other questions that take a long time to answer... one step at a time for now... pm me if necessary...

Ground loops can pick up noise other than mains hum. Very typical in a PC application is graphics card noise.

 

[thecheapseats]

Ground loops can pick up noise other than mains hum. Very typical in a PC application is graphics card noise.

yep... which is why I instructed the OP to turnoff/disconnect the computer... gotta' troubleshoot one set of problems at a time...

 

[dandeliongold]

not questioning your troubleshooting skills - but initially your issue was stated as ground loops? - yes?... is there other hash/noise as well?... to confirm (as a tone) ground loops are a slightly flat, low B-natural)... in reading your stuff again, it sounds like you have other noise issues besides a simple vanilla ground loop...

assumptions to start (just checking here)... there isn't any older balanced i/o anywhere in the system that might be pin3 hot instead of pin2 hot?- a goofy external patch-bay with miswired points? - ground-lifted a/c connections? - or an ancient console with transformers on the i/o?... like I said - just checking...

as is always the case, going from +4/bal to -10/unbal is always the problematic hookup - as there can be different wiring combinations (xlr to rca) that work - might sorta' work - or don't work at all - as @KSTR outlined above - depending on what kind of 'balanced' outputs your console really has as well as the impedance loading the outboard device presents to the console (too complicated to go into here)...

to start (yeah it's a hassle - but it's the only way to do this right) unplug each and every input and output cable from your console (obviously power-off your computer as well) - so that all you have connected is console->amp->speakers... power-up those three devices - slowly push you monitor-buss fader to normal listening levels (maybe a little higher) - and all you should hear (now) is thermal noise.. .no loops. etc...

now being methodical - attach each device one at a time that you know to be a problem (console bal-out->device unbal in - and device unbal out-> to console bal in)... listen... make notes re: the behavior (loops/noise, etc.), as well as the wiring of the send and return cables... got a loop?...if so, now try a differently wired xlr-f to rca-male (lift the cold pin from the rca shield - let it float, taped off so it doesn't short)... and llisten again... note differences (if any)...

ok that ought to keep you busy for a few days... questions about the process? - post here...

you had other questions that take a long time to answer... one step at a time for now... pm me if necessary...

Yes, there are a collection of noises depending on the configuration, but all can be modified or eliminated via changing ground relationships... EXCEPT the underlying buzz/hum (60 hz and 180ish hz [not 120!]). 180hz is always louder. This is also accompanied by a midrange motorboat fluttering sound in the background of that noise.

The other noises that are preventable are typical digital "hashing", high frequency whine, and varying characteristics of hiss, all of which are introduced when changing cable lengths, cable grounding configuration, or patchbay grounding configuraton (when patchbays are used).

yep... which is why I instructed the OP to turnoff/disconnect the computer... gotta' troubleshoot one set of problems at a time...

One thing that might be worth re-stating here: I did eliminate the ground noise (resulting in only clean noise floor) when turning off the computer, connecting a device direct to/from the USB interface (via TS cables, no usb connection), leaving other devices plugged in. (This is possible because the digital mixer that is my "console" can run from the USB hardware).

This may not definitively prove that there isn't interaction occurring with other devices, but it's hard not to look towards the computer as a source of ground noise.

The part that I have had less luck with is getting a clean signal with the computer on. The only success with that so far is transformer decoupling via an ASI ADGC2.

The other issue is that the USB interface is not the final interface anyways, it was just for testing. It's being used as an AD/DA. All AD/DAs are routed via MADI optical to/from my MADI FX PCIe card. In this case, turning off the computer is not possible for testing (the "console" is a digital mixer that runs on the PCIe card).

I have also tested with all devices unplugged from AD/DA except the unit in question (tested on PCIe card).

Only clean results are when computer is powered down, or not connected to interface digitally.

Maybe a good next test would be to disconnect all AD/DAs except one and test that. There could be something happening with the grounds of the other AD/DAs and how they relate to the PCIe card. There was definitely some increase/decrease in noise depending on if the AD/DAs were in the same rack as each other, or plugged into the same circuit as the computer (really bad noise when PC and AD/DAs were plugged into my Furman and were all racked together).

I'll test with one AD/DA next and see how that works. If no positive results, then starting from scratch might be on the table, if only to check for human error.

 

[thecheapseats]

wait a second - ok, I should have asked - but are you mixing/summing only itb (in the box)?... no console/analog summing? - in other words, can you NOT pass signal and create loops and noise (for troubleshooting purposes) without a computer?... and as well, are most of these bal-send to unbal-devices connected NOT at the same time? - in other words, only to cut tracks?... if so, a different troubleshooting method/order should be employed, starting with the computer...

quick query - are your computers connected to anything wan-internet when working with audio (e.g.,ethernet> telco-adsl or cable-modem coax, etc.)... think about it...

your number of conns is not huge - that you're going from bal to unbal and creating weird-wired loops with upper-harmonics is not helping - as well as (just guessing at this point) picking up some unshielded rf-induced noise... not impossible to deal with but it is complicated...

that you can isolate it at this point (sort of) is a partial start - as it is partially computer induced... it would be my goal if I were you to isolate what is computer induced and what isn't...

every sentence I write makes me want to ask more questions - I can't type as fast as i think - so I'll stop now...

in my world a computer is just another piece of outboard gear that is mitigated accordingly so it doesn't mess with my console's 96-plus analog inputs in remix, digital-stems from pro-tools, analog 2inch, dacs, a few hundred 3-point balanced tt-patch bay conn points for outboard - and appx sixty?(lost count) of unbal to bal-input points...

just a quick reaction - as I need to reread your comments and get a better picture...

 

[thecheapseats]

ugh - forgot to ask - are patch bays using some strategy of normalled / half-normalled / or some combination which might include unintentionally un-balancing your sends and returns?... or are they all straight thru?... if the former - eliminate them as 'reliable' points when you troubleshoot..

 

[Mr. Widget]

After trouble shooting, if it turns out that you do need isolation transformers, then I would suggest you contact Jensen and buy their OEM transformers. Transformer quality is critical and buying poorly designed or low cost transformers will affect the audio performance. Jensen's transformers are absolutely top notch. If you buy their OEM transformers and do the soldering and terminating, you will save a ton of money and have stellar performance.

Line Input | Jensen Transformers

Line Input Transformer Selector GuideFor eliminating hum and buzz in line-level signals, Faraday-shielded input transformers are the preferred solution. These transformers offer extended low-frequency and Bessel-tailored high-frequency responses, resulting in ultra-low time-domain distortion...

www.jensen-transformers.com

 

[thecheapseats]

After trouble shooting, if it turns out that you do need isolation transformers, then I would suggest you contact Jensen and buy their OEM transformers. Transformer quality is critical and buying poorly designed or low cost transformers will affect the audio performance. Jensen's transformers are absolutely top notch. If you buy their OEM transformers and do the soldering and terminating, you will save a ton of money and have stellar performance.

yes to that Jensen advice... as I mentioned to OP a few posts back... the second time I bought xfmrs from dean at his (then small) L.A. shop in '74 (showing my age here) - I was poor, starting out... and he was kind enough to give me a box of scrap metal sheets with my order... had no idea what these were but he said, "You'll need this someday"...he was right... it was several very large scrap sheets of mu-metal...

 

[antcollinet]

yep... which is why I instructed the OP to turnoff/disconnect the computer... gotta' troubleshoot one set of problems at a time...

Sorry if I misunderstood - but it sounded like you were saying here:

to confirm (as a tone) ground loops are a slightly flat, low B-natural

That a ground loop is only mains hum. Whereas in fact they can pick up and carry any noise source, eg from filter capacitor to ground, or magnetic coupling of any stray field into the loop.

 

[thecheapseats]

purpose was to differentiate the fundamental (without benefit of hearing the OPs problem) of a vanilla 60hz ground loop compared to the OP's description of what he was hearing - described as "digital hash" , then later "background hum" and "4db drop in hum" - sounds to me like two different kinds of audio trash... best way to ask was , 'what are you hearing?' and 'when does it occur?'...

unbalancing a balanced signal (several ways to do it wrong - never fun) was the original problem... after more OP's descriptions it's clear there is some computer interaction as well... assuming the integrity of any audio gear he as deployed, no leaky/failed caps, no spurious multiple-coils of a/c power cords - compared to simple poor a/c or signal grounding somewhere, - asking if he heard a 60hz tone seemed logical...

 

[dandeliongold]

wait a second - ok, I should have asked - but are you mixing/summing only itb (in the box)?... no console/analog summing? - in other words, can you NOT pass signal and create loops and noise (for troubleshooting purposes) without a computer?... and as well, are most of these bal-send to unbal-devices connected NOT at the same time? - in other words, only to cut tracks?... if so, a different troubleshooting method/order should be employed, starting with the computer...

quick query - are your computers connected to anything wan-internet when working with audio (e.g.,ethernet> telco-adsl or cable-modem coax, etc.)... think about it...

your number of conns is not huge - that you're going from bal to unbal and creating weird-wired loops with upper-harmonics is not helping - as well as (just guessing at this point) picking up some unshielded rf-induced noise... not impossible to deal with but it is complicated...

that you can isolate it at this point (sort of) is a partial start - as it is partially computer induced... it would be my goal if I were you to isolate what is computer induced and what isn't...

every sentence I write makes me want to ask more questions - I can't type as fast as i think - so I'll stop now...

in my world a computer is just another piece of outboard gear that is mitigated accordingly so it doesn't mess with my console's 96-plus analog inputs in remix, digital-stems from pro-tools, analog 2inch, dacs, a few hundred 3-point balanced tt-patch bay conn points for outboard - and appx sixty?(lost count) of unbal to bal-input points...

just a quick reaction - as I need to reread your comments and get a better picture...

My final "console" is the digital mixing software "TotalMix" which is controlled via the computer via a MIDI controller, but can be run off RME hardware that has an independent power supply. This allowed me to use the UFX+ (a USB device) as a test case. However, my typical set up is with the MADI FX card (a PCIe card) as the device running that software, acting as my digital mixer (router, monitor mixer, etc). I do have an analog console, but it is typically reserved for my multi-track reel-to-reel, or hardware utility mixing duties.

However, I can produce many, if not all of the noises with the computer off via my TT patchbays (depending on ground configuration), which is currently removed from the testing path.

The FX sends in question all are (currently) balanced send and return at the AD/DA, but unbalanced through the cable connecting to the device's input and output. As stated before, some amount of improvement was measured when using balanced cabling and shorting cold to ground at the devices I/O.

Good suggestion about checking ethernet. I do not think I had run any tests involving this. Will add to the list!

Thanks for the random thoughts/suggestions. They are all welcome to help me think more about the issue as well.

 

[thecheapseats]

I believe @KSTR gave you the seminal 'rane' pdf link (been around for decades) about bal-to-unbalanced and back the other way, connections... where to unbalance them, and how to unbalance them - both at the receive and send ends of the unbalanced device... again I'll repeat, make it sound wrong - then experiment with the float and tied cold pin(s) at the unbalanced ends - to possibly make it sound "not' wrong - if that's possible - it may not be and a xfmr solution may be necessary...

moving beyond that - do you have any kind of a grounding topology in place?...

 

[dandeliongold]

I believe @KSTR gave you the seminal 'rane' pdf link (been around for decades) about bal-to-unbalanced and back the other way, connections... where to unbalance them, and how to unbalance them - both at the receive and send ends of the unbalanced device... again I'll repeat, make it sound wrong - then experiment with the float and tied cold pin(s) at the unbalanced ends - to possibly make it sound "not' wrong - if that's possible - it may not be and a xfmr solution may be necessary...

moving beyond that - do you have any kind of a grounding topology in place?...

I believe I tested all the relevant bal-to-unbal and vice versa as per my previous post:

I've done the following tests:

1.) Device Out > TS > RME In. RME Out > TS > Device In (original configuration)
2.) Device Out > TRS > RME In. RME Out > TRS > Delay Unit In (cold shorted to sleeve/ground at the I/O of the Device) @KSTR
3.) Device Out > TRS > RME In. RME Out > TRS > Delay Unit In (ground opened at the I/O of the Device)
4.) Device Out > TRS > RME In. RME Out > TRS > Delay Unit In (100 ohm resistor shorting cold to ground of cable, sleeve of jack open) @AnalogSteph
(NOTE: 100ohms is the output impedance of the device according to specs)
4.) Device Out > TRS > RME In. RME Out > TRS > Delay Unit In (100 ohm resistor shorting cold to ground of cable, sleeve of jack connected to ground of cable and leg of resistor)

The only thing I haven't tried there that I still want to is creating an alligator test clip from the ground of a balanced cable. I just need more TRS cables now!

Can you clarify what you mean by grounding topology?

I have a Furman Elite 15 DM i Power Conditioner plugged into a dedicated circuit. The Furman powers my recording hardware and feeds matching powerstrips distributed throughout the studio. The computer is plugged into the dedicated outlet/circuit that the Furman is plugged into. As mentioned before, noise gets much worse when the computer is plugged into the Furman. I have a separate isolated circuit I am use for testing the computer on a different circuit, but the effect of moving the computer to that circuit is not reliable beyond a certain point (can remove digital hash, but not the hum I'm attempting to remove).

 

[thecheapseats]

is the delay device output ts or trs?... mfg?... if ts - why (except for #1) are you using trs? - or am I reading what you did incorrectly?... read the rane pdf... solder-up test cables... don't use the patch bay... minimal config... trust nothing or you're wasting your time...

"I have a separate isolated circuit I am use for testing the computer on a different circuit"...

ok... prob should have asked this earlier... are you in the u.s/n. america @ 120vac/60hz?...

(forget the bal/un bal issue for a moment)..

lose the furman for this troubleshooting - it's an unknown unless you've opened it up and confirmed it's working correctly - hey stuff breaks...

with the furman out of the equation, plug the computer into the same circuit as your audio equip for a moment, and create the hash config scenario...does the computer cause hash now?... if so good, sorta' - prob not furman...

but leave the furman out of the siuation as you continue - regardless of the outcome of the previous test... minimal config...

is this your home? - a single family dwelling that doesn't share power with another building?... service size? 100 amp or 200 amp panel?...

so... reconnect a/c - a separate circuit (breaker) for computer - and one for audio... are they on the same ac leg - or the opposite legs? do you understand what I'm asking or know how to determine this?...

if they are on opposite legs now - two diff breakers - create the hash config scenario...does the computer cause hash now?...

 

[KSTR]

The only improvement was a 4db drop in hum when using TRS with the cold shorted to ground. Beyond that point there was no further improvement.
Putting the computer on a different circuit has little or no affect on the noise.

Sorry to hear it didn't work out as expected.

Remote diagnosis is always difficult so my recommendation now would be to start completely from ground zero in systematic incremental fashion so the the point where hum/buzz troubles start is fully identifed.

Basic I/O: loopback from DA32 to AD32 (mounted in the same rack) with long TRS cable, the RME's digital connection with MADI optical. Double check with long TS, should be no noise as well. Besides listening with monitors/phones cranked I would suggest viewing the signal on a spectral analyzer (DAW plugin, or standalone tool like REW).

Check unbalanced source, one single pedal or whatever, with unbalanced cable: Short its input with a dummy plug and connect to AD32 with long TS cable, and use the planned supply of the pedal which ideally should be a nice 2-prong galvanically isolated supply, see below. If that is still hum-free, you know that a) the source itseft is hum-free and b) the pedals supply is floating enough to result in only very low mains leakage currents.
If not, you can test with temporarily modding the TS cable to a "hum dummy" cable (core connected to shield at pedal end and NOT connected to tip) if mains leakage current is the dominant factor. If so, you need balanced connection to AD32 and or better pedal supply.

Check unbalanced source with balanced cable as explained in my inital post. Short the source's input with a dummy plug and connect to AD32 with long TRS cable, and use the planned supply of the pedal. If that is hum-free you're good to go. If not, not much you can do as the source signal itself already is not hum-free, especially when the hum-dummy test above did pass. If the pedal supply is an AC supply or a high-ripple DC supply that could be a root cause.

Now connect the pedal's input with a long TS cable to the DA32. That still should be close to hum-free but of course I would recommend the special cable for servo-balanced out to unbalanced in. That really must give hum-free result, if not, back to square one, something went amiss.

Next, check further pedals and other unbalances devices in a similar fashion, individually, one at a time. After this it's a good idea to move on to the shortest possible cables as planned. Do some double-checks with these, of course.

Now comes the litmus test, use several pedals simultaneously. A big variable is the pedal supply. Ideally you have one single galvanically isolated supply per pedal, that's the gold standard. Shared supplies, especially when daisy-chaining the supply from one pedal to the next is a common source of hum/buzz/noise problems. Earth grounded supplies as well are not ideal. There might be need for some experimentation on that front.

At some point you need to bring in the UFX+, with the same strategies. Also, now would be the time to check what happens when the PC digital connections is copper, it should not make any difference.

It could well be that your noise problem is solved or at least tolerable with one or a small number of pedals but starts to accumulate once your hitting 10+ pedals, even when all devices passed the individual test and when used with best possible supply.... well, let's see...

 

[dandeliongold]

is the delay device output ts or trs?... mfg?... if ts - why (except for #1) are you using trs? - or am I reading what you did incorrectly?... read the rane pdf... solder-up test cables... don't use the patch bay... minimal config... trust nothing or you're wasting your time...

"I have a separate isolated circuit I am use for testing the computer on a different circuit"...

ok... prob should have asked this earlier... are you in the u.s/n. america @ 120vac/60hz?...

(forget the bal/un bal issue for a moment)..

lose the furman for this troubleshooting - it's an unknown unless you've opened it up and confirmed it's working correctly - hey stuff breaks...

with the furman out of the equation, plug the computer into the same circuit as your audio equip for a moment, and create the hash config scenario...does the computer cause hash now?... if so good, sorta' - prob not furman...

but leave the furman out of the siuation as you continue - regardless of the outcome of the previous test... minimal config...

is this your home? - a single family dwelling that doesn't share power with another building?... service size? 100 amp or 200 amp panel?...

so... reconnect a/c - a separate circuit (breaker) for computer - and one for audio... are they on the same ac leg - or the opposite legs? do you understand what I'm asking or know how to determine this?...

if they are on opposite legs now - two diff breakers - create the hash config scenario...does the computer cause hash now?...

The delay device has TS outputs. I have been using TS connections and moved to modified TRS connections as per the Rane PDF and other's suggestions (tying cold to ground at the delay device. This caused the 4db improvement in noise.

I'm in US. !20vac/60hz

Single family home, no shared power with other building. I could not find any information on the panel that tells me if it's 100 or 200 amps.

I'm not familiar with the term 'ac leg'. I do not know how to determine which ac leg a circuit is on.

I will test with the Furman removed as you've suggested. Thanks!

 

[dandeliongold]

Sorry to hear it didn't work out as expected.

Remote diagnosis is always difficult so my recommendation now would be to start completely from ground zero in systematic incremental fashion so the the point where hum/buzz troubles start is fully identifed.

Basic I/O: loopback from DA32 to AD32 (mounted in the same rack) with long TRS cable, the RME's digital connection with MADI optical. Double check with long TS, should be no noise as well. Besides listening with monitors/phones cranked I would suggest viewing the signal on a spectral analyzer (DAW plugin, or standalone tool like REW).

Check unbalanced source, one single pedal or whatever, with unbalanced cable: Short its input with a dummy plug and connect to AD32 with long TS cable, and use the planned supply of the pedal which ideally should be a nice 2-prong galvanically isolated supply, see below. If that is still hum-free, you know that a) the source itseft is hum-free and b) the pedals supply is floating enough to result in only very low mains leakage currents.
If not, you can test with temporarily modding the TS cable to a "hum dummy" cable (core connected to shield at pedal end and NOT connected to tip) if mains leakage current is the dominant factor. If so, you need balanced connection to AD32 and or better pedal supply.

Check unbalanced source with balanced cable as explained in my inital post. Short the source's input with a dummy plug and connect to AD32 with long TRS cable, and use the planned supply of the pedal. If that is hum-free you're good to go. If not, not much you can do as the source signal itself already is not hum-free, especially when the hum-dummy test above did pass. If the pedal supply is an AC supply or a high-ripple DC supply that could be a root cause.

Now connect the pedal's input with a long TS cable to the DA32. That still should be close to hum-free but of course I would recommend the special cable for servo-balanced out to unbalanced in. That really must give hum-free result, if not, back to square one, something went amiss.

Next, check further pedals and other unbalances devices in a similar fashion, individually, one at a time. After this it's a good idea to move on to the shortest possible cables as planned. Do some double-checks with these, of course.

Now comes the litmus test, use several pedals simultaneously. A big variable is the pedal supply. Ideally you have one single galvanically isolated supply per pedal, that's the gold standard. Shared supplies, especially when daisy-chaining the supply from one pedal to the next is a common source of hum/buzz/noise problems. Earth grounded supplies as well are not ideal. There might be need for some experimentation on that front.

At some point you need to bring in the UFX+, with the same strategies. Also, now would be the time to check what happens when the PC digital connections is copper, it should not make any difference.

It could well be that your noise problem is solved or at least tolerable with one or a small number of pedals but starts to accumulate once your hitting 10+ pedals, even when all devices passed the individual test and when used with best possible supply.... well, let's see...

I will add a loopback test with TRS and TS cables between the M32 to my list of tests.

As per your suggestion about using a dummy plug, I did follow the dummy plug tests via the Whitlock PDF to determine that the device itself or input to it are not the cause of the hum. It seems to be at the coupling of the output of the device and the input of the RME. I will retest this with the above considerations and variables.
(Note: I'm using two female TS jacks, with a known good 12" patch cable for convenience. The jacks are wired with 1kohm resistors as per the PDF. I'm assuming this shouldn't matter as long as nothing is shorted pins on the TS jacks.)

I'm not sure what a 'galvanically isolated supply' refers to in your comment, but I am using the supplied proprietary pedal with the device (Strymon TimeLine is my test device). Not to add too many points of data, but the hum I'm attempting to address is also present in my modular gear, which uses an entirely different power supply (3 prong built in supply vs the TimeLine's 2 prong wall wort)

 

[Mr. Widget]

I'm not familiar with the term 'ac leg'.

Virtually all residential wiring in the US comes in from the street as two opposing 120V AC legs. Across the two legs is your 240VAC, and half of the 120V circuits are on one leg and half are on the other leg.

If all of your devices are plugged into a single 20A circuit, then you don't need to worry about the two legs.

 

[thecheapseats]

Virtually all residential wiring in the US comes in from the street as two opposing 120V AC legs. Across the two legs is your 240VAC, and half of the 120V circuits are on one leg and half are on the other leg.

If all of your devices are plugged into a single 20A circuit, then you don't need to worry about the two legs.

yes true - however OP decided to move the computer to a second breaker circuit from the audio devices - to see if there was a difference (purely for testing purposes)... I was asking as results might only be meaningful if that second circuit was know to be on the opposite a/c leg from the audio gear...

 

[Philbo King]

Aside from the logical procedure called for in fixing ground loop problems... Aside from that, the reasonable course would be to try adding transformer(s) to one channel first and doing before and after measurements of the hum. Note I don't advocate this; finding and fixing the ground loop should be foremost.