Yet another Grounding Question-Separate 'Earth'


I emailed this question to Nsgarch since he gave advice on another thread respecting the separate grounding of a dedicated subpanel, but at the risk of making some of you read yet another grounding question, I decided to post it as well. Here goes:


My electrician has installed a separate subpanel for the audio system which is 'upstream' even of the main breaker panel in the house.
It will have several dedicated lines, each with a 20 amp breaker (Square D) running separate grounds to Hubbell Hospital Grade Outlets. I was concerned about the potential 'difference' among these separate lines- one will support mid-hi-freq. amps, others, the subwoofer amps, and a third, the lower powered front-end equipment (preamp, phono stage and TT- no digital). I do have one of those Granite Audio thingies which permits me to 'star ground' everything to a single point in the system, FWIW.
But, and here's the really critical question- my electrician has proposed a pair of separate ground rods about 10' from the main ground for the rest of the house electrical system, and in his view, the audio system subpanel would be grounded just to these new ground rods, not connected, by ground or anything else, to the rest of the house. In one of Nsgarch's postings on this subject, he indicated that there could be a differential in the two different panel groundings which could put current to the 'neutral' and create a shock risk. As I understood the advice, it was to make sure that the audio subpanel shares the same 'earth' ground as the rest of the house.
Could you comment?

=====================================
128x128whart
>>"It was my intention to have him set up a number of 20 amp dedicated lines from this new subpanel, strictly for the audio-only system, and rather than have him ground the receptacles at each wall box, to run separate ground wires back to the subpanel for each- not daisy chaining the receptacles to each other. I have ordinarily used Hubbell Hospital Grade receptacles"<<

New audio branch circuits should be dedicated circuits. Each with its own Hot, neutral, and equipment grounding conductors.

Nothing wrong with Hubbell Hosp grade receptacles. Just make sure they are the HBL8200H (15A) or HBL8300H (20A) type. Note the "H", that means non nickel plated. Hey do me a favor. When you speak to Mr. Hubbell will you confirm the gutts inside the 15A and 20A recepts are identical. And only the face plates are different to be in compliance with UL and NEC? I contend the gutts are the same, 20 amp rating.
==========

>>"The run from the panels up to the room is easily 75-100 ft., depending on routing."<<
>>>>>>

That would explain the new feeder and sub panel. The smallest wire I would of used is #3awg copper. If the electrician is indeed using #6awg, jmho, it ain't big enough, not for the distance for audio gear.
=========

>>"The only reason I hadn't considered another transformer for this system is that the one currently used in the video system is located in the room, inside one of the Mid-Atlantic racks, and I can hear it hum at idle, when all the video equipment is shut off."<<
>>>>>>>>>>

What manufacture is it? Is there possibly a midway point, or closer, to the audio equipment where the xfmr could be installed?
===========

>>" (When the video system is on, the fan noise masks that). I suppose one could set up an isolation transformer down in the basement, near the panels, but would that defeat the purpose if the outputs of the transformer then have to make that long run? Apologies in advance for my ignorance"<<
>>>>>>>>>

Not ignorance at all. Your logic makes sense. If it had to be in the basement, by the main electrical panel, imo it would just need to be a little larger, more KVA. Kind of like a bigger battery.
===========

>>"OK, went back down and checked- no new 60 amp breaker that I can see in the main panel, to support the new subpanel. Is it possible that the subpanel is wired as if an 'extension' of the main panel, and that he fed it, and its array of 20amp breakers from the main panel, but before any of the other breakers (other than say, the monster one at the top of the panel?) This would explain his statement about setting it up 'upstream' of the existing breakers (sic)."<<
>>>>>>>>>>>

Maybe he has not installed it yet. Or maybe you missed it. The breaker and wire could be larger than #6 with a 60A breaker. Just make sure one is installed after the the 200A main.
==================

Email this Audiogon member, Cincy_bob , he installed a 7.5 KVA Topaz xfmr with electrostatic shielding a while back . Ask him if his is noisey. I am sure he can give you some pointers.
==============
Jim
OK, here's the scoop: Richie, the electrician, completed the wiring of the new outlets to the room, as follows (BTW, I printed out this thread for him to read, and apart from a few questions about his competence, he explained to me why what he did met Code and was otherwise kosher): the 3 20 amp circuits are wired to the subpanel using #10 wire and each has a separate ground back to the subpanel. The subpanel ties into the main board at the 200amp breaker, but before any of the other breakers in the main panel. It shares the same ground as the main panel, but has an additional ground to earth in proximity to the house grounding rod. The subpanel is protected by its own 60 amp breaker, which then feeds the three 20 amp circuits. The subpanel is mounted adjacent to the main panel.
Since local supply houses did not have the Hubbell receptacles on hand, I bought a bunch of Porter's tweaked Hubbell receptacles.
Now, for the punchline. Perhaps a little less noise than when hooked up to the regular house outlets, but still suffering some grounding noise issues. Prior to installation of this new electrical subsystem, my best result was obtained by deriving all system power from the same set of outlets.(Remember, I don't draw a huge amount of current with this system).
Obviously, once I 'cheat' everything but the preamp, the grounding noise goes away, but I am still reluctant for safety reasons to lift the grounds on any of the equipment.

I have achieved quietude right now with the following arrangement- the Duo bass amps remain grounded through the 3 pin cord (Shunyata) plugged into a Hydra 2 which goes (with 3 pins) into one of the three new dedicated lines. (Each speaker has its own Hydra 2/Shunyata cord array, fully grounded, plugged into its own wall outlet right behind the speaker- both 'speaker ac' receptacles are on the same dedicated line- call that dedicated line "3").

The Lamm L2 line stage is plugged directly into the wall on dedicated circuit 2, no cheater. Ditto, the Steelhead phono preamp. Floating the ground of the STeelhead or not makes no discernible difference, so I have not cheated it.
The Audiopaxes are plugged into a Hydra 4/Anaconda which goes into dedicated circuit #1. Here, to get rid of the hum, I have to lift the ground on the amp cords going into the Hydra 4. (The Hydra 4 is using all 3 pins into the wall at dedicated circuit # 1 but viewing the Hydra as simply a wall receptacle extension, the amps really aren't
'grounded.')
I also tried plugging the Hydra 4/Audiopaxes (no cheater on the amp cords) into dedicated circuit #2, which powers the Lamm line stage, and no improvement.
I have hooked up the Granite Audio Ground Zero product, and have separate ground lines from the Duo amps, the Audiopaxes(cheated, as above), and the Steelhead all converging at the Ground Zero device. With some experimentation, this yielded a relatively quiet, hum free system, but if I ground the Ground Zero to a wall socket, using the supplied 'dummy' 3 pin plug, the hum is back.
For safety reasons, I'd still prefer to be able to avoid any cheaters. To sum up, right now, the only cheated plugs are on the Audiopax amps. (I assume that unplugging the Hydra 4 from the wall when I am not using the system makes no difference for safety purposes, since the amps are still connected to the rest of the system via the interconnect and speaker cables).
So, if you've hung in with me so far, it seems that there is almost no way around lifing the ground in this system somewhere to make it noise free. The grounding problem is an 'intercomponent' grounding difference, and not a bad ground from the wall. The grounding differential appears to be between the Audiopax amps and the amps in the subwoofers.
The only thing I didn't try was to also hook up the Duo subwoofer amps to the same dedicated circuit as the front end electronics- I could try that, which would mean the whole system is running on one 20 amp circuit- probably not a problem from a current draw perspective, but it would mean , if that worked, that I would have to have Richie rewire the outlets now on circuit 3 to tie into circuit 2 instead- I will test this temporarily before having him do that. It also means that having multiple dedicated lines for the same system is pointless, given the grounding differential between the circuits.(And, to the extent there is some subtle benefit to having more current on hand, despite the modest demands of my system, I am losing the benefit of that by forcing everything to feed off one 20 amp line).
With sincere apologies for the length of this post, I invite any thoughts or suggestions (:)!
oops> One last thought. Preamp fully grounded. Ground lifted on Audiopaxes. But, both components have separate ground wires to the Ground Zero device. Does this mean that the amps are effectively grounded via the preamp? Is that true even though the various connections on the Ground Zero are 'buffered' from one another?( ie, the Ground zero allows separate ground connections that interact with each other thru multiple position 'impedence' switches).
And, note that the Ground Zero is not itself separately grounded since if I plug its dummy ac plug into a wall outlet, the hum comes back.
Just to add to the confusion.:)
Whart,
You called your new dedicated branch circuits 1, 2 & 3. Is your new sub panel a 120/240 volt panel? Or is it a straight 120 volt panel? Does your circuit numbers match the factory breaker numbers on the panel front cover? If the sub panel is 120/240V are all the dedicated circuits on the same line, L1 or L2, in the sub panel but not one on one Line and the other two on the other Line? Audio equipment that is connected together by wire signal carrying ics should be fed from the same line. If not, this can be a source of noise.

For another source of noise, have you ever checked the AC plug orientation for the proper AC polarity of the primary side of the power transformers of the Amps and preamp? You will need a decent multimeter and a pair of tin snips. The snips, you will need to trim the wide blade on one of your ground cheaters. It will need to be trimmed enough so it will plug in either way into the electrical receptacle, outlet. If the ground cheater is the pigtail equipment grounding conductor type tape off the bare lug on the end. If it is the type with the metal equipment grounding strap either cut it off or tape it so it cannot touch the receptacle cover plate mounting screw. The screw is grounded.

Start with one of the Amps. The Amp must be totally isolated from the grounding system, of your electrical system. Disconnect the ics from it and the preamp. Plug the amp into the modified cheater plug and plug the cheater into the wall outlet. Turn on the amp. Set your multimeter to AC volts. Insert one of the test lead probes into the wall outlet ground hole, make sure you make good contact with the ground contact. Touch the other test lead probe to the back chassis of the amp. If the back is painted remove one of the screws and touch the probe to the bare threaded area of the chassis. You should now show an AC voltage on the meter. Write down this voltage measurement. Turn off the amp. Unplug the ground cheater and rotate it 180 degrees and plug the cheater back into the outlet. Wait a couple of minutes to let the caps in the amp to bleed off. Now turn the amp on. Connect the multimeter test probes in the same manner as before. Note the voltage reading. The **lowest reading** of the two readings is the correct AC polarity, phase relationship, for the primary winding of the power transformer of the Amp. If the plug orientation is correct the lowest AC voltage reading measured, the wider polarized blade of the plug for the Amp will match that of the neutral contact of the receptacle, (20 amp recept T slot).

Repeat the same test procedure for the other amp/s and the preamp.

http://www.boundforsound.com/tweak.htm