Better late than never to this discussion...
This is an area I have suffered through on many broadcast facilities. I
felt like "who died and left me as the grounding guru!" It happens if
you know a bit more than others about something, you get involved with
it. So I decided I needed to learn a lot more. Fortunately, I had access
to Ralph Morrison and attended Neil Muncy's "Pin One Problem" seminar in
the LA area. Neil didn't live long enough for us to have a very long
talk about isolated grounds and wiring practices.
Suffice it to say, that power systems need to be designed based on many
I was able to stop one potentially hazardous situation. There was a
facility with a microwave tower at one end of the building and the power
entrance was several hundred feet away and they wanted to run an
isolated ground for the equipment room those several hundred feet and
use that as the ground entrance for that room that included the
microwave gear feeding the tower. I don't care what you run several
hundred feet, it has significant inductance when thinking about lighting
discharges (which are very fast).
Ralph taught be about ohms-per-square ... for any material, it doesn't
matter what, if you have a square piece, the impedance (or at least the
resistance) will be constant if it's a square foot or a square mile,
though I wonder what happens when we have multiple wavelengths. I never
discussed that with Ralph. This brings up the niceties of a traditional
computer-floor-type signal reference grid. It's just a square piece of
material that has been "swiss cheesed," So its impedance is higher than
if it were a solid piece of coppoer, but no where near that of a single
One last story from the field: A facility in Burbank had lots of video
hum and we had isolated grounds with single ground conductors running in
the conduit (as required by code). The racks were bonded together and
there was a separate rack ground system as well.
What apparently was happening was that the ground wire was acting as the
secondary of a single-turn transformer (remember how well the Wen and
Weller soldering guns worked with a single turn secondary). This
injected hum into the receptacle grounds that was carried to the rack
ground. We could see 50 mV of hum between adjacent racks (which did
wonders to the unbalanced baseband video typically run between two
adjacent racks without differential inputs (at least back in those days).
The solution was bonding the isolated ground wires to the rack and the
conduit where it entered the rack. In that way, the generated voltage
was only induced in the conduit and didn't get into the rack. We
couldn't easily pull the green wires as that might have been a code
violation, but it wasn't a code violation to bond them at the far end.
Briefly, when I built my home studio here in Aurora, Ontario, in 2004, I
was concerned about the potential for lightning strikes. Our local
street light had apparently been struck in the past and the house nearer
the street light than mine suffered minimal damage, my house had some
stuff fried, and the neighbour on the other side (farthest of the three
of us from the transformer) had lots of stuff fried.
So my studio and workbench ended up being powered through a 240:120 V 10
kVA step down transformer, isolating the power from the incoming hots
and neutral. The secondary of my separately derived system was grounded
to the same cold water pipe incoming ground as the power system.
I ran Romex from the transformer to a panel in the studio. I wired the
panel single phase. There are nine branch circuits. Three of these feed
individual plug-mold strips on the rear of my equipment shelves. Three
feed wall receptacles around the room (including tape machines and
powered speakers). Two feed the console. One is on 24/7 as is one of the
plugmolds and one of the wall receptacles because I've added UPS's for
many of the pieces of equipment. The other console one is switched.
The ninth circuit feeds the workbench which has a GFCI receptacle wired
at the workbench.
This has worked very well. I used BX (AC) cable and although the
inspector would not let me run an auxiliary ground ring, I arranged the
BX so the console and wall receptacle feeds came down both sides of the
room (one BX per circuit) and then they crossed in the console. All BX
was anchored with metal clamps that effectively bonded two BX cables
together. This formed a ring around the rear of the studio with two
"horns" extending to the monitor wall.
On 8/16/2016 7:47 AM, Carl Pultz wrote:
> Interesting views, all. Thank you. And great to have input from Gary Galo, whose credibility for me is very high, as we both work in similar facets of audio. I haven't had a chance to try regeneration, but hope to eventually. For small monitoring systems, it is practical. Unfortunately not for larger facilities, as Corey points out.
> Some years ago I was surprised by a utility crew, which came by to install a transformer right at my drop. I think there was only one or two serving my city block, and one of the guys said the new unit should help stabilize the line. Of course I played some music as soon as they were done. Sadly, I couldn't convince myself there was a difference. It is shared with three other houses. Maybe if my gear was more sensitive to the AC, it would have benefited more from the transformer, or from the newly bonded ground and power connections. I may also have less interference than a resident of Tokyo suffers.
> We old-timers may be fully aware of this, but I always encourage audiophiles to have a dedicated circuit or two installed for their hifi. When I moved about 15 years ago, it was one of the first things I did, as my stereo was sounding like crap compared to what was going on in the old place. Living there in an older neighborhood, I was aware of the variation of sound quality day vs. night. That house was near by an antenna farm, so I had wired up shielded Romex with isolated grounds. Didn't go that far in the new place, but non-spliced runs on new breakers brought the magic back, with less variability than before. The service was 30 years old, so I tested and reinforced the ground straps. I couldn't measure voltage drop on the panel, but I wanted to be sure it wasn't contributing noise.
> Some systems prefer to have all components on the same AC circuit. I've found that's often the case with tube gear, or products such as Naim systems, which are very concerned with common grounds. Remember turning the plugs each way to find the best orientation? Can't really do that now, and it probably isn't necessary with modern designs. When I was using a power-hungry Bryston amp, it definitely liked its own AC line. My current Benchmark components seem happier sharing. They use switching power supplies. I wonder if Gary looked into the effect of the regenerators while auditioning the DAC2/AHB2 system.
> Anyway, I'm reminded of what I was told years ago when first getting serious about hifi: everything matters. If it matters to you. It can lead to madness. Or to joy.
> We all can either lament or fear hearing loss. But, reading about the research behind the MQA coding system has turned me on (within my limited understanding) to the idea that high frequency perception isn't the only thing our sense depends on. Fortunately, other factors that are not so subject to aging are maybe more important. Note that our Japanese colleague spoke of dynamics, imaging, and clarity, which are not primarily dependent on bandwidth. My older brother, a professional brass player, has significant loss of sensitivity, an occupational hazard. We were together just last week and I made many adjustments to his hifi. He heard pretty much all of it, just at a volume level that I could hardly stand. But concert-like levels are non-negotiable for macho trombone players!
Richard L. Hess email: [log in to unmask]
Aurora, Ontario, Canada 647 479 2800
Quality tape transfers -- even from hard-to-play tapes.