Not your average ground loop

[Jay Levitt]

I'm on a noise kick this week, so I've been running around lifting and
reattaching grounds, killing circuit breakers, and generally making a
mess of my studio to solve some noise problems.

I narrowed most of them down to improper connection of balanced and
unbalanced equipment (shocker, I know). I even discovered that the MOTU
MTP-AV grounds its MIDI IN ports (in violation of the MIDI spec, I might
add), and snipped some ground wires on my MIDI cables to solve -that-.
But one thing has me mystified.

I have a very-low-level 60-cycle hum on the outputs of my Roland A-90
keyboard. I have lifted the audio grounds; it's still there. I have
disconnected the MIDI cables to be sure I didn't miss something. There
IS no safety ground; it's a two-prong plug. There is absolutely no
ground anywhere near this thing. It's the floatingest 85-pound object
you've ever seen. It does have balanced outs, according to both the
manual and the rear panel silkscreening, although they are that magical
kind of output that can feed either balanced or unbalanced inputs, so as
I understand it that means there are no transformers involved. Yet when
I run it through a Countryman type-85 active DI, the hum goes away.

How can this be? Is the answer going to involve terms I have no hope of
understanding, like "capacitance", "inductance", or "analog"? Am I
already sorry I asked?

--
Jay Levitt |
Wellesley, MA | Hi!
Faster: jay at jay dot eff-em | Where are we going?
http://www.jay.fm | Why am I in this handbasket?

[[email protected]]

Jay Levitt > wrote:

> I have a very-low-level 60-cycle hum on the outputs of my Roland A-90
> keyboard. I have lifted the audio grounds; it's still there. I have
> disconnected the MIDI cables to be sure I didn't miss something. There
> IS no safety ground; it's a two-prong plug. There is absolutely no
> ground anywhere near this thing. It's the floatingest 85-pound object
> you've ever seen. It does have balanced outs, according to both the
> manual and the rear panel silkscreening, although they are that magical
> kind of output that can feed either balanced or unbalanced inputs, so as
> I understand it that means there are no transformers involved. Yet when
> I run it through a Countryman type-85 active DI, the hum goes away.

> How can this be? Is the answer going to involve terms I have no hope of
> understanding, like "capacitance", "inductance", or "analog"? Am I
> already sorry I asked?

What makes you think it's a grounding problem? My guess is the
keyboard itself is putting out low level AC hum. In other words
it's own power supply may be a tad under-filtered. I'm guessing
the reason the hum disappears with the DI is that the internal
hum appears the same on both of the balanced outputs. Therefore
if you feed into a true balanced DI, they are cancelled.
But if you use any unbalanced feed from either side of the
balanced out, you'll see (well, OK, hear) that hum.

There. Did I say, "capacitance", "Inductance" or "analog"?
I don't think so.

Benj

--
SPAM-Guard! Remove .users (if present) to email me!

[ryanm]

"Jay Levitt" > wrote in message
...
>
> I have a very-low-level 60-cycle hum on the outputs of my Roland A-90
> keyboard. I have lifted the audio grounds; it's still there. I have
>
It's in the keyboard, maybe x-former hum or somthing. Our key player
uses an A-90 and we have that hum in our live mix and have found no way to
get rid of it. Actually, his Hammond XK-2 is even worse, the A-90 just adds
an almost inaudible hum, but the XK-2 can be quite obnoxious. But I think
the XK-2 actually is a grounding problem, because it's worse at some venues
than others, generally varying directly with the apparent quality and age of
the wiring in the place.

> I run it through a Countryman type-85 active DI, the hum goes away.
>
I'm gonna have to try this. We are using a passive direct for his A-90,
I'll have to look at getting an active one to see if it helps.

ryanm

[Jay Levitt]

In article >,
says...
> What makes you think it's a grounding problem? My guess is the
> keyboard itself is putting out low level AC hum. In other words
> it's own power supply may be a tad under-filtered. I'm guessing
> the reason the hum disappears with the DI is that the internal
> hum appears the same on both of the balanced outputs. Therefore
> if you feed into a true balanced DI, they are cancelled.
> But if you use any unbalanced feed from either side of the
> balanced out, you'll see (well, OK, hear) that hum.

Oops - I guess I didn't mention where the the outputs were going to.
They're going to the monitoring section of a Mackie HUI, which has
balanced inputs, so I would have expected that power-supply noise would
be filtered out there if it appears on both sides. Conversely, I don't
-think- the Countryman DI is expecting a balanced input, so I'd actually
expect that only one side is being used.

In fact, now that I think about it, I connected the A-90 to the DI with
a mono (tip-sleeve) cable, so it couldn't have been a balanced feed.
However, in trying to duplicate it just now, I discovered that the
Countryman may not have actually made the hum go away, because the
Countryman adds its own, slightly different, 60-cycle buzz - one which
is present when the cable is hooked up to the keyboard, or to nothing at
all, but not to the vDrums module where it usually sits... that module
has a 3-prong power supply though.

--
Jay Levitt |
Wellesley, MA | Hi!
Faster: jay at jay dot eff-em | Where are we going?
http://www.jay.fm | Why am I in this handbasket?

[Jay Levitt]

In article >,
says...
> the A-90 just adds
> an almost inaudible hum,

Yeah, I wouldn't expect you could even notice it in a live situation -
it's a "if I turn the volume all the way up I can sort of hear it" kind
of hum. I'm more worried about it building up from layered tracks, plus
the general principle of the thing...

Sadly, the Countryman doesn't actually seem to have solved it. I'm not
sure why I thought it did.

--
Jay Levitt |
Wellesley, MA | Hi!
Faster: jay at jay dot eff-em | Where are we going?
http://www.jay.fm | Why am I in this handbasket?

[Kurt Albershardt]

Jay Levitt wrote:
>
> I have a very-low-level 60-cycle hum on the outputs of my Roland A-90
> keyboard. I have lifted the audio grounds; it's still there. I have
> disconnected the MIDI cables to be sure I didn't miss something. There
> IS no safety ground; it's a two-prong plug.

Have you put a high impedance AC voltmeter (DVM is fine) between it and
various other chassis, grounds, etc. to see what sort of potential
difference exists?

[Jay Levitt]

In article >, says...
> Have you put a high impedance AC voltmeter (DVM is fine) between it and
> various other chassis, grounds, etc. to see what sort of potential
> difference exists?

Have now. The ground output from the keyboard is a whopping 0.67Vrms
(0.70Vav) off from the mains ground, and has a strong 60Hz component.
That can't be a good thing.

I initially didn't see how this should matter, with balanced audio, but
I think I get it now. Let's see if I am understanding... the keyboard
is generating "balanced" output, but that balance is based on a faulty
ground reference. So, when the signal reaches the balanced input of the
HUI, it tries to eliminate common-mode noise, but hot minus cold
(referenced to proper ground) is no longer zero. Right? (Corollary:
You probably shouldn't try to generate balanced output with a two-prong
plug.)

I tried verifying my hypothesis with the meter, but considering that hot
and cold are in the tenths of a millivolt (compared to the keyboard's
ground), the difference is too small to see compared to proper ground.

This also suggests that if someone who knew what they were doing (i.e.
not me) modified the unit, they could change out the power supply and
plug, get a real ground reference, and solve this without even changing
the (presumably awful) output stage. Hmmm.

--
Jay Levitt |
Wellesley, MA | Hi!
Faster: jay at jay dot eff-em | Where are we going?
http://www.jay.fm | Why am I in this handbasket?

[Kurt Albershardt]

Jay Levitt wrote:
> In article >, says...
>
>> Have you put a high impedance AC voltmeter (DVM is fine) between it and
>> various other chassis, grounds, etc. to see what sort of potential
>> difference exists?
>
>
> Have now. The ground output from the keyboard is a whopping 0.67Vrms
> (0.70Vav) off from the mains ground, and has a strong 60Hz component.
> That can't be a good thing.

Right.


> I initially didn't see how this should matter, with balanced audio, but
> I think I get it now. Let's see if I am understanding... the keyboard
> is generating "balanced" output, but that balance is based on a faulty
> ground reference. So, when the signal reaches the balanced input of the
> HUI, it tries to eliminate common-mode noise, but hot minus cold
> (referenced to proper ground) is no longer zero. Right? (Corollary:
> You probably shouldn't try to generate balanced output with a two-prong
> plug.)

What most of us think of as "balanced" is more properly known as
"balanced and floating" meaining not referenced to ground--the
difference between high and low lines is all that is used to derive a
signal. So-called "impedance balanced" circuits and all sorts of plain
old bad design have turned "anything with three pins" into a "balanced"
jack on "pro" equipment.


> This also suggests that if someone who knew what they were doing (i.e.
> not me) modified the unit, they could change out the power supply and
> plug, get a real ground reference, and solve this without even changing
> the (presumably awful) output stage. Hmmm.

Could also be a bad cap in the power supply or any one of several other
places in the circuit (on either end.)

First thing to do is get rid of that potential difference between the
units: Time to go poking around a bit more with the meter, checking
inside the boxes and onto the ground planes of the boards, etc.

[Graham Hinton]

In article >,
Jay Levitt > wrote:


>I narrowed most of them down to improper connection of balanced and
>unbalanced equipment (shocker, I know). I even discovered that the MOTU
>MTP-AV grounds its MIDI IN ports (in violation of the MIDI spec, I might
>add), and snipped some ground wires on my MIDI cables to solve -that-.

Now you have two violations and those cables will come back to haunt you
later.
Snip the MTP internally, it is years out of warrantry anyway.


>I have a very-low-level 60-cycle hum on the outputs of my Roland A-90
>keyboard. I have lifted the audio grounds; it's still there. I have
>disconnected the MIDI cables to be sure I didn't miss something. There
>IS no safety ground; it's a two-prong plug. There is absolutely no
>ground anywhere near this thing.

So ground it.
Most Japanese consumer goods are made to have different power supplies
depending what part of the world they are being sold to.
Fit an IEC inlet, connect the two mains wires to live and neutral and then
ground the chassis. Then use a three pin IEC mains lead.
Alternatively, bolt an earthing terminal to the chassis and run that to
your safety earth.

>It does have balanced outs, according to both the
>manual and the rear panel silkscreening, although they are that magical
>kind of output that can feed either balanced or unbalanced inputs, so as
>I understand it that means there are no transformers involved.

It probably is electronic, but transformers can drive balanced or
unbalanced inputs. The action is the same, either side may be connected to
ground, but if you ground both something may blow up.

[Jay Levitt]

In article >, says...
> What most of us think of as "balanced" is more properly known as
> "balanced and floating" meaining not referenced to ground--the
> difference between high and low lines is all that is used to derive a
> signal.

Damn. I thought I followed, but now I don't. So if balanced audio
doesn't need a ground to do its hot-minus-cold math, then why does it
matter if the ground reference was off in the first place? Is it
essentially the same problem as on my multimeter - with a huge DC
offset, the difference between hot and cold gets (literally) lost in the
noise, thanks to some analog equivalent to the "significant digits"
concept?

Thinking about it numerically, which I realize may be a lousy analogy:
If I have a 0.5mV signal, balancing it gives me 0.25mV and -0.25mV. If
my ground reference is off by 600mV, I really have 600.25mV and
599.75mV. But at the far end, 600.25 - 599.75 is still 0.5mV, just like
0.25 - (-0.25) is.

> So-called "impedance balanced" circuits and all sorts of plain
> old bad design have turned "anything with three pins" into a "balanced"
> jack on "pro" equipment.

See? See! He said "impedance". One of these days, I have got to find
an electrical primer more advanced than "Tune In The World With Ham
Radio".


--
Jay Levitt |
Wellesley, MA | Hi!
Faster: jay at jay dot eff-em | Where are we going?
http://www.jay.fm | Why am I in this handbasket?

[Jay Levitt]

In article >,
says...
> Now you have two violations and those cables will come back to haunt you
> later.
> Snip the MTP internally, it is years out of warrantry anyway.

Actually, it's reasonably new; the MTP-AV is still the only 8-port unit
that'll do proper routing, so when I needed to switch from parallel to
USB I bought it again. Snipping internally would be a pain; I'd need to
desolder all the ports, snip the ground pin, and resolder. (The ground
trace isn't on the top layer of the board.) I'm lousy at soldering, but
I'll still probably do it eventually. Luckily, my cables have molded
plugs, so lifting ground involved slicing the sheath open - I will
*never* mistake these for a normal cable!

> Alternatively, bolt an earthing terminal to the chassis and run that to
> your safety earth.

That is an *excellent* idea, assuming they actually use the chassis as a
reference... I'll have to open it up and see.

--
Jay Levitt |
Wellesley, MA | Hi!
Faster: jay at jay dot eff-em | Where are we going?
http://www.jay.fm | Why am I in this handbasket?

[Kurt Albershardt]

Jay Levitt wrote:
> In article >, says...
>
>> What most of us think of as "balanced" is more properly known as
>> "balanced and floating" meaining not referenced to ground--the
>> difference between high and low lines is all that is used to derive a
>> signal.
>
>
> Damn. I thought I followed, but now I don't. So if balanced audio
> doesn't need a ground to do its hot-minus-cold math, then why does it
> matter if the ground reference was off in the first place? Is it
> essentially the same problem as on my multimeter - with a huge DC
> offset, the difference between hot and cold gets (literally) lost in the
> noise, thanks to some analog equivalent to the "significant digits"
> concept?

Not a bad analogy, really.



> Thinking about it numerically, which I realize may be a lousy analogy:
> If I have a 0.5mV signal, balancing it gives me 0.25mV and -0.25mV. If
> my ground reference is off by 600mV, I really have 600.25mV and
> 599.75mV. But at the far end, 600.25 - 599.75 is still 0.5mV, just like
> 0.25 - (-0.25) is.

Right, and if that offset is greater than the connected input
circuitry's ability to reject common mode noise (see CMRR) then you have
trouble.


Graham's suggestion is a good possibility--give those millivolts of 60
Hz a shorter path to ground than over your audio cable and you might
just be fine.

[Northamusi]

Hey Jay:
I had the same problem with my A90EX. My solution was to strap a 16-guage
copper from a chassis screw on the bottom of the A90 to the nearest unit with a
3-prong grounded AC cable -- in my case an Oberheim Matrix 6. If I recall
correctly Try it and I think your problem will disappear.

[Northamusi]

sorry.. think I hit enter by accident (damn fingers).

I believe that strapping a wire from the bottom of the A90 to a ground screw on
an ac receptacle or a nearby rack rail should also do the trick. Lemme know
what happens!
peace
Steve La Cerra

[Justin Ulysses Morse]

Jay Levitt > wrote:

> I'm on a noise kick this week, so I've been running around lifting and
> reattaching grounds, killing circuit breakers, and generally making a
> mess of my studio to solve some noise problems.

Good for you! I ought to get around to doing that. It's not like I
have any recording to do.

> I have a very-low-level 60-cycle hum on the outputs of my Roland A-90
> keyboard. I have lifted the audio grounds; it's still there. I have
> disconnected the MIDI cables to be sure I didn't miss something. There
> IS no safety ground; it's a two-prong plug. There is absolutely no
> ground anywhere near this thing. It's the floatingest 85-pound object
> you've ever seen. It does have balanced outs, according to both the
> manual and the rear panel silkscreening, although they are that magical
> kind of output that can feed either balanced or unbalanced inputs, so as
> I understand it that means there are no transformers involved. Yet when
> I run it through a Countryman type-85 active DI, the hum goes away.

First of all, transformers CAN drive balanced or unbalanced inputs
better than any other kind of circuit can (And they can RECEIVE
balanced or unbalanced signals equally well, too!).

I tend to suspect your Roland simply produces hum, aside from any
ground concerns. If you want to find out for sure, try disconnecting
EVERYTHING from it except for something to listen to (A pair of
headphones, ideally). If you do make a balanced audio connection from
the Roland to say a power amp, then I sugget using cables that have
their shields lifted at one end. Listen with and without a big fat
copper ground wire connecting the chassis of the Roland to the chassis
of the power amp. I'm only recommending this scenario for
troubleshooting purposes, not for normal use. After trying all these
things, if you ever manage to get rid of the hum while keeping the
audio, then it is in fact a grounding problem that can be solved. If
the hum never goes away no matter what, then quit worrying about
grounding. Then you can decide whether or not you want to look into
getting rid of the hum inside the Roland.

> Have now. The ground output from the keyboard is a whopping 0.67Vrms
> (0.70Vav) off from the mains ground, and has a strong 60Hz component.
> That can't be a good thing.

The reason why it's not a good thing: Think about the typical
situation you had before, where the Roland was not chassis-grounded,
except for the shielded balanced audio cables and MIDI cables
connecting its outputs to something else that is grounded. Now those
shields are providing a ground connection. Suppose there's a .6V
voltage difference between the Roland and the ground at the other end
of the audio cable. If your complex system of interconnects, equipment
racks, and power cable grounds somehow finds a return path back to the
Roland, then you can actually have current running through those shield
wires. If they're decent cables, the bulk resistance of those cable
shields will be well under an ohm, which means even your little .6V AC
ground noise will produce a significant amount of current through the
wire. This is the shield wire that's wrapped around the audio signal
wires to protect them from noise. Except now the shield is generating
an electromagnetic field that pulsates at 60Hz, inducing hum into the
signal wires. Since it's a balanced connection, this hum should get
cancelled out at the receiving end's differential amplifier (or
transformer). But the system isn't perfect, and you could very easily
be left with some low-level residual hum. Which is what you described.
The solution is to have ONE and ONLY ONE chassis ground connection on
each piece of equipment, so there's no circuit or "loop" for those
ground currents to follow. Get it?

ulysses

[Graham Hinton]

In article >,
Jay Levitt > wrote:

> Snipping internally would be a pain; I'd need to
>desolder all the ports, snip the ground pin, and resolder. (The ground
>trace isn't on the top layer of the board.)

You might be able to cut the trace around pin 2 with a scalpel or small
drill burr. Or pull the contact out of the moulding, it is the centre top
one.


>> Alternatively, bolt an earthing terminal to the chassis and run that to
>> your safety earth.
>
>That is an *excellent* idea, assuming they actually use the chassis as a
>reference... I'll have to open it up and see.

In some regions of the world it would be required to be earthed, it is just
that North America is seen as a "two pin" region as far as Roland are
concerned.
You can test continuity between the chassis and a connector 0V with a
multimeter. I would be surprised if it was not connected.

[Arny Krueger]

"Graham Hinton" > wrote in message


> In article >,
> Jay Levitt > wrote:

>> Damn. I thought I followed, but now I don't. So if balanced audio
>> doesn't need a ground to do its hot-minus-cold math, then why does it
>> matter if the ground reference was off in the first place?

> Your keyboard is floating and probably gone outside the voltage range
> that the input can reject.

Right. There's this concept I call "common mode rejection dynamic range",
that applies to modern transformerless balanced inputs. Strangely enough,
its almost never specified. IME it ranges from about 1500 volts for a good
transformer input to 3-5 volts for some "electronically balanced" inputs.
These days most balanced inputs are "electronically balanced", so go figure.

> The term "balanced" is used incorrectly these days. It used to be
> correct when valve equipment and microphones had output transformers
> with the center tap grounded, then the two signals were equal and
> opposite. However if you mix balanced/unbalanced equipment, eg with a
> jackfield, you can short out one half of the transformer winding and
> melt it. Most transformer outputs now are floating and fed into a
> differential input, which may be a transformer or electronic.

Arguably, the grounded center tap was always a bad idea. It put an
unnecessary premium on the matching of the windings on the input transformer
and the balancing of the device driving the input. One common *rule* for
modern balanced inputs is to NOT ground center taps or anything that looks
like one including center points in resistive dividers. This rule is often
broken by people who should know better.

> Electronic outputs are never floating, they have to be within the
> power supply rails.

Which are almost always someplace between +/- 12 and +/- 24 volts, mostly 15
or 18. Now for a real thrill, put a really long wire on your multimeter, and
measure the voltage between ground pins and chassis on outlets and
appliances around your house, particularly with the equipment in typical
use. Wait for the fridge to cycle...

> Often outputs are single ended, or
> pseudo-balanced, the important part is to use twisted pair cable and
> a differential input so the input "sees" the difference between the
> two pins on the output, even if one is 0V.

Right, because the odds are that the one that is supposed to be 0 volts
isn't exactly at the same 0 volts as some other relevant piece of equipment.

Here's my last attempt at analyzing the slings and arrows of the various
flavors of balancing:

http://www.google.com/groups?selm=NpqdnY0VTa4Gqa-iU-KYgw%40comcast.com

> The ground is still required for screening and safety though, which
> is why it is a bad idea to go round cutting cable screens. If there
> is a hum problem it can normally be solved at the input circuit.

....providing that common mode rejection dynamic range isn't exceeded. Gosh,
there has to be a tidier phrase to describe that concept!

>> Is it
>> essentially the same problem as on my millimeter - with a huge DC
>> offset, the difference between hot and cold gets (literally) lost in
>> the noise, thanks to some analog equivalent to the "significant
>> digits" concept?

> No, it is a "common mode" problem.

With a bullet!

> When equipment with a mains transformer is ungrounded it tends to
> float at half the mains voltage unless it gets a ground reference
> from one of the connected cables. Sometimes you can feel a tingle if
> you run your hands over it lightly.

Who HASN'T had an unexpected tingle. Once upon a time some guys I knew were
rolling some equipment racks around, and two of them touched and welded
themselves together on the spot! Ooops!

>> Thinking about it numerically, which I realize may be a lousy
>> analogy: If I have a 0.5mV signal, balancing it gives me 0.25mV and
>> -0.25mV. If my ground reference is off by 600mV, I really have
>> 600.25mV and 599.75mV. But at the far end, 600.25 - 599.75 is still
>> 0.5mV, just like
>> 0.25 - (-0.25) is.

> That is correct, but if the common mode signal is 60V, as it probably
> is in your case, the two input pins may not be able to reject it
> perfectly so you get the error difference as a signal.

Right. In my measurements, maximum common mode dynamic range can be in the
5-10 volt range. One mitigating influence is that equipment with a low input
impedance can actually bring the two chassis grounds closer together, if the
leakage impedance is high enough. But you don't want to try to take that to
the bank!

> Alternatively, if one output pin has a hum on it and the other does
> not you will be introducing the hum as a signal.

> The theory assumes that the signals at the output are perfect and the
> cancellation at the input is exact. Real life is never perfect, as
> you have found out, it is more how close you can get it to
> approximate it.

If you do things right you can get 40-60 dB hum rejection out of a good
balanced input. That can easily be the difference between trashy sound and
something that is subjectively quite clean.

[Jay Levitt]

In article >,
says...
> I had the same problem with my A90EX. My solution was to strap a 16-guage
> copper from a chassis screw on the bottom of the A90 to the nearest unit with a
> 3-prong grounded AC cable -- in my case an Oberheim Matrix 6. If I recall
> correctly Try it and I think your problem will disappear.

This seems to be the obvious solution, as several people have pointed
out.. unfortunately, in my case it didn't seem to work. There is an ATI
Disc-Patcher (balancing transformer) right next to my keyboard, so after
verifying that it ties its chassis ground screw to electrical ground,
and that the A-90 ties its chassis screws to its signal ground, I
connected the two with a 12G wire.

It actually made the 60 Hz hum louder, and added back the 445 Hz LCD
backlight hum that I was getting when my MIDI cable was grounded. (I
have a call in to the manufacturer on that; it's a medical monitor and
probably shouldn't be emitting such noise.) I disconnected everything
from the ATI to make sure I wasn't introducing a ground loop there; I
need to go double-check the patch panel and make sure the A-90's output
cables still have lifted shields, but if that's not it, I'm at a loss.
Guess I'll have to have someone take a look at the guts of the thing,
and/or decide to live with it as is.

--
Jay Levitt |
Wellesley, MA | Hi!
Faster: jay at jay dot eff-em | Where are we going?
http://www.jay.fm | Why am I in this handbasket?

[ShLampen]

In article >, Jay Levitt
> writes:

>t actually made the 60 Hz hum louder, and added back the 445 Hz LCD
>backlight hum that I was getting when my MIDI cable was grounded. (I
>have a call in to the manufacturer on that; it's a medical monitor and
>probably shouldn't be emitting such noise.) I disconnected everything
>from the ATI to make sure I wasn't introducing a ground loop there; I
>need to go double-check the patch panel and make sure the A-90's output
>cables still have lifted shields, but if that's not it, I'm at a loss.
>Guess I'll have to have someone take a look at the guts of the thing,
>and/or decide to live with it as is.
>

I know this sound very simplistic but, to be sure, go to the hardware store and
buy an AC tester. This is a 3-pin AC plug with some LED's on the back. When
you plug it in it show you that the hot is hot, the neutral is neutral, and the
ground is ground. (There's a liitle chart to show what it means if all the
lights don't light up.) You know, just because there's a third pin ground
doesn't mean it's working (or working correctly).

I used to carry one with me when I did remote broadcasts and it never ceased to
amaze me how terrible the wiring it, even in modern office buildings. It's $10
well spent.

Steve Lampen
Belden Electronics Division