On Sunday, April 28, 2019 at 1:19:56 AM UTC-4, wrote:
On Sat, 27 Apr 2019 21:43:45 -0700 (PDT), Its Me
wrote:

On Sunday, April 28, 2019 at 12:31:07 AM UTC-4, Bill wrote:


Why were the MOV to ground, instead of across the lines? Or a bigger value
Varistor to ground?

Because of the twisted pair, the danger isn't usually voltage spike across the pair, but rather the spike potential from the pair to ground. That's what we were trying to protect from. And what protection components on 66 punch blocks from back in the day did as well.

As far as the value, it's a bit of a tightrope. Too low of a value, and it's always firing and causing issues like we experienced. Too big of a value, and you may as well not have any protection on there at all. Even a transformer doesn't protect you, as it has an arc-over value. We thought the 180v parts would be OK, but we didn't realize that the lines would be as dirty as they were.


===

I'm wondering if you couldn't use an isolated op amp with differential
inputs to extract the signal. It would have to be totally isolated
from any ground reference to protect from the high voltage common mode
spikes. It would need a floating power supply of course, with the
output through an opto isolator or some such.

The perfect isolator for those common mode spikes is a transformer. It doesn't have any reference unless you give it one. You can certainly use an op amp to terminate and receive that twisted pair signal (and we did at times), but a large enough spike will take it out. There's no way to completely protect it. Once you start jumping air gaps inside the IC, or exceeding the reverse voltage spec on junctions in there, things start blowing up.

On Sunday, April 28, 2019 at 9:17:57 AM UTC-4, wrote:
On Sun, 28 Apr 2019 06:18:34 -0400, John H.
wrote:

On Sun, 28 Apr 2019 01:19:50 -0400, wrote:

On Sat, 27 Apr 2019 21:43:45 -0700 (PDT), Its Me
wrote:

On Sunday, April 28, 2019 at 12:31:07 AM UTC-4, Bill wrote:


Why were the MOV to ground, instead of across the lines? Or a bigger value
Varistor to ground?

Because of the twisted pair, the danger isn't usually voltage spike across the pair, but rather the spike potential from the pair to ground. That's what we were trying to protect from. And what protection components on 66 punch blocks from back in the day did as well.

As far as the value, it's a bit of a tightrope. Too low of a value, and it's always firing and causing issues like we experienced. Too big of a value, and you may as well not have any protection on there at all. Even a transformer doesn't protect you, as it has an arc-over value. We thought the 180v parts would be OK, but we didn't realize that the lines would be as dirty as they were.


===

I'm wondering if you couldn't use an isolated op amp with differential
inputs to extract the signal. It would have to be totally isolated
from any ground reference to protect from the high voltage common mode
spikes. It would need a floating power supply of course, with the
output through an opto isolator or some such.

---
This email has been checked for viruses by AVG.
https://www.avg.com

I'm wondering if we could talk about artillery and tanks for a while, with maybe a bit of C-4
discussion thrown in! :)

===

Sounds good, and I'm sure we'd all get a bang out of it. Is C4 what
they use in the combat simulation exercise during basic? I was right
next to a ring of sand bags when one of those charges went off. It
certainly gets your attention.

A cousin of mine gave my dad two artillery simulators years ago. I remember them being a plastic bottle almost the size of a drink can with a tube down the side that had a string you pulled. Then you had so many seconds until it went bang. Wow.

On 4/28/2019 9:38 AM, Its Me wrote:
On Sunday, April 28, 2019 at 1:19:56 AM UTC-4, wrote:
On Sat, 27 Apr 2019 21:43:45 -0700 (PDT), Its Me
wrote:

On Sunday, April 28, 2019 at 12:31:07 AM UTC-4, Bill wrote:


Why were the MOV to ground, instead of across the lines? Or a bigger value
Varistor to ground?

Because of the twisted pair, the danger isn't usually voltage spike across the pair, but rather the spike potential from the pair to ground. That's what we were trying to protect from. And what protection components on 66 punch blocks from back in the day did as well.

As far as the value, it's a bit of a tightrope. Too low of a value, and it's always firing and causing issues like we experienced. Too big of a value, and you may as well not have any protection on there at all. Even a transformer doesn't protect you, as it has an arc-over value. We thought the 180v parts would be OK, but we didn't realize that the lines would be as dirty as they were.


===

I'm wondering if you couldn't use an isolated op amp with differential
inputs to extract the signal. It would have to be totally isolated
from any ground reference to protect from the high voltage common mode
spikes. It would need a floating power supply of course, with the
output through an opto isolator or some such.

The perfect isolator for those common mode spikes is a transformer. It doesn't have any reference unless you give it one. You can certainly use an op amp to terminate and receive that twisted pair signal (and we did at times), but a large enough spike will take it out. There's no way to completely protect it. Once you start jumping air gaps inside the IC, or exceeding the reverse voltage spec on junctions in there, things start blowing up.



The vacuum chamber systems I used to design and build often had multiple
banks of quartz lamps inside used to heat optics to be coated to a
temperature that permitted ideal conditions for the evaporated materials
to condense. It was also to add energy to drive off any remaining
water vapor molecules that might still be on the optics. The lamps
ran on up to 240 vac, controlled by a proportional SCR controller.
Under vacuum conditions, remaining gas molecules on the optics or
fixturing often caused arcs to occur, very similar in nature to
mini-lightening strikes. Those arcs (to ground) would wipe out either
the fast acting (and expensive) SCR fuse or sometimes the SCR itself.

To minimize this we used a large, 1:1 (ratio) isolation transformer
that removed the steel chamber as the primary ground potential. It
solved most of the nuisance but expensive fuse blowing and provided some
extra protection for the SCR itself. Wasn't 100 percent effective
but reduced the blowing of the fuse to maybe once or twice a year.



---
This email has been checked for viruses by AVG.
https://www.avg.com

On Sat, 27 Apr 2019 21:18:08 -0700 (PDT), Its Me
wrote:

On Saturday, April 27, 2019 at 11:51:47 PM UTC-4, wrote:
On Sat, 27 Apr 2019 18:59:16 -0700 (PDT), Its Me
wrote:

On Saturday, April 27, 2019 at 8:48:44 PM UTC-4, wrote:
On Sat, 27 Apr 2019 17:37:48 -0700 (PDT), Its Me
wrote:

On Saturday, April 27, 2019 at 5:46:28 PM UTC-4, wrote:
On Sat, 27 Apr 2019 13:31:59 -0700 (PDT), Its Me
wrote:

On Saturday, April 27, 2019 at 4:00:01 PM UTC-4, wrote:
On Sat, 27 Apr 2019 11:21:11 -0700 (PDT), Its Me
wrote:

On Saturday, April 27, 2019 at 1:44:28 PM UTC-4, wrote:
On Sat, 27 Apr 2019 09:59:10 -0700 (PDT), Its Me
wrote:

On Thursday, April 25, 2019 at 10:02:15 AM UTC-4, wrote:


"Flop around"? Each conductor is securely attached to insulators.
They are not physically twisted in a bundle, they just get an
electrical twist by having the phases change positions from pole to
pole. They call it transposition. It allows the lines to be balanced.
(Relative inductance to each other and to the environment around them)

Oh, I thought you were saying that each *leg* consisted of three twisted conductors. I've never seen that but who knows what FPL might be doing.

Transmission lines overhead are bare wires. They are twisted strands
but that is just to increase flexibility and skin effect. The twist I
was talking about is how the lines get "transposed" on the poles every
half or so where they all get swapped around


Those solar panels aren't part of the grid, and have no "sync". Sunlight isn't 60 Hz.

The inverter will be grid tied so the transport to the house would be
in sync.

Of course, but it wasn't the feed to the house that I was talking about.

You think there will be some voltage induced into the panels?
I am not seeing that on anything that is terminated in any way. In
this case the panels would be terminated into the inverter and if
there was any 60hz present it would simple be chewed up and spit out
by the inverter anyway. After all it would be "in sync" with the grid
anyway.

I don't know, I've never played with it, but it can induce some noise. Probably not enough in that situation to do anything at all.

Years ago I installed some comm equipment in a train yard. We had built some interface cards for some comm circuits that was just a 600 ohm twisted pair that ran along the track for miles until it hit a transmitter shack, sometimes with a handset in there to talk back to dispatch. The interface card had a transformer and some audio switching circuits, and was protected by a couple of MOV's rated at 160-180 volts, one for each side of pair going to ground.

When we installed the new equipment and started testing, some of those circuits had a horrible amount of noise and buzzing on them. Doing some troubleshooting with a butt set, I discovered that if our interface card was disconnected, the noise went away. Then I hung an oscilloscope on one side of the pair, then the other. There was a HUGE 60hz component coming in on some of the circuits, in some cases around 200 volts. Since the twisted pair was doing what it does, you didn't hear that on the circuit (when you induce a signal on a twisted pair, the same signal is present on both wires.
Since you only care about the difference in potential between the two wires, the signal stays intact.). It wasn't until our interface card was connected that the MOVs saw that voltage from each wire to ground and turned on, unbalancing the circuit and causing the noise to be audible.

Turns out that the pairs left the yard and were run on power poles, occasionally in some cases dropping down to fence posts on the right-of-way, then back up to the posts. They picked up the induced signal from the power lines and brought it right back to our interface cards. We clipped out the MOVs, and all was good. We had to depend on primary lightning protection to protect the equipment. The MOVs were secondary.

More likely you were simply seeing the ground shift induced by using
earth as part of the return path. That is inevitable with wye
distribution, even if you do have a neutral conductor. Current does
not take the path of least resistance, it takes ALL paths. The voltage
drop on distribution lines is imposed on the earth. When we were
investigating blown interface cards at a college campus I found up to
35 volts between "ground" on the various buildings. We found a number
of solutions depending on the topology and geography but some
customers just turned it over to the phone company, using leased phone
lines between buildings that were a hundred yards apart. The other
solution was copper, bonding "grounds" together, usually from machine
frame to machine frame directly. We had to call it a "drain" and use
black wire to avoid NEC problems.

Nope, this was voltage induced into long runs of unshielded, twisted pair wires. Across the pair you didn't see it, but compared to ground you (properly) did. It's basic electronics 101. Unbalance the pair, which the MOVs did, and now the differential between the conductors is significant and you can hear the induced noise.

It's exactly why RS-232 data is only good for 50ft (or with special, low capacitance cable up to 1,500 feet), and RS-422 is good for nearly 5,000 feet. RS-232 is single ended and is compared to ground (which you can't shake with induced noise), while RS-422 (or 485) is twisted pair. Hit the 232 with a 25 volt spike, and you just changed a 0 to a 1. But with 422, you are looking for a 5 volt difference between the wires. You don't care if they are 0 and 5 volts, or 20 and 25 volts, the data is still intact. Well, until it hits 1000 and 1005 volts, then you've blown up the receiver chip. But the data was still good!

I am not going to argue with you but I was a physical Planning rep in
a place with 200 thunderstorms in a typical year and I know what works
and what doesn't. I have seen a lot of well credentialed EEs, usually
from up north, fail miserably with their theories. The proof is always
in the magic smoke that escapes when they guess wrong.
We went from a couple "lightning" calls a day to a couple a year once
we started implementing our strategies and ignoring all of the "ground
loop" loonies.

That's nice and all, but we aren't talking about thunderstorms. I've worked with many non-engineering types that thought they knew what they were talking about, but many of their practices were flawed, and it showed through in the end. Your attitude isn't uncommon. It's similar to class warfare, eh?

Bottom line, I was dealing with induced AC on a twisted pair that was enough to turn on 180v MOVs. You can talk about ground differentials all you want, but that had absolutely NOTHING to do with what was happening in that situation.

Have a good evening!

You said the line to line was nominal so where was the MOV referenced?
To "ground"?

Copied from above:

"But with 422 (*twisted pair*), you are looking for a 5 volt difference between the wires. You don't care if they are 0 and 5 volts, or 20 and 25 volts, the data is still intact. Well, until it hits 1000 and 1005 volts, then you've blown up the receiver chip. But the data was still good!" You aren't comparing it to ground.

So, yes... the voltage across the MOVs was referenced to ground. That's why there was no problem until the MOV was introduced to the circuit. The difference in potential between the balanced wires is what's important to signal integrity. The diff between the pair and ground isn't, until there is a large enough potential that it causes a problem, as it did with the MOVs.

But back to the original discussion... there was enough induced AC from power lines to turn on 180v MOVs. I could see it on a scope. It could be measured on a voltmeter. It manifested itself by having a huge amount of noise and hum on previously quiet comm circuits until the MOVs were introduced. The induced AC from parallel power lines caused the problem.

Not sure what you are arguing against... this is precisely the type of real world application experience you are always saying trumps those dumb engineers. But argue on...

I am just trying to find out if you are talking about voltage between
the twisted pairs or you are talking about a common mode voltage.

I thought my explanation would have made that clear. Again, it is a voltage induced into the pair, both wires, which is a common mode voltage. Shift out of tech mode into engineering mode.

I am also curious about how much current we are talking about. Would
it light a light bulb?

I have seriously doubt if the measured situation would have resulted in enough current to run a light bulb. That would have required hooking the load to the office end and the field end of those wires, miles away. Or grounding the far end, and then seeing what you got that way compared to ground. That's how I saw the 60Hz voltage component. But it was enough to fire those MOVs hooked up to ground.

I can wave a scope probe in the air and se
voltages. Attached to a piece of wire it can be a really big voltage.

You won't see 200v P-to-P, 60Hz sine waves, unless you are under those FPL lines behind your house.

When you are talking about really long wires, maybe you should be
taking guidance from a Telco or CableCo Demark box.

These are not owned by a telco. Railroads years ago owned their own comm circuits. For example, the old AT&SF RR had their own analog microwave that ran from Chicago to San Bernardino, CA. You could make a call from Chicago to CA without ever hitting Ma Bell. Pretty much all now either own their own fiber buried in their right-of-way, or the have rights to some dark fiber strands that someone else has buried in that ROW.

In any case, those leased lines from Ma Bell are nearly all gone. They priced copper out of existence, and it's all now IP. Things have moved on since Windows XP.

It sounds like the railroad should have looked at how the telco deals
with miles of wire hanging in the air. The MOV thing confused me. They
are for transients, not a steady state voltage gradient. The MOV
switching on and off may have been where the noise was coming from.
I wonder what a Telco carbon protector would have done.

On Sat, 27 Apr 2019 21:43:45 -0700 (PDT), Its Me
wrote:

On Sunday, April 28, 2019 at 12:31:07 AM UTC-4, Bill wrote:


Why were the MOV to ground, instead of across the lines? Or a bigger value
Varistor to ground?

Because of the twisted pair, the danger isn't usually voltage spike across the pair, but rather the spike potential from the pair to ground. That's what we were trying to protect from. And what protection components on 66 punch blocks from back in the day did as well.

As far as the value, it's a bit of a tightrope. Too low of a value, and it's always firing and causing issues like we experienced. Too big of a value, and you may as well not have any protection on there at all. Even a transformer doesn't protect you, as it has an arc-over value. We thought the 180v parts would be OK, but we didn't realize that the lines would be as dirty as they were.

You might have done better with 500k or so between both legs and
ground (one meg across the signal lines, center tapped to ground).
That phantom voltage might have just disappeared. That is essentially
what a Telco carbon protector is but the resistance is lower.

On Sun, 28 Apr 2019 01:19:50 -0400,
wrote:

On Sat, 27 Apr 2019 21:43:45 -0700 (PDT), Its Me
wrote:

On Sunday, April 28, 2019 at 12:31:07 AM UTC-4, Bill wrote:


Why were the MOV to ground, instead of across the lines? Or a bigger value
Varistor to ground?

Because of the twisted pair, the danger isn't usually voltage spike across the pair, but rather the spike potential from the pair to ground. That's what we were trying to protect from. And what protection components on 66 punch blocks from back in the day did as well.

As far as the value, it's a bit of a tightrope. Too low of a value, and it's always firing and causing issues like we experienced. Too big of a value, and you may as well not have any protection on there at all. Even a transformer doesn't protect you, as it has an arc-over value. We thought the 180v parts would be OK, but we didn't realize that the lines would be as dirty as they were.


===

I'm wondering if you couldn't use an isolated op amp with differential
inputs to extract the signal. It would have to be totally isolated
from any ground reference to protect from the high voltage common mode
spikes. It would need a floating power supply of course, with the
output through an opto isolator or some such.

Before Opto Isolators our line receivers that were taking inputs "from
the wild"" used input transformers. Common mode noise or stray voltage
was lost in the translation. A voltage transient that exceeds the
voltage rating of the LED in the isolator will still blow it up tho.
Fiber has made most of this obsolete.

On Sun, 28 Apr 2019 08:23:21 -0400, "Mr. Luddite"
wrote:

On 4/28/2019 6:18 AM, John H. wrote:
On Sun, 28 Apr 2019 01:19:50 -0400, wrote:

On Sat, 27 Apr 2019 21:43:45 -0700 (PDT), Its Me
wrote:

On Sunday, April 28, 2019 at 12:31:07 AM UTC-4, Bill wrote:


Why were the MOV to ground, instead of across the lines? Or a bigger value
Varistor to ground?

Because of the twisted pair, the danger isn't usually voltage spike across the pair, but rather the spike potential from the pair to ground. That's what we were trying to protect from. And what protection components on 66 punch blocks from back in the day did as well.

As far as the value, it's a bit of a tightrope. Too low of a value, and it's always firing and causing issues like we experienced. Too big of a value, and you may as well not have any protection on there at all. Even a transformer doesn't protect you, as it has an arc-over value. We thought the 180v parts would be OK, but we didn't realize that the lines would be as dirty as they were.


===

I'm wondering if you couldn't use an isolated op amp with differential
inputs to extract the signal. It would have to be totally isolated
from any ground reference to protect from the high voltage common mode
spikes. It would need a floating power supply of course, with the
output through an opto isolator or some such.

---
This email has been checked for viruses by AVG.
https://www.avg.com

I'm wondering if we could talk about artillery and tanks for a while, with maybe a bit of C-4
discussion thrown in! :)



Heh. I learned a long time ago that anyone who claims to understand the
physics of lightning and can reliably predict it's effects should be
taken with a grain of salt. :-)


That was our take on it but we did start to understand what worked.
The gang at State Farm (Winter Haven) came up with a lot of good stuff
that had absolutely zero engineering support. They just tried things
and just built on concepts that showed promise. The "drain" wire came
from them and flew in the face of theory about ground loops. There
were also NEC problems.
We started playing with ferrite beads, building on their bonding
ideas. Our thinking is if we could delay the transient with a choke
long enough for the bond wire to bleed it off we would not have any
real damage. Is it true to engineering concepts, nope, or so I was
told.

I had Holiday Inn asking me why we were the only ones in Florida who
were not losing the registers in the pool bar in thunder storms. We
fixed it with an 8 ga copper wire between the machine frames (inside
and out to the pool) and a fist full of ferrite beads.

On Sun, 28 Apr 2019 09:26:43 -0400,
wrote:

On Sun, 28 Apr 2019 08:23:21 -0400, "Mr. Luddite"
wrote:

On 4/28/2019 6:18 AM, John H. wrote:
On Sun, 28 Apr 2019 01:19:50 -0400, wrote:

On Sat, 27 Apr 2019 21:43:45 -0700 (PDT), Its Me
wrote:

On Sunday, April 28, 2019 at 12:31:07 AM UTC-4, Bill wrote:


Why were the MOV to ground, instead of across the lines? Or a bigger value
Varistor to ground?

Because of the twisted pair, the danger isn't usually voltage spike across the pair, but rather the spike potential from the pair to ground. That's what we were trying to protect from. And what protection components on 66 punch blocks from back in the day did as well.

As far as the value, it's a bit of a tightrope. Too low of a value, and it's always firing and causing issues like we experienced. Too big of a value, and you may as well not have any protection on there at all. Even a transformer doesn't protect you, as it has an arc-over value. We thought the 180v parts would be OK, but we didn't realize that the lines would be as dirty as they were.


===

I'm wondering if you couldn't use an isolated op amp with differential
inputs to extract the signal. It would have to be totally isolated
from any ground reference to protect from the high voltage common mode
spikes. It would need a floating power supply of course, with the
output through an opto isolator or some such.

---
This email has been checked for viruses by AVG.
https://www.avg.com

I'm wondering if we could talk about artillery and tanks for a while, with maybe a bit of C-4
discussion thrown in! :)



Heh. I learned a long time ago that anyone who claims to understand the
physics of lightning and can reliably predict it's effects should be
taken with a grain of salt. :-)


===

Yes, Ben Franklin is lucky to have survived his kite flying
experiment. Right here in SWFL is a good place to do research since
we have so many thunderstorms in the summer. In 15 years we've lost
one tree and a bunch of electronic devices. A neighbor a few blocks
over had their boat struck at the dock. It fried about $25K in
electronics.

I haven't lost a thing here since we hardened the place in the 80s.

On Sun, 28 Apr 2019 12:38:01 -0400, wrote:

On Sat, 27 Apr 2019 21:43:45 -0700 (PDT), Its Me
wrote:

On Sunday, April 28, 2019 at 12:31:07 AM UTC-4, Bill wrote:


Why were the MOV to ground, instead of across the lines? Or a bigger value
Varistor to ground?

Because of the twisted pair, the danger isn't usually voltage spike across the pair, but rather the spike potential from the pair to ground. That's what we were trying to protect from. And what protection components on 66 punch blocks from back in the day did as well.

As far as the value, it's a bit of a tightrope. Too low of a value, and it's always firing and causing issues like we experienced. Too big of a value, and you may as well not have any protection on there at all. Even a transformer doesn't protect you, as it has an arc-over value. We thought the 180v parts would be OK, but we didn't realize that the lines would be as dirty as they were.

You might have done better with 500k or so between both legs and
ground (one meg across the signal lines, center tapped to ground).
That phantom voltage might have just disappeared. That is essentially
what a Telco carbon protector is but the resistance is lower.


===

Telcos have been using carbon protectors for close to 100 years.
Designed by Bell Labs, and built by Western Electric, they had a lot
of serious engineering and experience behind them. Of course all of
that pre-dated solid state electronics and their sensitive little
gates.

---
This email has been checked for viruses by AVG.
https://www.avg.com

On Sunday, April 28, 2019 at 12:32:21 PM UTC-4, wrote:
On Sat, 27 Apr 2019 21:18:08 -0700 (PDT), Its Me
wrote:

On Saturday, April 27, 2019 at 11:51:47 PM UTC-4, wrote:
On Sat, 27 Apr 2019 18:59:16 -0700 (PDT), Its Me
wrote:

On Saturday, April 27, 2019 at 8:48:44 PM UTC-4, wrote:
On Sat, 27 Apr 2019 17:37:48 -0700 (PDT), Its Me
wrote:

On Saturday, April 27, 2019 at 5:46:28 PM UTC-4, wrote:
On Sat, 27 Apr 2019 13:31:59 -0700 (PDT), Its Me
wrote:

On Saturday, April 27, 2019 at 4:00:01 PM UTC-4, wrote:
On Sat, 27 Apr 2019 11:21:11 -0700 (PDT), Its Me
wrote:

On Saturday, April 27, 2019 at 1:44:28 PM UTC-4, wrote:
On Sat, 27 Apr 2019 09:59:10 -0700 (PDT), Its Me
wrote:

On Thursday, April 25, 2019 at 10:02:15 AM UTC-4, wrote:


"Flop around"? Each conductor is securely attached to insulators.
They are not physically twisted in a bundle, they just get an
electrical twist by having the phases change positions from pole to
pole. They call it transposition. It allows the lines to be balanced.
(Relative inductance to each other and to the environment around them)

Oh, I thought you were saying that each *leg* consisted of three twisted conductors. I've never seen that but who knows what FPL might be doing.

Transmission lines overhead are bare wires. They are twisted strands
but that is just to increase flexibility and skin effect. The twist I
was talking about is how the lines get "transposed" on the poles every
half or so where they all get swapped around


Those solar panels aren't part of the grid, and have no "sync". Sunlight isn't 60 Hz.

The inverter will be grid tied so the transport to the house would be
in sync.

Of course, but it wasn't the feed to the house that I was talking about.

You think there will be some voltage induced into the panels?
I am not seeing that on anything that is terminated in any way. In
this case the panels would be terminated into the inverter and if
there was any 60hz present it would simple be chewed up and spit out
by the inverter anyway. After all it would be "in sync" with the grid
anyway.

I don't know, I've never played with it, but it can induce some noise. Probably not enough in that situation to do anything at all.

Years ago I installed some comm equipment in a train yard. We had built some interface cards for some comm circuits that was just a 600 ohm twisted pair that ran along the track for miles until it hit a transmitter shack, sometimes with a handset in there to talk back to dispatch. The interface card had a transformer and some audio switching circuits, and was protected by a couple of MOV's rated at 160-180 volts, one for each side of pair going to ground.

When we installed the new equipment and started testing, some of those circuits had a horrible amount of noise and buzzing on them. Doing some troubleshooting with a butt set, I discovered that if our interface card was disconnected, the noise went away. Then I hung an oscilloscope on one side of the pair, then the other. There was a HUGE 60hz component coming in on some of the circuits, in some cases around 200 volts. Since the twisted pair was doing what it does, you didn't hear that on the circuit (when you induce a signal on a twisted pair, the same signal is present on both wires.
Since you only care about the difference in potential between the two wires, the signal stays intact.). It wasn't until our interface card was connected that the MOVs saw that voltage from each wire to ground and turned on, unbalancing the circuit and causing the noise to be audible.

Turns out that the pairs left the yard and were run on power poles, occasionally in some cases dropping down to fence posts on the right-of-way, then back up to the posts. They picked up the induced signal from the power lines and brought it right back to our interface cards. We clipped out the MOVs, and all was good. We had to depend on primary lightning protection to protect the equipment. The MOVs were secondary.

More likely you were simply seeing the ground shift induced by using
earth as part of the return path. That is inevitable with wye
distribution, even if you do have a neutral conductor. Current does
not take the path of least resistance, it takes ALL paths. The voltage
drop on distribution lines is imposed on the earth. When we were
investigating blown interface cards at a college campus I found up to
35 volts between "ground" on the various buildings. We found a number
of solutions depending on the topology and geography but some
customers just turned it over to the phone company, using leased phone
lines between buildings that were a hundred yards apart. The other
solution was copper, bonding "grounds" together, usually from machine
frame to machine frame directly. We had to call it a "drain" and use
black wire to avoid NEC problems.

Nope, this was voltage induced into long runs of unshielded, twisted pair wires. Across the pair you didn't see it, but compared to ground you (properly) did. It's basic electronics 101. Unbalance the pair, which the MOVs did, and now the differential between the conductors is significant and you can hear the induced noise.

It's exactly why RS-232 data is only good for 50ft (or with special, low capacitance cable up to 1,500 feet), and RS-422 is good for nearly 5,000 feet. RS-232 is single ended and is compared to ground (which you can't shake with induced noise), while RS-422 (or 485) is twisted pair. Hit the 232 with a 25 volt spike, and you just changed a 0 to a 1. But with 422, you are looking for a 5 volt difference between the wires. You don't care if they are 0 and 5 volts, or 20 and 25 volts, the data is still intact. Well, until it hits 1000 and 1005 volts, then you've blown up the receiver chip. But the data was still good!

I am not going to argue with you but I was a physical Planning rep in
a place with 200 thunderstorms in a typical year and I know what works
and what doesn't. I have seen a lot of well credentialed EEs, usually
from up north, fail miserably with their theories. The proof is always
in the magic smoke that escapes when they guess wrong.
We went from a couple "lightning" calls a day to a couple a year once
we started implementing our strategies and ignoring all of the "ground
loop" loonies.

That's nice and all, but we aren't talking about thunderstorms. I've worked with many non-engineering types that thought they knew what they were talking about, but many of their practices were flawed, and it showed through in the end. Your attitude isn't uncommon. It's similar to class warfare, eh?

Bottom line, I was dealing with induced AC on a twisted pair that was enough to turn on 180v MOVs. You can talk about ground differentials all you want, but that had absolutely NOTHING to do with what was happening in that situation.

Have a good evening!

You said the line to line was nominal so where was the MOV referenced?
To "ground"?

Copied from above:

"But with 422 (*twisted pair*), you are looking for a 5 volt difference between the wires. You don't care if they are 0 and 5 volts, or 20 and 25 volts, the data is still intact. Well, until it hits 1000 and 1005 volts, then you've blown up the receiver chip. But the data was still good!" You aren't comparing it to ground.

So, yes... the voltage across the MOVs was referenced to ground. That's why there was no problem until the MOV was introduced to the circuit. The difference in potential between the balanced wires is what's important to signal integrity. The diff between the pair and ground isn't, until there is a large enough potential that it causes a problem, as it did with the MOVs.

But back to the original discussion... there was enough induced AC from power lines to turn on 180v MOVs. I could see it on a scope. It could be measured on a voltmeter. It manifested itself by having a huge amount of noise and hum on previously quiet comm circuits until the MOVs were introduced. The induced AC from parallel power lines caused the problem.

Not sure what you are arguing against... this is precisely the type of real world application experience you are always saying trumps those dumb engineers. But argue on...

I am just trying to find out if you are talking about voltage between
the twisted pairs or you are talking about a common mode voltage.

I thought my explanation would have made that clear. Again, it is a voltage induced into the pair, both wires, which is a common mode voltage. Shift out of tech mode into engineering mode.

I am also curious about how much current we are talking about. Would
it light a light bulb?

I have seriously doubt if the measured situation would have resulted in enough current to run a light bulb. That would have required hooking the load to the office end and the field end of those wires, miles away. Or grounding the far end, and then seeing what you got that way compared to ground. That's how I saw the 60Hz voltage component. But it was enough to fire those MOVs hooked up to ground.

I can wave a scope probe in the air and se
voltages. Attached to a piece of wire it can be a really big voltage.

You won't see 200v P-to-P, 60Hz sine waves, unless you are under those FPL lines behind your house.

When you are talking about really long wires, maybe you should be
taking guidance from a Telco or CableCo Demark box.

These are not owned by a telco. Railroads years ago owned their own comm circuits. For example, the old AT&SF RR had their own analog microwave that ran from Chicago to San Bernardino, CA. You could make a call from Chicago to CA without ever hitting Ma Bell. Pretty much all now either own their own fiber buried in their right-of-way, or the have rights to some dark fiber strands that someone else has buried in that ROW.

In any case, those leased lines from Ma Bell are nearly all gone. They priced copper out of existence, and it's all now IP. Things have moved on since Windows XP.

It sounds like the railroad should have looked at how the telco deals
with miles of wire hanging in the air. The MOV thing confused me. They
are for transients, not a steady state voltage gradient. The MOV
switching on and off may have been where the noise was coming from.
I wonder what a Telco carbon protector would have done.

Carbon protectors are for lightning strikes, and generally fire around 500v.. They are primary protection, but wouldn't have done anything in this case.