An unusual Oscilloscope phenomenon

John Larkin a écrit :

the

we

Probably. Lots of people don't understand this and I've seen this kind of mistake (opened shield) recommended in lots of highly regarded EMC books.

But that's really nice for us. (Joerg?)

Oh yeah? You young punks have it easy. Back in my time, we didn't have plastic chairs. If we wanted to make sparks, we had to rub two cats together.

only if you bolted the camera to your head...

At the receive (instrument) end, the ungrounded end of the coax shield becomes the hot side of a resonant "vertical" antenna. Any voltage there becomes 100% common-mode signal on the inner twisted pair.

What's wrong with a little current in the shield? The signal pair isn't affected. But volts, or tens of volts, of common-mode RF can be nasty.

John

Just hit the top of your head with your fist. You'll see the typically underdamped transient response of your visual tracking system.

John

When I was a kid, my brother built a Heathkit TOOB amp; in the low level circuits, they only ever grounded one end of the shields.

Cheers! Rich

Yeah! But shhht, don't spill the beans here, at least not until I am retired :-))

the

we

No, which is precisely the reason to ground only one end of the shield.

Cheers! Rich

Inside a steel chassis pan, it wouldn't matter. But then Heathkit was hardly an authority on signal management.

John

the

It could become really, really nasty if the cable length happens to be close to 1/4, 3/4 and so on wavelength of the offending RF signal. If the shield current is of concern it can be broken by a wee ferrite.

Probably the litmus test for RF stuff like this is when people schlepp it to Otis Street and see if it can survive Sutro Tower.

the

Now wait a minute, they engineered darn good stuff. The old HW100/101 transceivers were among the best ever made back in the tube days. Still got one. No noises or buzzes whatsoever and its performance in the vicinity of strong signals is stellar. The only not so good thing was the two rubber pulleys because they cooked out too fast, but you could keep a stash of them in the drawer.

I think I'll retired after you, but this kind of misconception is sufficiently widely spread that we don't have to worry :-)

They had more volts in them as well

Oh, I am not worried at all :-)

But I am concerned about what happens when guys like you and I are in their 90's, the hands are shaky, the mind ain't what it used to be and the senior living place doesn't allow a Tektronix mainframe in the rooms. The next generation doesn't have anough analog engineers.

Yup. Once people get fixated on "ground loops" they are forever lost to reason.

Ditto high-speed "return currents."

John

Bad focus but shows the concept, when I was a teenager I made almost everything in wood, nicely polished, several layers of laquer, some more polishing:

I usually spent a lot more time on the enclosures than on the electronics but the stuff is still working.

it

the

Or a nice wideband ESD spark. Then the receive-end differential signal might see a kilovolt p-p common-mode ring.

If

Right. That's easy to add as needed.

It's terrible here. Scope traces are gross fuzz, and all sorts of opamp front-ends like to rectify stuff.

You should see the spectrum of a 10" clip lead.

We did do a useful simulation of a system that has a superconductive magnet/front end on one side of a room and some ADCs on the other side. We convinced ourselves that we need to go differential.

John

it

the

Toob amps didn't rectify RF as well as silicon does.

John

Won't matter. We'll all be deep into communal despair :-(

...Jim Thompson