Transmission Line Demonstrators

All right, here's the real one...

Bob

MIT's transmission line lab had a line made with an iron core wound with the center conductor to give milliseconds of transit time. Made it down-right trivial to observe wave behavior, delays, termination effects, etc.

...Jim Thompson

Rope?? Nah - a Slinky! ;-)

What happens when you start a Slinky going downstairs, on an up escalator? Does it stay in one place forever? ;-)

Cheers! Rich

Send them to France. No one would notice. :(

How does that demonstrate traveling waves? A.P. French, still at MIT, has a book out on fundamentals of wave motion for going on 40 years, chock full of stuff like that designed at MIT...It's a good read anyway.

I made one of those myself once. I don't remember where I saw the one I was emulating. Maybe that Carl and Jerry story played a part.

For the center torsional member I used piano wire. The cross pieces were from welding rod about 10" long. I spaced them maybe 1/4 or 3/8 " apart. I made up some kind of jig to hold things in alignment and space the next wire for soldering. When assembled I had a wooden backbone with metal pieces on the ends to pull the center wire taught. It worked as described in the article.

You can short the end by clamping the last rod. For variable loads, I had a wire attached to the end of the last rod. The other end of this wire had a disc on it. The disk goes in a glass of water for load. Changing the size of the disk changes the loading. For fine tuning, you can move the load attachment point in or out along the last rod. Loaded just right, there is no reflection.

It was cool, but big and easy to bend if you bumped it. Kind of a pain after the first day.

Back around 1990 I was trying to visualize in my head how standing waves really happen and not having much success. I wrote a PC program with the basic transmission line formulas and a graphic display of the waves propagating. You could turn on or off the display of the three waves: forward, reflected, and sum. Each had a different color for its sinusoid. You could specify the value for normalized termination at the end point, and optionally make the line lossy or not.

The result was very obvious and very unexpected when I watched what was happening and how it showed up on the transmission line. I also added a feature where you could 'probe' the signals at different places along the line and see, in another window, the wave you would see if you put a scope there on a real line.

I thought it was very helpful to my understanding so I showed it to an instructor. He conferred with another teacher and they decided my program was broken. It didn't really look like that.

I set up a real system with a bunch of short sections of coax and real loads, shorts, etc on the far end. Every signal I measured with a scope at points along the line looked exactly like what my program was displaying with the 'probe' feature. I gave up showing it to people.

In the last few years, I thought about rewriting the program for current Windows computers and posting it somewhere, but I never got around to it.

Yeah, I did something very much like that in Scilab. I'm sure it would run faster if optimized as a stand-along program, but it was useful. Too bad the instructors got the wrong answer. :-) I've run across some like that from time to time, too.

I think it would be a nice contribution if you brought the program up to date and made it available. It would be nice if the excitation could be pulses or steps in addition to sinusoids, though dealing with line loss as a function of frequency could be a challenge.

Cheers, Tom

Well, you watch the waves travelling down the line...reflecting off discontinuities... In the case of a perfect reflection, you can watch the travelling wave start out, and the standing wave build as a result. If the termination is another line of a different impedance, you can watch the wave start on the first line, and excite a travelling wave in the second as well as a standing wave in the first.

Cheers, Tom

But are you talking about a torsion wave?

Yes, of course: it was stated as a torsion transmission line. So to watch that, you watch the ends of the rods. Without the rods, it's pretty hard to see. See the url that Robert posted for pictures.

Cheers, Tom

Rich Grise snipped-for-privacy@example.net posted to sci.electronics.design:

Maybe it is a "race" condition.

I've seen these made with a brass rod with smaller brass pointers brazed to it (ASCII art below):

| | | | | | | | | | | | | | +--+--+--+--+--+--+--+--+--+--+--+--+--+ wave direction | | | | | | | | | | | | | |

The wave is initiated by imposing a torque on the main shaft and the angular displacement can be seen by viewing the pointers end-on (with a stroboscope*).

The main shaft is supported as required by bearings or bushings. The ends can be left free or clamped down to simulate different terminations. Weights and other devices can be added to simulate lumped components. The impedance can be adjusted by the stiffness of the main shaft and the mass of the pointers. Lines with different characteristics can be coupled.

Hmm that wasnt the demonstrator I found, but that seems like it does what I'd like it to do. And you're more to understanding what I meant. I didn't necessarily mean a demonstrator as in a rope and explaining reflections and harmonic oscillations, more to do with electrical/ electronics.

theory;

found?

Mark,

Thanks for the reply, the idea has crosssed my mind but I need to be able to demonstrate to an extent the debounces that happen in a mismatched circuit I don't suppose you have an idea of how I'd do that? As for the products I was talking about:

They're more along those lines.

try some of Len Lyes kinetic sculptures.

I have seen a number of these, at (IIRC) the Brewster Gallery in New Plymouth, NZ. Blade (1965) is brilliant..... as soon as I saw it, I thought "transmission line" followed by "I wish we'd done an electromagnetics field trip here"

Universe is even cooler. And they all sound awesome too.

I couldnt find any MPEGs of these sculptures , but here are some of the water whirler....

Cheers Terry

I just stumbled onto a web applet that is similar to the old program I wrote.

That's just the sort of thing I was looking for when I decided to write my own so I could visualize how the forward and reflected waves summed to create the measurable standing wave.

The main thing I see missing from the applet is loads with reactance to display phase shift between forward and reflected waves. (It does show

No line loss either.

That's an interesting demo. The key to making it working is to enter a Reflection Coefficient between -1.0 (driving a short) and 1.0 (driving an open). The page says to enter a value between 0 and 1, but this doesn't allow for termination impedance below 50 ohms.

Bob

Or you can enter the termination impedance rather than the reflection coefficient. I never thought of entering a negative reflection coeffiecient but I entered terminations in the 0-50 range.