Simple transmission line question

[Robert Baer]

If I have a source terminated transmission line (think of a length of
coax with a 50 ohm resistor to ground at the source.) and it goes into
an open circuit. (say into an unterminated 'scope input.) Then is
there a reflection from the unterminated end? And if so, the reason
that there is no ringing observed is that the source termination soaks
up the return signal. Or does the source termination make the
transmission line 'happy' regardless of what's on the other end?

Thanks,
George H.

Depends on your definition / configuration.
Config #1: generator is 50 ohms and you consider that as "source
termination" --> signal bounces back, one sees it at generator.
Config #2: generatro is 50ohms, then a 50 ohm resistor ("source
termination") in series to drive coax -->signal bounces back and all
looks kosher at coax source end.

 

[Robert Baer]

Are you driving the coax with a voltage source with 50 Ohms "to
ground", or is the source impedance 50 Ohms?

An open-terminated (50 Ohm impedance) line driven by 50 Ohms _source_
would reflect NOT-inverted.

Thus observed at the source end of the line, you'd see half your pulse
height initially then popping up to full pulse height upon reflection
return. (If I remember my freshman-level transmission line material
correctly

Likewise a shorted line will cancel the input pulse after full
out-and-back transit time... think short pulse generation techniques.

...Jim Thompson

Check; case #1 in my posted list.

 

[Mark]

to the op.

as long as you are sending the signal into the cable at one end taking
the signal out at one end (no taps etc) then you need to terminate the
cable at only one end, either end will do...

when terminating the cable on the source end as you described, there
will be reflections on the cable all along the way, but it is just as
you said, the Rx end will have a clean waveform. There will be no RE
reflection because the source is terminated.

Mark

 

[[email protected]]

(physics students) Some can hardly drive a 'scope. I've spent some
time on the phone explaining triggering....

Appendix A on "how to drive a scope" in Art of Electronics, 2nd edition,
by Horowitz and Hill looks like a reasonable student guide to me. I
must warn you that I have never tried it on, or seen it tried on, real
live physics students. Because of when it was written, it's for
analog scopes.

Standard disclaimers apply; I don't get money or other consideration
from any companies mentioned.

Matt Roberds

 

[Phil Allison]

"George Herold"

So I have a digital pulse and I want to send a monitor signal out to a
'scope. But the digital logic doesn't have enough poop to drive 50
ohms,


** FFS - do you have a wide band scope probe ??

That job is just what they are designed to do.

The only alternative is an active probe.

Unterminated co-axial cable has a null ( ie zero impedance ) at the 1/4 wave
length frequency - about 50MHz for a 1 meter length.




.... Phil

 

[brent]

If I have a source terminated transmission line (think of a length of
coax with a 50 ohm resistor to ground at the source.) and it goes into
an open circuit.  (say into an unterminated 'scope input.) Then is
there a reflection from the unterminated end?  And if so, the reason
that there is no ringing observed is that the source termination soaks
up the return signal.  Or does the source termination make the
transmission line 'happy' regardless of what's on the other end?

Thanks,
George H.

Here is an interactive tutorial about how reflection works:

http://www.fourier-series.com/rf-concepts/reflection.html

 

[mike]

Yeah but then you have to tell people to terminate the line. This way
it's 'mindless'.
(Instruments for students.)

George H.

Based on my experience with new engineers, they need to be taught stuff
like this.
Hiding the "gotchas" isn't doing them any favors.
Last thing I'd want is a Mindless Engineer.

 

[Jasen Betts]

Thanks Jon, (and Wimpie) In this case the source doesn't have enough
current to drive the 50 ohms.

Switcing to 75 Ohm or higher impedance coax would seem like a good
first step then.

 

[whit3rd]

I wish! (physics students) Some can hardly drive a 'scope. I've
spent some time on the phone explaining triggering....

The current generation digital scopes are TERRIBLE from a human-factors design perspective,
In my generation, students who had experience with television sync
twiddling were handling analog trigger controls on the front panel,
with one-and-only-one display mode running just left of the knobs.
Except for left-handers, this was an easy task.

Unlabeled 'multipurpose' knobs, mixed mode (letters-on-screen/trace-on-screen,
cursors-on-screen/pointers-on-screen/colorcodes) readouts,
menu hierarchy, and 'automatic' features make a modern digital scope a
morass for a struggling newbie. Us oldsters get trapped, too.

 

[[email protected]]

Unlabeled 'multipurpose' knobs, mixed mode (letters-on-screen/trace-on-screen,

cursors-on-screen/pointers-on-screen/colorcodes) readouts,

menu hierarchy, and 'automatic' features make a modern digital scope a

morass for a struggling newbie. Us oldsters get trapped, too.

If you ever had to do complicated scope data capture with no end to the number of conditions upon which the capture should be conditioned, you would appreciate a modern scope more, and maybe realize it's still not enough. Thesame goes for modern logic analyzers, never enough counters.

 

[Bill Sloman]

to the op.

as long as you are sending the signal into the cable at one end taking
the signal out at one end (no taps etc) then you need to terminate the
cable at only one end, either end will do...

But if you are really anxious to minimise reflections, "double termination" at both ends can be worth the extrra effort, and the halving of the signal amplitude involved. Balanced pairs of signal can give you back the lost amplitude.

when terminating the cable on the source end as you described, there
will be reflections on the cable all along the way, but it is just as
you said, the Rx end will have a clean waveform. There will be no RE-
reflection because the source is terminated.

Except that no termination is ever perfect. Figure on a +/-2% tolerance on the cable impedance. I was once offered cables that were custom made to a more tightly specified tolerance, but I've never seen anything like that from a distributor.

 

[Bill Sloman]

1% terminators were common in the '70s. .1% are common now.

1% termination resistors sort of make sense. 0.1% don't unless you've got peculiarly precisely constructed coaxial cable/transmission line. The tolerance on the characteristic impedance of standard coaxial cable is about +/-2%.

http://www.belden.com/docs/upload/Precision-Video-Cables-Part-1.pdf

Terminating something like that with a O.1% tolerance terminating resistor would be a complete waste of money.

 

[mike]

But if you are really anxious to minimise reflections, "double termination" at both ends can be worth the extrra effort, and the halving of the signal amplitude involved. Balanced pairs of signal can give you back the lost amplitude.


Except that no termination is ever perfect. Figure on a +/-2% tolerance on the cable impedance. I was once offered cables that were custom made to a more tightly specified tolerance, but I've never seen anything like that from a distributor.

I designed a source-terminated pulse generator.
The only way we could guarantee the customers wouldn't return it for
"transient response out of spec" was to ship a tight tolerance cable
with it.

 

[George Herold]

[...]
If the source is a 5 volt step generator with a 50 ohm source
impedance, the initial wave will be 2.5 volts. When it hits the open
end, it will double to be a clean 5 volt step. A 5 volt wave will then
zip back to the generator end and die there when it hits, and it's all
over. Clean 5 volt step at the far end, and the generator voltage sees
100 ohms (50 internal+50 line) as the transient load, 50 mA peak
current.
LT Spice has transmission lines. If you string a bunch together, you
can see what happens at intermediate points.

I'll have to try it.  Does LT spice have directional couplers?  It'd
be nice to see what's coming and going....

Directional couplers in spice are trivial! Basically, it's nothing
more than a Wheatstone bridge. Replace one of the sides by the tline
and apply twice the signal to the top of the bridge. The reflection
appears across the bridge. Below an LTspice example to play with.

Jeroen Belleman
==================================================
Version 4
SHEET 1 880 680
WIRE 32 112 -320 112
WIRE 224 112 32 112
WIRE 32 144 32 112
WIRE 224 144 224 112
WIRE -320 224 -320 112
WIRE 32 240 32 224
WIRE 128 240 32 240
WIRE 160 240 128 240
WIRE 224 240 224 224
WIRE 336 240 224 240
WIRE 576 240 432 240
WIRE 32 272 32 240
WIRE 336 272 272 272
WIRE 496 272 432 272
WIRE 496 288 496 272
WIRE 576 288 576 240
WIRE 272 304 272 272
WIRE -320 336 -320 304
WIRE 336 352 272 352
WIRE 400 352 336 352
WIRE 32 368 32 352
WIRE 272 368 272 352
WIRE 400 368 400 352
WIRE 224 384 224 240
WIRE 576 384 576 368
WIRE 160 432 160 240
WIRE 224 432 160 432
WIRE 272 464 272 448
WIRE 400 464 400 448
FLAG 272 304 0
FLAG 496 288 0
FLAG 32 368 0
FLAG 576 384 0
FLAG -320 336 0
FLAG 272 464 0
FLAG 400 464 0
FLAG 336 352 reflected
FLAG 128 240 incident
SYMBOL res 16 128 R0
SYMATTR InstName R1
SYMATTR Value 50
SYMBOL res 16 256 R0
SYMATTR InstName R2
SYMATTR Value 50
SYMBOL res 208 128 R0
SYMATTR InstName R3
SYMATTR Value 50
SYMBOL tline 384 256 R0
SYMATTR InstName T1
SYMBOL res 560 272 R0
SYMATTR InstName R4
SYMATTR Value 1meg
SYMBOL voltage -320 208 R0
WINDOW 123 0 0 Left 2
WINDOW 39 0 0 Left 2
SYMATTR InstName V1
SYMATTR Value PULSE(0 2 10n 1n 1n 10n)
SYMBOL e 272 352 R0
SYMATTR InstName E1
SYMATTR Value 1
SYMBOL res 384 352 R0
SYMATTR InstName R5
SYMATTR Value 1
TEXT 80 40 Left 2 !.tran 200n- Hide quoted text -

- Show quoted text -

Cool, Thanks. I'll have to think a bit before I understand how that
works... no don't tell me! I'll ask if I get stuck.

George H.

 

[George Herold]

something like this?

Version 4
SHEET 1 880 680
WIRE -192 64 -304 64
WIRE 176 64 -112 64
WIRE 352 64 256 64
WIRE 480 64 352 64
WIRE 720 64 576 64
WIRE 720 80 720 64
WIRE 480 128 480 96
WIRE 576 144 576 96
WIRE 176 160 96 160
WIRE 320 160 256 160
WIRE 720 176 720 160
WIRE 96 224 96 160
WIRE 96 288 96 224
WIRE 176 288 96 288
WIRE 352 288 352 64
WIRE 352 288 256 288
WIRE 0 384 -32 384
WIRE 176 384 0 384
WIRE 320 384 320 160
WIRE 320 384 256 384
WIRE 416 384 320 384
WIRE 464 384 416 384
FLAG 0 464 0
FLAG 416 464 0
FLAG 96 224 0
FLAG 720 176 0
FLAG -304 144 0
FLAG 480 128 0
FLAG 576 144 0
FLAG -32 384 forward
FLAG 464 384 reverse
SYMBOL ind2 160 176 R270
WINDOW 0 32 56 VTop 2
WINDOW 3 5 56 VBottom 2
SYMATTR InstName L1
SYMATTR Value 100µ
SYMATTR Type ind
SYMBOL ind2 272 48 R90
WINDOW 0 5 56 VBottom 2
WINDOW 3 32 56 VTop 2
SYMATTR InstName L2
SYMATTR Value 1µ
SYMATTR Type ind
SYMBOL ind2 272 272 R90
WINDOW 0 5 56 VBottom 2
WINDOW 3 32 56 VTop 2
SYMATTR InstName L3
SYMATTR Value 100µ
SYMATTR Type ind
SYMBOL ind2 160 400 R270
WINDOW 0 32 56 VTop 2
WINDOW 3 5 56 VBottom 2
SYMATTR InstName L4
SYMATTR Value 1µ
SYMATTR Type ind
SYMBOL res -16 368 R0
SYMATTR InstName R1
SYMATTR Value 50
SYMBOL res 400 368 R0
SYMATTR InstName R2
SYMATTR Value 50
SYMBOL res 704 64 R0
SYMATTR InstName R3
SYMATTR Value 1000
SYMBOL voltage -304 48 R0
WINDOW 0 -56 -5 Left 2
WINDOW 123 0 0 Left 2
WINDOW 39 24 124 Left 2
SYMATTR InstName V1
SYMATTR Value SINE(0 1 10e6 100n 0 0 1)
SYMBOL tline 528 80 R0
WINDOW 3 -5 -93 Top 2
SYMATTR InstName T1
SYMATTR Value Td=500n Z0=50
SYMBOL res -96 48 R90
WINDOW 0 0 56 VBottom 2
WINDOW 3 32 56 VTop 2
SYMATTR InstName R4
SYMATTR Value 50
TEXT 64 112 Left 2 !k1 L1 L2 1
TEXT 56 336 Left 2 !K2 L3 L4 1
TEXT -242 488 Left 2 !.tran 1.5u

-Lasse- Hide quoted text -

- Show quoted text -

Nice, Thanks. But that's going to take even more thought than the
circuit Jeroen posted.
(I can get lost going through those transformers :^)

George H.

 

[George Herold]

Appendix A on "how to drive a scope" in Art of Electronics, 2nd edition,
by Horowitz and Hill looks like a reasonable student guide to me.  I
must warn you that I have never tried it on, or seen it tried on, real
live physics students. Because of when it was written, it's for
analog scopes.

Standard disclaimers apply; I don't get money or other consideration
from any companies mentioned.

Matt Roberds

Grin, as the saving goes, "you can lead a horse to water but you can't
make him read."

George H.

 

[George Herold]

Based on my experience with new engineers, they need to be taught stuff
like this.
Hiding the "gotchas" isn't doing them any favors.
Last thing I'd want is a Mindless Engineer.

Sure... but you can only teach one (maybe two) thing(s) at a time. In
this case there's lots of other stuff going on so that just giving
them an 'easy' monitor pulse is not that bad.

In retrospect I should have done as Joerg (and maybe others) suggested
and added more poop. A real line driver would not have been much
more.

George H.

 

[George Herold]

The current generation digital scopes are TERRIBLE from a human-factors design perspective,
In my generation, students who had experience with television sync
twiddling were handling analog trigger controls on the front panel,
with one-and-only-one display mode running just left of the knobs.
Except for left-handers, this was an easy task.

Unlabeled 'multipurpose' knobs,  mixed mode (letters-on-screen/trace-on-screen,
cursors-on-screen/pointers-on-screen/colorcodes) readouts,
menu hierarchy, and 'automatic' features make a modern digital scope a
morass for a struggling newbie.    Us oldsters get trapped, too.

Well don't be too hard on the latest generation of DSO's. They are
much better than those earlier generations that were all buttons and
maybe one knob. Those were impossible to use.
I really like having colored traces that match the color on the gain
and offset knobs... that is nice.

I guess the one button I'd like to remove is the autoset. (I'd rather
have them think!)

George H.

 

[Robert Baer]

Exactly, 56R is a better match, but even your resistor precision gets a 5% tolerance too. Just pretend you don't see the 10mV overshoot, it's just monitor front end parasitics ....

On PCB layout, we used to use two 100 ohm 1/8W carbon comps flat
against the ground plane spread 60 degrees from the RF connector.
Carbon comps were typically 2 percent low with a one percent spread.
40 years ago.

 

[George Herold]

Here's a cheap 51 ohm carbon film resistor:

https://dl.dropboxusercontent.com/u/53724080/TDR/51R_setup.JPG

and its TDR step response

https://dl.dropboxusercontent.com/u/53724080/TDR/51R_TDR.JPG

The main error seems to be from the length of the resistance element,
which is over in about 200 picoseconds. Then it rings at 9.6 GHz for a
bit.

Plenty good up to maybe 1 GHz. Carbon comps are probably better.

--

John Larkin         Highland Technology, Inc

jlarkin at highlandtechnology dot comhttp://www.highlandtechnology.com

Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators
Custom laser drivers and controllers
Photonics and fiberoptic TTL data links
VME thermocouple, LVDT, synchro   acquisition and simulation- Hide quoted text -

- Show quoted text -

Neat, and if you stick a metal film in there does it take longer? (I
assume 50 ohm MF's are spiral cut.)

George H.