Searching \ for '[EE] Transmission Line or Induction effect?' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/index.htm?key=transmission+line
Search entire site for: 'Transmission Line or Induction effect?'.

Exact match. Not showing close matches.
PICList Thread
'[EE] Transmission Line or Induction effect?'
2011\12\15@093850 by Electron

flavicon
face

Howdy,
say I have a signal generator that is generating low frequency pulses with very quick
rising/falling times, connected directly to my oscilloscope, and (through a BNC splitter)
to a couple of 1 meter long wires (one for signal the other for ground), open for now.

I can see a nice waveform, which looks as it should.

Now I add a capacitor to the end of the 1 meter cable (not a coaxial cable, just 2 wires),
and of course the waveform will look distorted, because the capacitor is loading the 50 ohm
output of the signal generator.

BUT if I set the 2 Gsps oscilloscope to measure peak, I can see that just after the rising
and falling slopes, now that I have added the capacitor there have appeared voltage spikes.

It doesn't surprise me, but what I cannot be 100% sure of is if these spikes are due to
transmission line effects (probably they are), or by inductive effects (of the cable and/or
of the capacitor)?

Cheers,
Mario

2011\12\15@101223 by Dave Tweed

face
flavicon
face
Electron wrote:
> say I have a signal generator that is generating low frequency pulses with
> very quick rising/falling times, connected directly to my oscilloscope, and
> (through a BNC splitter) to a couple of 1 meter long wires (one for signal
> the other for ground), open for now.
>
> I can see a nice waveform, which looks as it should.
>
> Now I add a capacitor to the end of the 1 meter cable (not a coaxial cable,
> just 2 wires), and of course the waveform will look distorted, because the
> capacitor is loading the 50 ohm output of the signal generator.
>
> BUT if I set the 2 Gsps oscilloscope to measure peak, I can see that just
> after the rising and falling slopes, now that I have added the capacitor
> there have appeared voltage spikes.
>
> It doesn't surprise me, but what I cannot be 100% sure of is if these spikes
> are due to transmission line effects (probably they are), or by inductive
> effects (of the cable and/or of the capacitor)?

Short answer: Yes.

Longer answer: A transmission line is simply a pair of conductors with
distributed inductance and capacitance, so in the end, you really have to
consider the lumped effects and the distributed effects together.

For a good introduction, see Robert Lacoste's column in Circuit Cellar -- in
particular:
#223 February 2009, p. 64, Microstrip Techniques
#225 April 2009, p. 50, Time Domain Reflectometry - Detect and Measure
  Impedance Mismatches
#227 June 2009, p. 52, High-Speed Signal Transmission - From Eye Diagrams to
  Preemphasis and Equalization
#245 December 2010, p. 50, Parasitic Components - When Capacitors Behave Like
  Inductors
#253 August 2011, p. 64, RF and High-Speed PCBs - How to Avoid Basic Mistakes
I followed up on this topic myself in my EQ quiz in issue #226.
Questions #1 and #2 directly address the situation you're asking about.
#226 May 2009, p. 13, HTML <http://www.circuitcellar.com/eq/226/226.html> Test
  Your EQ
-- Dave Twee

2011\12\15@113333 by Sean Breheny

face picon face
I agree with your basic point, Dave, but there is a distinction
between a finite lumped-element model and a real transmission-line
model. The former has no true delay since it has a finite number of
poles whereas the latter has a true delay. I think it is likely in
this case that Mario could determine which kind of effect he is seeing
by looking at the timing of the voltage overshoot versus the rising
edge to see if it corresponds to the expected propagation delay. Also,
if he arranges the wires in a well-defined rectangular loop, then he
could use an inductance formula to estimate the inductance of the
wires and construct a physically-small lumped-element circuit which
provides the same inductance and the same capacitance and see if he
gets the same effect.

Sean


On Thu, Dec 15, 2011 at 10:12 AM, Dave Tweed <spam_OUTpicTakeThisOuTspamdtweed.com> wrote:
> Short answer: Yes.
>
> Longer answer: A transmission line is simply a pair of conductors with
> distributed inductance and capacitance, so in the end, you really have to
> consider the lumped effects and the distributed effects together.

2011\12\16@024611 by Electron

flavicon
face

or I will try to double the length of the wires and see on the scope
screen how the spike changed (I should expect it to become twice as
wide).


At 17.33 2011.12.15, you wrote:
{Quote hidden}

>

More... (looser matching)
- Last day of these posts
- In 2011 , 2012 only
- Today
- New search...