http://opalkelly.com/

Author Topic: My Time Domain Reflectometer (a la w2aew)  (Read 8387 times)

0 Members and 1 Guest are viewing this topic.

Offline android

  • Regular Contributor
  • *
  • Posts: 129
  • Country: au
My Time Domain Reflectometer (a la w2aew)
« on: July 23, 2015, 12:15:44 AM »
Thanks to w2aew's excellent YouTube videos on TDR, I've finally got around to building my own version.  :)

I was going to use through hole, but ended up using SMT parts so it looks pretty clean. I tested it with a 5m length of RG59 and measured a 41 ns round trip which seems about right (although by my calculation that would make the velocity factor about 0.81 ...I don't know whether that's normal or not).

Lecturer: "There is no language in which a double positive implies a negative."
Student:  "Yeah...right."
 

Offline AF6LJ

  • Supporter
  • ****
  • Posts: 2365
  • Country: us
Re: My Time Domain Reflectometer (a la w2aew)
« Reply #1 on: July 23, 2015, 12:25:32 AM »
Nice;
I have been meaning to build one of those.
Sue AF6LJ
Test Equipment Addict, And Proud Of It.
 

Offline mtdoc

  • Super Contributor
  • ***
  • Posts: 2681
  • Country: us
Re: My Time Domain Reflectometer (a la w2aew)
« Reply #2 on: July 23, 2015, 05:02:14 AM »
Awesome! I like the googly eyes.

I made one a while back - dead bug style.  A fun and usefull project. Kudos to Alan!
 

Offline tggzzz

  • Super Contributor
  • ***
  • Posts: 5035
  • Country: gb
    • Having fun doing more, with less
Re: My Time Domain Reflectometer (a la w2aew)
« Reply #3 on: July 23, 2015, 10:16:05 AM »
I'm currently playing around with a homebrew TDR that can measure stub lengths with a resolution of ~40mm. Since it is dirt cheap (£40 IIRC) it does have some limitations, but might be useful nonetheless.

I'm slowly documenting it, and will publish it in the not too distant future, unless higher priority interrupts occur.
There are lies, damned lies, statistics - and ADC/DAC specs.
Gliding aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less
 

Offline android

  • Regular Contributor
  • *
  • Posts: 129
  • Country: au
Re: My Time Domain Reflectometer (a la w2aew)
« Reply #4 on: July 23, 2015, 07:17:11 PM »
Awesome! I like the googly eyes.
...best practice is to keep an eye on cable terminations  ;D

I'm currently playing around with a homebrew TDR that can measure stub lengths with a resolution of ~40mm.
...sounds good. I'd be interested to see the outcome. How do you get 40 mm resolution? That's about a 165 ps interval for my RG59 cable.

While I was shopping for SN74C14 chips I stumbled across this interesting chip: LTC6991 TimerBlox: Resettable, Low
Frequency Oscillator
. It has a 1 ns rise time (compared to the 74C14's 4 ns -  but I don't know how important that really is) and a really simple circuit. Only 20 mA drive current though.



Lecturer: "There is no language in which a double positive implies a negative."
Student:  "Yeah...right."
 

Offline tarantula

  • Contributor
  • Posts: 14
  • Country: de
Re: My Time Domain Reflectometer (a la w2aew)
« Reply #5 on: July 28, 2015, 06:38:55 AM »
Why not Jim Williams' avalanche pulse generator?

http://www.jensign.com/avalanchepulsegenerator/index.html
 

Offline w2aew

  • Super Contributor
  • ***
  • Posts: 1584
  • Country: us
  • I usTa cuDnt speL enjinere, noW I aR wuN
    • My YouTube Channel
Re: My Time Domain Reflectometer (a la w2aew)
« Reply #6 on: July 28, 2015, 06:59:02 AM »
Looks good guys - but you should really add a power supply filter cap (like 100nF or so) directly across the supply pins of the chip - especially since you don't have a ground plane. 

...and it does cause me pain to see the trace shown on the Siglent scope, while the 300MHz Tek 2465 is still just below it, unpowered  :'(
======================================
YouTube channel: http://www.youtube.com/w2aew
FAE for Tektronix
 

Online tautech

  • Super Contributor
  • ***
  • Posts: 8807
  • Country: nz
    • Taupaki Technologies Ltd.
Re: My Time Domain Reflectometer (a la w2aew)
« Reply #7 on: July 28, 2015, 07:06:57 AM »
...and it does cause me pain to see the trace shown on the Siglent scope, while the 300MHz Tek 2465 is still just below it, unpowered  :'(
:-DD

Maybe it's  :-BROKE
Avid Rabid Hobbyist & NZ Siglent Distributor
 

Offline android

  • Regular Contributor
  • *
  • Posts: 129
  • Country: au
Re: My Time Domain Reflectometer (a la w2aew)
« Reply #8 on: July 28, 2015, 07:22:36 AM »
Why not Jim Williams' avalanche pulse generator?
/avalanchepulsegenerator/index.html[/url]
...didn't know about it. Might give it a try though, thanks.

you should really add a power supply filter cap
...it's there alright (1206 ceramic)....plus a larger 4.7uF ceramic too.

...and it does cause me pain to see the trace shown on the Siglent scope, while the 300MHz Tek 2465 is still just below it, unpowered  :'(
...I did test it on the Tek too but I don't really trust it because it's uncalibrated since I fixed it (dead attenuater and U800). Still has some strange behaviour but I've already spent too much on it to bother opening ì up again. At least I don't need to worry about the EEPROM fading...it doesn't have the type that causes all the pain.
Lecturer: "There is no language in which a double positive implies a negative."
Student:  "Yeah...right."
 

Offline Warhawk

  • Frequent Contributor
  • **
  • Posts: 268
  • Country: 00
    • Personal resume
Re: My Time Domain Reflectometer (a la w2aew)
« Reply #9 on: July 28, 2015, 07:43:33 AM »
Why not Jim Williams' avalanche pulse generator?
/avalanchepulsegenerator/index.html[/url]
...didn't know about it. Might give it a try though, thanks.

you should really add a power supply filter cap
...it's there alright (1206 ceramic)....plus a larger 4.7uF ceramic too.

...and it does cause me pain to see the trace shown on the Siglent scope, while the 300MHz Tek 2465 is still just below it, unpowered  :'(
...I did test it on the Tek too but I don't really trust it because it's uncalibrated since I fixed it (dead attenuater and U800). Still has some strange behaviour but I've already spent too much on it to bother opening ì up again. At least I don't need to worry about the EEPROM fading...it doesn't have the type that causes all the pain.

Hi, I use my version of Jim's avalanche pulse generator and it's a great thing for TDR. You can even recognize reflections on all T adapters and so on. Very educational. Build yourself a one ;)

Offline tggzzz

  • Super Contributor
  • ***
  • Posts: 5035
  • Country: gb
    • Having fun doing more, with less
Re: My Time Domain Reflectometer (a la w2aew)
« Reply #10 on: July 29, 2015, 02:02:27 AM »
I'm currently playing around with a homebrew TDR that can measure stub lengths with a resolution of ~40mm. Since it is dirt cheap (£40 IIRC) it does have some limitations, but might be useful nonetheless.

I'm slowly documenting it, and will publish it in the not too distant future, unless higher priority interrupts occur.

For more information, see https://entertaininghacks.wordpress.com/2015/07/27/poor-mans-homebrew-tdr-with-4cm-resolution-part-1/

Here's a taster of its capabilities: a cable with a tee junction at 3m with an open stub. The two impulses are for a 19cm open-circuit stub and a 29sm open-circuit stub. The different lengths are clearly visible.
« Last Edit: July 29, 2015, 02:04:37 AM by tggzzz »
There are lies, damned lies, statistics - and ADC/DAC specs.
Gliding aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less
 

Offline android

  • Regular Contributor
  • *
  • Posts: 129
  • Country: au
Re: My Time Domain Reflectometer (a la w2aew)
« Reply #11 on: July 29, 2015, 06:54:10 PM »

a cable with a tee junction at 3m with an open stub. The two impulses are for a 19cm open-circuit stub and a 29sm open-circuit stub.

You've lost me already  :( ...I know it must be really simple, but what is meant by "open-circuit stub"?

I've just been measuring unterminated lengths of coax.
Lecturer: "There is no language in which a double positive implies a negative."
Student:  "Yeah...right."
 

Offline vk6zgo

  • Super Contributor
  • ***
  • Posts: 3602
  • Country: au
Re: My Time Domain Reflectometer (a la w2aew)
« Reply #12 on: July 29, 2015, 08:04:52 PM »

a cable with a tee junction at 3m with an open stub. The two impulses are for a 19cm open-circuit stub and a 29sm open-circuit stub.

You've lost me already  :( ...I know it must be really simple, but what is meant by "open-circuit stub"?

I've just been measuring unterminated lengths of coax.

"Open circuit Stub" is RF "slang" for sections of open circuit transmission line connected across the main line.---they are commonly used  for filtering of unwanted signals.

For instance,a section of open ended line connected in such a manner will look like a short circuit across the main line at the frequency at which it is a 1/4 wavelength.(normally,the unwanted frequency)

To the wanted frequency,it will look like an open circuit,& will effectively be invisible to it.(you can have "short circuit stubs",too,& combinations of both)

The TDR,being wideband,sees it as an impedance mismatch,& will indicate how far down the cable it is.
 

Offline Melt-O-Tronic

  • Regular Contributor
  • *
  • Posts: 182
  • Country: us
  • Brilliant with a slaughtering iron in my hand!
Re: My Time Domain Reflectometer (a la w2aew)
« Reply #13 on: July 30, 2015, 12:00:34 AM »
Short Answer:
Open circuit stub = unterminated lengths of coax.  In the context of this thread, they are exactly the same.

More Blah Blah Blah:
In other contexts, an open circuit stub implies an unterminated length of coax whose electrical length is 1/4 wavelength (or multiple of that) of some frequency.  A stub that is 1/4 wavelength at 7.2 MHz, for example, will greatly attenuate signal components at that frequency and also multiples of that frequency to a lesser degree when it is connected to a transmission line with a tee adapter.

W2AEW has a bunch of really enlightening YouTube videos that touch on the subject.  Oops.  As the OP, you are obviously already aware of W2AEW.
« Last Edit: July 30, 2015, 12:10:28 AM by Melt-O-Tronic »
 

Offline vk6zgo

  • Super Contributor
  • ***
  • Posts: 3602
  • Country: au
Re: My Time Domain Reflectometer (a la w2aew)
« Reply #14 on: July 30, 2015, 03:30:33 PM »
Short Answer:
Open circuit stub = unterminated lengths of coax. 
Or any other type of transmission line.
A "stub"usually implies an additional section of line connected across the main transmission line.
Thus,a single piece of line attached to the TDR for the purpose of testing that line would not normally be termed a"stub". 
Quote
More Blah Blah Blah:
In other contexts, an open circuit stub implies an unterminated length of coax whose electrical length is 1/4 wavelength (or multiple of that) of some frequency.
Again "or any other type of transmission line" applies.
Of course,any length of line is a 1/4 wavelength at some frequency!
Quote
  A stub that is 1/4 wavelength at 7.2 MHz, for example, will greatly attenuate signal components at that frequency and also multiples of that frequency to a lesser degree when it is connected to a transmission line with a tee adapter

Only at odd multiples!!
At even multiples of the frequency it is 1/4 wavelength at,an open circuit "stub" will look like "just that"--an open circuit!!
At other,unrelated frequencies,it will look like an impedance which is more reactive than anything else,tending towards XL or Xc,depending on its length
 

Offline Gavin Melville

  • Contributor
  • Posts: 16
Re: My Time Domain Reflectometer (a la w2aew)
« Reply #15 on: August 02, 2015, 08:57:46 PM »
Once you get your head around it, a step (leading edge of longish pulse) works just as well as a short pulse, and much easier to make.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf