Poor man\'s TDR...

[Martin Brown]

Can anyone think of a cunning quick and dirty way to do TDR on modest
length 3 - 10 m cables to look for breaks ideally with resolution 10cm
or better and without using an expensive fast sampling scope?

That means return pulse timing accurate to 30ns or better.
(in the ballpark of what modern CPU internal timers can do)

--
Martin Brown

 

[Arie de Muijnck]

On 2023-04-27 11:05, Martin Brown wrote:

Can anyone think of a cunning quick and dirty way to do TDR on modest length 3 - 10 m cables to look for breaks ideally with resolution 10cm or better and without using an expensive fast sampling scope?

That means return pulse timing accurate to 30ns or better.
(in the ballpark of what modern CPU internal timers can do)

Most cable breaks can be found by measuring the capacitance of the wire from each end against all others shorted together.

Arie

 

[Don Y]

On 4/27/2023 2:05 AM, Martin Brown wrote:

Can anyone think of a cunning quick and dirty way to do TDR on modest length 3
- 10 m cables to look for breaks ideally with resolution 10cm or better and
without using an expensive fast sampling scope?

The inexpensive meters use a nominal capacitance/unit length.
Some allow this to be trimmed or blindly overridden.

If all of your cable is from the same supplier, you may
just want to use some known length to \"calibrate\" such
a meter.

[I\'ve not checked the absolute accuracy of mine. But, know that
all the cables are \'store bought\" so unlikely to be 11.237 ft long...]

That means return pulse timing accurate to 30ns or better.
(in the ballpark of what modern CPU internal timers can do)

 

[Martin Brown]

On 27/04/2023 11:19, Arie de Muijnck wrote:

On 2023-04-27 11:05, Martin Brown wrote:
Can anyone think of a cunning quick and dirty way to do TDR on modest
length 3 - 10 m cables to look for breaks ideally with resolution 10cm
or better and without using an expensive fast sampling scope?

That means return pulse timing accurate to 30ns or better.
(in the ballpark of what modern CPU internal timers can do)


Most cable breaks can be found by measuring the capacitance of the wire
from each end against all others shorted together.

Thank you both for the suggestion of capacitance meter. That\'s easy
enough... Even without knowing the capacitance per unit length I can
measure from each end and use the ratio to guess where the break is.

Its an XLR mike cable belonging to the group that did a free gig at our
village hall that went noisy and I promised to take a look at it. The
shield has broken somewhere along its length and I want to find where.

(I was hoping for a broken or dry joint on one of the XLR connectors)

--
Martin Brown

 

[Lasse Langwadt Christensen]

torsdag den 27. april 2023 kl. 13.49.15 UTC+2 skrev Martin Brown:

On 27/04/2023 11:19, Arie de Muijnck wrote:
On 2023-04-27 11:05, Martin Brown wrote:
Can anyone think of a cunning quick and dirty way to do TDR on modest
length 3 - 10 m cables to look for breaks ideally with resolution 10cm
or better and without using an expensive fast sampling scope?

That means return pulse timing accurate to 30ns or better.
(in the ballpark of what modern CPU internal timers can do)


Most cable breaks can be found by measuring the capacitance of the wire
from each end against all others shorted together.
Thank you both for the suggestion of capacitance meter. That\'s easy
enough... Even without knowing the capacitance per unit length I can
measure from each end and use the ratio to guess where the break is.

Its an XLR mike cable belonging to the group that did a free gig at our
village hall that went noisy and I promised to take a look at it. The
shield has broken somewhere along its length and I want to find where.

(I was hoping for a broken or dry joint on one of the XLR connectors)

an multimeter on the screen and massaging the cable from end to end
might find it

 

[Don Y]

On 4/27/2023 4:49 AM, Martin Brown wrote:

On 27/04/2023 11:19, Arie de Muijnck wrote:
On 2023-04-27 11:05, Martin Brown wrote:
Can anyone think of a cunning quick and dirty way to do TDR on modest length
3 - 10 m cables to look for breaks ideally with resolution 10cm or better
and without using an expensive fast sampling scope?

That means return pulse timing accurate to 30ns or better.
(in the ballpark of what modern CPU internal timers can do)


Most cable breaks can be found by measuring the capacitance of the wire from
each end against all others shorted together.

Thank you both for the suggestion of capacitance meter. That\'s easy enough...
Even without knowing the capacitance per unit length I can measure from each
end and use the ratio to guess where the break is.

Its an XLR mike cable belonging to the group that did a free gig at our village
hall that went noisy and I promised to take a look at it. The shield has broken
somewhere along its length and I want to find where.

Oh, that\'s easy! Just cut off a few inches and check to see if the
shield is still \"open\". Repeat until you\'ve found the section containing
the break! ;-)

> (I was hoping for a broken or dry joint on one of the XLR connectors)

 

[Uwe Bonnes]

Don Y <blockedofcourse@foo.invalid> wrote:

Oh, that\'s easy! Just cut off a few inches and check to see if the
shield is still \"open\". Repeat until you\'ve found the section containing
the break! ;-)

(I was hoping for a broken or dry joint on one of the XLR connectors)


Enhance to binary search and start cutting in half, then quarter, etc ;-)

--
Uwe Bonnes bon@elektron.ikp.physik.tu-darmstadt.de

Institut fuer Kernphysik Schlossgartenstrasse 9 64289 Darmstadt
--------- Tel. 06151 1623569 ------- Fax. 06151 1623305 ---------

 

[Martin Brown]

On 27/04/2023 16:00, Uwe Bonnes wrote:

Don Y <blockedofcourse@foo.invalid> wrote:

Oh, that\'s easy! Just cut off a few inches and check to see if the
shield is still \"open\". Repeat until you\'ve found the section containing
the break! ;-)

(I was hoping for a broken or dry joint on one of the XLR connectors)


Enhance to binary search and start cutting in half, then quarter, etc ;-)

The idea is to leave them with a lead long enough to be useful...

--
Martin Brown

 

[Don Y]

On 4/27/2023 8:05 AM, Martin Brown wrote:

On 27/04/2023 16:00, Uwe Bonnes wrote:
Don Y <blockedofcourse@foo.invalid> wrote:

Oh, that\'s easy!  Just cut off a few inches and check to see if the
shield is still \"open\".  Repeat until you\'ve found the section containing
the break!  ;-)

(I was hoping for a broken or dry joint on one of the XLR connectors)


Enhance to binary search and start cutting in half, then quarter, etc ;-)

The idea is to leave them with a lead long enough to be useful...

Of course. The (intended) humor was in imagining folks
considering how such an approach COULD work (e.g., if you
start at the \"wrong\" end, you could conceivably end up
with a 2 inch cable -- if the break was that close to the
far end!)

 

[Don Y]

On 4/27/2023 8:17 AM, Don Y wrote:

On 4/27/2023 8:05 AM, Martin Brown wrote:
On 27/04/2023 16:00, Uwe Bonnes wrote:
Don Y <blockedofcourse@foo.invalid> wrote:

Oh, that\'s easy!  Just cut off a few inches and check to see if the
shield is still \"open\".  Repeat until you\'ve found the section containing
the break!  ;-)

(I was hoping for a broken or dry joint on one of the XLR connectors)


Enhance to binary search and start cutting in half, then quarter, etc ;-)

The idea is to leave them with a lead long enough to be useful...

Of course.  The (intended) humor was in imagining folks
considering how such an approach COULD work (e.g., if you
start at the \"wrong\" end, you could conceivably end up
with a 2 inch cable -- if the break was that close to the
far end!)

Said another way, it\'s like the carpenter cutting the legs
on the table: \"Damn! STILL too short!\"

 

[Clive Arthur]

On 27/04/2023 10:05, Martin Brown wrote:

Can anyone think of a cunning quick and dirty way to do TDR on modest
length 3 - 10 m cables to look for breaks ideally with resolution 10cm
or better and without using an expensive fast sampling scope?

That means return pulse timing accurate to 30ns or better.
(in the ballpark of what modern CPU internal timers can do)

Connect a high voltage supply across the two shield ends and use your
thermal imaging camera to find the arc. Or use a portable radio to find
the arc.

[It may sound silly, but I have a cheap eBay HV arc generator PCB and a
TIC both within arm\'s reach.]

--
Cheers
Clive

 

[John Larkin]

On Thu, 27 Apr 2023 10:05:29 +0100, Martin Brown
<\'\'\'newspam\'\'\'@nonad.co.uk> wrote:

Can anyone think of a cunning quick and dirty way to do TDR on modest
length 3 - 10 m cables to look for breaks ideally with resolution 10cm
or better and without using an expensive fast sampling scope?

That means return pulse timing accurate to 30ns or better.
(in the ballpark of what modern CPU internal timers can do)

Resolving 10 cm needs a roughly 60 ps TDR rise time. The scope would
need to have a bandwidth around 5 GHz.

There are cheap VNAs that can do FFT tricks to emulate TDR.

Or as suggested, measure capacitance.

Most cable shields are leaky; something could be done with that. Or
mag fields. Or even thermal imaging.

 

[John Larkin]

On Thu, 27 Apr 2023 12:49:06 +0100, Martin Brown
<\'\'\'newspam\'\'\'@nonad.co.uk> wrote:

On 27/04/2023 11:19, Arie de Muijnck wrote:
On 2023-04-27 11:05, Martin Brown wrote:
Can anyone think of a cunning quick and dirty way to do TDR on modest
length 3 - 10 m cables to look for breaks ideally with resolution 10cm
or better and without using an expensive fast sampling scope?

That means return pulse timing accurate to 30ns or better.
(in the ballpark of what modern CPU internal timers can do)


Most cable breaks can be found by measuring the capacitance of the wire
from each end against all others shorted together.

Thank you both for the suggestion of capacitance meter. That\'s easy
enough... Even without knowing the capacitance per unit length I can
measure from each end and use the ratio to guess where the break is.

Its an XLR mike cable belonging to the group that did a free gig at our
village hall that went noisy and I promised to take a look at it. The
shield has broken somewhere along its length and I want to find where.

(I was hoping for a broken or dry joint on one of the XLR connectors)

Connect one shield end to a signal generator and slide a scope probe
along the outside. The break should be obvious.

 

[John Larkin]

On Thu, 27 Apr 2023 16:38:26 +0100, Clive Arthur
<clive@nowaytoday.co.uk> wrote:

On 27/04/2023 10:05, Martin Brown wrote:
Can anyone think of a cunning quick and dirty way to do TDR on modest
length 3 - 10 m cables to look for breaks ideally with resolution 10cm
or better and without using an expensive fast sampling scope?

That means return pulse timing accurate to 30ns or better.
(in the ballpark of what modern CPU internal timers can do)


Connect a high voltage supply across the two shield ends and use your
thermal imaging camera to find the arc. Or use a portable radio to find
the arc.

Just look for where the smoke is coming out.

[It may sound silly, but I have a cheap eBay HV arc generator PCB and a
TIC both within arm\'s reach.]

 

[Martin Brown]

On 27/04/2023 18:09, John Larkin wrote:

On Thu, 27 Apr 2023 12:49:06 +0100, Martin Brown
\'\'\'newspam\'\'\'@nonad.co.uk> wrote:

On 27/04/2023 11:19, Arie de Muijnck wrote:
On 2023-04-27 11:05, Martin Brown wrote:
Can anyone think of a cunning quick and dirty way to do TDR on modest
length 3 - 10 m cables to look for breaks ideally with resolution 10cm
or better and without using an expensive fast sampling scope?

That means return pulse timing accurate to 30ns or better.
(in the ballpark of what modern CPU internal timers can do)


Most cable breaks can be found by measuring the capacitance of the wire
from each end against all others shorted together.

Thank you both for the suggestion of capacitance meter. That\'s easy
enough... Even without knowing the capacitance per unit length I can
measure from each end and use the ratio to guess where the break is.

Its an XLR mike cable belonging to the group that did a free gig at our
village hall that went noisy and I promised to take a look at it. The
shield has broken somewhere along its length and I want to find where.

(I was hoping for a broken or dry joint on one of the XLR connectors)

Connect one shield end to a signal generator and slide a scope probe
along the outside. The break should be obvious.

That ought to work pretty well - thanks.
>

--
Martin Brown

 

[Martin Brown]

On 27/04/2023 13:24, Lasse Langwadt Christensen wrote:

torsdag den 27. april 2023 kl. 13.49.15 UTC+2 skrev Martin Brown:
On 27/04/2023 11:19, Arie de Muijnck wrote:
On 2023-04-27 11:05, Martin Brown wrote:
Can anyone think of a cunning quick and dirty way to do TDR on modest
length 3 - 10 m cables to look for breaks ideally with resolution 10cm
or better and without using an expensive fast sampling scope?

That means return pulse timing accurate to 30ns or better.
(in the ballpark of what modern CPU internal timers can do)


Most cable breaks can be found by measuring the capacitance of the wire
from each end against all others shorted together.
Thank you both for the suggestion of capacitance meter. That\'s easy
enough... Even without knowing the capacitance per unit length I can
measure from each end and use the ratio to guess where the break is.

Its an XLR mike cable belonging to the group that did a free gig at our
village hall that went noisy and I promised to take a look at it. The
shield has broken somewhere along its length and I want to find where.

(I was hoping for a broken or dry joint on one of the XLR connectors)


an multimeter on the screen and massaging the cable from end to end
might find it

I tried that first - sadly no joy.

Peculiar thing for me was that despite it only being an earthing shield
the sound was badly compromised with a lot of crackling and popping on
the two balanced lines. XLR audio gear isn\'t my thing at all but I can
solder a lot better than whoever last put the lead together!

I saw no signs of a short or path to either signal line both were well
isolated from the shield >10M.

--
Martin Brown

 

[John Larkin]

On Thu, 27 Apr 2023 19:12:20 +0100, Martin Brown
<\'\'\'newspam\'\'\'@nonad.co.uk> wrote:

On 27/04/2023 18:09, John Larkin wrote:
On Thu, 27 Apr 2023 12:49:06 +0100, Martin Brown
\'\'\'newspam\'\'\'@nonad.co.uk> wrote:

On 27/04/2023 11:19, Arie de Muijnck wrote:
On 2023-04-27 11:05, Martin Brown wrote:
Can anyone think of a cunning quick and dirty way to do TDR on modest
length 3 - 10 m cables to look for breaks ideally with resolution 10cm
or better and without using an expensive fast sampling scope?

That means return pulse timing accurate to 30ns or better.
(in the ballpark of what modern CPU internal timers can do)


Most cable breaks can be found by measuring the capacitance of the wire
from each end against all others shorted together.

Thank you both for the suggestion of capacitance meter. That\'s easy
enough... Even without knowing the capacitance per unit length I can
measure from each end and use the ratio to guess where the break is.

Its an XLR mike cable belonging to the group that did a free gig at our
village hall that went noisy and I promised to take a look at it. The
shield has broken somewhere along its length and I want to find where.

(I was hoping for a broken or dry joint on one of the XLR connectors)

Connect one shield end to a signal generator and slide a scope probe
along the outside. The break should be obvious.

That ought to work pretty well - thanks.

A sig gen and a cheap scope can measure femtofarads, or even
attofarads.

 

[Joe Gwinn]

On Thu, 27 Apr 2023 12:49:06 +0100, Martin Brown
<\'\'\'newspam\'\'\'@nonad.co.uk> wrote:

On 27/04/2023 11:19, Arie de Muijnck wrote:
On 2023-04-27 11:05, Martin Brown wrote:
Can anyone think of a cunning quick and dirty way to do TDR on modest
length 3 - 10 m cables to look for breaks ideally with resolution 10cm
or better and without using an expensive fast sampling scope?

That means return pulse timing accurate to 30ns or better.
(in the ballpark of what modern CPU internal timers can do)


Most cable breaks can be found by measuring the capacitance of the wire
from each end against all others shorted together.

Thank you both for the suggestion of capacitance meter. That\'s easy
enough... Even without knowing the capacitance per unit length I can
measure from each end and use the ratio to guess where the break is.

Yes. This is the way the US Phone Companies did it, but for detecting
shorts with a Wheatstone bridge.

Capacitance came later. They also could measure distance to wet cable
(copper wires, pulp paper insulation, lead cable jacket with small
hole allowing water ingress). I don\'t recall the exact method, but I
read it in a book published by AT&T in 1953.

Its an XLR mike cable belonging to the group that did a free gig at our
village hall that went noisy and I promised to take a look at it. The
shield has broken somewhere along its length and I want to find where.

(I was hoping for a broken or dry joint on one of the XLR connectors)

It would be a good bet for a studio cable - they almost always break
at a connector.

Joe Gwinn

 

[Tabby]

On Thursday, 27 April 2023 at 10:05:38 UTC+1, Martin Brown wrote:

Can anyone think of a cunning quick and dirty way to do TDR on modest
length 3 - 10 m cables to look for breaks ideally with resolution 10cm
or better and without using an expensive fast sampling scope?

That means return pulse timing accurate to 30ns or better.
(in the ballpark of what modern CPU internal timers can do)

Breaks normally happen in the 6\" at one end, and you can usually tell which end has had a harder life just by looking. You don\'t always need testgear.

 

[Clifford Heath]

On 28/04/23 03:06, John Larkin wrote:

On Thu, 27 Apr 2023 10:05:29 +0100, Martin Brown
\'\'\'newspam\'\'\'@nonad.co.uk> wrote:

Can anyone think of a cunning quick and dirty way to do TDR on modest
length 3 - 10 m cables to look for breaks ideally with resolution 10cm
or better and without using an expensive fast sampling scope?

That means return pulse timing accurate to 30ns or better.
(in the ballpark of what modern CPU internal timers can do)

Resolving 10 cm needs a roughly 60 ps TDR rise time. The scope would
need to have a bandwidth around 5 GHz.

There are cheap VNAs that can do FFT tricks to emulate TDR.

Yes - the NanoVNA does it.