SELTing for magnet wire break detection/location?

[Stealth]

While working as tech-support for one of the local alternative ISPs dealing with standard, POTS and DSL lines, one of the things we learned to utilize for fault detection is a so-called SELT - a Single-End Line Test that supposedly emits a signal and through some method determines the distance where the wire might have a break or a short. In that way, we could determine whether the break occurred at the customer premises, somewhere along the underground/above-ground cable or at the distribution center. The readings are accurate up to a 10m resolution, which is just fine for an average 1.5 km line.

What occurred to me is - does anyone utilize a similar method when winding pickups? Commonly a SELT works over twisted pairs, though if memory serves, I have been getting fairly correct readings even if one of the lines in the TP got snapped or unwound. This way, you might be able to utilize the same method to find if a coil has snapped and where it snapped.

Last time I checked at my university (as my secondary BSc studies had me involved in telecom work, my primary being DSPs), what they showed us as a definitive instrument was a Fluke meter used specifically for testing CAT 3, 5, 6 cables, so I'm not sure if any other meters have that option, but I recall that specific Fluke meter also had a secondary slave unit, effectively performing a DELT (Dual-End Line Test).

Any ideas if this works? I'm reading through this entire MEF f-11 forum from the back towards the front and I only got to page 42. I haven't seen the method mentioned anywhere, so I figure it might work - but it might not, as well.

[David King]

I'll bet a SELT works by measuring capacitance between the two lines. The longer the line the higher the capacitance. You would need to know the cap per foot of that particular cable and you'd need to clip onto both conductors of a twisted pair...

If the cut end of a wire could bounce back a reflection you could measure the time it takes for a signal to travel both directions.

Why would you want to know where the break is when 99% of the time you're just going to cut the coil off the bobbin and rewind anyway (or toss the pickup) .

[peskywinnets]

We used to use a Time Domain Reflectometer to ensure the taps on 10mbps thick wire ethernet cable were sound...each 'bee sting' (tap) into the main carrier cable would reflect back a bit of energy ...each one looking like a spike on the display - so long as all the taps showed up a spikes...evenly spaced, we knew the taps were sound.

It used a pulse of energy down the copper & waited for the reflection (then calculated how far away the break was). The same TDR could also be used to detect physical breaks in copper cable - I'm not so sure that such technology could be used with pickups as the TDR doesn't like sharp bends & pickups have those in spades!

[David Schwab]

Interesting idea.

I think you can check the inductance to look for shorts and stuff. I don't own an LCR meter yet, but I think that's a common use for them.

Perhaps someone will chime in that uses them.

[SkinnyWire]

It's an interesting topic for discussion, but I'd agree with a previous statement alluding to it's marginal use with pickups. For the most part, you're going to repair breaks between the hookup and magnet wire. If you can find on within a few 10s of turns, maybe. Deep in the coil? You're probably just going to cut and rewind.

Maybe it would be a useful diagnostic, if it works with pickups, that you can use to advise for repair or rewind?

[chevalij]

It used a pulse of energy down the copper & waited for the reflection (then calculated how far away the break was). The same TDR could also be used to detect physical breaks in copper cable - I'm not so sure that such technology could be used with pickups as the TDR doesn't like sharp bends & pickups have those in spades!

I was shown how to use one to tell the length of wire on a spool. It may be wrong, but I was told that it was basically useless on spools less than 2" in dia. as it would see the curve as a kink an give a false reading.

[Possum]

...

The only thing that will tell if you have a short in a coil is the Extech LCR meter. It won't tell you where it is just that something is wrong by unusually high AC resistance and only that reading. I've had coils that DC resistance read correctly or close to correct and yet the AC resistance was nearly double what it should be. I do know there is some kind of meter I think for the autmotive industry that will spot shorts in coils, I've seen them on Ebay rarely and think maybe they are an older technology....

[Joe Gwinn]

The only thing that will tell if you have a short in a coil is the Extech LCR meter. It won't tell you where it is just that something is wrong by unusually high AC resistance and only that reading. I've had coils that DC resistance read correctly or close to correct and yet the AC resistance was nearly double what it should be. I do know there is some kind of meter I think for the autmotive industry that will spot shorts in coils, I've seen them on Ebay rarely and think maybe they are an older technology....

Growler.

[David Schwab]

Growler.

Dave might get grumpy, but you don't have to resort to calling him names.

[SJE]

It used a pulse of energy down the copper & waited for the reflection (then calculated how far away the break was). The same TDR could also be used to detect physical breaks in copper cable - I'm not so sure that such technology could be used with pickups as the TDR doesn't like sharp bends & pickups have those in spades!

SELTs uses reflection in exactly the same way and as you say sharp bends ain't it's strong point!

Also since SELTs is usually used for AWG28 and bigger (solid copper cores) I'd be even more surprised if it could be used for AWG42 and beyond.

[peskywinnets]

Growler.

How does that expression go again?....

"The US & UK .....two countries separated by a common language!"

The word has a different meaning over here & the first 20 seconds of this video should give you an idea....

http://www.tinyurl.com/yhm5nqa

[salvarsan]

How does that expression go again?....

*"The US & UK .....two countries separated by a common language!"

Oi.
In the New World, I believe winnet is called 'dingleberry' (or its plural).

The etymology of the Mexican term 'pendejo' is closely related and similarly located.
In other news, the New World use of 'root' as a verb applies to hogs digging for food and has long historical precedent. *Down Under in the even newer world, 'root' as a verb is* equivalent to 'shag'.
I know of no slang for the term Time Domain Reflectimetry.


Ain't them Innernets great?
-drh

[Possum]

....

I think Growler is a brand name? ............rrrrrrrrrrrrrrrrraawwwwwwwwwwwwwRRRRRGGGGHHH HHH! Sorry, I had a burrito for lunch

[Joe Gwinn]

How does that expression go again?....

"The US & UK .....two countries separated by a common language!"

The word has a different meaning over here & the first 20 seconds of this video should give you an idea....

YouTube - Bo' Selecta! - Series 3 - Lorraine Kelly's Bloopers {Part 1}

My oh my. Definitely not an American English idiom.

Growler (electrical device) - Wikipedia, the free encyclopedia

[Stealth]

I think Growler is a brand name?

Also, an F/A-18G electronic warfare variant.

Anyway - it's true, SELT doesn't react well to sharp bends as it screws up the measurements - my only gripe is that I can't borrow a Fluke from anyplace.