Hello,

I would like to know a method about how to do it but without opening or damaging the cable, that have, of course with a coating of insulated plastic material.

Is it possible ? AFAIK, NCV only works with AC, not for audio cables.

The objective is to break and resoldier the cable at that exact point.

Regards,

Roger

 

You can try measuring the resistance then trying to bend the wire at certain places and see if you see a change in resistance. Sometimes the wires will touch momentarily when bent, then break again.
You can also look for wear at some point along the cable.
With some cables you can use a sharp sewing needle to pierce the insulation, but with a shielded audio cable that might be too hard to do because the shield surrounds the inner wire(s). This would probably only work if the shield itself was bad.

If not anything else, cut the wire into 2 halves, determine which half is bad. Then cut that into 2 halves, determine which half is bad. Then cut that into 2 halves, determine which half is bad. Eventually you will end up with maybe a 1 foot piece that is bad and the rest good. You then have to solder all the pieces back together again.
If you had an 8 foot cable then you would end up with two 4 foot cables one bad one good. Cutting the bad half in half would give you 2 feet good and 2 feet bad. Cutting the bad part in half would give you 1 foot good and 1 foot bad. Soldering the pieces you would have one 4 ft one 2ft and one 1 foot good, which would reduce the total length to 7 feet. If you really need another 6 inches then cut that last foot in half and solder the good 6 inch piece to the 7 foot piece giving you 7.5 feet.

Of course the other thing you can do is buy a new cable

 

Try a capacitance meter, between core and screen.
The capacitance from each end to the break should be proportional to the distance to the break.

 

Are you sure the break isn't in one of the connectors?

 

A time-domain reflectometer could detect the position of the fault. Much more expensive than a new cable though.

 

A time-domain reflectometer could detect the position of the fault. Much more expensive than a new cable though.

It is not that expensive if you know how to do it.
You send a fast rising edge or very narrow pulse down the cable and look for the reflection at the sending end on a fast oscilloscope. Do the same from the other end and find the percentage from the sum of the two delay times.

 

Well if we want to get into expensive equipment why not X ray tomography

 

Are you sure the break isn't in one of the connectors?

Or at the end of the connector strain relief, at the point the cable can start to flex freely again?

 

Time Domain Reflectometry isn't at all expensive....if you know where to shop! This is a find from a couple weeks ago.

 

I built a TDR setup with a function generator and an oscilloscope. Sent a square wave through the cable and looked for the step reflection using the scope. The step delay depends on the length of the cable. I tested it out on several different length known good cables and came up with a spreadsheet from which I used the trendline to find where the break was on the cable I needed to test. This got me within a foot of the break.

I also located buried cable in a similar fashion. I used an AM radio to locate the cable that I'd connected to my function generator.

 

Time Domain Reflectometry isn't at all expensive.... if you know where to shop!

I've not seen that linked TDR kit available at less than ~$1k. So just a tad more expensive than a new audio cable (unless, of course, this cable is an oxygen-free palladium-plated gold-cored audiophile-beloved type)?

 

I've not seen that linked TDR kit available at less than ~$1k. So just a tad more expensive than a new audio cable (unless, of course, this cable is an oxygen-free palladium-plated gold-cored audiophile-beloved type)?

TDR wouldn’t work on that sort of cable because it is perfect: 100% of the power is transferred from the source, so there could be nothing left to reflect.

 

A time-domain reflectometer could detect the position of the fault. Much more expensive than a new cable though.

It is not that expensive if you know how to do it.

TV station night shift: Young, Restless, and Bored.

Back in the 70's I built one out of a Fast Schottky TTL hex inverter. In a bundle of video cables 250 feet long in wiring trays suspended in the space above an acoustic tile drop ceiling, one cable caused a small reflection in the signal. IOW, a "straight'" piece of cable caused a ghost. We theorized that there was a kink in the cable. The circuit (plus a 500 MHz scope) narrowed the location to withing around 2 feet, good enough to find in on the first try.

ak

 

I use a different method, which is to apply a voltage to the two ends of the broke conductor, and raise the voltage until there is an arc across the break. That arc burns a small hole and so there is a short bit of the cable damaged. It is best done outside where the cable can be stretched out and watched.
I have used it on power cables and "RG#" type coaxial cables, but not on audio cables, But it should work the same.
Extreme caution is mandatory because the high voltage is dangerous.
The method is cheap, fast, and accurate, but rather crude.

 

TV station night shift: Young, Restless, and Bored.

Back in the 70's I built one out of a Fast Schottky TTL hex inverter. In a bundle of video cables 250 feet long in wiring trays suspended in the space above an acoustic tile drop ceiling, one cable caused a small reflection in the signal. IOW, a "straight'" piece of cable caused a ghost. We theorized that there was a kink in the cable. The circuit (plus a 500 MHz scope) narrowed the location to withing around 2 feet, good enough to find in on the first try.

ak

During the 1970's I was examining crosstalk on 18 inches of PCB. I used a (I think it was a Tektronix 475A??) oscilloscope and generated the pulses (a few nSec wide) by avalanching a couple of ordinary small signal transistors. (Generally it involves applying excessive collector voltage to a transistor with maybe 10 ohms base-emitter; breakdown causes a very fast rising edge. That edge then triggered another transistor to produce the falling edge. This design is not recommended for long life of the transistor but I do not remember that transistor failure was a problem for the test circuit.) It would take me some time to figure out the pulse generator design which I did almost 50 years ago, but perhaps there are some examples posted around. Certainly the pulse generator is cheap, and if you have the oscilloscope then the whole effort would be cheap. Otherwise I like the proportional capacitance method mentioned above.

 

This engineer published the video explaining how yo do it.

 

Certainly the TDR method is valid, and IF the time measurement is accurate enough, and IF the propagation velocity is uniform in the cable, and IF the measurement is done from both ends, then probably the location of the open can be quite close. And in a long cable that can work rather well, if there is no other choice, and close is good enough.
The "arc-across-the-open" method is exact and does not require an oscilloscope or a fast pulse generator.
I realize that the use of a higher voltage certainly requires quite a bit of caution and careful attention to what one is doing. But the time that the high voltage is present will be less than a minute, so it is not such a long time to focus one's attention on what one is doing.
AND the safety procedures for working with the high voltage using this scheme are neither complex nor confusing.