How would you go about repairing a data cable that does not work anymore?
It has two end connectors and it is not cut in half or anything.

I would first cut the ends and see if the pins of the connectors and the wires inside the cable are making connection. If the connections are good, then the fault must come from the cable itself.
Thus I would cut the cable in two and check for continuity. Thus, one half of the cable should be broken, while the other half hopefully good. Then cut the bad half again in two and check again for continuity.

And I do this until I find the location of the fault. Or maybe if I am satisfied with a shorter cable, I would just use the good half, maybe the good half plus the other good half and stop the cutting.

Is there a better method?

 

hi,
Cables usually fail close to the connectors.
Remove the cable connector clamp, unscrew the connector cover.
Inspect the solder pins for any broken connections.
LIGHTLY pull on each wire in turn, if the wire is broken close to the connector, the pulled wire will come out.
Repeat for both ends.
Which type of cable is it.?
Good Luck.
E

 

post 2 is great for broken connections

the cable you should be able to check for continuity with a multimeter

my question - is it worth spending the time to repair it?

data patch cables - cat4/5/6/sata/esata/hdmi/ribbon etc are cheaper to buy new than to repair.

from my experience, cables are not the issue, sockets are normally at fault, bent pins, dry/cold joints/oxidised contacts

 

Is there a better method?

There are ways to inject an oscillating signal and search for where it stops radiating. Problem: Data cable isn't rated for 120 VAC so the cheap electricians pen style detector is a risky method. You have already defined the, "half interval search", but the ends are almost always the culprit. I have seen exactly one cable that was broken in the middle in 40 years of doing this kind of repairs.

 

Is there a better method?

A Time Domain Reflectometer could indicate where the fault is, but you need to know the velocity of propagation for the cable.

 

hi,
Cables usually fail close to the connectors.
Remove the cable connector clamp, unscrew the connector cover.
Inspect the solder pins for any broken connections.
LIGHTLY pull on each wire in turn, if the wire is broken close to the connector, the pulled wire will come out.
Repeat for both ends.
Which type of cable is it.?
Good Luck.
E

Some cable used in automotive engineering when testing ECUs through HIL simulation. I do not know its particular name. I don't think it has one. One end is a D-Sub connector, while the other end is a Lemo connector.

Good method. i did not think about pulling out the wires.

 

post 2 is great for broken connections

the cable you should be able to check for continuity with a multimeter

my question - is it worth spending the time to repair it?

data patch cables - cat4/5/6/sata/esata/hdmi/ribbon etc are cheaper to buy new than to repair.

from my experience, cables are not the issue, sockets are normally at fault, bent pins, dry/cold joints/oxidised contacts

So to look first at the connectors. Thanks.
It is worth spending time, why not. It is a good cable.

 

Thanks for answers. So the connectors might be the problem with high probability.

Now the question is. Which connector from the two connectors to cut first and verify?

Maybe just try my luck.

 

Now the question is. Which connector from the two connectors to cut first and verify?

A TDR will eliminate the guesswork.

 

A TDR will eliminate the guesswork.

I don't have a TDR.

 

I don't have a TDR.

Do you have an oscilloscope?

 

A TDR will eliminate the guesswork.

I'm surprised you can buy one for less than $200, but you can.
http://www.testequipmentdepot.com/megger/tdr/cfl510f.htm

My advice is to quit agonizing and look inside an end connector.
Even if the problem is at the other end, looking twice still costs less than $200.

 

I'm surprised you can buy one for less than $200, but you can.

If you have a scope, you can make one with a suitable hex inverter, a handful of R's, C's, and misc hardware.

If the connectors aren't meant to be opened, knowing which one was bad would be an advantage.

I wish I had one 35 years ago when I had some contractors install multiple 400' runs of cable from a computer room to offices. The gorillas pulled for all they were worth and damaged a number of the ends. Fortunately, the connectors could be disassembled. But it was a hassle having to do both ends on multiple 7 conductor, 25 pair, and 40 pair cables.

 

A Time Domain Reflectometer could indicate where the fault is, but you need to know the velocity of propagation for the cable.

Since we're still beating this horse, I would mention that you can find the propagation speed by measuring one of the good wires.

 

I saw that the connectors were meant to be open. I opened them and one was bad. The pins of the connector weren't making contact with the wires inside the cable no more. But I fixed it.