Whats the best technique to use to locate a sheath fault for a cable thats put inside of a conduit. The Insulation resistance between the metal screen and earth fails and reads a low value of 200kOhm to 500kOhm. The cable not being in direct contact with earth ground soil makes it difficult to locate using conventional fault locating methods, therefore does anybody have any experience with this special situation?
Is the resistance voltage independent ? It may not be a single point fault but a long length of compromised insulation with damp ingress.
The problem is that the resistance is very high compared to the transmission line impedance of the wire and the conduit acting as a coax, so TDR may well struggle.
Sometimes DC testing can be done. If you can use a significant current to induce a known voltage gradient down the conduit (or the armour) and then isolate the cable shield (or the conduit) at both ends and measure the voltage to which it drifts (some fraction of the conduit voltage like 50% hopefully) You may be able to use one of the cable cores to create or measure the far end voltage. (Varley method and Murray method come to mind as the official names - not sure off-hand which is which, but the same idea just where you inject and where you measure varies. .)
Such tricks can be a bit error prone if for example there are parallel paths or buried joints and steps in size so the resistance per unit length is not constant. A 'floating' supply is needed, not one grounded on one side...
Is the resistance voltage independent ? It may not be a single point fault but a long length of compromised insulation with damp ingress.
The problem is that the resistance is very high compared to the transmission line impedance of the wire and the conduit acting as a coax, so TDR may well struggle.
Sometimes DC testing can be done. If you can use a significant current to induce a known voltage gradient down the conduit (or the armour) and then isolate the cable shield (or the conduit) at both ends and measure the voltage to which it drifts (some fraction of the conduit voltage like 50% hopefully) You may be able to use one of the cable cores to create or measure the far end voltage. (Varley method and Murray method come to mind as the official names - not sure off-hand which is which, but the same idea just where you inject and where you measure varies. .)
Such tricks can be a bit error prone if for example there are parallel paths or buried joints and steps in size so the resistance per unit length is not constant. A 'floating' supply is needed, not one grounded on one side...
