Hello,

I just got a Megger TDR2000/3 and have read about a million documents on interpreting the wave analysis to identify an open, short, split, etc on the cable under study but I cant find for the life of me on how to actually connect the device??

It has a terminal with 2 crocodile plugs (Red and Black). Where do I connect these? if i got a cable do i connect one to one end and the other to the other end? what if the cable is 900m long and i dont have acces to both ends?

Example: I suspect a earth fault in one of the three phase cables coming from an MCB and finishing at a motor. The cable is 500m long and underground. Where do I connect the plugs to analyse the red, white and blue phases?

Thanks!!!!!

 

you would connect it between two different conductors in your cable. The high voltage would then be detecting leaks, or faults in the insulation between those two conductors. In your case one of the conductors is the 'earth' or ground and you are testing insulation between the conductors and earth.

 

It has a terminal with 2 crocodile plugs (Red and Black). Where do I connect these? if i got a cable do i connect one to one end and the other to the other end? what if the cable is 900m long and i dont have acces to both ends?

You don't need access to both ends, that's the beauty of a TDR. You attach the leads to both conductors where you have access, e.g. for coax, red to the center conductor and black to the shield. The TDR will send a pulse down the wire and you analyze the reflection to determine what the fault is and where it's located.

 

Having never used a TDR...do you need a reference cable of the same type to compare for the length to the fault?

Ken

 

Having never used a TDR...do you need a reference cable of the same type to compare for the length to the fault?

No. The characteristic impedance and the velocity of propagation are used to determine fault type and location.

 

No. The characteristic impedance and the velocity of propagation are used to determine fault type and location.

Is that determined by the TDR instrument, or set by the operator?

Ken

 

Go to www.electronicdesign.com and navigate to "Ideas for design", search for "Cable Radar". A great article (from 1998) on the subject. I built and use(d) the circuit. It works amazingly well (and yes Virginia, there is still a use for those old 74 series ICs!). E

 

Is that determined by the TDR instrument, or set by the operator?

The operator sets the width of the pulse, which needs to be set appropriately for the approximate distance to the fault. The input pulse and the reflection are viewed on the screen. You have to look up the velocity of propagation for the cable being tested.

A scope and pulse generator are all you need to make one. I've been trying to get a plug-in for my Tek scope, but they're not readily available (at a good price). I've been meaning to build one using some 74AC14. You can't use any old inverter and I saw a DIY using the 74AC14.

 

The operator sets the width of the pulse, which needs to be set appropriately for the approximate distance to the fault. The input pulse and the reflection are viewed on the screen. You have to look up the velocity of propagation for the cable being tested.

A scope and pulse generator are all you need to make one. I've been trying to get a plug-in for my Tek scope, but they're not readily available (at a good price). I've been meaning to build one using some 74AC14. You can't use any old inverter and I saw a DIY using the 74AC14.

Is there a table somewhere for the OP to find the velocity factor for 3-phase power wire, assuming through a metal conduit (or maybe not)?

Ken

 

Is there a table somewhere for the OP to find the velocity factor for 3-phase power wire, assuming through a metal conduit (or maybe not)?

Probably...

TDR assumes a uniform characteristic impedance. If you know the dielectric constant of the cable, you can calculate propagation velocity.

Here's a link I found from a wire manufacturer: http://www.picwire.com/technical/velocity_factor.php

 

Probably...

TDR assumes a uniform characteristic impedance. If you know the dielectric constant of the cable, you can calculate propagation velocity.

Here's a link I found from a wire manufacturer: http://www.picwire.com/technical/velocity_factor.php

Those are nice uniform shielded cables. The OP is dealing with 3-phase power cable, that is likely just 3 or 4 X-gauge wires pulled through a conduit.

Ken

 

Those are nice uniform shielded cables. The OP is dealing with 3-phase power cable, that is likely just 3 or 4 X-gauge wires pulled through a conduit.

That's going to be a problem for him because TDR's don't work well if impedance isn't uniform.

 

That's going to be a problem for him because TDR's don't work well if impedance isn't uniform.

That's what I was wondering.

Ken

 

Thanks guys! SO if I was to test my motor 3-phase cable, I would disconnected the end connected to the motor and test the W wire using B as reference, then R using B as reference, then B using R or W as reference? How do I know if the wire with the short to ground is the cable been tested or on the reference cable? (ie. Connect red plug to White wire (cable under test), Black plug to blue wire, but blue has a short to ground)

 

You need to test between all combinations of pairs of conductors, including ground or shield. Any unterminated pairs of conductors with no fault should give a similar response. Any pair with a short will give a pulse inverted from the open cases. Your unit appears to be calibrated in meters, so either you enter VOP or it sets on its own. I couldn't download the manual without creating an account. An accurate VOP is necessary to obtain an accurate distance.

 

You need to test between all combinations of pairs of conductors, including ground or shield. Any unterminated pairs of conductors with no fault should give a similar response. Any pair with a short will give a pulse inverted from the open cases. Your unit appears to be calibrated in meters, so either you enter VOP or it sets on its own. I couldn't download the manual without creating an account. An accurate VOP is necessary to obtain an accurate distance.

Thnaks dl324! I do need to set the VOP of the cable, the rest is done automatically (cable impedance and gain). The problem I had was one that phase A of a 3-phase cable connecting a motor had been cut and had been touching a metal cable tray, hence causing a ground fault. It took us a while to find the fault as we did not have a TDR atm. If I was to go and try to find this fault now and did pair of phase A&B, B&C, C&A, C&B would I have been able to find this fault? or would I have had to wire the phase A and the actual metal tray to find the short (see picture attached)? What would I have seen if i did phase A&C?

Thanks!!

 

The problem I had was one that phase A of a 3-phase cable connecting a motor had been cut and had been touching a metal cable tray, hence causing a ground fault. It took us a while to find the fault as we did not have a TDR atm. If I was to go and try to find this fault now and did pair of phase A&B, B&C, C&A, C&B would I have been able to find this fault? or would I have had to wire the phase A and the actual metal tray to find the short (see picture attached)? What would I have seen if i did phase A&C?

You would have had to also check each wire to the cable tray because that's where the discontinuity was. It's probably good that you found it before obtaining a TDR; only to find out that it's performance was unsatisfactory on cable that didn't have a uniform characteristic impedance.