Single Core Armoured Cables Underground Cables' Failure

Harmonics induced by the drive?

_{wild guess.}

Thank you very much. The question is how can the harmonics cause such as a severe damage? Do you have a theory to back your reason?

If all 3 cables failed at one location, with fault to the armor, I would suspect that there was some damage done to the cable when it was installed. It passed testing at that time, and operated for a while (you didn't tell us how long ago the cable was installed), until some stress caused deterioration of the weak point in the cables to failure. It may have been only one phase that failed, but since you said your protection did not operate properly, the fault may have been severe enough to damage all 3 cables.

Take it for what it's worth - it's an educated guess without seeing the actual cables, location, etc.

Peter, Thank you very much for your contribution. The cables have been installed for 13 years. I also suspected that the cables might have sustained a damage during installation but I have a problem with this reason as how can all three get damaged - how did this damage get deteriorated with time. The three cables and the armour were sort of melted together. From the damage it does not show that there were any mechanical impact of any kind which suggests that there were no any damage during installation time.

Confusing. You write "armour was not earthed in both ends" then later "generator earth fault protection did not.. operate...because armour was not earthed".

It would seem sensible to earth armour at source (generator) end only.

Was armour earthed at VSCF only or not anywhere?

There has been a lot of correspondence in CR4 about cable failure at junction boxes where armour is bonded phase to phase - unbalanced load current causes unequal armour/sheath voltages so interconnections circulate current.

How do you know the cable is not overheated by load current? Was "temperature rise by resistance" ever done? Or temperature indicating tapes (e.g. thermindex) at intervals? You write there is a concrete trench - usually cables are put IN the Trench (in air) not, as you wrote "UNDER the trench".

Thank you very much 67Model. Not earthing the Armour at both ends was a mistake. The plan was not to earth at both ends to avoid overheating due circulating currents but to earth only one end. The earth fault protection would not see this fault as the earth fault currents would not flow through the armour. The secondary side of the transformer is connected to the VSCF but the amour is earthed on side only- the transformer side.

There was no any overload noted at any time during the operation of the machine. The trench is covered by soil but at the area where there is a movement of heavy equipment - this part is concreted to avoid damaging the cables due to heavy equipment movement.

If I had to guess, I'd say that the cable was surrounded by concrete that was either very rocky or perpetually wet (or both) and probably damaged during installation. Over time any ground motion would cause the rocks and/or water to penetrate the shield, then the dielectric strength of the insulation was comprised and it failed phase to phase.

Thank you very much RAMConsult. The strange thing about the cables being damaged during installation is that - how can all the three cables get damaged at one place. The concrete is the top part of the trench for which I find difficult to see how can this have impact on the cables to cause such a severe damage.

Is it possible that because the armour was not earthed at both ends could cause such a damage I wonder. If so how - what is the theory?

How can all three cables be damaged in one place?

I do not claim experience with cable failures but I suggest the following...

1. By not earthing armour, instead of a fault in one core causing an earth fault, it had to be a phase to phase fault - so immediately two cores are damaged, not (hopefully, it depends how fast the protection/breaker work) one.
2. You have not described the armour material and cross-section area compared to conductors, but it is likely a phase to phase fault is a bigger current - in other words, a bigger arc welder let loose on your cable.
3. The arc welder is rapidly going to damage the third cable (3 phase fault) which then contributes current - about 800 Amp full load, 2500 amp short circuit?

Any fault will begin with high localized leakage current, which leads to local heating, more leakage, till "run-away" occurs. Local heating can accelerate failure in adjacent cables.

You mention the cable is about 13 years old. This is less than the life which would be expected usually. Burying a cable is an uncertain business - the heat conductivity of soil varies from poor to very low, even compared to installation in air. Soil which is OK initially may dry out - water tables can fall, especially since parts of the world are becoming drier. It is not surprising that when a cable is critical, like for generation, people resort to cables in air in a concrete duct or back-filling a trench with material of known thermal conductivity.

I recommend you make efforts to make sure your cable is not hotter than you expect. Attach a temperature sensor to the cable at where it failed, bring it out to a box on the surface, so that you can instal a recorder or measure by hand periodically. Do a loop resistance test on the hot cables or sheaths if you can [it should be possible to use a high impedance DC current source applied to the unearthed ends which will not be affected by the AC sheath potentials] - it will not detect a hot spot, but will show if there is an overall problem. If the cable is overstressed, it may fail again soon, it is better to be prepared with cable repair kits and cable replacement cost/programme than be just reacting to faults.

A summation of how much this fault cost, total man-hours x cost per hour to find and repair the fault plus lost generation value may help to show the value of anticipating/avoiding future failures by measuring what you have.

I don't know if the area is subject to much equipment movement but poor construction and back-filling can doom your cables to failure due to a phenomena known as "treeing"...

http://en.wikipedia.org/wiki/Electrical_treeing

Occurrence and causes

Electrical treeing first occurs and propagates when a dry dielectric material is subjected to high and divergent electrical field stress over a long period of time. Electrical treeing is observed to originate at points where impurities, gas voids, mechanical defects, or conducting projections cause excessive electrical field stress within small regions of the dielectric. This can ionize gases within voids inside the bulk dielectric, creating small electrical discharges between the walls of the void. An impurity or defect may even result in the partial breakdown of the solid dielectric itself. Ultraviolet light and ozone from these partial discharges (PD) then react with the nearby dielectric, decomposing and further degrading its insulating capability. Gases are often liberated as the dielectric degrades, creating new voids and cracks. These defects further weaken the dielectric strength of the material, enhance the electrical stress, and accelerate the PD process.

Over time, a partially conductive, branching 3D tree-like figure is formed within the dielectric. The tree can grow to the point that it eventually causes complete electrical failure of the dielectric. This has been a long-term failure mechanism for buried polymer-insulated high voltage power cables.

That's the idea. GA

It only looks like all 3 punched out at the same time. With this tree phenomenon causing a runaway situation that leads to failure of one insulator, the failure of the others adjacent is caused by the first failure in a quick cascade. Only one failed for a reason and the others were cooked collaterally.

Possibilities:

1. Damage during installation

2. Insulation damge due to soil movement over a period

3. Weight resting on the cable - the concrete covering / floor

You may review your current installation & recommend improvements.

What is the (type/material) armour used?.

Dear qarbao,

You have stated that,

1) the cables were laid in trefoil manner,

2)cables are about 13 years old,

3) armoures were not earthed at both ends,

4) thecables were under the trenches which was concreted with the thickness of the concrete about 200mm.

Now, considering the cable is XLPE or PVC insulated cable, just think, the trefoil configuration might be disturbed during concreting work, resulting in residual electromagnetic fluxes are generating due to flow of load currents which might intercept the embeded iron rods inside concrete(and it might happened at the particular point i.e. 85 metre from trafo the embeded iron rods were coming closer towards the cables) which might cause continuos heating of cable insulation in addition to the normal heating due to load flow. The temperature withstand limit of XLPE cable is about 90 deg. celcius( for PVC cable it is about 70 deg. celcius ). It is estimated that for each 8 to 10 deg. celcius rise of temperature of insulation over it's designated temperature, the life of the insulation becomes half. As such due to this overheating the insulation of the cable failed and the above fault occured.

Thanks,

Manindra.

1)Check wheather you have laid the cable with a phase shift of 120 degree. if you lay the cable straight for a long distance , the harmonics in the cable will be high .

2) Since the cable has been coverred with concrete , derating factor will be high .

3) Due to high hormonics and derating factor , the heat generation will be high . that might have leaded the cable for melting .

4) Earth fault of the cable can be detected in the following condition

4a)when fault current flow through ground via armour .

4b) when there is a unbalance in the cable due to falut in a phase .

5) since you didn't connect the armour with ground the fault current doesn't have a path to flow . that is the reason why the cable has mented and failured.

Hope this will help you .

Regards ,

Sathish kumar.N

8056028522