Hi-potential testing on HV cables 5

[Deansharafi]

Hi All,
I know that Hi-pot testing on HV paper insulated and screened cables is no longer recommended in the industry, but I still see it on some standards (ex. Australian Standard). What is the case for you who live in Europe or the US. Do you still do this test on HV cables?
For XLPE cables it is pretty clear that no Hi-pot testing must be carried out, but when it comes to paper insulated cables, it is still a matter of ambiguity.
I think imposing a pressure of 3 times rated voltage to a cable can be destructive itself, and it is a good reason why it shouldn't be done, however I still see it in the standards.
Your input is appreciated.

 

[jghrist]

There is no problem with dc hi-pot tests on paper insulated cable or with tests on new solid dielectric cable. The problem that has been identified with testing aged solid dielectric cable is that the reversal of the electric field during a discharge (when the cable fails the test at one point) may increase the size of existing insulation voids and water trees on other parts of the cable.

 

[busbar]

From IEEE Std 400: DC testing has been accepted for many years as the standard field method for performing high-voltage tests on cable insulation systems. Whenever dc testing is performed, full consideration should be given to the fact that steady-state direct voltage creates within the insulation systems an electrical field determined by the geometry and conductance of the insulation, whereas under service conditions, alternating voltage creates an electric field determined chiefly by the geometry and dielectric constant (or capacitance) of the insulation. Under ideal, homogeneously uniform insulation conditions, the mathematical formulas governing the steady-state stress distribution within the cable insulation are of the same form for dc and for ac, resulting in comparable relative values; however, should the cable insulation contain defects in which either the conductivity or the dielectric constant assume values significantly different from those in the bulk of the insulation, the electric stress distribution obtained with direct voltage will no longer correspond to that obtained with alternating voltage.

 

[BJC]

I don't think I would hypot another cable. To much stress on an old cables. Mostly it's good for deciding when your cable fail ( during the test -not at some inoportune time) I am recommending the PArtial Discharge test to clients. The problem with it is only one company is doing it.
One advantage is the cables don't have to be de-energized for the test. Here's an artical, You can find the name of the company that's doing it and I'm sure they send a salesperson to see you.

http://ecmweb.com/ar/electric_putting_hipot_pasture/index.htm

 

[jbartos]

Suggestion: References:
A.
The 2003 NETA Acceptance Testing Specifications
7. INSPECTION AND TEST PROCEDURES
7.3.3 Cables, Medium- and High-Voltage (cont.)
2. Electrical Tests
B.
ANSI/IEEE 48, IEEE 386
Insure that the maximum test voltage does not exceed the limits for terminators specified therein, or in manufacturer’s specifications.

Notice that """3 times rated voltage""" appears to be too high for installed HV cables.

 

[davidbeach]

jbartos, if you are going to quote the NETA Acceptance Testing Specifications (as opposed to the NETA Maintenance Testing Specifications), your statement about "3 times rated voltage" is not supported for most cables. The Acceptance Testing DC voltages are all either more than 3 times, or only a few volts less than 3 time. The AC Acceptance Testing voltages are 3 times rated through 15kV cable then drops to about 2 times rated. Maintenance testing would use lower voltages for both.

 

[gordonl]

Many companies have stoped DC Hi-Pot testing of aged dielectric insulated cables, the southwire site has some info on the problems with doing so.

See EPRI report TR-101245 "Effect of DC Testing on XLPE Cables"

AEIC CS5 and CS6 state "after installation testing , DC testing is not recommended"
(Most manufacturers say the same thing)

IEEE 400 notes that "DC testing of may be detrimental to cables that have been subjected to long periods of exposure to moisture"

In Canada I know the province of Quebec still requires DC hipot testing for worker safety before re-energization.

This applies to extruded cables in wet enviroments. AC low frequency hipot test units are becoming available and the draft IEEE 400.2 covers test levels and durations.

 

[jbartos]

Comment on davidbeach (Electrical) Feb 22, 2004 marked ///\\jbartos, if you are going to quote the NETA Acceptance Testing Specifications (as opposed to the NETA Maintenance Testing Specifications), your statement about "3 times rated voltage" is not supported for most cables. The Acceptance Testing DC voltages are all either more than 3 times, or only a few volts less than 3 time. The AC Acceptance Testing voltages are 3 times rated through 15kV cable then drops to about 2 times rated.
///Please, notice that the High Voltage (HV) definition includes voltages:
1. Greater than 100kV and equal or less than 230kV for a Class of Nominal System Voltages
IEEE Std 100-2000 Dictionary...
2. High Voltage Aluminum Sheathed Power Cables are for maximum phase to phase rms AC voltage above 72500V to 242000V.
IEEE Std 100-2000 Dictionary...
3. Etc.\\ Maintenance testing would use lower voltages for both.

 

[Alwaysexciter]

I have had bad experiences testing older paper insulated lead jacketed cables in the past. I had one job where the customer's engineer insisted on a 3X DC hipot on all his 15+ yr old feeders. Needless to say several of them failed and cost him an extensive amount of money. Without hipot testing these cables may have lasted another 15+ yrs.

Remember, A 15 kV rated cable on a 13.2 kV system will only normally be stressed to about 7200 volts to ground (Wye System)

The problem with the paper seemed to be that when it failed, it failed rapidly with little warning in the way of a gradual increase in leakage current. This makes it difficult for the technician to stop the test in time to prevent irreversible damage from occurring.

Since then I have strongly recommended only testing older paper insulated cables to a max. DC of 1.41x rating. (Equivalent to AC peak to Peak)

If using a 25HZ AC tester I wouldn't go above cable rating if the paper insulated cable is over 10 yrs old.

This is only my preference and is not in any spec that I know of.

 

[jbartos]

Suggestion: The cable manufacturer recommendation for the existing older cable might be followed since the cable manufacturer knows the cable structure and quality best.

 

[benlanz]

As others have said the old DC highpot test on aged cables is a no no. Actually low levels of DC, like a megger at <1Uo, for a short period of time (less than 5min) are harmless. On PILC and new extrudes cables it is fine, but what are you going to learn? The cable can hold DC at xkV for x amount of time. Great! I can cut half way throught the insulation of a XLPE cable and it will pass the DC test, yet turn the AC on and it will rapidly fail. What you need to do is perform the exact test that is done in the cable factory when they look for defects. 60Hz partial discharge test. For more information please see the website

http://www.imcorptech.com.

 

[busbar]

I disagree with benlanz on the value of DC overpotential testing for newly installed medium-voltage extruded-dielectric cables. It is a well demonstrated, very good way to expose substandard splice/termination {workmanship} quality. I has been known to put testers and splicers at odds on the jobsite.

Changes in DC-microampere-range leakage current in are very revealing. AC-milliampere leakage currents are almost useless.

 

[DanDel]

Hipot Acceptance testing is definitely warranted to verify any stress cones and splices, but as for Maintenance testing, use PD or even a PI test with a megger.
There has been too many abrupt failures of used cables during hipot testing to risk any cable already in service unless there is a backup cable already pulled in or the circuit can be left de-energized without a serious equipment power outage while a new cable is installed.

 

[alehman]

See thread238-86054

 

[benlanz]

Actually there is no serious disagreement with the busbar chap! He is correct in that if there is a large amount of conduction in an accessory at DC test with a sensitive ammeter (pA, nA, uA) will detect the defect. Unfortunately many DC tests are performed as a ‘withstand’ test rather than a diagnostic where the cable’s response to the voltage stress is measured (e.g. busbar’s suggestion). Again unfortunately most modern constructions of accessories do not reveal high levels of conduction when installed incorrectly. The vast majority of defects in accessories (especially solid dielectrics) do emit partial discharge under stress. Refer to EPRI’s “Estimation of Future Performance of Solid Dielectric Cable Accessories” Report number 0420002. Partial discharge can not only be detected, but it can be located with modern equipment. Thus is it one of the best performance indicators. Please refer to IEEE specifications.

IEEE 404 for joints
IEEE 48 for terminations
IEEE 386 for separable connectors.

There are many diagnostic methods that might give an indication that something is wrong with a cable accessory, but in a new installation what does this mean? Is one going to replace the entire cable because someone installed one accessory incorrectly? A 60Hz partial discharge test will locate the defect and with IEEE specifications you can be confident with conservative judgment.

 

[jghrist]

Hi-pot tests on new MV cable installations will reveal gross installation defects by causing a failure at the defect. Hi-pot tests can be done by most contractors experienced in UG distribution work.

While I agree that partial discharge testing is a desirable installation test, is it cost-effective? How much would it cost to contract a testing service for a test?

 

[benlanz]

What is a gross installation defect? During the EPRI project 0420002, joints where installed with thier housing sliped off the insulation shield by 1". They held 2Uo voltage stress with 8hr, 70 degrees C load cycling, and moist earth conditions for 4 months! In fact none of the severe typical workmanship defects failed during the study! Many defects take years to fail. This is long after the contractor warranty is up. I can refer to many other cases, but I think you understand the point.

Partial discharge testing can be very cost effective in comparsion to the investiment of new cable. When compared to hipot testing, well we are comparing apples and oranges.

 

[electricpete]

Can someone explain more about partial discharge testing of cables?

Does it require installation of a permanent pd-monitoring capacitor as is done for motors?

Or is it a temporary TVA-type probe used to scan near the cable? If so sounds labor intensive and limited in scope.

Does anyone other than cable oem's use pd monitoring for cables?

How much usefulness do Iris motor pd capacitors have in monitoring pd in the power cable feeding the motor?

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[alehman]

PD testing can be done on-line:

http://www.electro-test.com/Literature PDFs/eti Cable Services - Partial Discharge 2003-2.pdf

or with permanent equipment:

http://www.eatonelectrical.com/unsecure/cms1/BR02713001E.PDF

http://www.eatonelectrical.com/unsecure/cms1/TD02718003E.PDF

http://www.eatonelectrical.com/unsecure/cms1/TD02718001E.PDF

 

[benlanz]

On-line PD testing is your only choice while you can't shut down a process. When you can shut down a process, during a plant shutdown every 3 to 5 years, it is good to do thorough off-line PD test. Have you every heard the old adage, you don't get anything for nothing?

On-line testing has its limitations because:
-it is an uncalibrated test
-only 3% or fewer defects in cable insulation appear at operating voltage or below.

Calibration is the basis of every cable factory base line test. The only base line your cable is likely to have!! If a test is not calibrated the test system has know idea what size of signal it is detecting or how much attenuation the signal has experienced. An uncalibrated system might tell you that no signal is detected and that nothing is wrong when the true is the signal is so attenuated that it cannot reach the sensor! If your detection system cannot detect 5pC and is not calibrated to the cable under test (the basis of all IEEE specifications) the value of the test is limited.

On 2500km of extruded cable tested only 3% of 960 defects were found in the cable insulation. Why you ask? The explanation is intuitive. One question, How long does extruded material last under constant PD conditions (in an electrical tree –the most common failure causing defect in extruded cable). The answer is not an exact number of course, but most experience shows typically minutes, maybe hours, definitely not more than a week. What is the likelihood of an on-line test with a test time of 15minutes to detect a defect? Very small (3%??).

If we look to the medical world for an analogy we can look to the EKG test vs. the Cardiovascular (CV) Stress test . The EKG measures the CV system at rest while you lay on a table. The accuracy of this test is about 50%. If you take the same patient and put them on a treadmill and put them through a stress routine the accuracy is 70%. Why? The answer is simple the CV stress test put the patient through the transient situations that our CV system experiences every day!

If you want to thoroughly test your cable, use an off-line test that can present the cable the tranisient situations that it is going to experience during is life time. If you can risk just testing the accessories you can go with an on-line test, but know the limitations of any technology you decide to use.

Accessories can typically handle PD at operating level for months, if not years, according to one of the latest 3 year EPRI studies (“Estimation of Future Performance of Solid Dielectric Cable Accessories” 2003)

If you just want to know what is going on with your cable accessories and you have an access to each of the accessories, On-line test will tell you something about the integrity of each joint or termination.

I see by some of your reference that some of you are doing some research! I think that is fantastic. There is so much confusion in the cable testing industry right now we need people to get to the truth. Some of you have recommended reading articles that are written by a few of my colleague who were speaking of technology that was state of the art 15 years ago! Others of you have referenced a 1 year old copy of the DTE Cable|wise on-line technology that is being sold as being 98 percent accurate! Reportedly this is on the basis of a few laboratory experiments performed in an EMC shielded room. Someone is going to get a black eye with that pitch. If you want answers look to well documented, statically significant field trials controlled by third parties. If any of you are interested in such studies I recommend you contact me directly ben.lanz@ieee.org. There are no magic wands or black boxes that give you a red or green light. If the technology it sounds too good to be true, it probably is.

Cheers!