Testing of medium voltage solid dielectric cable. 2

[wiretwister]

Recently while the contractor was Hi-Pot testing some newly installed 15 KV solid dielectric(EPR) cable, I noticed that he was getting high readings that I don't quite understand.
The runs of cable were 3-1/c in the same duct about 4000 ft.in length with 12 X 3 Raychem splices and Raychem terminators on both ends.
Test voltage was 56 KV (80% of factory test voltage Ref:IEEE 404)for the Raychem splices not the cable.
In over 45 years of experience, I have not witnessed a test that the readings were out of range like this.
Could it be in the splices?
Has anyone had the same problem here recently?

 

[wareagle]

What were the out of range readings? What would have been the acceptable reading?

 

[wiretwister]

The readings were in the range of 60ua after 15 min.
Normal readings for this type of cable and environmental conditions would be about 15 ua max.
Thank You for responding...

 

[PWR]

Were the measured currents in each phase nearly the same? Did the currents continue to decrease during the 15 minute hold at 56kV? Given the cable characteristics you have, what level of current do you calculate should be present? While 60 micro amps sounds high to me as well, it may be explainable as something other than poor workmanship in the field or at the factory.

 

[wiretwister]

The measured currents were all about the same.
The current did have a downward trend.(ie:1-15 min)
The cable under test had a insulating level of 133%.
What gets me there were two other complete circuits and the test results were similar.
Poor workmanship is out of the question, I personally know the contractor and his splicing workman.
Thanks PWR

 

[LWS]

We recently encountered a similar condition. Cause was dirty elbow terminators (hence leakage current). You didn't describe the type of Raychem terminators (Raychem didn't use to make elbows) BUT if you have elbows, I'd suggest removing them and leaving bare cables in the clear to repeat your hi-pot. Lastly Raychem terminators can also track excessively on the exterior. Try cleaning the exterior if its terminators you have.

As an aside, I'm on the IEEE Insulated Conductor Committee Working Groups revising standards for field testing of extruded cables (like yours). Fact is that gross defects can pass a DC Hi-pot test and its become more and more clear that it is close to a useless test. I'd recommend you procure a VLF Hi-Pot test. It's GO/NO-GO - you're not measuring anything - just a witstand test at rated test voltage - See IEEE 400.2.

 

[wiretwister]

Thanks LWS,
I'm with you in your statement that gross defects can go undetected in a DC Hi-Pot Test.
Just like the "Treeing Effect" was not discovered in cross-linked pe for several years after that kind of cable was on the market and installed all over the country.
My main concern, is that new products entering the market may exhibit higher leakage current with the antiquated test proceedures we have now in place.
wt

 

[PWR]

I feel compelled to reply to the statement by LWS that a DC hipot test is "close to useless". Leaving aside the fact that he contradicts himself by describing a problem that was detected by a DC hipot test (improper elbow installation), it is important to distinguish between maintenance testing and acceptance testing. I have advised against DC hipot testing of cables for maintenance purposes for at least 20 years and would do so irrespective of whether or not it shortens cable life due to it's effect relative to the treeing phenomenon. However, for acceptance testing, I have personally seen it detect inumerable problems which could then be corrected without the need to replace new cable. As far as it's inability to detect some types of "gross defects", please let me know when someone discovers any test procedure that does not have limitations.

 

[LWS]

My statement with regard to the DC Hi-Pot test may have been slightly extreme but not much. At the outset, let me concur with [COLOR=red yellow]PWR[/color] that any test has its limitations including the VLF Hi-Pot I suggested. I think most would agree that we’d like to use a test that has the least limitations, where possible. VLF has long proven itself in Europe, and is now gaining acceptance in the USA just because of that fact.

Given: An acceptance test SHOULD detect [highlight]gross defects.[/highlight] If it does not, what may we infer from that? I conclude that it does not meet the criteria of an acceptance test and, therefore, close to useless. That it can pick up the “grossest of defects”, to me, is not quite the same as picking up “gross defects”.

For the benefit in part of [COLOR=red yellow]wiretwister[/color] and in part to respond to [COLOR=red yellow]PWR's[/color] observations and, at the risk of being lengthy, here are a few relevant excerpts from latest IEEE Std. 400: [italic](Note: For more detailed info on VLF Testing, please consult IEEE 400.2 – IEEE Guide for Field Testing of Shielded Power Cable Systems Using Very Low Frequency (VLF))[/italic]

IEEE 400 (2001) - IEEE Guide for Field Testing and Evaluation of the Insulation of Shielded Power Cable Systems:

3. Definitions

For the purposes of this Guide, the following terms and definitions apply. IEEE 100., The Authoritative Dictionary of IEEE Standards Terms and Definitions, Seventh Edition [B30]5 should be referenced for terms not defined in this clause.

3.1 acceptance tests: A field test made after cable system installation, including terminations and joints, but before the cable system is placed in normal service. The test is intended to further detect installation damage and to show [highlight]any gross defects[/highlight] or errors in installation of other system components. (IEEE Std 48-1996. and IEEE Std 404-2000.).

4.2 Summary of direct voltage testing

DC testing has been accepted for many years as the tandard 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. As conductivity is generally influenced by temperature to a greater extent than the dielectric constant, the comparative electric stress distribution under dc and ac voltage application will be affected differently by changes in temperature or temperature distribution within the insulation. Furthermore, the failure mechanisms triggered by insulation defects vary from one type of defect to another. These failure mechanisms respond differently to the type of test voltage utilized. [highlight]For instance, if the defect is a void where the mechanism of failure under service ac conditions is most likely to be triggered by partial discharge, application of direct voltage would not produce the high partial discharge repetition rate that exists with alternating voltage. Under these conditions, dc testing would not be useful.[/highlight] However, if the defect triggers failure by a thermal mechanism, dc testing may prove to be effective. For example, dc can detect the presence of contaminants along a creepage interface.

In the case of joints and accessories, their dielectric properties may differ from that of the cable with regard to conductivity. This may result in a dc stress distribution at the interfaces between the cable and the accessory that is very different from the stress under ac voltage. A careful examination of the system is necessary prior to a dc test in order to avoid difficulties.

Testing of cables that have been service aged in a wet environment (specifically, XLPE) with dc at the currently recommended dc voltage levels (see IEEE P400.1?) may cause the cables to fail after they are returned to service (see Fisher, et al. [B23], and Stennis, et al. [B48]). The failures would not have occurred at that point in time if the cables had remained in service and not been tested with dc (see Eager, et al. [B21], and Srinivas, et al. [B47]). [COLOR=red yellow]Furthermore, from the work of Bach, et al. [B7], we know that even massive insulation defects in extruded dielectric insulation cannot be detected with dc at the recommended voltage levels.[/color]

After engineering evaluation of the effectiveness of a test voltage and the risks to the cable system, high direct voltage may be considered appropriate for a particular application. If so, dc testing has the considerable advantage of being the simplest and most convenient to use. The value of the test for diagnostic purposes is limited when applied to extruded insulations, but it has been proven to yield excellent results on laminated insulation systems.

5. Direct voltage testing

5.5 Summary of advantages and disadvantages

Some of the advantages and disadvantages of dc testing are listed below.

5.5.1 Advantages
[ul]
[li]Relatively simple and light test equipment, in comparison to ac, facilitate portability.[/li]
[li]Input power supply requirements are readily available.[/li]
[li]Extensive history of successful testing of laminated dielectric cable systems and well-established
data base.[/li]
[li]It is effective when the failure mechanism is triggered by conduction or by thermal consideration.[/lii]
[li]It is effective on interface problems of joints and terminations and surface problems of terminations.[/li]
[li]Purchase cost is generally lower than that of non-dc test equipment for comparable kilovolt output.[/li]
[/ul]
5.5.2 Disadvantages
[ul]
[li][COLOR=red yellow]It is blind to certain types of defects, such as clean voids and cuts.[/color][/li]
[li]It may not replicate the stress distribution existing with power frequency ac voltage. The stress
distribution is sensitive to temperature and temperature distribution.[/li]
[li]It may cause undesirable space charge accumulation, especially at accessory to cable insulation
interfaces.[/li]
[li]It may adversely affect future performance of water-tree-affected extruded dielectric cables.[/li]
[li]age current readings may have wide variations due to atmospheric conditions and lack of control of charges at termination lugs.[/li]
[/ul]
8.3 VLF testing with sinusoidal waveform

8.3.3 Advantages and disadvantages

8.3.3.1 Advantages
[ul]
[li]Cables are tested with an ac voltage up to three times the normal phase to ground voltage. After initiation of a partial discharge, a breakthrough channel at a cable defect develops very rapidly.[/li]
[li]Due to continuous polarity changes, dangerous space charges do not develop in the cable insulation.[/li]
[li]Test sets are transportable, and power requirements are comparable to standard cable fault locating equipment.[/li]
[li]The VLF test can be used on extruded as well as on fluid impregnated paper type cable insulations.[/li]

The VLF test with sinusoidal waveform works best when eliminating a few defects from an otherwise good cable insulation. The VLF test is used to fault the cable [highlight]defects without jeopardizing the cable system integrity.[/highlight] When a cable passes the recommended 0.1-Hz VLF test, it can be returned to service.[/li]
[li]Due to the sinusoidal regulated waveform and to the highest electrical tree growth rate as compared to the cosine-rectangular waveform, electrical trees will be initiated at a defect within minutes.[/li]
[li]The test voltage level and waveform is defined as RMS voltage and is completely independent of the cable length.
[/ul]

VLF test sets with 0.1-Hz dissipation factor (tan ) measurement capability for identifying cables with highly degraded cable insulations are available and can be used with a 0.1-Hz withstand test. This test is described in 8.4.

8.3.3.2 Disadvantages.

When testing cables with extensive water tree damage or ionization of the insulation, VLF withstand testing alone is often not conclusive. Additional tests that measure the extent of insulation losses will be necessary.

Limitations are the maximum available test voltage of 57 kV rms and the maximum capacitive load of approximately 3 mF at 0.1 Hz (30 mF at 0.01 Hz). The total charging energy of the cable has to be supplied and dissipated by the test in every electrical period. This limits the size of the cable system that can be tested. A long testing time must be seen as an inconvenience rather than as a limitation.