Cable hi-pot withstanding time, why 15 minutes?

[madMAX]

nobody said anything about doing what you feel. A gang test is a perfectly viable option unless specs say otherwise.

IEEE NETA ANSI and other standards exist for a reason. The 15 minutes has a purpose. Just test according to the spec and record your results. If your just looking to do whatever you feel like then what's the point of standards and procedures.

[madMAX]

LMAO, so true.

You mean I can't just wave that NETA card over a broken breaker and it won't fix itself????

[ARNOLD ZIFFEL]

I grew up in the utility world and was taught to "gang test". Nothing in NETA specs ever told me to test individual phases only. Acceptance tests are pass/fail only. As per NETA-

"If no evidence of distress or insulation failure is observed by the end of the total time of voltage
application during the test, the test specimen is considered to have passed the test."


When contracted to perform Tan Delta VLF testing as a diagnostic test, of course you test individual conductors to assess splice, termination, and overall insulation condition when compared to the associated conductors.

And in case you are wondering I passed Level IV and did not have to resort to cheating.

[ArchIsChompa]

tie all three cables together, do a gang test, and roll out early.

Definitely this. Saves a lot of time, granted one of your cables isn't testing bad.

[benlanz]

Definitely this. Saves a lot of time, granted one of your cables isn't testing bad.

Assuming we are talking about solid dielectric cable systems, you should test the phases independently with a insulation resistance test. Since the actual resistance value has little meaning the best you can do on a 3 phase system is to compare the phases. I recommend a measurement after stable voltage or <1 minute. A VLF withstand test can be performed with all three phases connected since there is no insulation performance measurement.

The major issue I have with this discussion, is that we are splitting hairs. An insulation resistance and VLF tests typically detect less than few percent of issues that do not meet the minimum performance requirements as per the cable and accessory manufacturers' QC specification. Just keep in mind that, while you can be reasonably confident that you will defect a failure (insulation rupture/short) in the cable system, these test are nearly helpless to find the vast majority of cable reliability issues. If you really need to determine a cable system's reliability, you are going to need a completely different type of test.


Ben Lanz
Past Chair IEEE 400

[lovetacycle]

IEEE 400 is a go, no-go test. It states, "The system is required to withstand the specified voltage for the specified time duration. This test will normally reveal gross imperfections due to improper field handling, such as excessive bending or airgaps between the insulation and shield interfaces. While performing a high-voltage dc test, it is common to monitor insulation leakage current and/or insulation resistance. The guide notes that a decrease in current with time is generally a practical criterion for acceptance." 15 minutes was chosen to make rule out improper installation, air gaps, etc.. It was deemed that in this amount of time a problem will manifest itself.

[ies0900012]

Hi-potential can be performed on new cables, before it can be commissioned or put online, right. Is it still practical and or advisable to subject an old or in operation power cable to hi-pot testing? Will it not degrade or shorten it's life span?

[BrickSalad]

With regards to VLF hipot, I've got an excerpt from my company's literature (paraphrased of course) that might bear some light on the 15 minutes requirement:

The results of field tests on over 15,000 XLPE cable circuits show that 68% of recorded failures happened within 12 minutes, 89% within 30 minutes, 95% within 45 minutes, and 100% after an hour. (Moh[B17])

(I'm guessing here that these tests were run for an hour, hence the 100%.)

So it actually makes sense to test a cable *longer* than 15 minutes if you really want to catch all of the cable faults. There is a balance between practicality and perfection however, and it seems like most companies go for 15 minutes. Detects 3/4 of defective cables without taking an unreasonably long amount of time. We sometimes run 1/2 hour or even full hour tests on cables at my company, but usually the customer only wants the standard 15 minute test.

Hope that helps!

[VallyOrent]

Hi everyone..i am a new user here. As per my knowledge its true that you should wait the 15 minutes and also you should plot the readings that you take. I think you will observe the trend of the results, it will reduce quickly in the beginning and then start to level out at the end of the 15min period.

[epzjamin]

15 minutes because some faults wont show right away and take some time under load