PRODUCTS SHOWCASE

t h e m a g a z i n e f o r p o w e r , t e l e c o m a n d c a t v u t i l i t i e s

SHOWCASINGSHOWCASING

MARCH 2002

 www.utilityproducts.com

MARCH 2002

 www.utilityproducts.com

Test & Measurement

Utility Vehicle Accessories

Underground & Overhead

Distribution

Test & Measurement

Utility Vehicle Accessories

Underground & Overhead

Distribution

5kV TestingHow Much Is Enough?

5kV TestingHow Much Is Enough?

 

nally been treated as a drying operation can become a rewind job.

An insulation tester (commonly referred to as a “Megger,” although that is in fact aregistered AVO trademark) overcomes these problems. Rigorous limiting of output cur-rent by a resistor network in the test circuitmeans that under the lowered resistance as-sociated with deteriorated insulation, the testvoltage will automatically drop to non-inju-rious levels. And an insulation test, instead of a simple pass/fail indication, provides anaccurate reading of the resistance value.There can be no “near misses.” If the reading

he plethora of insulation test methodsand procedures that appear in the lit-erature and product descriptions canseem bewildering to someone tryingto establish a testing regime or meet

an application. A thorough explanation of all of them, of course, is well beyond thescope of a single article, but for a conve-nient organizing principle, one can refer tothe Law of Diminishing Returns. That isto say, no single test can tell everythingabout a particular section of in-sulation or its associated apparatus,so that additional tests may be re-quired to isolate a problem or make a determination. But the-ory must be balanced against practicality in order to avoid thewasted time and expense of mak-ing more and more tests for fewer and fewer added results. A proce-dure issued to customers by a lead-ing manufacturer of electriclocomotives, GM Electro Motive Di-vision, illustrates this concept very well.

Locomotives returned to shop for rou-tine scheduled maintenance were tradi-tionally tested by either/or 1 kV Megger‚insulation tests or high-potting, typicallyat 1080 VAC. While these tests caughtmost problems, the practice did not catchthem all. There persisted annoying and costly failures after some locomotives werereturned to service. Why?

High-potting is basically a crude test.Many rely on it because its dramatic re-sults (failure light and buzzer) tend to in-spire confidence in the test, but therein liesthe shortcoming. A high-pot makes a pass/fail determination based on a trip set-ting for either breakdown or leakage cur-rent, or both. This may be set by theoperator, or it may be factory-set by themanufacturer. It gives only an indicationof the condition of the test item (in thiscase, a locomotive) at the time of the test.It has no predictive value. If the insula-tion has a pinhole that arcs current to theframe, the high-pot indicates failure, and the problem is recognized. But if, say,

moisture ingress has seriously lowered theresistance of the insulation to a value justabove the trip setting, it’s a case of “a miss isas good as a mile to a blind horse.” The lo-comotive goes back into service, but may failafter the first heavy rain. Furthermore, be-cause a high-pot will output higher currentsand sustain an arc, it can further damage theinsulation. A failure that could have origi-

is above bare minimum but close to unac-ceptable, further cleaning and drying maywell restore it to a condition that will keepthe equipment running satisfactorily well intothe next scheduled maintenance period.

Then why did EMD still experience fail-ures in the field? Because an insulation tester is commonly thought of as a 1 kV piece of test equipment. Many experienced and ca- pable operators have used insulation testersfor years without going above a 1 kV test.This is, of course, generally adequate. Butthere are problems that a 1 kV test may notsee. Pinholes and cuts in insulation posed a big problem for EMD. Depending on their  positioning with respect to ground, a giventest voltage may or may not “pull” an arc.Remember, air is an effective insulator, sothe more voltage that is applied, the greater the separation that can be arced. Carbontracks and similar physical damage will alsoallow greater amounts of current at higher voltages. Such damaged spots may “pass”at 1 kV, but rapidly deteriorate to breakdownwhen put back into service. Does this meanthat testing must be repeated at ever higher voltages? Fortunately, there’s a practicallimit, as EMD discovered through rigorousexperimental testing.

They conducted an extensive and rigorousstudy of tests at increasing voltages from 1kV up to 5, and were able to document thateach increase in test voltage revealed addi-tional problems in the tested insulation.Fortunately, by the time they reached 5kV, they’d achieved “close to 100% suc-

cess” in recognizing problems. As a re-sult of the study, EMD hasrecommended the implementation of a5 kV testing regime to anyone servic-

ing their products.The selection of test voltage is a critical

first step in applying an insulation test. For most common situations, the familiar lower voltages will work just fine. But fortu-nately, as the EMD study has revealed, for 

 bigger equipment and more demanding re-quirements, enormous test voltages are notthe order of the day. A convenient compro-mise between the theoretical and the practicalhas been “proofed out” at 5 kV. ■

5kV TestingHow Much Is Enough?

BY  JEFF JOWETT,SENIOR A PPLICATION ENGINEER, AVO INTERNATIONAL 

TEST & MEASUREMENT

Questions?For answers and to locate your

nearest Distributor call1-800-723-2861 today!

Reprinted with permission from Utility Products Showcase © 2002.