Vacuum Interrupter - Field Emission Causing Inconsistent Test Results?

[saladhawks]

At the utility where I work, a new 27 kV gas insulted switch with vacuum interrupters is exhibiting inconsistent test results across A, B, and C phases. Here are the test details as completed by utility:
1) All phases passed 1-Minute 40 kV AC Hi-Pot Testing.
2) B Phase exhibited a decrease in IR during DC Step Voltage Test (2 to 10 kV DC for 1 minute duration over 5 minutes).
3) Compared to A and C Phases, B Phase exhibited inconsistent test results for both Polarization Index (10 kV DC) and Power Factor Tests (10 kV AC).
4) B Phase exhibited an increase in leakage current during 60 kV DC Test (15 minute).

Switch was sent back to vendor, who completed the following tests:
1) All phases passed 1-Minute 40 kV AC Hi-Pot Testing.
2) B Phase exhibited a higher leakage current during 60 kV DC Test (15 minute) compared to A and C phases. B Phase leakage current did not increase during test.

Vendor has advised that switch is ok and suitable for installation. Vendor advised that higher leakage current during 60 kV DC Test across B phase was likely the result of field emission phenomenon, whereby a microscopic sharp feature exists that is being observed during testing as increased leakage current. Vendor has advised that utility 10 kV tests (DC step voltage, polarization index, and power factor) do not provide any useful information on state of vacuum interrupters and that AC Hi-Pot test should be the ultimate means of testing integrity of vacuum interrupters.

Does vendor explanation sound reasonable?

 

[dpmac]

Do you have the ability to use the DC Hipots Guard Circuit to see if you can isolate any components?
IE: Guard each component (insulators, standoffs, etc) out of the test circuit to see if test results improve.
This may allow you to detect a bad component.
Is this "microscopic sharp feature" supposed to be on the contact face of the B Phase interrupter?
Does it effect the contact resistance test results of B phase?
If its so microscopic you may be able open and close the breaker under a small load and get rid of it.
I find it hard to believe the 10kVdc test would vary much between phases. 60kVdc I can understand.

 

[ausphil]

Is there one single gas compartment that feeds all three phases, or is it individual gas compartments? has the gas been checked?

Are the tests being performed across the open break or with the breaker closed? I assume that at least the DC tests were.

When you say the leakage current was increased, are we talking about 10% increase, 100% increase? Also, from what value was the increase, was it 10uA or 1mA. If the current value was negligible to start with, the fact that it has increased (by whatever the % is) may not be a big issue.

Have you performed partial discharge tests on the unit? I would measure PD with the breaker closed from the phase to earth (ie hv stressing the whole vacuum break unit plus all the insulated bars), and across the break, possibly from both sides to determine if there is any particularly harmful discharges. If there is any "microscopic sharp feature" these tests should show them up, especially if it is the cause of the increased leakage current (sort of like the "tip up test" of old, where the losses from partial discharge activity used to manifest themselves in the power factor test when you got to sufficiently high test voltages)

ausphil

 

[DTR2011]

Generally speaking the AC Hypot Test is the accepted test for vacuum bottle integrity tests. It is a Go-No Go test. With that said, there are a few things you may also want to consider.

I am assuming this is some kind of pad mount switchgear - with dead front connections. Have you checked you test connectors (elbows) that you are using? Have you tried to swapping connections? When you connect the test equipment to the test elbows, have you compensated for corona issues? Additionally is the DC Hypot source well regulated? Any change in source voltage can affect the readings. Often times DC HyPot units are fed with a voltage regulator to stabilize the readings.

Based on NETA standards, if the switch passes the 1 minute insulation resistance and 1 minute AC HyPot test, it is good to go.

The manufactures reply that the switch is OK sounds reasonable. The rest sounds like a polite way of saying its OK.

 

[Zogzog]

Your vendor is misonformed. Hipot tests, AC or DC, are simply a go-no go test and tell you little about the remaining life of the vacuum bottle. We conduct Magnetron Atmosphere Condition (MAC) assesments on all vacuum breakers and starters we service in our shop and on many field testing jobs.

MAC testing is based on the Penning discharge principle which in a nutshell states when a high voltage is applied to open contacts in a gas and the contact structure is surrounded with a magnetic field, the amount of current (ion) flow between the plates is a function of the gas pressure, the applied voltage, and the magnetic field strength. By using this test method we are able to determine the pressure in the interuptor and therefore predict remaining life expectancy.

Keep in mind that VI's have roughly a 20-30 year design life, if your VI's are close to that age or over you may be on borrowed time.

Now, the results you are getting (I assume you are refering to the open pole (Or Vacuum integrity test)) is typical for a VI nearing end of life, we have seen this occur often at 20-10% of remaining life and will recommend replacement of the VI's at that point because failure is imminent.

I have attached a white paper on the subject, let me know if you have any questions.

 

[dpmac]

Zog,

The OP stated this is a new unit.

Your statement "Now, the results you are getting (I assume you are refering to the open pole (Or Vacuum integrity test)) is typical for a VI nearing end of life, we have seen this occur often at 20-10% of remaining life and will recommend replacement of the VI's at that point because failure is imminent." Please explain which results you are referring to that indicate imminent failure.

From what I have seen of the MAC test it is used solely for Vacuum Interpretors alone and gives no indication as to the integrity of any other components in the system. Please correct me if I am wrong.

The possibility may exist that an insulation component in the primary path of B Phase may be questionable and its not the VI at all.

 

[Zogzog]

Missed the "new" part but could still be a bad VI, maybe losing vacuum. I was refering to his phase B results showing an increase in leakage current, a clarification of how these tests were performed by the OP would clear this up.

Really the point of my post was addressed to this " Vendor has advised that utility 10 kV tests (DC step voltage, polarization index, and power factor) do not provide any useful information on state of vacuum interrupters and that AC Hi-Pot test should be the ultimate means of testing integrity of vacuum interrupters."

And yes, MAC testing is just for the VI itself and is just one of many tests that should be done for overall evaluation of the equipment.