Video of hi-pot tester in action posted

[LB3HC]

Hi. I have posted a video of my new hi-pot tester taken during a high voltage breakdown test of a std. air variable transmitting capacitor.

http://www.youtube.com/watch?v=KhYFOk3-qXo&feature=player_profilepage

You can see some pictures here: http://www.lb3hc.net/hipot?nggpage=2

73 LB3HC

[WD5JKO]

Pretty neat test fixture! The subject of hi-pot testing might be categorized into "destructive", and "non-destructive". For example hi-potting a transformer to the point where an arc might jump from winding to winding, or winding to core will prove to be destructive much like an automobile "crash test".

I wonder what your thoughts are on how to "hi-pot" in a non destructive way. The air variable being tested in your video might accumulate arcing pits and voids with each successive arc. I guess a lot depends on the peak current allowed once the arc is initiated.

I have been around HV testing over my career at voltages up to 400,000 volts DC. Here we would sneak up the voltage until a corona point was illuminated, maybe a single spot. An AM broadcast receiver nearby would also pick up on the corona discharge. Given time with corona active, enough ozone is concentrated locally and a giant arc occurs. Often the point of corona would be a sharp point, or a burr in the surface finish. Other times the surface would be coated with an insulating paint such that the local charge buildup would not equalize.

The rule of thumb with air dialectic is sometimes said to be 10KV / inch. That assumes the surfaces are flat and parallel. Two nails facing each other will arc between the points at a much lower voltage. Add a little ozone in there (from corona) and the breakdown voltage is steadily lower as the ozone accumulates. Sometimes the air is replaced with Sulfur hexafluoride since this allows greater volts per inch capability.

The subject of hi-potting is complex one. Any perspective you have on my comments would be appreciated.

Jim
WD5JKO   

[LA6TPA]

Hi Marius,

I am building a hi-pot tester to. It’s about 95% finished.
But there is one problem that I discovered in my initial testing here. Problem is that the meters are erratic.
I guess when meters are mounted I a near pure Plexiglas environment there is a static buildup that makes the meters behave like crazy, I think.

Any suggestion….??

[Ed/KB1HYS]

Antistatic spray on the meters.  Sometimes the clear plastic faces can pick up enough charge to affect the needles swing. 

[LA6TPA]

Antistatic spray on the meters.  Sometimes the clear plastic faces can pick up enough charge to affect the needles swing. 

Thanks. I guess this spray acts as a "grounding" screen.

[WD5JKO]

Meters stuck by static charge can also be remedied by rubbing a little water on the meter face with your thumb. Once the charge is gone, then your OK until you rub the meter face again with dry fingers.

My earlier post remains unanswered. Hi-Pot testing till the break-point is found is a lot like automobile crash testing. In both cases, the device being tested is destroyed. I was hoping to get insight on how to sneak up on the break-point without destroying the device under test. Maybe have an AM broadcast receiver nearby listening for corona noise that may occur just prior to the fatal punch through arc?

Jim
WD5JKO

[LA6TPA]

Here is a revised circuit.
Moving the uA meter over to the HV- line removed the erratic meter movements.
Moving the meter also cut my internal leak current in half. Now down to 3uA @ 11kV

Also put in a new current limiting resistor on the HV- line making the tester a lot safer for the operator.
Now it should not be possible to get any currents higher than 300uA outside the enclosure, not between the HV+ and HV- terminals and not between HV terminals and mains/ground level.
Ideally I would prefer a transformer with an isolated secondary but I could not find one that suited.

Geir
LA6TPA