Breakdown voltage of multiple materials

[lepp1892]

I need to find an equation to figure out the breakdown voltage of layers of materials. In my case it will be air then glass then an inert gas.

Any one have any info on how to get this.

Thanks.

 

[Hobin]

Start here. It has a nice table you may find useful.

 

[lepp1892]

Yes, I saw that table. I was just wondering how to add up materials to get the voltage through all 3 layers. I was told that I can't just add up the voltage of each seperatly.

 

[sophiecentaur]

This is a difficult problem because I believe that polarisation of an insulator (dielectric) can upset the voltage gradient over the whole gap, leaving you with more volts than you expected across one of the other gaps.
This is also a problem when you try to put high voltage capacitors in series to increase the overall working voltage. You certainly can't afford just to 'add' the breakdown volts together.

 

[lepp1892]

So are you saying that there is no direct equation to do this? How about something to give me a general idea?

 

[sophiecentaur]

This is a bit too specialist for me to help you.
One way round it could be to see what standard practice is in HV applications.
Are we talking kV or MV?
If you were just take one of the layers (the glass) and work to that, would that give you an awkward value for the spacing.

 

[lepp1892]

We are talking in kV. If I were to take one of the layers away. I would want to take away the air layer since it is the smallest. The glass is the most important layer for me.

 

[sophiecentaur]

Sorry. I dindn't mean take a layer away - I just meant that you should do your calculations using (taking) the glass and neglect any extra protection the other layers would give you. (They can't make matters worse, can they.)

 

[lepp1892]

The layers of air/gas do add insulation to the voltage. It requires much more voltage to go through all 3 compared to just the glass.

 

[sophiecentaur]

Of course they do but, if you can't work out how much they contribute to the insulation, you can do the 'pessimistic' thing and assume they aren't there. That will put you on the safe side.
In any case, if you are dealing with only a few kV, it should be easy enough to get hold of thick enough glass. If you are dealing with more substantial voltages then I would suggest that you must get better acquainted with the practices of HV engineering before you go near it. It is high risk stuff.

 

[lepp1892]

Alright I will try to calculate the voltage breakdown of just the glass and then compare the results with testing.

What is the equation for finding the voltage breakdown for that?

 

[sophiecentaur]

Use the data in 'that link'.
As long as there are no sharp edges, which could cause breakdown earlier than the table would suggest, just take the V/m for breakdown and multiply that by the thickness of the glass. That will indicate the voltage for glass of that thickness to breakdown.

I am pleased you mentioned Testing. That's the way to go!

 

[lepp1892]

Well I have already been doing some testing but I was trying to find calculations to verify my results.

 

[sophiecentaur]

It would be interesting to find out how closely your test results follow that theoretical figure.
btw, you may need to be careful about the routing of your cables and the insulation on the rest of the circuit if you want to be sure of your results.

 

[lepp1892]

Also, any idea how compareable the window glass in that link is to Simex glass tubing?

 

[lepp1892]

So if I go by the glass number there it should break through at about 12 kV. I didn't get a result till 33-38 kV.

 

[sophiecentaur]

OK. Then how big an air gap was there? Working on 3kV/mm in the table, we could estimate that the rest of the gap (air plus gas) could have been about 7 or 8 mm?

Looks like, from the fact that you are actually doing experiments, you could actually come up with some pretty useful results yourself and actually see the effect of glass on its own / glass and air / air on its own - all for various thicknesses.

 

[lepp1892]

Alright well I found some data on the glass I am using. It's not real helpful since in the few spots I found info the dielectric strength ranges from 16-30 kV/mm. The gap between the electrode and the glass is very small (almost touching) and the gap from the edge of the glass is about 3.5mm.

I have tried just the air and just the gas and have gotten pretty good results compared to the data I have found stating what the dielectric strengths are.

 

[sophiecentaur]

That's very satisfying. Well done.

Regarding the gap between electrode and glass, the geometry could be affecting the field locally if the electrode is small. The simple approach assumes that the field is uniform (between flat plates but the field is higher where there are small objects involved. (Corona will strike at way below the value suggested by the 'bare fact' of breakdown voltage / distance)