Reproducible spark gaps / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

Reproducible spark gaps

1 2 next

Move Thread

LAN_403

Dr. Slack

Tue Jun 13 2006, 10:10PM

Registered Member #72 Joined: Thu Feb 09 2006, 08:29AM
Location: UK St. Albans
Posts: 1659

There was a thread a few weeks ago asking which was the better TC protector, Terry reckoned his was better as "some people don't get their safety gaps set right". As I'm about to fire up my TC for the first time, I had gap setting floating around in my mind. It's easy enough to make a gap with bits of bent wire, or adjustable screws, but the variation of the electrodes and the uncertainty of the setting makes them very variable, you need to adjust them every time you make them.

How about a gap where the electrode shape is pre-defined, so not a variable, and the gap is set automatically, so is also not a variable? Would this make a "build to recipe and forget" gap?

The two electrode materials and methods I thought of were a) ball bearings and b) straight wires. Ball bearings have the attraction of being very smooth, relativiely cheap, nice looking, and easy to set by sandwiching them between two pieices of insulator, say GRP or Perspex, with two holes drilled in each (drawing or photo may follow). Their disadvantage is making the electrical connection to them. Wires would be crossed rather than parallel, this gives a well-defined point of closest approach, and allows them to be spaced by mountiing one each side of a piece of insulator with a single hole drilled through it. The thickness of the insulator is the spacing of the gap. A stack of thin insulators can be used to adjust the gap to whatever size is required.

Here are some calibration measurements I made on the crossed parallel wires gap. I was only able to go to 12kV with the gear I had. I noticed that breakdonwn occured initially at quite a low voltage, which then rose to a plateau for any given spacing. This behaviour has been documented before, and is apparently due to the discharge burning off microscopic high spots at which the field is highest.

These readings are voltage for which "seasoned" electodes will usually breakdown, but sometimes not for a second or two, versus the spacing gap. I used straight copper wire of diameter 1.2mm, this was the earth wire from some 2.5mm "twin'n'earth".

0.5mm 3.8kV
1.0mm 6.6kV
1.5mm 8.0kV
2.0mm 9.7kV
2.5mm 10.5kV
3.0mm 11.5kV
3.5mm 12.5kV

The temperature was 26C, the humidity felt high, perhaps 70%, I don't know how much either of these may affect the breakdown voltage.

I would be interested if other people could repeat these measurements, mainly as a practical test of whether the method is indeed as repeatable as I think. It might be nice to calibrate gaps to higher voltages. Note that gaps with points would not be expected to be nearly as reproducible, due to the uncertainty of the electrode radius.

HV Enthusiast

Wed Jun 14 2006, 12:21AM

Registered Member #15 Joined: Thu Feb 02 2006, 01:11PM
Location:
Posts: 3068

Humidity is a big factor, although it has opposite effect as most think. As humidity increases, the breakdown voltage of air actually increases as well. However, you may get more coronal discharge (i.e. spitting, etc...) because water will attach to dust making sharp points on hv electrodes etc...

For a close-gap, humidity may actually have an opposite effect since the dust / water particles may decrease distance between the gaps.

Terry Fritz

Wed Jun 14 2006, 12:26AM

Registered Member #393 Joined: Tue Apr 18 2006, 12:30AM
Location:
Posts: 297

Hi,

There are "standard" spark gaps generally using charts like this:

But spark gaps tend to foul and burn with hard use. It might work fine at first, but firing 120 times a second at hundreds of amps will wear it out over time. If it was not for wear, it would be easy to make a "standard" spark gap or a common one with well known charateristics that were repeatable. In some low current low repetition rate applications is works very well and there are HV voltmeters that use standard sphears with micrometer distance settings.

Drilling the super hard steel in ball bearings is not easy. You have to file a flat spot to get it started and use probably cobalt bits on it. I have tried but never have been very successful.

In my case, I live at 5500 feet so spark distances are all messed up anyway.

Cheers,

Terry

Dr. Slack

Wed Jun 14 2006, 07:41AM

Registered Member #72 Joined: Thu Feb 09 2006, 08:29AM
Location: UK St. Albans
Posts: 1659

Thanks for the link to that table Terry. I had assumed something like that was to be found, but you know how it is when you're looking for something! I hadn't realised that humidity could work the "wrong" way round. The feeling of stickiness made me check three times rather than twice as I moved about my PSU!

I was assuming use in saftey gaps, which fire rarely enough to stay pristine, as opposed to working gaps which get toasted.

I hadn't for one moment contemplated drilling ball bearings, which is why I assumed connection would be tricky or untidy. I was assuming a wire tip touching the ball through the alignment hole in the plastic support. Not efficient enough contact for a TC working gap, but good enough for a safety or Marx gap. In fact a double row of ball bearings would make a very nice set of Marx gaps, all in a perfect line for UV illumination.

The data in the table illustrates my points quite well. The smallest repeatable electrode size tabulated is for 1" diameter balls, not many of us would use saftey gaps with electrodes that size - though I guess the heads of coach bolts would be quite a reasonable approximation, even if not as smooth as a ball. Also tabulated is points (what is the tip radius of a point, I suppose a "standard dress-making pin" might be fairly consistent?). These two sizes bracket the range in which most (all?) amateurs are going to be working, and the ratio of spark lengths is typically 3:1 between balls and points. I wonder where 1mm wires are in that range?

The mutiple aims of the post were to suggest a robust gap geometry which is stable, settable and repeatable, and uses materials the typical amateur has to hand. We all have straight copper wire to hand! One important question is, is the surface finish on commercial solid Cu wire consisitent enough to get repeatable results when used like this? Perhaps seasoning the gap makes any surface finish varaition irrelevant? The other aim was to cross check my measurements.

GODSFUSION

Wed Jun 14 2006, 11:54AM

Registered Member #157 Joined: Sun Feb 12 2006, 08:00PM
Location:
Posts: 76

I looked but couldent find what i wanted. Does anyone know the breakdown voltage of those flash tubes in disposables? I had tried a closed gap inside a broscilliate(sp) T pieceThe two parallel ends of the T has the electrodes and the tail piece of the T had dessicant in a wire screen ball. I really diddent see much effect, but it was trial and error and electrode wear was a prob.

- Wayne -

...

Wed Jun 14 2006, 06:18PM

Registered Member #56 Joined: Thu Feb 09 2006, 05:02AM
Location: Southern Califorina, USA
Posts: 2445

the flash tubes in cameras run with a low presure xenon fill, so I doubt that your rig would be even close... In any case, they breakdown arround 500ish volts (for the little ones), if you want I can try and see what they break down at... I suppose a string of flash tubes would work as a saftey gap, but not all tubes are the same....

Sulaiman

Wed Jun 14 2006, 08:23PM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3140

I use 1" ball bearings as spark gap and chicken-stick end.
A steel bolt can be welded to rhe ball bearing 'head first'
leaving a ball on a threaded shaft.
With a messy but unimportant weld visible.

P.S. Steel, being ferromagnetic, is a poor choice for a sg tc.

Cesiumsponge

Thu Jun 15 2006, 01:08AM

Registered Member #397 Joined: Wed Apr 19 2006, 12:56AM
Location: Western Washington
Posts: 125

Drilling ball bearings is usually a practice in fustration without special tooling. Tool steel files and cobalt HSS drills are usually hardened around the same ballpark range as purpose-ball bearings (~57-61Rc). I've tried once and it was basically an effort in futility as the ball bearing and drill/file worn one another down and you couldn't get any decent life out of the tooling and it left a sloppy job. Diamond impregnated files and tile drills or carbide tooling could do the job but they aren't as easy to find (and more expensive). Tapping is also an issue too as generally available hardware store taps (cobalt/HSS) are usually right in the same ballpark range (actually slightly less so they won't be as brittle). You'd probably have better luck drilling "balls" that aren't meant for bearings, therefore aren't hardened. Either that, or you can anneal the ball bearings if they can be.

Any benefits of hardening probably go out the window once you start striking arcs and annealing the metal anyhow. The only benefits I can see using ball bearings is that they are usually grade 25 or higher (sphericity of .000025") which means they are highly uniform and spherical, but such tolerances aren't necessary once they start eroding.

Variations of drawer pulls or brass spheres (which can be had pre-tapped) are probably the simplest mechanical and monetary solution while still staying in the range of hardware store availability (brass tapped spheres might have to be obtained from a lighting/lamp specialty store in quantity) . They needn't be perfectly spherical either. Haefely (manufactuer of probably over 50% of all high voltage impulse generators being used commercially) use obloid spark gaps with a larger radius on the faces than it does on the edge. These can be seen here on page 6 and 7 of the pdf file

You can see they are very similar to some obloid drawer pulls that are commonly available. The greater face radius allows you to run less extreme spark gap distances as well. They do look a tad goofy though. It would be difficult to get a community to standardize on one particular pullknob. As for brass tapped spheres, they come defined with threading and diameter and they should be the same for anyone. The only downside is brass isn't exactly long-lasting as an electrode material at even reasonable power levels.

Here is the cheapest e-store I have found to date for these
Turned balls:

Turned half-balls

Steve Conner

Thu Jun 15 2006, 10:29AM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

Maybe we should do a 4hv bulk buy of brass balls from that online store. The price goes right down for 10+.

"4hv.org, the forum with the most balls."

Dr. Drone

Thu Jun 15 2006, 04:58PM

Registered Member #290 Joined: Mon Mar 06 2006, 08:24PM
Location:
Posts: 1673

1 2 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob