[bobfrancis1980]

Hello, I have been unable to find spacers online that have a small shallow channel for o-rings on the top and bottom of the spacers.

At first I thought about using a 3d printer to print the spacers with channels in them with ABS but the 3d printing forum I reached out to said that was a bad idea, that prints aren't water/air tight and they would need supports on the bottom where the print is attached to the bed which would leave a rough surface there.

Now I am wondering about buying the spacers and trying to either use a hole saw or a cnc router to cut the insets in the spacers. My big worry is finding a way to hold the spacers in place so they don't move while cutting the shallow insets with either a drill press or a cnc router.

I would appreciate any advice and experience you guys could share.

Thank you

Bob

[peteeng]

Hi Bob - What are the dimensions you are trying to make? Peter

[deadlykitten]

pls share some images

[bobfrancis1980]

I wanted to use these o-rings:
33mm OD 27mm ID 3mm
https://www.oringsandmore.com/buna-9...minimum-5-pcs/

I have since considered using a 1" thick piece of HDPE cutting the insets on one side of the board, flip it over, cut the insets on the fresh side, cut the 27mm hole in the middle, and cut the outside dimensions of the spacer.

I need about 20-22 of them.

[pippin88]

What is this for?
What is a 'spacer'?
Why does it need orings in the end?
How precisely do the orings need to be aligned in relation to 1. the spacer and 2. each other?

[joeavaerage]

Hi,
HDPE is an absolute pain to machine, it forms long strings rather than chips. It's hard to get it to cut cleanly.

My go to material for machining plastic parts of good dimensional tolerance and (comparative) ease of machining is Acetal. Even with Acetal I use flood cooling, mainly to flush
chips out of the cutzone. Allowing chips to be re-cut is a surefire way to a bad finish, swarf welding itself back together and sticking to the tool.

Craig

[joeavaerage]

Hi,
I bought some PTFE recently to make some molds. Eye wateringly expensive, but the epoxy molded part pops out perfectly with absolutely no surface release pre-treatment,
and that is the whole point of using PTFE in the first place.

Forms nice chips, still used flood cooling, but with a sharp tool leaves a silky smooth finish. In short it cuts like butter.....beautiful!.

Craig

[bobfrancis1980]

I was going to try and build a replica of this Marx generator:
https://www.researchgate.net/publica...Marx_Generator

The spacers make up the vertical air chamber where the spark gap electrodes will be. I am going to use compressed air to blow through the air chamber and I was probably going to use transformer oil to insulate the rest of the marx outside the vertical air chamber (in their version they used SF6).

A spacer would be in between an individual marx stage and then in between each stage.
Only the inset on the top and bottom of the spacer for the o-ring needs to be cut cleanly. The central hole and outside of the spacer's smoothness is not important.

I was thinking of using HDPE for the top and bottom sheet of each stage as well.

I will look into Acetal, thanks for the suggestion

Bob

[joeavaerage]

Hi,

The central hole and outside of the spacer's smoothness is not important.

Some years ago I built an electrostatic precipitator which required a DC supply of up to 75kV. Quite a challenge.

Surface texture is very important, any 'furry bits' on a surface is a path for surface discharge. If you don't know about surface discharge, don't worry, you're going to find out.
I used transformer oil, but its not easy to get. Even if you can buy the oil, it still has to be de-watered begore you can use it. The power companies whom have the plant to do it are very crabby about
selling small quantities, as every time they open a container of dewatered oil there is the risk of contamination by atmospheric moisture.

The last piece of the puzzle is that should you get a surface discharge you will inevitably end up with 'soot' dispersed in the oil, and its insulation value tanks! One spark and it's all over.

High voltage is a challenge. Read up on 'High Voltage Creep'.

Craig

[deadlykitten]

hello please, when i asked for photos, i was not after o-rings

but the thing you try to assamble, so to figure out part's shape, etc, as they seem chalenging somehow

I was going to try and build a replica of this Marx generator:

Some years ago I built an electrostatic precipitato

what is an electrostatic precipitator ? and what's the deal with orings in it ? why won't any plastic work, as long as you machine it with care ?

One spark and it's all over

do you have a video or something ? you can't make such promises

[joeavaerage]

Hi,

what is an electrostatic precipitator ?

An electrostatic precipitator is a device that charges smoke particles in exhaust plume (corona dischare.....just short of an outright arc -over, say 25kV per inch) and thereby accelerate the smoke
particles (by virtue of the electric field) to the side of the flue where they can be caught. It's a very common practice industrially to 'clean' an exhaust plume. In my case I was using it to reduce the smoke output of a
domestic log fire. It works, and remarkably well, about a 90% reduction under ideal conditions. Having a 75kV power source in a domestic situation is problematic as is the
electrical noise of the corona discharge.

do you have a video or something ? you can't make such promises

Regrettably no, it was about 22-25 years ago now. Initially I had the high voltage transformer (25kV_peak) wound on concentric glass formers, and the whole lot immersed in transformer oil. The container
was a 1l plastic container, so it was hardly big. The power input to the transformer was about 100W, for a corona current of about 1mA at 50kVDC. The DC output came from a tripler
circuit. The tripler itself gave me quite a bit of grief until I came to understand high voltage creep....and most importantly how to control it.
It worked pretty well, but eventually atmospheric water contaminated the oil....and there was a flashover. Thereafter you could only take the output up to about 15kV before it would flash over,
degrading with every flash over event.

If you get a flash-over or surface discharge event over a plastic surface te plastic degrades to carbon particles, which promote repeat flash-over events. In short once its flashed over
its going to do it again, to the extent the device is unusable.

Just digging through my junk and I still have the transformer in its little plastic case, I'll take some photos of it tomorrow. In fact I've just found the trplier as well. The tripler is
potted in epoxy. The diodes are ceramic and about 15mm long, but they would surface discharge immediately they were powered with 10kV or so despite the diodes being rated for 45kV!!!
The surface flash-over beats you every time. Once it was potted the surface discharge problem was all but solved. If you were not careful, and the epoxy did not tightly adhere
to the surface of the component it would leaves a micro gap between the cured epoxy and the component....and you guessed it a flash-over event was just waiting to happen,
and now it was internal where you had no chance to repair it.

I did get around to molding an 'end stack' transformer, winding and vacuum potting the transformer in polyurethane. This was to be a production design. I do not have it anymore,
nor the ferrite cores, which is a distinct bummer, because I had to get those in from Belgium....and they were bloody hard to get. I found a pic on YTube, they call it a 'crossover winding'
but it's what I call a stacked winding.

High voltage has a way of testing your patience.! You seem to have a lot of arc-overs and that sort of thing before you get to the right spot.

Craig

[joeavaerage]

Hi,

why won't any plastic work, as long as you machine it with care ?

Yes, that is true, most plastics behave reasonably well under high voltage. Having said that there are distinct favorites industrially. Polyurethanes and epoxies (filled and unfilled) are the go-to materials for potting.
Polyimides are for higher temperatures but cost more.

Many plastics can be used for coil forms including Polyesters, UHMWPE, Acteals and Nylons If you don't mind paying extra, then PTFE is highly recommended and if
money is no object the glass filled PEEK is the 'bees' knees'.

No matter what material you choose as a coil former you must ensure that your potting compound sticks to it otherwise, you'll have issues with discharges 'burrowing through the interstitial cracks'.
For that reason, I chose to mold polyurethane to make a coil former, and then use the same polyurethane to pot the transformer when done. This seemed to ensure the best potting compound-to-former
bond with the least propensity for discharge paths to be set up.

Craig

[joeavaerage]

Hi,
these are some pics of the stuff I made. The transformer was a first attempt, and as you see the coils were formed on concentric glass tubes. The primary was made of heavy
rectangular copper wire and is self supporting. The ferrite cores I recycled from a TV flyback transformer. The whole lot was immersed in transformer oil. The container was
closed but not sealed and eventually atmospheric moisture got in and buggered up the insulation value of the oil....but it did work for quite some weeks before the oil degraded
and we started getting flashovers.

If you want to ask why I did not use a TV flyback transformer/Trippler its because the polarity of the High Voltage was incorrect for an electrostatic precipitator.

The Trippler was just a regular string of HV diodes and capacitors and then potted. I tried using my ordinary laminating epoxy but for this purpose it exothermed too badly
and did not work.....the trippler flashed over internally in fairly short order. The one pictured is potted in an Araldite product made for casting. It was not characterised for
HV potting, but it has worked fine. I did vacuum degas the epoxy once poured. You really need to be careful not to entrain air bubbles in your potting mixes, they are
a source of flash over points.

After this time I went to an Electrolube product that is a two pot polyurethane made especially for potting, and bloody good it is too. It comes in 500g sachets
that you mix without opening the outer most sachet, thus you could mix the two components WITHOUT entraining any air, great idea. It has quite a low viscosity, especially if you pre-warm
to 40⁰C, so it molded well and infused really well. I did vacuum degas once poured, but I probably could have skipped degassing if I wanted. Vacuum degassing and vacuum infusion
are two great techniques for getting best possible results from potting compounds.

My vacuum pump, back then at least, was single stage and quite good enough to vacuum degas and vacuum infuse, it was not good enough to really thoroughly dry parts.
For this you need a good two stage vacuum pump. What I did was wash the parts in isopropyl alcohol as a dewatering fluid and then evaporate the residual alcohol and hopefully
any moisture the remained under the best vacuum I could get, about 1 Torr at the time. I have access to a better pump these days, about 20mTorr, and that does a better job of drying.

Craig

[deadlykitten]

If you want to ask why I did not use a TV flyback transformer

pfff ... joe, i have no clue with such things, i am out of them what made you think i would ask such a question ? haha

ohh, got it, obviosly, question was not for me ... sometimes i wonder where's my head at

[joeavaerage]

Hi,
the district in which I live has a propensity to form an 'inversion layer' in the atmosphere at night resulting in high airborne pollution levels over the winter months.
The local government was introducing emission standards for domestic log fires with the consequence that nearly all of the existing installations were to be banned.

A friend and I had an idea that an electrostatic precipitator would solve that problem and allow people to continue using their existing log fire while reducing the smoke output
and comply with the clean air goal. In fact it worked quite well, under good conditions about 90% capture, which would be within range of the clean air goal that was proposed
at that time.

The real issue is that as a domestic installation an electrostatic precipitator is a bit of a stretch, The device needs to be able to produce about 75kVDC. A 'zap' from one of these things
will put you off high voltage for life.....they are VICOUS!!!. Even assuming it could be made fool-proof that no-one could or would get zapped, there is that problem of corona discharge,
somewhere in the region of 500uA to 1.5mA. This is in effect pure electrical noise. I guessed it was dubious that I would be able to persuade the New Zealand version of the FCC that the electrical
noise was acceptable. These two objections largely scotched the project....not that it was not fun to play, experiment and learn about high voltage. It is a long and steep learning curve.

Craig

[bobfrancis1980]

Would you suggest not bothering with using an insulating fluid in the Marx generator?
I have come across this bipolar Marx generator that provided +500kV and -500kV but it had a slow pulse rate:
https://www.electricstuff.co.uk/marxthree.html

I just figured transformer oil would be better than air for reducing corona even if the transformer oil is 'dirty'.

[joeavaerage]

Hi,

Would you suggest not bothering with using an insulating fluid in the Marx generator?

You will have to use something otherwise the components will flash over long before you get to your desired voltage.

I just figured transformer oil would be better than air for reducing corona even if the transformer oil is 'dirty'.

Air is for breathing, but has no place as an insulating material in dense high voltage equipment, it breaks down and flashes over quick smart. By all means use transformer oil, but be aware
that its not easy to get, it degrades markedly on exposure to the atmosphere and should an arc occur it becomes polluted very rapidly whereon its no good anymore. All the big power companies
have huge machines that filter/recycle/recover transformer oil. They run continuously and the rejuvenated transformer oil is guarded jealously and never exposed unnecessarilly to the atmosphere
and not because its dirty, but the atmosphere will 'dirty up' the oil!

What is this project for?

If you just want to generate a high DC voltage then a high voltage transformer and a tripler or quadrupler would work. To whit I produced up to 75kV and continuous currents up to 1.5mA (at 50kV).
If its sparks you are after then the DC discharge from a tripler or quadrupler has one hell of a 'crack'. On a good day I could get a full blooded 'bang' as a spark leaped 4-6 inches. You don't want to
take that hit personally, it hurts like 40 b_a_s_t_a_r_d_s!

Craig

[bobfrancis1980]

The Million volt marx page did not mention arc over, just corona. I was thinking of providing a high current DC source in order to make up for wattage lost due to corona.
I only plan on running the generator at 35 seconds a time for actual testing.

I need the high voltage DC generator because I want to test the Biefeld-Brown effect at much higher voltages than has been done in the public record.
I initially wondered if I could DC bias a Tesla coil but it seemed wore complicated than building a high pulse rate marx generator.

[joeavaerage]

Hi,
what do you suppose arc-over is? Let's say a given component, say a resistor, when it has 5kV on it there is a modest corona discharge, take the voltage up a bit more, say 8kV and the
corona discharge doubles or triples, take the voltage up a bit more and then.....bang...arc over. Arc over occurs when a corona discharge becomes that strong that the corona forms a conductive
plasma channel, a spark by any other name.

The point at which a corona discharge forms a conductive plasma is purely a function of the properties of the material or medium, or even a surface over and through which the plasma forms.
All that you'll do by having a higher current capable source is make the arc over that much more spectacular.

I wanted to make an electrostatic precipitator. The charging electrode was to be a wire down the center of a six-inch flue. I determined that at 75kV the charging electrode would arc over, or spark, or whatever you want to call it
would happen under nearly every environmental condition. But I did not want it to arc over, I wanted the charging electrode to emit electrons in a corona discharge and those electrons would land on the smoke particles in the flue
and cause them to have an electric charge. The trick was to get the voltage as high as possible, but WITHOUT inducing arc-over. That is to say I want to maximize the corona current but a little less than the arc-over voltage.
This happy medium worked out about 50kV to 60kV most days and most condition, and the corona current required to meet the objective was from about 500uA to about 1.5mA At around 2mA I found that arc over was
imminent, and it was time to back off the voltage a little.

The idea of using a high current supply to accommodate the 'losses' caused by corona is a fallacy, as you will soon hit the point where a conductive plasma channel is formed, ie a spark. If you want to avoid a spark at
that particular location you have to increase the insulation, or creep distance, or something, to lower the corona current BEFORE it arcs over.

Have you had much experience with high voltage? If not I suggest getting a 'waste spark' ignition coil off a car and experiment with it. It will handily give you 20kV, and all it requires is a 12V switching
circuit of 5A or so in the primary. You can have a lot of fun with these coils. Put a tripler on the output of the coil and you could get 50kV easy and maybe as much as 75kV. It will be very VERY VERY
instructive about corona discharge verses conductive plasma channels. Lots of fun cheaply....just the ticket. High voltage is something that you work up to, you do not get 1MV out of the box, start at a few kV
and work upwards. There is a lot to learn. My experimentation stopped when I got to about 75kV....who knows what it takes to get to MV's.

Craig

[bobfrancis1980]

The million volt Marx generator I linked to seemed simple enough. You can now buy 30kV resistor, instead of soldering small voltage resistors in series like he did. His Marx used 25kV capacitors and a DC input voltage. I was going to use 24kV. You can get cheap ZVS boards on ebay that take in 12-30V DC, run it through a flyback transformer and it boosts the voltage 1000x.

I run two 12V batteries in series to the ZVS board, into the admittedly flaky flybacks that burn out too quickly and get 24kV out. I have played with some Marx generators from ebay that are made in China: https://www.ebay.com/itm/124638951259