Plasma cleaning- DIY HF HV source?

[Conrad Hoffman]

I've read that plasma cleaning can improve epoxy joint bonds. Is this as easy as running an HF HV arc over the surface? What's the simplest  generator? Something based on a TV flyback coil, a Tesla coil, a neon transformer or something else? "Have wire, will wind."

[MadTux]

Used one at school some years ago.

As far as I understood it, 2 plates inside a vacuum chamber at reduced pressure, so that plasma can form, HV-HF between these plates and the piece you want to clean, inbetween.

It's basically reverse sputtering, gas ions that knock out impurities from the surface you want to clean, instead of depositing material.Oxygen plasma also reacts with these impurities. So you get a super grease/oil free surface that has a very little material sputtered away from the surface, ie roughness at microscopic level. So, ideal conditions for epoxy, if you can afford and fit into a vacuum chamber.

[LaserSteve]

Google "Oxygen /Argon  Plasma Needle" that does not need the  chamber, you can build a wand.  I made one about five years ago for  a Professor, 13 Kv pulsed DC  at a few mA / few kilohertz.. With a series ballast resistor to protect the PSU.

For the 9" diameter maybe 16 inch long plasma cleaning  chambers at work we use 1-2 Torr of O2 or Ar and 13.56 Mhz RF at up to 1.5 Kw  on a glass chamber with a large copper coil around it with a  commercially made automatic matching/tuning coil box.   A starting spark from a handheld tesla coil is some times needed, but the resonant coil usually provides its own ignition.

Steve

[coppercone2]

for what?

if you have metal, just heat it to 300-400C and it will literately burn any oils or anything on the surface. Steel will blue in this condition.

You might ruin the temper and stuff, but for certain applications you can just use a torch even better if you sand it and then heat it in a inert atmosphere to the 400C temperature. Wrapping it in some aluminum foil or steel foil might be good enough. Then you have like 1 hour to work.

Keep in mind though, some heavy oils will boil at like above 500C. If its in air obviously your gonna torch everything. Flame is like a plasma or is plasma.

if you are actually using a stone I wonder if you can burn the stone first to clean it, burn the surface, then grind with the clean stone to get your epoxy scratched up surface thats oxide free.

But for sensitive parts etc, you can find the youtube channel 'applied science' he does it to some microscope slides and stuff IIRC. Pretty sure its home made.

[Conrad Hoffman]

This is for small metal and plastic parts that can't be heated too much. I found the handheld units like the Piezobrush, but they seem to be crazy expensive. I may try some experiments in a small vacuum chamber with just DC, a ballast resistor and operating in the corona region. If it looks effective, I'll see about building something portable for use in air. Thanks!

[coppercone2]

He uses a tuning network for his

[coppercone2]

I also had a thought, I wonder if epoxy would be stronger if you mixed it on fired ceramic or something plasma cleaned. I don't renember if the loctite syringe style mixers come packed in a medical style wrapping to make sure atmospheric vapor does not get into them. I wonder if the mechanics of the epoxy can make it so a film ends up building up ontop of the epoxy. Or if plasticizers can form a film on the plastic syringe mixer thing.

I usually use medical cups that are in a big bag, I figure they are pretty clean, for jb-weld etc.

How much does a Piezobrush cost? I never saw that before.

LOL, I wonder if a 'violet wand' would work.

[Conrad Hoffman]

Drifting OT, I mix a lot of epoxy for critical joints. The medical cups are fine. My guess is even if they have some mold release, it doesn't come off or contaminate the epoxy. Always mix by weight on a good scale, never by volume. Stir with a stainless steel rod about five times more than you think is enough. There will still be films of unmixed epoxy on the walls and bottom of the cup. Pour or scrape the epoxy into a second cup and mix again. Degas it by vacuum for most any epoxy, or by centrifuge (only for unfilled epoxies). Figure out how to get your substrate clean. The best joints will be on etched surfaces, but the etchants necessary are usually things I don't want to use. I usually clean with Alconox and rinse with DI water using an ultrasonic cleaner. Wiping with most available alcohols will leave things worse than when you started. Not sure about other stuff. The water test is really important. If the water doesn't sheet, the surface isn't clean. The guy in the video doesn't really do it right. Here's an example of what plasma cleaning can really accomplish-



FWIW, I've tested a lot of epoxies, including JB Weld. The stuff is really good for filling defects and sands and machines well, but you can do way better for adhesion. Industrial products like (Henkel) Hysol 0151 or the Epotek products like 353ND (needs high temperature cure) or 301 can do better. Filled epoxies can have improved mechanical properties, but may not "stick" any better. Whatever you use, keep it off your skin. People can become sensitized to epoxy components and if it happens, you might never be able to use it again, or even be in a room where it's being mixed.

[CatalinaWOW]

I'll second the recommendation on care.  I now break out in hives with any direct contact to a wide variety of epoxy resins.  Fortunately my sensitivity hasn't reached the vapor level. 

I used to think gloves and other precautions were too much trouble.  Wish I hadn't felt that way because now I have to take even more tedious precautions.

[coppercone2]

Do you have any tips for epoxy storage?

The main reason I use consumer brands is because they screw up in the containers.

Superglue or methyl acrylate glues I never had a problem with, and I keep that in a jar with dessicant. All kinds of epoxy screwed up on me, including hysol brand. My MGchemicals silver epoxy is still good, even after a year. I am gonna be cursing hard when expensive silver shit goes bad though. 

I never thought to degas it in a vacuum chamber, I will try this.

I use JB weld because it comes in a metal tube and I can pick it up in 5 minutes if I need it. I think it lasts longer then the loctite dual syringe ones from home depot etc. If I go with a product that expires (inconsistently) that I can't easily buy in person quick, its gonna end up pissing me off, and I can't justify it, because epoxy IMO is a good tool for weird unpredictable repair or assembly jobs that just come up randomly... and you can't really justify waiting or spending big money on those jobs IMO. If I need to use epoxy, I need it right there and now. Different if you run a production line or have a specific project in mind or have a big budget where you can afford to keep the good stuff on hand.

MY mind would change if there was some known good way to store it that is not a bunch of hearsay. Or if they sold hysol brand in little metal toothpaste tubes so you can keep a few in stock.. those dual-syringe ones are just too big for low scale use.

If you got one of the big ones, and transfered it into smaller evacuated containers, would you get the closed product shelf life?

[Conrad Hoffman]

Storage can be a problem and I try to use the stuff before it goes "bad". If it crystallizes you can heat and stir both parts to get it back to good condition. That advice is right from some manufacturers. I don't like dual syringes at all. They can give inaccurate mix ratios and are subject to cross-contamination. I've purchased huge industrial ones and repackaged the stuff in plastic squeeze bottles, but it's a PITA and life wasn't that long. Nor am I a fan of 5-minute epoxy. It tends to be softer and not as strong, though the Loctite industrial one they sell in the big home stores is pretty good. If you can find regular cure "2-ton", that's decent. It's hard to find good consumer epoxy. Best bet is commercial stuff on Amazon. I've never used Araldite as it's uncommon here, but they seem to make some good stuff and it appears common everywhere but here.

[LaserSteve]

Most Atmospheric plasmas are quite cold. We were using them on polymer nanofiber mats with no damage, just surface modification to enhance reactivity.

     Someplace there is a picture of a much younger LaserSteve holding his finger in the plasma (just Argon)
With no ill effects while standing on a insulated mat.

Steve

[coppercone2]

what exactly happens to the surface because of plasma?

Is it just thermal cleaning or is there mechanical deformation or some kind of chemical change (oxidation?)? I had it explained to me that in a RF plasma you can kinda think about it like a field basically causing the plasma molecules to physically beat/peen the surface kinda, based on the RF field frequency, in a harmonic motion, but with gas rather then a hammer.

The possibility being to combine one of those wands with a gas tank or nitrogen generator possibly to increase activation (like a nitrogen generator from a airless soldering iron or medical oxygen gen or just welding tanks)?

And conrad hoffman, have you ever degassed the re-stored epoxy under vacuum to see what would happen? I mean does it magically just fuck off once you break the manufacturers seal, or can it be resealed properly some how? If you can seal it into a bunch of small containers that are like 0.5cm x 3cm it would be great actually. I have a feeling I will just get a buncha covermyass bullshit from loctite. I am not making human organ replacements.

Also for non-optical requirements, how does sand blasting perform compared to plasma? Especially if you cook the sand first to degrease it and run some through the sand blasting gun to clear out the internal buildup of crap, and you run the sandblasting gun from real clean dry air not a shady shop compressor?

The plasma seems like a ultra-painintheass source unless you are dealing with something cosmetic or optical because it smells bad, high voltage, expensive, kinda can't tell if it is working without testing...

Also, won't sandblasting generally be better because you cut into the surface physically, which is important for glues? I was always told to scuff stuff up anyway. I figure if you rough it up with like 80 grit sand paper, then sandblast that down so its not that jagged, you get a combination of high surface area and good dimensional fit (if its just sanded with low grit paper you get mountains your trying to glue).

I am interested in this technology but I am trying to find out why it would be useful other then from stuff that really has to look super pretty and can't be masked with tape for some reason, because its just such a massive step up from what I am used to with glues.

To use air correctly it seems like you would need to use a maybe a stainless tube connected to your regulator that you can decouple from the tank and also cook. The rubber hose seems like it would cause a problem. Since it's not being used for actual cleaning and just light surface work, you can make your own simple sand feeder you can also clean (see youtube for simple tutorials, you can make them with simple tools like files and drills, no need for complex gun systems).

[Conrad Hoffman]

One of my many holy grails is good epoxy bonding to 303 stainless steel. When I test, I always get adhesive failure- the glue stays on one part or the other and pulls away from the metal. I routinely sandblast (glass bead actually, with Trinmix #4) followed by water based cleaning. A water test on the stainless is never as good as I think it should be. What you want is cohesive failure, where the glue itself comes apart with some still adhering to the metal. I can do it with many other materials, including 400 series stainless, but not the 303. I'm guessing etching with a dichromate would improve it, but that's not stuff I want around, thus the interest in plasma. Unfortunately the application doesn't justify the several thousand dollars for the hand-held devices.

All I can really say about the life issue is when you can't get it to recrystalize or the color changes or it doesn't cure to a hard glassy material with no films or stickiness, it's time to dump it. When in doubt test.

[LaserSteve]

Um, a surplus photocopier drum charging power  supply, a ceramic ballast resistor, two tungsten welding  rods, a bit of plastic,  two microscope slides as plasma guides, plus welding grade gas tanks,  with used regulators, were not that expensive. Plasma Pencils are trivial to build.   I will admit a liter to two liters per minute of Argon can get expensive on a production floor.



Steve

[coppercone2]

can you post plans?

I seriously think it would be like a super popular project. Everyone hates gluing plastic shit.

I wonder if I can just clean stuff with a plasma cutter ! I think it would be a bit much even on the lowest setting though.

-type of electrodes (what doping? radioactive?)
-what kinda plastic, teflon?
-how to setup the plasma guides
-anticipated pressure
-V/I of PSU
-What current limiting value (I guess this is standard for arc generation though)
-angle of electrodes? Sharp or domed?
-general layout of parts
-any special inlet geometry for the gas etc

and, is it a finicky bastard like a atomic hydrogen torch ?

[coppercone2]

am I getting the idea right?



What is the power supply? I see printers go from like 1-12kV.

[Conrad Hoffman]

Heck, I've already got oxygen for the torch! I see commercial units that run at 13.56 MHz. Is there something magic about that frequency, or is there some unlicensed band there where the EMI won't get you sent to jail? If I use a chamber in the corona region, can I just use DC and a ballast resistor?

[coppercone2]

with a plasma jet like that I have no idea. A metal plasma cutter works in the KHz I think.

For a plasma chamber I heard (from someone) it had some effects on the momentum of the atoms but there is no real magic. 13.56 MHz and its multiples are basically used for this stuff. You get 27MHz and 40MHz too. They account for distortion and harmonics in the legislation.

Are those 13.56MHz units powered by square waves?

I am pretty sure a power supply for a copy machine is DC. A molecular hydrogen welding torch runs on 60Hz and it does something kinda like what I see the pictures of the plasma pencil on google do.


Can you use this to clean wounds before applying a bandaid? Or before popping a zit? Wash your hands in the stuff?

I am kinda thinking, once you strike the arc, you have a bunch of ionized stuff. When you have flow you start to get diffusion. I don't think it matters how you make the arc, just if the plasma will get to the outlet before it cools enough. Is dripping dye into a fluid stream a reasonable way to think about this?

Is anyone else planning on substituting showers for plasma baths?

I am imagining with DC you get a kinda steady arc that you adjust gas flow to diffuse a bit of the plasma to reach some kind of linear steady state wheras with AC you might be PWMing plasma doses into the argon stream and reigniting it, maybe, depending on the time constants. I think it might be actually some what possible to figure this out on paper using basic equations maybe.

I might do this but I don't have any tig rods or copier HVPSU. I do have some spare teflon, argon and microscope plates. I do have MOT, 1.2kV lab supply and maybe some monitor transformers. Is tig rod really necessary at these power levels? Can't I use hardened steel like a sharpened drill rod or something?

[coppercone2]

I managed to liberate a boiler ignition transformer. I dont know if its a autoformer or isolated but i can try it through a variac

Does the ignition transformer have a current limiting resistor built in since its made for arcs?

[LaserSteve]

17 milliseconds between pulses is too long to sustain "soft" plasma in air.  I use pulsed DC at few Khz from the output of a short time constant voltage multipler in the photocopier charging psu. You reduce the gas flow to ignite the plasma in my design, then increase the pressure to extend the plasma out of the chamber. You need a thin Townsend like streamer, not an arc.


Mots and furnace transformers have an air gap in their magnetic core to provide some crude current limiting by preventing saturation. A BIG TRANSFOMER like that is too powerful for small soft plasmas. And they will get you killed if you try them for this application at atmospheric pressure.

Steve

[Conrad Hoffman]

If I want to try this with stuff off the bench (don't have any HV RF generators or flybacks hanging around), let's say I've got low current DC supplies up to maybe 1.5 KV or so. I've got power oscillators up to 20 kHz. I've got low power oscillators to 1 GHz. I've got wire and semis. I've got medium sized ferrite cores. I've got a PVC vacuum chamber with a rubber stopper and wires fed through. What would be the easiest thing to throw together? This is just way out of my wheelhouse.

[coppercone2]

Yea furnance transfor.ers hurt like hell.

It wont work if i find the right current limiting value? I dont like the idea of a pen either at such voltages, i was imagining more like a benchtop forge.

I never looked at the frequency responce of a boiler transformer but from the sound of things it might not be good enough.

Maybe a higher powered plasma would be useful for cleaning metal and ceramics and the softer one for plastics ?

[coppercone2]

I freaking hate flybacks. I found one too but i dont want to deal with that crap.

[Conrad Hoffman]

Updating an ancient thread, we did buy a Plasmaetch Plasma Wand, two actually. Way too expensive, but they do a nice job. Some of our parts are metal and you have to use a special nozzle, either that activates through glass, or one that shoots an argon/plasma jet. It's easy to wreck the unit if you pull too much current an overheat it. Don't ask how I know. A dramatic illustration of how well they can work is to take a piece of Teflon sheet and try to write on it with a Sharpie. Doesn't work. Then hit it with the Plasma Wand and try again- you'll be able to write on it. I doubt you could bond to it, but I've been pretty successful at getting epoxy to adhere to Delrin and similar things. The guts of the unit are here- https://www.tdk-electronics.tdk.com/en/373562/tech-library/articles/applications-cases/applications-cases/cold-plasma-from-a-single-component/1109546