Andrew, this is great news and hopefully it will help me get up more enthusiasm.
But time is a problem.
As to reversing the wires - well it worked as it was supposed to do.
And the cathode will be connected to the negative (-) wire so it makes sense to me.
The + gas ions were attracted to the mangets, hit the AL and heated it.
Perhaps a much thicker piece of metal such as a thick disk will melt or evaporate more evenly.
This is more like the commercial targets. I really want to deposit Cr for this application of mine. And I want it to stick really really well and thought it just may be worth a trial of sputtering. Hence the interest. Which increased when I found out that the magnetrons, which sounded complicated, were a lot simpler than I had expected when I researched it.
I have not seen videos - only pictures and theoretical stuff. But it seems almost too simple to work.
I note in most real world applications , they use water cooling and it seems from your experiments that it will be needed.
Again from what I have read, there is a plasma pattern and voltage which is far more effective at sputtering. Maybe with a magnetron this is less critical because the pattern goes around in an endless ring.
The real issue with evaporative coaters is a much higher clean vacuum. If this can be got to work well it means a diff or turbo pump will not be needed.
But I find it hard to accept that at the milder vacuum pressure where the Plasma is working, the mean free path is long enough for the metal ions do travel and deposit as a perfect shiny film.
Have to run.
Peter.
Peter,
I have been meaning to try something very similar but just keep running
out
of time.
Are the magnets coupled by a magnetic path such as iron underneath ?
I saw that in the youtube videos also but my setup did not allow for it so
I did not do it. The round magnet is about 3/4" thick and the N35 magnet
is about 1/4" thick so I could not really put a bar behind them. I did
stack up 2 of the N35 magnets so that made them stronger ( I presume ) but
a single one was already far stronger than the round magnet so I doubt it
made a difference.
The big issue was the polarity of the leads going to the various parts of
the assembly. At first I attached what indicated to be the (+) lead to the
magnet and target alumum foil. I dangled what indicated to be the (-) lead
above the magnets a few millimeters. At 600 vdc the copper wire that
indicated (-) began to glow red and promptly melted. No aluminum foil
sputtering was going on anywhere. There was a really cool green plasma
flowing in the test chamber but only till the copper wire melted. Do you
know how hot you have to get copper to melt it ! ! ! ! !
So I swapped the leads on the magnets. I know the magnets are supposed to
be cathodes and that means (+) but I don't understand electricity and it
hates me so I figured why not try. I put the lead that indicates (-) on
the magnet and dangled the (+) lead above the the assembly and at 450 VDC
the plasma lit up and that ring of magic happened and sputtering of
aluminum commenced. This time the magnets got hot and the wire did not.
So once again I have no idea what direction electrons fly or crawl or
teleport but if you cant make sputtering happen with the leads one way,
then try the other way.
Did you check if Al was being deposited ?
No. This was a test to see if the concept worked at all. The aluminum
left the surface of the aluminum foil so it went somewhere. I will see if
I cant put a microscope glass slide in there today and deposit some
aluminum onto it.
I did discover that the rate of erosion went up dramatically when I turned
off the little vacuum pump. There was still vacuum in the little can but
the overall pressure went up a bit. That made the plasma very angry. I am
guessing it is like back filling with a bit of argon ( putting in some gas
that can be ionized ) as the instructions say to do. I will try that with
the big chamber as my little can really is not set up for this kind of
stuff. It keeps shorting across the inputs as they are only a few inches
apart.
Oh I used a few layers of aluminum foil as the start up sometimes blast
holes in the foil. With three layers the plasma arcs dont make it to the
magnet.
Why choose Al for sputter testing. It is supposedly very slow to sputter
so
something else should work a lot better ?
Because you cant get copper foil at Piggly Wiggly :_)))))) I swiped some
aluminum foil from the wife and used it to protect the magnets. If she had
copper foil then I would have tried that. My guess as to why aluminum
sputters slower is that the heat created by the plasma is dissipated very
quickly into the aluminum mass. As the aluminum targets are tied to the
magnets and the magnets are cooled my guess is that the aluminum target is
cooled and a lot of the heat that would be going into making aluminum gas
is being drawn off by the heat flow characteristics of the metal itself.
But that is just a guess.
I did notice on the eroded aluminum foil that there are dozens of tiny
little balls of aluminum all over the eroded patches. My guess is that the
aluminum was liquifying on the surface of the aluminum foil and surface
tension was pulling it into a ball before evaporation eroded the ball.
Again, just a guess. But there are no aluminum balls on the shiny parts of
the foil.
I seem to remember that you have an evaporative Al setup so this seems
unnecessary for Al coated mirrors.
Yes I do. I have 9 sets of 4 tungsten coils arranged on a 48" inch circle
( with one in the center ). I can vaporize all the aluminum I want as long
as I put in 100 amps at 12 vac into 4 of the coils in series. That works,
but I have run the 9 sets in series and only coat a section of the mirror
at a time. If it all works great then I get realy good coatings and they
stick down good. But if only one set of coils is unhappy for any reason, I
get a bad coating. So I want to switch to plasma deposition like the big
boys use.
I am building a rotary system that will allow me to rotate the whole 65"
dia mirror holder on the door of the chamber so that the mirrors rotate in
front of the plasma discharge bar ( yet to be built ). That way the plasma
sputter bar can be a few inches away from the mirror and the sputtering
will have the best chance of sticking to the mirrors. Again, this is how
the big boys do it. I just have my mirrors mounted on a vertical wall so
all my stuff has to flow horizontally. The big boys suspend their mirrors
above the rotating plasma sputter bar or suspend the rotating plasma
sputter bar above the mirror. Either way it is a long bar that covers the
radius of the mirror and the coating is done by rotating the bar till it
covers the whole mirror.
Hummmmmm now that I think of that, I wonder how they account for the
change in angular velocity and the different rates at which square inches
of uncoated glass is passing under the sputter bar. There must be more
sputtering going on as you go outward on the radius to make up for the
increasing amount of glass you have to coat. Or the sputter section would
have to be pie wedge shaped in some way. Do you have any idea how they do
this.
I will post my successes and failures as I learn how to do this. I am
hoping I can build a bunch of cheap small plasma coaters and just line them
up in a row for my spray bar. I dont see why this will not work but I am
sure there is a reason that I just dont know.
I have two Hign Vact setups. One will deposit (evaporate) Al on a mirror
surface but the vacuum is not clean enough for good adhesion.
I had thiughtg of revamping this to sputtering hoping the plasma will
glean
as well and promote better adhesion. And a magnetron setup seems a good
way
to speed things up.
So I now have this oilless setup which I want to experiment with as well.
Unfortunately, while I have partly built a chamber with leadins etc it is
not finished and I am very bust with other work at the moment. Some of
what I want to do is quite small.
Please post anything else you do and hopefully in a few weeks I may have
something of my own to add.
Peter Smith
---
This email is free from viruses and malware because avast! Antivirus
protection is active.
---
This email is free from viruses and malware because avast! Antivirus protection is active.
