In vacuum, noone can hear you fail…


So, you want to experiment with rarefaction…But, do you know what you want to do? because not all voids are equal!

First, an introduction to pressure units for vacuum:

micron(of mercury) < torr <atm

You can switch between those using scientific notation, both up or down, but at some point, one unit becomes more convenient than the others.

Let’s talk about pressure levels:

Numbers have been rounded up a bit for ease of visualization.

Beyond this point, that’s ultra high vacuum, wich is a completely different can of worms. Everything outgases, you have to use weird materials, take into consideration the surfaces of the chambers/funnels, etc…not where I want to go.

So, where have I been?

1.- One pump to rule them all? Yeah, you wish.

That was my first bout with vacuum. I managed to buy an oil free, swiveling piston, vacuum pump, wich, in my naivety, I expected to allow me to do many kinds of experiments. For 200 euros, it performed, well…as good as a swiveling piston can do (see chart). However, achieving one tenth of atmospheric pressure, was just good enough to well, vacuum bag resin and make hot water boil.

2.- Sputter that shit, baby!

Fast forward some years, and I found myself at the other end of a high voltage stick, so to spark speak. I wanted to do big shit, well, literally, small shit (more on that, on another post), but to do that, I need to build some equipement first. On the vacuum side, I ended up buying a refrigeration repairman pump.

As far as I understood, being two stage, it would take a while, but it would attain a decent vacuum level. So there I went, built a nice, fancy vacuum chamber and connected it to the pump through a custom adapter (1/4SAE to M8 with o-ring seat). Once 500V where applied, I was greeted by a nice purple glow, so, there I was, staring at my first plasma ever!

But the vacuum was not low enough, wich allowed sparks to form with the leftover ionized gas. When that happens, it`s a short circuit inside the chamber that will fuck up your power supply, like a coil discharge, and also vaporizes chamber wall material wich will deposit elsewhere.

Applying a magnetic trap to it, focused the plasma, but after thinking about it, I realized the magnet was too far from the chamber (10mm thick plate), making for a weak plasma trap:

Spark on the right corner as blue light. Diffused plasma trap in two rings.

So, a recess was machined in the plate to decrease the gap to 4mm, wich did the trick:

 Top, plasma ring trap, bottom, residual plasma from, still, not good enough vacuum.

After poking around, If I started with a cool pump (room temperature), I managed a vacuum wich I guess was about 500/300 microns or whereabouts.  After a while, the pump performance decreased noticeably (in terms of plasma quality), trying to help by coupling my vacuum bagging pump to it’s exhaust, made absolutely no difference.

Next day’s plasma, looking redish from copper deposited in the chamber walls:

Note how much bigger the residual plasma is (also, plasma trap looks wonderful)

Good enough for what the pump was built, it didn`t work for what I wanted to do. I could probably have managed with an Argon atmosphere, but damn, loosing power to just make gas glow (in the rest of the chamber) well…that doesn`t cut it fo me. Also, it meant that the mean free path for sputtered atoms, was very short, so instead of embed themselves in the sample, they just condensed on top, so you could rub them off with a finger.

That’s mostly copper oxide, wit a resistance of about 100Ω per cm.

Also saw some interesting effects on the plasma from aplying more potential to the plates (plasma compression?):

After some more tests, like raising the positive plate with a “virtual anular plate”:

It became clear that this wasn’t going to get much better, and having to wait a month for pieces to try an Argon atmosphere, I just decided it wasn’t worth, so fuck it, I went and sold the pump in a pawn store.
There was no much point in storing it, because I already had one wich doesn’t use oil, wich is less messier than those “cheap” refrigeration pumps.

3.- Shit just got real:

Upon thinking my frustration out, I looked for secondhand pumps on wallapop, because you know, sometimes if one digs, finds glod.

An ad gave me good vibes. The pump had KF-16 high vacuum flanges, not 1\4 SAE fittings, wich was a good sign. Also, whoever had stored it, had had the provision of covering both flanges (inlet and exhaust) with proper KF seals, not just leave them open or covered with film. @ 150€, (175 with shipping) it looked like a well mantained bargain. Upon contacting the seller, he commented it had been used to make Neon signs, wich require a good vacuum. With that, I decided I had to have it. (also, the next best thing was in the 500€ range)

Once I got it, I immediately changed it`s oil, wich, in the pantone chart of degradation, looked bad, but not dead pump bad.

With new oil, the pump started, but after a couple seconds, it would stall. Thinking about denser residues from storage, I charged the pump with a 50\50 kerosene-oil mix, to try to clean it up a bit, and ran the pump. It ran fine for a few moments, then disliked it even more than just oil. Maybe I had a dud, altough the seller, who seemed confiable enough, told me the pump ran fine.

I must add that, altough the pump was rated for 220\380v three-phase, the motor came, by the looks of it, wired for a cheater capacitor to go from sigle phase to three pase. At that point, and nothing to loose, I decided to dismantle the pump completely.

I didn`t know what to expect, so I just went along, loosing bolts, carefully poking around trying to not break anything, and using a nylon mallet if I had to push harder.

To my amazement, it came apart easily enough, and looked deceptively simple. While doing it, I found the pump was completely drenched in oil on the inside, and thought that that was what stalled the motor. After all, the pump is a compressor, and oil is not compressible, right?

The pump was so old, it still used cork seals for vibration dampening (not the high vacuum seals)

So, once the pump was completely apart, it was given an all nigher in a KH7 bath, (except the pieces with plastic seals, wich where given like a three hour rinse.

Note how the rotors and rotor howsings are either empty or not towching anything, to prevent damage.

Then, proceeded to reassembly it, wich, to my delight, was absolutely uneventful.

Oil was painted on each sealant ring and metal sliding surface (it has bronze bearings backed with plastic oil seals, and also, everything is sumbersed in an oil bath), and the pump looked fantastic. There where some deceptive bits I didn`t understood at that moment, but became clear later on.

So, with the pump completely clean and reassembled, I charged it`s oil bath and…BAM…stalled motor again.


4.- Franken-pump.

Just for the heck of it, I decided to connect the pump to the lathe in some perilious yet controlled way, with a flexible coupling! don’t worry. I had either reassembled it wrong (highly doubted that) or had some problem, OR, most likely, the motor could only run on three pase, or was a dud.


See that??? That’s 60 microns right out of the bat! A-Mazeing!!! (pump is designed to run at 1400 rpm, however, my lathe just does 1100)

So, what now? Well, it was clear I had to replace the motor. Given that it looked like it had a standard ring connection, I looked out for electric motors on my country ebay, because in my new town, noone seemed to know where to buy an electric motor. Found some interesting prices, and just out of curiosity, looked where that shop was, and amazingly, they where just around the corner!, so to speak. (60km)

Funnily enough, my new motorbike backpack paid itself at that moment, because it had a recess where I could tuck the motor for transport, so the person in the shop could see what we where dealing with.

Got it halfprice new from a nearby seller. (new ones have a horrid text on the sides, so, double win)

To the shop I went, and to my dismay, the motor HADN’T a standard connection. Changing it for a more powerful, or different one, wasn’t much of an option without heavy machining, altough I stated that I had the tools to do it, need it be.

But, you know when a shop owner really, I mean, REALLY listens to you and your problem? I could see his brain trying to work a solution for me. And indeed he worked it out. The motor was fine, however, using a capacitor to cheat a three-phase motor, reduces it’s performance dramatically. Once that was established, he raised the “wait for it finger” and came back with a Variable Frequency Driver.

However I got afraid that the VFD was going to be a lot more expensive than the motor (wich they normally are, btw). However, that was secondhand (with warranty) and just 80€, wich was less than a new motor, however cheap, and zero machining.

Fun thing, he must have caught my eye on being slightly above average intelligence, because just said to me that the manual for it was freely avaliable and I could program it without problems, even after me saying that I had never programed one, nor had much knowledge. There was no doubt in his mind about me NOT screwing up the machine.

So, there I went back, fully loaded in my motorbike with an electric motor (6Kg motor + 3Kg VFD) and my hopes really high. (motorbiking with such weight in your back is very tiring, HOWEVER, my backpack just sits on the passenger seat of the motorbike, therefore relieving me from the load, nice, isn’t it?)

Went home, looked at the manual, scratched my head, looked at the manual some more, understood it has codes and such, programmed it to run at 50Hz, (55 max.), tried the motor without pump to confirm direction of rotation was good, and crossed my fingers:


That’s a fantastic vacuum! On further tests, the pump settled between 40 and 50 microns, wich is an awesome vacuum, nonetheless. Mind you, at 700 microns you can already start a turbomolecular and at 200 microns, you can start an oil diffusion pump, both for very high vacuum, so, I am ecstatic.

So, what now?

There is an O-ring I really would like to replace (altough I don’t expect it to make any difference, but well, replacing o-rings is always good) and I am waiting for some materials to build all the HVAC plumbing (everything KF-16), so, until then, everything is well stored and ready.

See ya!

2 responses to “Suck-cess!

  1. Thank you so much for explaining such detailed work on the vacuum project!

    Keep it up!! :)

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s