Vacuum

Well in that case

I am wondering if anyone has ever improvised things such a kickback flasks ,washing bottles, oil pump exhausts/filters ?
Currently I'm designing a system for my lab and some good suggestions are very welcome.

This is the next pump I added to my vacuum station, around a grand used rebuilt, good to 0.1 micron. So I have a single stage rotary for rough to around 70 microns, a new 2 stage to one micron, and If I can work the plumbing out I will add the 0.1 micron Alcatel. It runs very quietly. For now I think I will just clean it and do an oil change, test it to see how well it does. Will take some plumbing work to keep a nice short, large diameter route into the base of my bell jar platform but worth it. Best I could do before this one was around 1 micron. This hobby is expensive, a grand to move the decimal point by one place from 1 to 0.1.

The Alcatel only had a few days worth of run time with the oil kept clean by changing, then was stored for a few years. Kind of wondering if I should buy the rebuild kit before I start using it because it was stored (but never run) in a desert sandy location. Every time I went to that storage room sand was across the floor from blowing in under the door. I worry that there could be microscopic particles of sand in there even though I flushed it out and put new oil in. The exhaust muffler only had a cloth plug in it to keep dirt in the air out. Pic below is just off the web, when I get around to it I'll take a pic of the one I have here (identical to the advertising pic below).




[Edited on 12-29-2010 by IrC]

Quote: Originally posted by IrC

On the subject from another thread I must add my Plexiglas problems were insurmountable. I was buying it from various local hardware stores. Several attempts and every time I had problems with cracking with the slightest pressure, and epoxy failing to adhere. I am convinced this is due to the formulation and/or manufacturing of the Plexiglas now coming from China.

I don't pump down any chemical experiments, mostly physics type experiments. Looked at the price and it is high for that oil. For anything chemistry related I have a few Gast diaphragm types which would survive corrosion better and in general are inexpensive pumps. Thinking about removing the top pump as it barely does 150 microns and I need room for large tubing for the Alcatel. Besides I have learned through experience it is better for two pumps to be closer together in maximum vacuum capability.

Below are a few pics of my system. I built it for getting down in the 55 micron range and providing various plumbing to allow gasses in for laser work. Wanting to play with the Farnsworth Fusor idea I decided to find a way to include the Alcatel pump. That thing is 10 times less noisy than even my new two stage pump. Very smooth even if it could use a touch up paint job. I cannot measure the vacuum with it, will have to get some better gauges as it pegs everything I own. Looking at Pirani gauges and looking at plans to build a Capacitance Vacuum Gauge, still undecided.

Finished my top but am waiting for a freon leak checker (off ebay) as I cannot get below 100 microns due to a leak somewhere. I could not think of an easier way to look for tiny leaks. I still wonder if extruded Plexiglas is the reason. Next I want to build a cage around the bell jar just in case.

One last thing. Does anyone know where to get user info for Alcatel? Damn hard to find, with the name connected to Lucent in a completely unrelated technology. I am looking for oil level in the site glass information. What is full, over, under, etc., and off or running when you look in bubble in sight glass. Refilled it after flushing and is centered in the red circle on, raising to top of glass off. Looked natural that way, took exactly 1 Qt to reach this level. No idea for sure and with these things going for up to $700 on ebay broken - as is, I sure do not want to make any mistakes. Worth much more in perfect running condition as this one is. I never use my high quality pumps until I have already reached the 150 micron range just to keep them in top condition.



















[Edited on 1-3-2011 by IrC]

{edit}I only saw your reply mid typing this, so I realise you've already got some of it covered and that the neoprene is for the chamber seal. Neoprene is not a nice material at all. It's far too weak for hard chemical exposure, and it is porous yes. Whenever I think of it, I always get a mental image of a wetsuit. What are you thinking of doing in the chamber? You could start with natural rubber, silicone or polysulphide sealants. I'm not sure how they fair up in terms of numbers, but I suspect they'll all be better than neoprene. You could also try levelling the base of the bell jar and using a high vacuum grease instead. Some searching around can net you a desiccator that has a ground flange for use with grease. If you find the pressure isn't dropping how it should, my first port of call would be the copper fittings. If they're compressions, they are likely leaky and a soldered option may be the way to go.{/edit}

This is an old, but very nice to read, manual on vacuum, showing how the guys began working on it back in the day - rated A* by John

Hi IRC,

I have taken one of those Alcatel's to bits, screw by tiny easy to loose part.

If the pump has been barely run, I doubt it will desperately need a rebuild and you'd be best off checking what pressure it's reaching first.

As I have mentioned before, physics is far more demanding on vacuums. A prime component of the vacuum it creates is the oil in there, as it can end up full of solvent and water. Of coarse, if there is some 100C or below boiling thing in the oil, it will be trying to boil off and get out the exhaust as you run the pump; so the pressure won't drop to it's full extent until all of that is gone (vacuum pump oils were developed to have very low vapour pressures so they don't start off gassing under vacuum in the pump head). If you've just got the pump, and have a gauge handy, try connecting the gauge to it and leaving it running overnight to see if the vacuum increases over 12h or so. It will drop a little normally anyway as the pump head warms up, but if it keeps dropping all across the 12h period, there'll be something in the oil. It can take absolutely to get moisture out, so the gauge may just sit there and not move. Even lighter oils will create a problem, especially as they will likely have a higher specific heat capacity and so will take longer to boil off.

Mechanical dial gauges are okay for chemistry, but not much help for physics, because they can't pick up the changes very well as they approach the maximum of a rotary - the typical dial gauges can't pick it up at all actually, as most have at best 1mBar divisions, and the rotary can theoretically reach ^-3 mBar.

I was going to say, if it's been in the desert, that may be a good thing due to the heat and lack of humidity. But it depends where and how it was stored and what it was used for I expect.

Looking at the pump, I see signs of rust on the external components. That suggests it's been somewhere fairly damp, as I've not seen the rust on mine and 'grey drizzle' is the predominant weather system in the UK.

If you look at the photo I've included at the end, you can see the insides of these pumps (Alcatels) look a lot like a sandwich, with blocks that slide together. You can also see, they're CAKED in rust. This is quite normal in my experience, and is a sign that damp air has been running through it - e.g. people have been leaving the pump running in a workshop as they get something ready, or they've been using it to boil off moisture.

Opening a pump and finding it looks like this is not a great experience, but it's also less of a problem than it looks. The actual chambers inside are usually still shiny. The outside of the pump head rusts because the film of oil drips off it when not in use and the pump heads are often cast iron.

When you say neoprene seal, do you mean the one around the oil box (the orange thing)? That shouldn't actually be exposed to the vacuum, it's more like a collection pan that the oil sits in.

I would say you first goal should be a gauge so you can actually CHECK the pressure it's reaching, which is real important for physics work. If you're predominantly interested in lower pressures, there are a number of HVAC gauges you can buy that use the convection / thermal conduction properties of gas for measurement.

These are usually a lot cheaper than the pirani's in the link, but will read to 0.01mBar and can swap between units on the display. In my experience, they start reading at about 25 - 50mBar. Being able to swap between units so easily is helpful when every person you speak to seems to have their own preference for Pascals, PSI, microns, torr, mBar, ATM and so on.

If you search eBay for 'digital vacuum gauge', you'll see the HVAC ones come up at about $80 to $160.

You are right to save that pump for physics and use the diaphragms for chemistry. Provided they can get down to aspirator pressures, that's all that's needed for distillation, filtration and so on, and they are far easier to make chemically inert.

I believe quite a lot of semiconductor labs are now ditching their rotaries for alternative rough pumps (oil free) so they don't have to think about oil back streaming into the ultra high purity silicon as it's being processed - which is a problem for those guys as 5 minutes of reading about what they get up to demonstrates.... they need it to be clean, super, duper clean to an atomic scale. They've been using turbo pumps for quite a while over diffusions for the oil reason, but it's only been more recently that the vacuum companies have started coming up with suitable replacements for the rotaries that usually back them.

I have a turbomolecular pump (as soon as the guy gets back about paying for it), but the leap in complexity of parts and costs is gigantic from chemistry to that. For instance, normal seals aren't okay, and you have to start using bits of indium wire, copper gaskets, massive ISO flanges to avoid choking, the gas behaves in a none fluid manner, bake out ovens, metal hoses, at least two different (expensive) gauges. It's normal for an ultra high vacuum system to take a week or two of constant baking and pumping before it actually reaches the full vacuum; due to boil off inside the system and the time it takes for the remaining molecules to randomly bounce into the diffusion or turbo pump's intake.

Random is the word here, as they no longer behave like the gases we can suck out of a straw or with a rotary. Part of the reason for the bake out is to encourage things to boil off but also to increase the velocity of the molecules left in the chamber, so they are more likely to make a collision with the walls that will bounce them at an angle towards the high vacuum intake (because are making more collisions per unit time than they were at a lower temperature).

Due to the way the motion of the gas changes at this pressure, these pumps can't really be connected to the system by hoses, as it's effectively like trying to hit ping pong balls through a window tens of meters away - almost all of them will miss and bounce back. This is why diffusion and turbo pumps have such wide openings on the top and go on the bottom of the chamber. The length of the connection between the pump and the chamber is also important - it usually shouldn't be more than the flange is wide.

Being able to establish an ultra high vacuum has been one of the drawbacks of the new silicon technologies, like quantum structures. The huge amount of time involved in pumping the chamber down makes it slow and costly, and risks a lot if something goes wrong with a wafer in there.

Making your own ion gauge was something I was thinking about as well, but didn't put too much thought into it as I still have lots and lots of other bits that need collecting before it's going to reach ion gauge pressures. You can't switch on ion gauges and such until the pressure is right down. I think they tend to use iridium wire (a platinum, but with more mechanical strength to avoid warping of the electrodes / distortion / noise).

I would suggest you also ignore the ion gauge stuff until you have a system approaching a level that will need one - which means a diffusion or turbo pump and all the related components / seals. You need to be using metal components for that, plastics and rubbers off gas too much as you enter that region of pressure.

Calibrating a DIY ion gauge will also be an interesting challenge, as what do you have to calibrate it against?

I was thinking my first bit of high vacuum physics would likely be PVD, and sputtering some mirrors - for a laser / telescope.

If the pump has been used to move cylinder gas, it's could just be rusty on the outside. Even if there is rust on the pump head, provided the actual chambers are clean and free of sand, it'll still be okay.

Any thoughts, I'd be interested to chat,
John

When moisture strikes.


[Edited on 20-1-2011 by peach]

I'm not sure, but I know tube fanatics will pay serious money for some of that stuff.

A top end hi-fi from these guys is going to set you back a good chunk of a million quids, or more. A lot of it is full of silver, since it's ever so slightly better at conducting*** or uses cobalt / mumetal for the magnetics

Prices? If you have to ask, you're too poor. Get on with listening to your filthy 'dubstep' and 'rock and roll'.

I love the guy in the videos. It's what I'd like to be doing at his age. He's not just messing around painting kits, he's got all his stuff together and is doing something really well. Something I haven't seen other people doing.

***During the development of the first nuclear weapons, the US government had everyone hand over their silverware to build the coils for the separator.

[Edited on 24-2-2011 by peach]

Yes that was amazing as hell... Also yes tubes these days are stupid expensive. Russia only recently started again making tubes. They had stopped after the fall much to the dismay of audiophiles. People were recycling used tubes from old radar equipment for audio systems for a while.

Oh and rock and roll sounds amazing through tubes. So does dubstep I might add . Tubes generate harmonics not heard in transistors, or rather in a transistor it distorts with random interference. In a tube the distortion is harmonic to the sounds going through it.

This seriously makes me want to make my own tubes as well. I buy them when I get a commission for a new amp. Peach is as usual correct. This slab of black walnut and the tubes shown will be combined with transformers (pure silver wire wound toroids, that's getting to be the price of the tubes themselves) to produce a tabletop tube amp. Price tag, over 1000 dollars for a simple design.

Oh it's not for me , I agree. You also missed the freaky dangerous voltages required on said monster tubes. I didn't choose the tubes I just build what the customer wants. In this case it's not going to run as a pentode, single ended triode instead.

I add though that compared to transistor sound... I'll take tubes from a crashed aircrafts avionics

Oh god, pole pigs. Never screwed around with one of them, yet. The large ground transformers from say a school, yes. But not neighbourhood powering monsters. Tesla coil guys are in that category of people I respect even if I have more knowledge in general (not you, I have friends that are like my life, insane). It's like working with venomous snakes. Your still alive when you really shouldn't be, kudos

Hey, your into tesla coils. What is the attraction? I mean super insane high voltages(duh), but is there a quest for science behind it ? Have we not yet achieved some kind of dimension warping energies that are being sought ? Or is it just 12 foot arcs for the fun of it ?

The Sandia Z machine not withstanding, which may actually be tapping some as yet unknown energies.

A - dimension warping energies that are being sought
B - super insane high voltages
C - just 12 foot arcs for the fun of it

A - What I tell people

B, C - the secret truth

However it may be I just like toys.








[Edited on 2-25-2011 by IrC]

Quote: Originally posted by starman

Do you think its worth the money for the one with the pirani sensor(inficon about $190)?

Starman,

You will likely find most of your high boiling points drop to around 100, depending on the pump, the oil, your condenser and how well sealed up it all is. You can afford to go down to around 30C before a tap water condenser will start entirely loosing the distillate. It is better to be a few tens of degrees higher to be sure, particularly if you live somewhere hot, where the distillate flask could warm up to 30C. A pump with a ballast on it is handy for avoiding that, so you can raise the pressure a little if the thing isn't condensing properly. If it doesn't have one, you can always add it.

The problem with rotaries is volatile solvent. E.g. removing it all post reaction you'd need a dry ice / LN2 trap or it'll go straight through the condenser and on to the pump if it's not been ballasted up a significantly higher pressure.

I've listed the rough minimum pressures you can use with a tap water cooled condenser for a few different volatile solvents here, if you'd like to have a peruse to get an idea.

Brass should be fine, but I'd tape up the threads since it costs tens of pence for a roll and will only help with blocking any remaining gaps in the threads.

[Edit]Watson is correct that Pirani's go wonky as the pressure goes up. For air and nitrogen, they start doing it around 100mBar, but are still somewhere near linear up to 1 atm. Above 1mBar, they're scrambled if the gas is significantly different to air or nitrogen; e.g. argon, or whatever you're distilling.

Those HVAC gauges, I've taken one apart and they do use a hot element. The one in mine also had a little thermistor of some description poking into the side of the tube the element was in. So I would assume they are single wire Pirani's, which behave in a similar way.

It depends a lot on what you're distilling, what information you need to know from the distillation and what else is in there. If you're just trying to clean up something where you know the majority of it is your material of interest, you don't necessarily need a gauge to begin with, just a stable temperature on the thermometer. If you're trying to split close boiling points of not well known mixtures, and use the temperature as evidence, you do need a gauge. It also needs to be one that displays an actual number, not a region of pressure. You can check rough purities, for example, without a gauge, based on whether or not the temperature sits still in the first place and if so, how long it does it for compared to the rest of the distillation. You can also add ballast to raise the BP so it'll condense better without a gauge, but again, you can't then work out what is coming through without knowing the pressure, so you'd have to know before hand or split samples and test it afterwards by some other means.

If you're distilling something that boils at 400C at 1atm, you can manipulate the temperature by around 20-30C whilst staying in the linear region of a hot gauge. But for wider variations, it's unlikely to be happy.

Unfortunately, the range from distilling volatile solvents to high boilers is within the range where you could really do with more than one gauge. So, unless you need to know the temperature and pressure simultaneously, you may be better thinking around it by other means. Quite a few things can be overcome by ideas like contras below. Set the temperature you want to work at, drop the pressure until it appears. Or, if you're splitting something and know the substrate boils cooler than the product, discount the first fraction through.

Here's an example of a Pirani going off the rails.


[Edited on 6-3-2011 by peach]

Quote: Originally posted by Lord Emrone

I regularly dry ammonium nitrate (completely dissolved in water) on my stove, but it's inside and produces a lot of moisture in my house ... Maybe I should drive the water out with vacuum drying (outside), but I'm not really into chemistry ... Is there a relatively cheap and easy way to vacuum dry or does it require a lot of studying ? Is this pump appropriate for my purposes ? http://www.aircoshop.nl/accessoires/vacuumpomp this with a spherical-shaped flask and tube is all I need then ?

Quote: Originally posted by Lord Emrone

... but I'm not really into chemistry ...

Quote: Originally posted by Picric-A

Quote: Originally posted by Lord Emrone   ... but I'm not really into chemistry ... If your not really into chemistry what are you using the Ammonium Nitrate for?

Last week I was at my favorite surplus lab equipment store and I picked-up a brand new vial of Beckmans vacuum grease. it's listed in every lab supply catalog at around $50 and up. Only paid five bucks for it! God I love that place!



Anyway my point is this... for a while, I have been using for my ground glass joints some "silicone dielectric compound" from Canadian tire at $7 a tube. That's the stuff used to shield electrical contacts in automobile connectors from moisture and oxidation.

I submitted both compounds to a series if tests, including concentrated HNO₃, H₂SO₄, HCl, NaOH and a few solvents like Heptane and Isopropanol and it didn't react to any of them. I tested both greases on the ground glass surface of my dessicator under vacuum and it didn't try to creep off or suck in. Both compounds are remarkably similar. So far, the only difference is the outrageous price of the laboratory grease...

I'm always baffled by the price of labware and equipment from the official big-name lab suppliers... How do they get away with such outrageous prices? Maybe that's why meds from big pharmaceuticals are so stupidly expensive.

P.S. Sigmhumm-Aldrhumm sells the dessicator in the pic above for something like $600. I paid $25 for it, including a one-pound bottle of blue drierite!

So in conclusion, if you want to save a few bucks on vacuum grease, in Canada, go to Canadian Tire, or in the US to your local NAPA auto parts store and get the silicon dielectric compound!

Robert

How to best use RZ2.5 Rotary

Quote: Originally posted by trb456

So far, I have been unable to prevent emulsifying the pump oil, which I understand means water vapor and general badness. [...] In my last test, I used a cold trap in cold water (iced with salt--I'm not ready for liquid nitrogen), and the cold trap did have some condensation, but clearly not enough to prevent the white oil. It seems, based upon my limited reading, that this type of pump would prefer an inert gas, which I am not ready to provide.