Hi,

I would like to make a small oil diffusion/vapor jet pump out of st/stl. preferably a two stage pump based on an existing working design.

and i was wondering if anybody here has any dimensioned drawings that i could base it on or could point me in the right direction to any books with details in.

I did manage to find a few drawings online but they all seem to be for large 4" pumps

Any help would be appreciated
Ross

Our vacuum aluminizer at the Chevy Chase Community Center's amateur
telescope making workshop has given me quite an education over the past
few months or so.

We began finding that the pressures would not go down to the required
levels for a good coating, last year -- but intermittently; sometimes it
worked.

(How can we tell? Discovered that unless the pressure on the Varian Cold
Cathode gauge goes below about 8 or 9X 10 ^ -5 Torr, then the coating is
somewhat transparent - you can see lights in back of the mirror. We were
only getting it to about 1.5 x 10^-4 torr, which one would be forgiven
for thin\king are not too far apart, but it was too far for success.
Plus, in the Strong Procedures in Expewrimental Physics and elsewhere
tab les are given for mean free path at various vacuum pressures; at
1x10^-4 torr the MFP for nitrogen at 0C is about 60 cm, or 2 feet, and
that's about the distance between our filament and the mirror surface.
So the gauge and the book and the mirror coatings roughly agree. By the
way, the scotch-tape test shows no pull=off of aluminum in either case)

Tightened up hose clamps on the rubber hoses, then found that sometimes
cuts the hose! So, doubled up on clamps and don't tighten them so tight
- lesson imparted, lesson learned. Also replaced hoses made by me,
earlier, from auto parts store with proper thick vacuum rated hoses;
David H cleaned out the mechanical vacuum pump, replaced the oil, fixed
a number of unsafe electrical features. Sam C added a bleeder valve from
mechanical pump and added a senson, discovered that the two of the
sensor gauges were contaminated and thus giving false readings. The
sensor leading to the bell jar was full of aluminum flakes and the one
leading to the back of the diffusion pump was full of DP oil. Was told
NOT o clean them out with compressed air because it would destroy the
filament.

Discovered bleeder valve itself was leaking, so soldered in two new ones
in series for reliability in the line from the mechanical Sargent pump.

David H removed a mechanical passthrough and examined it, began fixing
it, and decided that since there was no need for it in the first place,
it would be better to plug it with a plumbing part.

John P discovered that the 'monnkeyshit' (technical nickname, I'm told,
for Apiezon Q vacuum putty) that was keeping the home-made high-voltage
passthroughs not leak, had failed. Located and ordered and received and
installed new commercially made ones for about $90-120 each at Lesker,
learned about 1/2 national pipe thread angles and threading facts,
ordered enough O-rings (in lots of 100 each, from McMC) to keep us
supplied for many lifetimes. David H tried warming up the hard wax used
to make the homemade ones that Jerry S and Bob B had made many years ago
that had sprung leaks; if David's method works, and the passthroughs
from Lesker fail, then we have possible backups, as well as some of that
monkeys##t....

Friday night, John helped clean and re-assemble and tighten and pump
down the system to test it, as I was helping other folks make
telescopes. It's working fairly well, but it would only go down to 1.1 x
10^-4 torr, and it needs to be into the 10^-5 torr range in my experience.

I suspect we still need to outgass the entire system for quite a few
hours until it's back working properly , but also that cleaning out the
diffusion pump is going to be needed as well. Then rebuild the
high-voltage ion plasma geneator, using our copious photographs and our
carefully kept containers of parts as guidance.

Without the assistance and example of Bob B, Jerry S, David H, John P,
and Sam S, Mike F, Alan T, Bill R, and Bill B, and Jean-Paul R, and
others, I could not have gotten as much done as we did. I'm sure I've
left out a lot of steps and gotten stuff wrong, but I would like to
thank all of those kind souls.

Guy Brandenburg, Washington, DC


============================

Search eBay for Freon Detectors, many from about $20.00. Haven't tried it but sounds good.

Don Black.

On 28-Jan-14 2:27 PM, Vladimir Chutko wrote:

Drew,

What is the sensitivity of that method? Long time ago I successfully used halogen leak detector to find leaks in high vacuum systems pumped down with a leak to 10-3 - 10-5 torr. The special sensor has been installed in the fore-line at the mechanical pump inlet flange and the vacuum chamber and pumping system were blown with freon or other halogen contained gas (even cigarette smoke works fine). All other was exactly like you described. It was very inexpensive ( as far as I remember less than $1000 a brand new one) small portable device and its sensitivity was high enough - for vacuum coating applications it worked no worse than helium leak detector.

Now I can't find such a device anywhere on sale...

Regards,

Vladimir Chutko


On 1/27/2014 9:53 AM, Andrew Aurigema wrote:

I came up with one but it is not fast or environmentally friendly.

Get yourself a digital Freon ( halogen ) sniff detector off Ebay ( $20 ). Put a sock over the output of your vacuum pump so the discharge does not spit oil vapor. Put the detector in the discharge stream so it is sniffing. Set it to most sensitive. pump down your system to get vacuum in there. I go down to 5 torr. Just crack your vent valve so that there is a tiny leak of air into the chamber...... so there is actually some air flowing into the chamber and that air can be pumped out via the pump. The detector should be silent thru all this pumping as there is no halogen gas in the discharge.

To test the detector, put a plastic bag over the vent and put a puff of "caned air" in the bag. Compressed air is really Freon 134a ( a halogenated gas ) so in a few seconds the halogen gas will be sucked into the system via the valve you have cracked and exhausted via the pump. The detector will go crazy for a few seconds. Just put a short blast of the gas in the bag and let the system run for a minute to clear out the chamber of test gas.

Eventually ( like in 30 seconds ) the detector will be silent again. Retest to make sure your system is working. Let clear 30 seconds each time to clear the chamber ( with fresh leaked in air ). Now you are ready to test. Put the compressed air baggy over any part you are suspecting and give that new suspect volume a blast of caned air. If your detector goes off 30 seconds later you have a very very large leak.

It is not fast, but it is a very good way to find leaks.

Every so often check your test by putting a puff of caned air into the calibrated leak and make sure the detector goes off.

Any brand of "caned air" will work. Just look on the can to see that it is a halogen gas. Check your freon detector against the can before you start to get a feel for the amazing sensitivity of the machine. Any freon sniffer will work so dont be fooled into buying an expensive one.

Hope it helps.

Drew in sunny Florida

On Mon, Jan 27, 2014 at 11:18 AM, Guy Brandenburg <gfbrandenburg@...> wrote:

Anybody have any nice cheap easy reliable leak/detection methods other than
1. Deduction by logic by compartmentalizig
2. Stethoscope
3. Squirting acetone at suspected joints to see if pressure rises
?

Guy Brandenburg
so full of hilarious errors... ;-}}

---

This email is free from viruses and malware because protection is active.

I came up with one but it is not fast or environmentally friendly.

Get yourself a digital Freon ( halogen ) sniff detector off Ebay ( $20 ). Put a sock over the output of your vacuum pump so the discharge does not spit oil vapor. Put the detector in the discharge stream so it is sniffing. Set it to most sensitive. pump down your system to get vacuum in there. I go down to 5 torr. Just crack your vent valve so that there is a tiny leak of air into the chamber...... so there is actually some air flowing into the chamber and that air can be pumped out via the pump. The detector should be silent thru all this pumping as there is no halogen gas in the discharge.

To test the detector, put a plastic bag over the vent and put a puff of "caned air" in the bag. Compressed air is really Freon 134a ( a halogenated gas ) so in a few seconds the halogen gas will be sucked into the system via the valve you have cracked and exhausted via the pump. The detector will go crazy for a few seconds. Just put a short blast of the gas in the bag and let the system run for a minute to clear out the chamber of test gas.

Eventually ( like in 30 seconds ) the detector will be silent again. Retest to make sure your system is working. Let clear 30 seconds each time to clear the chamber ( with fresh leaked in air ). Now you are ready to test. Put the compressed air baggy over any part you are suspecting and give that new suspect volume a blast of caned air. If your detector goes off 30 seconds later you have a very very large leak.

It is not fast, but it is a very good way to find leaks.

Every so often check your test by putting a puff of caned air into the calibrated leak and make sure the detector goes off.

Any brand of "caned air" will work. Just look on the can to see that it is a halogen gas. Check your freon detector against the can before you start to get a feel for the amazing sensitivity of the machine. Any freon sniffer will work so dont be fooled into buying an expensive one.

Hope it helps.

Drew in sunny Florida

On Mon, Jan 27, 2014 at 11:18 AM, Guy Brandenburg <gfbrandenburg@...> wrote:

Anybody have any nice cheap easy reliable leak/detection methods other than
1. Deduction by logic by compartmentalizig
2. Stethoscope
3. Squirting acetone at suspected joints to see if pressure rises
?

Guy Brandenburg
so full of hilarious errors... ;-}}

?

I came up with one but it is not fast or environmentally friendly.

Get yourself a digital Freon ( halogen ) sniff detector off Ebay ( $20 ). ?Put a sock over the output of your vacuum pump so the discharge does not spit oil vapor. ?Put the detector in the discharge stream so it is sniffing. ?Set it to most sensitive. ?pump down your system to get vacuum in there. ?I go down to 5 torr. ?Just crack your vent valve so that there is a tiny leak of air into the chamber...... so there is actually some air flowing into the chamber and that air can be pumped out via the pump. ?The detector should be silent thru all this pumping as there is no halogen gas in the discharge. ?

To test the detector, put a plastic bag over the vent and put a puff of "caned air" in the bag. ?Compressed air is really Freon 134a ( a halogenated gas ) so in a few seconds the halogen gas will be sucked into the system via the valve you have cracked and exhausted via the pump. ?The detector will go crazy for a few seconds. ?Just put a short blast of the gas in the bag and let the system run for a minute to clear out the chamber of test gas. ?

Eventually ( like in 30 seconds ) the detector will be silent again. ?Retest to make sure your system is working. ?Let clear 30 seconds each time to clear the chamber ( with fresh leaked in air ). ?Now you are ready to test. ?Put the compressed air baggy over any part you are suspecting and give that new suspect volume a blast of caned air. ?If your detector goes off 30 seconds later you have a very very large leak. ? ?

It is not fast, but it is a very good way to find leaks. ?

Every so often check your test by putting a puff of caned air into the calibrated leak and make sure the detector goes off. ?

Any brand of "caned air" will work. ?Just look on the can to see that it is a halogen gas. ?Check your freon detector against the can before you start to get a feel for the amazing sensitivity of the machine. ?Any freon sniffer will work so dont be fooled into buying an expensive one. ?

Hope it helps. ?

Drew in sunny Florida

On Mon, Jan 27, 2014 at 11:18 AM, Guy Brandenburg <gfbrandenburg@...> wrote:

?

Anybody have any nice cheap easy reliable leak/detection methods other than
1. Deduction by logic by compartmentalizig
2. Stethoscope
3. Squirting acetone at suspected joints to see if pressure rises
?

Guy Brandenburg
so full of hilarious errors... ;-€}}

?

Someone proposed that I investigate how much it would cost, both time and energy, to add on the capacity to do overcoating (i.e silicon monoxide or dioxide or MgFl or whatever) to a late-1960s Electrodynamics corp vacuum chamber with bell jar that right now just does bare aluminum.

I know my predecessors tried and failed; they were a lot better at electronics than me. I've looked it up in the past in various printed and online sources dating back to the 1950s and I'll admit I'm intimidated. I don't recall any details though. Sounded like the hardest part is determining when you have the right thickness, and you have to use all kinds of weird and strange tricks do determine that thickness indirectly...

Anybody done this? How hard is it? What extra devices did you find that you needed?

Thanks so much for your help in the past. (Our chamber has been doing pretty well for a while now.)

?

Guy Brandenburg, Washington, DC?


============================