Procedure to shut down a system without a gatevalve to the diffusionpump?

#3669

Hello guys!

I have finally built together a small simple high vacuum system with a diffusionpump and the vacuum only goes up from little under 1 milliTorr to 1.2 Torr in 2 days when all valves are shut and that is fine with me so it is holding the vacuum ok as i see it, might also be some degas involved.

I want to test the diffusionpump ASAP so i can start making small mirrors but i dont know if the shutdown procedure i think of is right to do without a gatevalve.

The diffusionpump is connected directly to the chamber the rest is built with one valve to its discharge and one valve to the chamber and these two are then connected to a tee and then the this is connected to the backing pump with a bellow tube and thats it.

To start it all is of course very easy but to shut it down with only two valves is something i want to be sure on if it works as i think of to do so without getting oil into the chamber or anything.

This is what i think might happen and/or will work and when i shut it down it all has to do with the diffusionpump as i see it:

As long as it is hot it pumps and thus must have the backing pump going all the time i guess or could the discharge valve be closed right away or as soon as it is only a little cooler? The very small amount of air in the bottom wont cause any oil to go into the chamber since this air will try to go up into the chamber and draw any oil from the oil curtain with it?

Or should i carefully monitor the discharge pressure and as soon as it starts to go down i shut the valve since i then see this to indicate that the dp is so cool that it cant hold down the air in the bottom anymore so air is then starting to be sucked back through the pump into the chamber from the discharge?

Will any of these scenarios work or will it pull back oil into the chamber or doesnt it matter since there might be so little air in the lower part of the diffusionspump that there wont be any problems at all no matter what i choose to do?

Im very concern of this so it wont go south very quickly all of a sudden, so any help with this would be highly appreciated so i know if this might work with only 2 valves and no gatevalve to the DP.

Thanks a lot!

Regards Henry

#3670

Henry,

You done a great job! Please don't destroy it trying to use difpump without gate! There may be a lot of scenarios how to use it without gate, and on practice they all have  negative results, sooner or later you run into a great troubles! You can try to estimate all risks, but IT NEVER WILL WORK FINE! Neither when it starts up, nor when it shut down. Don't forget that oil vapor is a serial killer for coatings, especially for metal ones.

I guess your pump doesn't has 500 mm input flange. So go to any online auction and you'll find cheap (sometimes very cheap) gate valve with manual control. I also would look for water cooled baffle. Install gate valve at least and enjoy! Trust me, you'll save more money, time and your own nervous with any gate valve!

Good luck!

Vladimir

Show quoted text

On 7/23/2011 4:24 PM, henry_3507 wrote:
Hello guys!

I have finally built together a small simple high vacuum system with a diffusionpump and the vacuum only goes up from little under 1 milliTorr to 1.2 Torr in 2 days when all valves are shut and that is fine with me so it is holding the vacuum ok as i see it, might also be some degas involved.

I want to test the diffusionpump ASAP so i can start making small mirrors but i dont know if the shutdown procedure i think of is right to do without a gatevalve.

The diffusionpump is connected directly to the chamber the rest is built with one valve to its discharge and one valve to the chamber and these two are then connected to a tee and then the this is connected to the backing pump with a bellow tube and thats it.

To start it all is of course very easy but to shut it down with only two valves is something i want to be sure on if it works as i think of to do so without getting oil into the chamber or anything.

This is what i think might happen and/or will work and when i shut it down it all has to do with the diffusionpump as i see it:

As long as it is hot it pumps and thus must have the backing pump going all the time i guess or could the discharge valve be closed right away or as soon as it is only a little cooler? The very small amount of air in the bottom wont cause any oil to go into the chamber since this air will try to go up into the chamber and draw any oil from the oil curtain with it?

Or should i carefully monitor the discharge pressure and as soon as it starts to go down i shut the valve since i then see this to indicate that the dp is so cool that it cant hold down the air in the bottom anymore so air is then starting to be sucked back through the pump into the chamber from the discharge?

Will any of these scenarios work or will it pull back oil into the chamber or doesnt it matter since there might be so little air in the lower part of the diffusionspump that there wont be any problems at all no matter what i choose to do?

Im very concern of this so it wont go south very quickly all of a sudden, so any help with this would be highly appreciated so i know if this might work with only 2 valves and no gatevalve to the DP.

Thanks a lot!

Regards Henry

------------------------------------

Yahoo! Groups Links

#3671

Vladimir,

Thanks for the answer!

Yes i would like a gate or butterfly valve there absolutely and i have searched. The only place i know is Ebay and the only one i found that would fit, and most likely pretty much the only one, goes for over 240$ incl. shipping. And 240$ is out of my economy since a lot of money have been spent on the rest, but sure cheap for a valve like that. Acceptable for me would be in the 50$ range incl. shipping but a NW50 gate or butterfly valve at that price would unfortunately most likely not exist as i see it.

The simple system i have is actually from a mass spectrometer incl. the chamber, diffusionpump and Bayard-Alpert tube with its controller but missing all the connection tubes, backingpump and so on that i had to look for during several month now to get.

The bits and pieces of this mass spectrometer had been all taken apart, not by me, and put in a boxes so i dont know if all the things to the diffusionpump were there like a valve but i think it was since in the other boxes what that belonged together were together. The only thing that was missing were one half of a little special composite sealant to the diffusion pump and that was originally the only bit i needed to put the DP to the chamber the first time, now i have bought a aluminum from Ebay. If there had been used an isolation valve i see as this would have been in this box.

The pump used for this mass spectrometer is a small cute Edwards EO50/60 and in the box there where no isolation valve what so ever but everything else that directly belonged to the chamber which make me think that it has been built without a such. Mass spectrometers are of course very sensitive to foreign gases and materials so it wont disturb the real measuring but the most likely lack of an isolation valve makes me think that due to the design of the chamber and the small pump used this wasnt necessary to use.

Just above the pump inlet there is several fins acting as a oil baffle sloping down to the DP for the oil to go back to it and the chamber it self is built like an L turned 90 degrees to the right where the pump then is hanging in the left bottom sort of speak. This makes it 100% optical dense so no oil at all should be able to get into the chamber even without a isolation valve to the diffusion pump.

But anyone would more then gladly tell me wrong about this so i know because this is only a theory of mine, i havent been into high vacuum before.

I might be able to put up some picture on it if needed if there might be something that is unclear or so.

Regards Henry

#3672

Henry,

Actually for 50 mm flange you can use any 50 mm flange valve, but you're right - it is practically impossible to find something for $50 including shipping... Just shipping cost may be more than $50 depending on your location.

Regarding your idea to use L type elbow - well, it may work.  At least theoretically it must work in the molecular range of gas flow, i.e. at pressures something like below 5x10^-4 - 10^-3. What happens at higher pressures, in viscous range,  - I can't say, all depends of temperatures and oil vapor pressure. I have no such experience, sorry. Hope your backing pump will pump down oil vapor. But don't forget there may be strong oil back migration via the pump and other components walls.

OK, you can try to use L shape elbow as a gate. But I would also cool it down with water pipe welded or soldered outside along its and insert inside additional optical dense plates (2-3) to work as a baffle.

Best regards,

Vladimir

Show quoted text

On 7/24/2011 5:48 PM, henry_3507 wrote:
Vladimir,

Thanks for the answer!

Yes i would like a gate or butterfly valve there absolutely and i have searched. The only place i know is Ebay and the only one i found that would fit, and most likely pretty much the only one, goes for over 240$ incl. shipping. And 240$ is out of my economy since a lot of money have been spent on the rest, but sure cheap for a valve like that. Acceptable for me would be in the 50$ range incl. shipping but a NW50 gate or butterfly valve at that price would unfortunately most likely not exist as i see it.

The simple system i have is actually from a mass spectrometer incl. the chamber, diffusionpump and Bayard-Alpert tube with its controller but missing all the connection tubes, backingpump and so on that i had to look for during several month now to get.

The bits and pieces of this mass spectrometer had been all taken apart, not by me, and put in a boxes so i dont know if all the things to the diffusionpump were there like a valve but i think it was since in the other boxes what that belonged together were together. The only thing that was missing were one half of a little special composite sealant to the diffusion pump and that was originally the only bit i needed to put the DP to the chamber the first time, now i have bought a aluminum from Ebay. If there had been used an isolation valve i see as this would have been in this box.

The pump used for this mass spectrometer is a small cute Edwards EO50/60 and in the box there where no isolation valve what so ever but everything else that directly belonged to the chamber which make me think that it has been built without a such. Mass spectrometers are of course very sensitive to foreign gases and materials so it wont disturb the real measuring but the most likely lack of an isolation valve makes me think that due to the design of the chamber and the small pump used this wasnt necessary to use.

Just above the pump inlet there is several fins acting as a oil baffle sloping down to the DP for the oil to go back to it and the chamber it self is built like an L turned 90 degrees to the right where the pump then is hanging in the left bottom sort of speak. This makes it 100% optical dense so no oil at all should be able to get into the chamber even without a isolation valve to the diffusion pump.

But anyone would more then gladly tell me wrong about this so i know because this is only a theory of mine, i havent been into high vacuum before.

I might be able to put up some picture on it if needed if there might be something that is unclear or so.

Regards Henry

------------------------------------

Yahoo! Groups Links

#3673

Vladimir,

I think you misunderstood me a little. The whole chamber which is flat in contruction but long and where the diffusionpump is connected to in one end looks like an L that is turned 90 degree clockwise so its a natural baffle by it self. There has also been made an internal oil baffle to block out even more oil vapor if needed so its pretty much a double oil baffle. However it doesnt seem like any of this has been cooled more then by convection but i might put something on to do this since it cant hurt.

Thanks for your input and help!

Regard Henry

#3674

Henry,

Yes, I see now... The pictures would be great to avoid misunderstanding. Anyway, If your cool down your baffle and walls around it, it just will work better.

Best regards,

Vladimir

Show quoted text

On 7/24/2011 7:26 PM, henry_3507 wrote:
Vladimir,

I think you misunderstood me a little. The whole chamber which is flat in contruction but long and where the diffusionpump is connected to in one end looks like an L that is turned 90 degree clockwise so its a natural baffle by it self. There has also been made an internal oil baffle to block out even more oil vapor if needed so its pretty much a double oil baffle. However it doesnt seem like any of this has been cooled more then by convection but i might put something on to do this since it cant hurt.

Thanks for your input and help!

Regard Henry

------------------------------------

Yahoo! Groups Links

#3675

Vladimir,

Yes that true i should have put them up before. I put up some pictures on it now if perhaps anybody else want to comment on this or so.

Yes i think ill go with to cool it down in some way, it wont hurt.

The first picture is the whole system, and when i take of the lid to the left far left that is the oil baffle that sits just above the diffusionpump it looks like the other two pictures. It feels like its a good contruction on preventing oil from entering the chamber.

Just copy and past in the address bar.

Regards Henry

Henry,

Yes, I see now... The pictures would be great to avoid misunderstanding. Anyway, If your cool down your baffle and walls around it, it just will work better.

Best regards,

Vladimir

开云体育

Procedure to shut down a system without a gatevalve to the diffusionpump?