Foreline valve and trap - what is neccesary?

[ChristofferBraestrup]

Hi all!

I'm in the progress of assembling a turbo pump high vacuum system, and I'm currently debating with myself whether or not I should bother having an isolation valve and/or foreline trap between my rotary vane pump and my turbo.

I've worked on a great many mass spectrometers which had neither. Even fairly large two-turbo TOF-HRMS systems.

This is a small vacuum system, so no direct roughing path is implemented, all evacuation is through the turbo.


The two problems the two foreline instruments are trying to achieve, as I see it is:

A) Foreline trap: catching oil mist

B) Isolation valve: "Backflow of oil" into turbo


A is a legitimate concern, but one that can be solved in a simpler way. As far as I can tell, the longer the connection between backing pump and turbo, the less oil migration.

B sounds strange in my ears. When would it be advantagous to isolate the turbo from the roughing pump? One is supposed to start both turbo and roughing pump at the same time, and at shutdown, the turbo is allowed to spin down to half speed, then a bleed valve is opened on the turbo or chamber to vent slightly. Where in this scenario could oil be sucked back into the turbo?

The EXT255H turbo manual specifies that there should be a valve, but honestly I don't see its use.

Has anyone made a turbo-based vacuum system without cutoff/foreline filter, can confirm the stable operation of these? Thanks in advance!

--Chris

[John Futter]

Putting a much longer hose on the forline will slow pumping speed dramatically
not putting a valve on the roughing pump outlet risks oil suck back into turbo -- it is very difficult to remove all the oil from a turbo that has suffed suckback - this usually requires a factory turbo rebuild
especially on small turbos that have greased bearings as the oil washes out the grease

[ChristofferBraestrup]

Longer pump-down time is definitely true. The installations I've seen this on typically runs indefinitely, so plenty of time to reach ultimate foreline pressure.

not putting a valve on the roughing pump outlet risks oil suck back into turbo -- it is very difficult to remove all the oil from a turbo that has suffed suckback - this usually requires a factory turbo rebuild
especially on small turbos that have greased bearings as the oil washes out the grease

Yeah, so I've been taught as well, but litterally every single MS application I've seen that was turbo-based has omitted this, and still achieving 10^-7 mbar without an issue - and oil would be VERY detectable in a mass spectrometer.

The procedure for pumping down an MS system I've learnt is as follows, for a system consisting of [bleed valve]->[MS chamber]->[Turbo]->[roughing pump]->atmosphere:

1. chamber is vented to atmosphere (ie. after maintainance.)
2. turbo and roughing pump turned on at the same time
3. bleed valve closed. system pumps down.

that's it.

for shutdown:
1. turbo is turned off. spins down over ~20 min
2. when speed reaches down to something like 60%, bleed valve is opened
3. when turbo has spun down all the way, roughing pump is turned off and chamber can be opened.


I'm not even completely certain where in this procedure you'd squeeze in opening or closing an isolation valve. Much less if it's needed. That said, I'm not risking my turbo before I'm certain!

[Maciek Szymanski]

From my experience the fore line isolating valve is used in the situation when the turbo is kept under the vacuum. I have some systems without gate valve between the chamber and the turbo, and at the system shutdown the chamber with turbo is isolated from the roughing line. With dry (scroll) pumps just isolation valve is enough, for oil vane pumps it’s usually a combined isolation-vent valve directly on the pump. It allows the fore line to be under vacuum while the pump can be vented.
I’ve observed that keeping as much system as possible under vacuum improves dramatically the final pressure and reduces the pump down time.



The small turbo with the isolating valve at the roughing line connected to the scroll pump. The small chamber is isolated with another valve from the main experimental chamber and kept under vacuum together with the turbo:

3F3F9481-FDA0-4BC0-A2BE-6C5C55A49236.jpeg (117.42 KiB) Viewed 5538 times

The rotary vane pump fitted with an oil trap and the combination valve. The big knob closes the foreline passage. The small one allows venting of the roughing pump to atmosphere. Thos way the roughing line can stay under vacuum while the pump is vented without a risk of oil backstreaming:

ADE314AA-CAB2-4481-8A29-A11C2DA06054.jpeg (134.04 KiB) Viewed 5538 times

[Jerry Biehler]

I have never had a pump suck back into the foreline, not saying it cant happen, it just has not happened to me, they have internal checkvalves which would be about just as reliable as putting another valve in series. You could just use a length of line with a greater volume than how much the pump holds and make sure the turbo is higher.

But if you are running a relatively modern turbo you may have a turbo-drag setup which means you can run these things from a dry pump (diaphragm) and not worry at all.

[Richard Hull]

In the end, when all the smoke clears, if used properly and in sequence, there is no such thing as too many isolation valves! You just have to have the right valves for the job and the bucks to afford good ones.

Richard Hull

[ChristofferBraestrup]

I've had a chat with a MS guy, and it turns out Jerry was right on the money. A lot of rotary vane pumps has a check valve that will catch oil spraying back.

That said, risking a turbo (and hours of cleaning) to avoid buying a valve is risking a lot of money for not that much money. I'm gonna get a valve to be sure.

thanks for helping clear that up!

[ChristofferBraestrup]

Speaking of foreline isolation valves,

The vast majority of affordable ones are some strange mix of pneumatic/electrically actuated, like the Pfeiffer EVB 025 PA

Has anyone attempted to get these working? Could one, in theory, remove the acting motor, and just put a crank on the screw that moves the bellows? Or is it the pneumatics that holds it open/closed?

[Jerry Biehler]

Not a strange mix, thats just your average valve, air operated. They use a solenoid to supply air pressure to a piston which either opens or closes the valve. Technically you could convert one to manual but it is more trouble than it is worth. And those are generic valves. These are true pump isolation valves, they just require power when the pump turns on to operate:

https://www.idealvac.com/files/manualsI ... Valves.pdf

[ChristofferBraestrup]

Ah, I see. I'll scratch that idea then.

I think I'll try and find a hand-cranked bellows-type elbow valve for isolation.

I did learn that my rotary vane pump has an isolation solenoid that closes when the pump is switched off, so I guess one could shut down the vacuum system that way, starting by turning off the roughing pump, then the turbo.

Still a little sketchy i feel.

[Dennis P Brown]

Until you get a proper valve, the simple solution is to leave the mechanical pump running - turn off the turbo; then allow it to spin down (assuming you have no mid-point vent built into the turbo body.) After it stops, turn off the mechanical pump and let air into the system. No oil vapor back flow issues (just water vapor but for most fusor's, that issue isn't too big a problem if you first run the system down to the 10^-5 torr range before opertaion.)

[ChristofferBraestrup]

Okay so this is exactly what I wanted to do in the first place. I thought the higher vacuum once the turbo slows would 'suck' oil back from the roughing pump even though it kept running.

And I do have a mid-point turbo vent valve, both solenoid and manual.

[John Futter]

you should also have an activated alumina trap in the roughing line to stop oil vapor from the backing pump backstreaming into the turbo wwhen at high vac with no leaks

[ChristofferBraestrup]

I should, yeah, but that is less critical as far as I know. The magnitude of oil is a trillion times lower than sucking half a litre oil back into the pump!

I wouldn't be terrified about pumping down the system without a foreline trap.