Bell and whistles diffusion pump based vacuum system.

Every fusor and fusion system seems to need a vacuum. This area is for detailed discussion of vacuum systems, materials, gauging, etc. related to fusor or fusion research.
User avatar
Maciek Szymanski
Posts: 154
Joined: Wed Nov 20, 2019 7:31 pm
Real name: Maciek Szymański
Location: Warsaw, Poland

Bell and whistles diffusion pump based vacuum system.

Post by Maciek Szymanski » Mon Oct 26, 2020 10:09 am

For some time I’m on the quarantine because my wife has positive covid test (she is ok). To use the time productively I’ve decided to build a complete versatile high vacuum pump system at my garage. I would like to share the building progress and show all the components, as I think it may be beneficial to the newcomers getting into high vacuum world.
The system will be based on the polish Unitra-Unima components. No often to come by outside my country, but for other hand they are quite typical. Almost all the parts I’m using come out of dumpsters at my work (yes, I’ve access to very interesting scientific dumpsters :-)).
The system may look quite complex at first look, and it’s possible to get vacuum without more than half of the hardware I’m using. For other hand this layout is based on the technological setup used in commercial appliances like sputtering systems etc. so it is almost the most flexible and comfortable diffusion pump based system layout.
The main highlights are:
  • It’s possible to vent and pump the chamber many times without need to shut down restart the diffusion pump.
  • The oil back-streaming is minimal.
  • There is no complicated shutdown procedure - just switch off and go home.

E9FF600F-C0AD-44C5-AB46-4558D620740D.jpeg
The diagram of the system: 1. Two stage rotary vane pump. 2. Foreline vacuum canister. 3. Oil diffusion pump. 4. Cooled baffle. 5. Diffusion pump isolation gate valve. 6. Chamber pump valve. 7. Diffusion pump forevacumm valve. 8. Chamber vent valve. 9. Foreline isolation valve and rotary pump vent combination valve. 10. Oil trap. 11. DN100 chamber flange. 12. Cooling water circulation pump. 13. Water reservoir. 14. Foreline pirani gauge. 15. Chamber pirani gauge. 16. Bayard–Alpert high vacuum gauge.

The whole system will be mounted on the wooden trolley and any desired chamber can be connected to the CF100 flange at the top (11).
The gate valve (5) allows to isolate the diffusion pump at any time. So it’s possible to vent the chamber at wish with the valve (8) and keep the diff pump running. When the chamber needs to be pumped again the forevacuum valve (7) is closed and chamber pump valve (8) opened to pre-evacuate the chamber. The vacuum canister (2) keeps the back pressure for safe operation of the diffusion pump. As the chamber pressure reaches the desired level the valve (6) is closed and valve (7) opened again. Then the gate valve (5) may be safely opened to pump the chamber to the high vacuum by the diff pump. As the chamber is pre-evacuated by the rotary pump there is no oil vapor burst from the diff pump.
The vacuum canister also simplifies the shutdown procedure - it’s enough to just close all the valves and turn off the heater. The backing vacuum will be held by the canister during the cool down of the boiler. It’s however desired to keep the water circulation to shorten the cool down time.
The combination valve (9) allows to isolate the foreline from the rotary pump (1) and vent the pump itself, to prevent oil flow into the vacuum port.
“Begin at the beginning," the King said, very gravely, "and go on till you come to the end: then stop.” ― Lewis Carroll, Alice in Wonderland

User avatar
Dennis P Brown
Posts: 2113
Joined: Sun May 20, 2012 2:46 pm
Real name: Dennis P Brown
Location: Glen Arm, MD

Re: Bell and whistles diffusion pump based vacuum system.

Post by Dennis P Brown » Mon Oct 26, 2020 10:33 am

In college, a very long time ago (in a galaxy far, far ... ;) not really) I built an identical system for coating dielectric films as well as metals like Al. Such a system (the ability to seal off the DP and pump the chamber) is very useful and a picture series would be wothwhile as both instruction and general knowledge. I am wondering what type of fittings do you plan on using?

User avatar
Maciek Szymanski
Posts: 154
Joined: Wed Nov 20, 2019 7:31 pm
Real name: Maciek Szymański
Location: Warsaw, Poland

The PDO300 diffusion pump

Post by Maciek Szymanski » Mon Oct 26, 2020 11:05 am

The heart of the system is PDO300 oil diffusion high vacuum pump. This is water cooled, fractioning type pump. The basic specifications are as follows:

Code: Select all

Pumping speed               300 l/s
Ultimate pressure         <10E-6 Tr
Backing pressure limit     2,5E-1 Tr
Heater power               500 W
Oil quantity               100 cm3
EAED1BD1-3716-4697-9B68-AEBFC53BC595.jpeg
Overall view of the pump: 1. Foreline connection (KF25). 2. Cooling water outlet port. 3. Cooling water inlet port. 4. Inlet flange (CF100). 5. Cooling jacket. 6. Thermal switch (“pump ready” signal). 7. Boiler.

4D8602AB-D92B-4056-AB5F-8236E511719D.jpeg
Heater unit. The heater is put on the bottom of the pump and held with a single central bolt, so the receptacle can be set in any orientation.

0A6ED6C2-EF53-4084-B98C-D8B16D99CA20.jpeg
View into the pump’s throat with the jet stack in place. Note the special I13 rings visible in the backing port. Those rings together with the port cooling are the measures to prevent the back-streaming into the foreline.

91DC84A1-1AB1-4721-AF6A-EBD93493A2B4.jpeg
All the components of the jet stack disassembled.

C4145154-760A-402D-8D6C-92D48403FC94.jpeg
View of the boiler bottom with the jet stack removed.

1164EAF1-061D-4D4E-B690-A66B6E007E93.jpeg
View of the jet stack assembled inside the pump body.

7E02C263-2289-4D7B-A795-DAA93A131920.jpeg
Cross-sectional view of the pump from the user manual. Note the funny detail - the real pump is a “mirror” version compared to the drawing.
“Begin at the beginning," the King said, very gravely, "and go on till you come to the end: then stop.” ― Lewis Carroll, Alice in Wonderland

User avatar
Maciek Szymanski
Posts: 154
Joined: Wed Nov 20, 2019 7:31 pm
Real name: Maciek Szymański
Location: Warsaw, Poland

The OWF300 cooled baffle

Post by Maciek Szymanski » Mon Oct 26, 2020 11:59 am

The OWF300 baffle is designed to be directly bolted to the PDO300 diffusion pump flange. As the pump flange has the rotating retaining ring the pump and baffle orientation can be adjusted as needed.

B3D04552-22C6-47F6-BF4F-F34A8DB5CF6F.jpeg
The OWF300 cooled baffle. From both sides it has CF100 flange with 8 M8 threaded holes and accepts the standard CF100 centering ring with an o-ring seal.

The baffle consist of two sets of coper deflectors and a coiled tube for cooling water circulation. The deflectors are arranged in a way that there is no direct (“optically clear”) path for the particles passing. So the oil vapor hitting the deflectors is cooled down and condensate. The deflectors of course also affect the pumped gas so the baffle limits the pumping speed to some extent.
This baffle is intended to be water cooled using the pump’s cooling circuit (it should be the firs component the cold water enters), but there are also ones cooled with the liquid nitrogen or a peltier module. But even with moderate water cooling the oil contamination of the chamber side of the system is greatly limited.

AA751EB3-8AF6-4F9E-8E39-767A16C03397.jpeg
There is no direct path through the baffle. Even despite pumping speed degradation it’s a worth addition to any diffusion pump system.
Last edited by Maciek Szymanski on Mon Oct 26, 2020 7:37 pm, edited 2 times in total.
“Begin at the beginning," the King said, very gravely, "and go on till you come to the end: then stop.” ― Lewis Carroll, Alice in Wonderland

User avatar
Maciek Szymanski
Posts: 154
Joined: Wed Nov 20, 2019 7:31 pm
Real name: Maciek Szymański
Location: Warsaw, Poland

The ZP100ES gate valve

Post by Maciek Szymanski » Mon Oct 26, 2020 12:24 pm

The ZP100ES gate valve is the solenoid operated valve which connects to the CF100 flange. The pump side flange has rotating ring so it’s orientation can be adjusted as desired. It has also 4 KF25 ports for foreline and gauge connections. The Choke plate is actuated by the clever kinematic system so it can lift and rotate with a single solenoid pull. In the open position the choke is parallel to the flow direction so almost full 100mm passage is open.

C5465FB4-CBFF-46EE-9BC2-8ED3BC0B8A99.jpeg
The ZP100ES valve. The solenoid coil uses 24 V DC to hold, but needs a short pulse of 220 V DC to open.

0B377324-14FB-480D-BB32-757D778B81BE.jpeg
Side view. The chamber side KF25 ports are visible.

A4D8C639-B989-4369-9B7B-78052C7A2098.jpeg
View into the open valve. This is view from the chamber side down the diffusion pump. In this orientation when the chamber is vented the atmospheric pressure pushes the choke against the seal, protecting the diffusion pump form leaks.
Last edited by Maciek Szymanski on Mon Oct 26, 2020 1:16 pm, edited 1 time in total.
“Begin at the beginning," the King said, very gravely, "and go on till you come to the end: then stop.” ― Lewis Carroll, Alice in Wonderland

User avatar
Maciek Szymanski
Posts: 154
Joined: Wed Nov 20, 2019 7:31 pm
Real name: Maciek Szymański
Location: Warsaw, Poland

The forevacuum distribution

Post by Maciek Szymanski » Mon Oct 26, 2020 12:46 pm

As I’ve not found the original Unitra vacuum canister in any dumpster I was forced to build my own:

E14E0E4C-B50F-4DEB-8406-C003A62B70B9.jpeg
The vacuum canister. It’s welded from a piece fo SS tube and standard caps. The flanged port connects to the outlet of the diffusion pump. The two CF25 ports are for the pirani gauge and the control valve.

The two foreline control valves (6. and 7. on the diagram) are of the ZK 25E type. This are plunger 90° solenoid valves. The plunger is gland sealed, but there were also produced “ZWK” valves with bellows sealed plunger. The valves have CF25 fittings and operate with the same combination of 220 and 24 V DC as gate valve.

8F231EB3-5399-48E9-8F89-958E8E2B2A2A.jpeg
The ZW 25 Valve. There is KF25 centering ring with seal and KF25 clamp for connections. As well a bellows compensator.

CD7796E1-D7FC-4EC2-A539-5B2A64514DFB.jpeg
The vent valve. It’s a small plunger type valve with a mesh filter. The solenoid is operated by 220V AC.
Last edited by Maciek Szymanski on Mon Oct 26, 2020 1:07 pm, edited 1 time in total.
“Begin at the beginning," the King said, very gravely, "and go on till you come to the end: then stop.” ― Lewis Carroll, Alice in Wonderland

User avatar
Maciek Szymanski
Posts: 154
Joined: Wed Nov 20, 2019 7:31 pm
Real name: Maciek Szymański
Location: Warsaw, Poland

The assembled high vacuum part of the system.

Post by Maciek Szymanski » Mon Oct 26, 2020 1:00 pm

The whole high-vacuum part is now assembled on the cart.

B3682B12-0306-42DB-AD6C-BA69C040B3D3.jpeg
The cart with the opening for the chamber CF100 flange. The lower shelf will hold the rotary pump and cooling water tank.

E7A17CE3-FF12-4DB1-B326-7CABA6A927DB.jpeg
The high vacuum assembly. Items as on the diagram: 2. Foreline vacuum canister. 3. Oil diffusion pump. 4. Cooled baffle. 5. Diffusion pump isolation gate valve. 6. Chamber pump valve. 7. Diffusion pump forevacumm valve. 8. Chamber vent valve.

71292BB0-3B4C-44DF-9CE0-44E4CB7B264C.jpeg
“Begin at the beginning," the King said, very gravely, "and go on till you come to the end: then stop.” ― Lewis Carroll, Alice in Wonderland

User avatar
Richard Hull
Moderator
Posts: 12915
Joined: Fri Jun 15, 2001 1:44 pm
Real name: Richard Hull

Re: Bell and whistles diffusion pump based vacuum system.

Post by Richard Hull » Mon Oct 26, 2020 2:12 pm

Great start on a superb vacuum system. Thank you for sharing your progress as you build the system. A very clean looking diffusion pump. All good vacuum systems make use of numerous valves. I consider a minimum of 2 bellows valves are needed on any fusor system. 3 such valves would be better. Unfortunately, such valves are expensive and hard to come by, surplus, in good shape for the average vacuum system neophyte.

Richard Hull
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
Retired now...Doing only what I want and not what I should...every day is a saturday.

User avatar
Artem Artemov
Posts: 41
Joined: Sat Sep 21, 2019 8:57 pm
Real name: Artem Artemov
Location: Moscow, Soviet Union

Re: Bell and whistles diffusion pump based vacuum system.

Post by Artem Artemov » Mon Oct 26, 2020 4:50 pm

It reminded me of the Soviet VUP-5 installation. With this installation, my acquaintance with vacuum technology began. But I noticed that the entire structure is attached to OSB (perhaps you have this material called differently). I don't see any metal stiffeners. Will it support a lot of weight? OSB may be stratified due to the high temperature of the diffusion pump!
Attachments
A795F1FA-21F7-48E6-98BD-BF12D2D2B1B3.jpeg
A4FFEE65-A988-4799-AED4-7B8AF843A0E5.jpeg
17423D79-E958-43E4-9A81-13698928CCC0.jpeg
8A52B8C6-94D5-406B-9D0D-7C92DF538043.jpeg

User avatar
Maciek Szymanski
Posts: 154
Joined: Wed Nov 20, 2019 7:31 pm
Real name: Maciek Szymański
Location: Warsaw, Poland

Re: Bell and whistles diffusion pump based vacuum system.

Post by Maciek Szymanski » Mon Oct 26, 2020 5:11 pm

Very nice setup. I like the chamber design. The electronic unit on top is a pump system controller or a part of the process equipment? The original Unitra system looks like this:

AE48944E-AA8B-4EC5-AB8D-483D8B4C4C20.jpeg
And yes, the shelf’s are made of 18mm OSB, but they are reinforced with frames of 20x40mm pine wood. It withstand my weight without any visible defection. And the only hot part is the heater and boiler - the parts connected to the board are cooled.

BTW - I’ve got a very nice Soviet made diff pump baffle cooled with peltier module. The pity is that the flange do not fit the polish equipment and the peltier requires more than 20 amps. So it nods some machining and a decent buck converter PSU.
“Begin at the beginning," the King said, very gravely, "and go on till you come to the end: then stop.” ― Lewis Carroll, Alice in Wonderland

Post Reply

Return to “Vacuum Technology (& FAQs)”