372596180

168

RIKEN Accel. Próg. Rep. 24 (1990)

V-2-16. All-Metal Gate Valve with an RF Contact

S. Yokouchi, K. Yagi,* K. Yoshida, * T. Ohbayashi, * and S.H. Be

10->1-760 Torr

less (han IX 1 <H°Torr l/s

SUS304

Be-Cu

Alalloy

5~7 kgf/cm²G

150°C

JVIS 003 NB 100 Bakcable Flangc

^GN21 5 Kgf/cm²1

Fig. 3. Valve seal part. Fig. 4. Test system.

We manufactured an all-metal gate valve with an RF contact on trial. Table 1 gives main specifica-tions of the gate valve. The gate valve at the open position is shown in Fig. 1. The present valve body was madę of stainless-steel, but that to be used in the practical SR ring is madę of aluminum-alloy except a part of valve seats and disks. A techniąue is employed, which provides an enclosed evacuated space between double seals, thereby reducing the

Table 1. Main specification.

Pressure rangę Leak rato

body and valve seal parł* Materials

Main parts RF contact Others

Comprcsscd air pressure Bakcout temperaturę Flangc dimension

• Differcntial pressure on the valvc seal is 1 aim in either direction

Fig. 1. Outside view.

Fig. 2. Amplification mechanism of sealing force.

pressure difference across the seals and decreasing the leak ratę of the seals. In the open position, the RF contact which has the same geometry as the beam chamber can bring about a direct electrical connection between two body flanges. Applying the thrusting force of an actuator to cam parts, which are connected to a carrier and two disks by a link mechanism (see Fig. 2), the sealing force between the seats and disks coated with silver is amplified by F/F₀>1. The further amplification of the sealing force is obtained by pushing the disks with a smali arc to tapered seats polished like a mirror (see Fig. 3). The deformation amount of the valve disks and seats due to the sealing force is within an elastic limit.

To investigate the leakage reliability at seals, we actuated the valve 1,000 times using the test system as shown in Fig. 4. The results show that the performance is not yet satisfied completely because there are a leak of the order of 10^-8 Torr/s^_1 over several times during actuations of these first 1,000 times. However, the present leak ratę can be reduced to less 1 X 10^-10 Torr-/-s^_1 by evacuating the enclosed space between double seals.

To improve fluctuation of the leak ratę to 1 x 10“¹⁰ Torr-/-s~\ we are planning an investigation of the relation between the sealing pressure and the leak ratę.

* Osaka Yacuum, Ltd.