beautypg.com

5 crystal fail inhibit, 6 crystal life and starting frequency – INFICON XTM/2 Thin Film Deposition Monitor User Manual

Page 89

background image

4 - 5

IP

N 07

4-

18

6S

XTM/2 Operating Manual

4.5 Crystal Fail Inhibit

In many coating plants the crystal fail output relay closure is given major
importance and causes the entire system to shut down. This can cause
problems when the crystal is changed as part of the normal reloading
procedure. This potential conflict is resolved by utilizing the crystal fail inhibit
input; refer to

section 2.5.4 on page 2-18

. When this input is activated the

crystal fail relay will not close on crystal fail. The front panel messages and
instrument operation still work normally. The operator may now change the
crystal and verify that it is operating without inducing a major process
interruption.

4.6 Crystal Life and Starting Frequency

Crystal life is displayed as a percentage of the monitor crystal's frequency shift
relative to the 1 MHz frequency shift allowed by the instrument. This quantity is
useful as an indicator of when to change the monitor crystal to safeguard
against crystal failures during deposition. It is normal to change a crystal after
a specific amount of crystal life (% change) is consumed.

It is not always possible to use a monitor crystal to 100% of crystal life. Useful
crystal life is highly dependent on the type of material being deposited and the
resulting influence of this material on the quartz monitor crystal. For well
behaved materials, such as copper, at about 100% crystal life the inherent
quality, Q, of the monitor crystal degrades to a point where it is difficult to
maintain a sharp resonance and therefore the ability to measure the monitor
crystal's frequency deteriorates.

When depositing dielectric oroptical materials, the life of a gold, aluminum or
silver quartz monitor crystal is much shorter; as much as 10 to 20%. This is due
to thermal and intrinsic stresses at the quartz-dielectric film interface, which are
usually exacerbated by the poor mechanical strength of the film. For these
materials, the inherent quality of the quartz has very little to do with the monitor
crystal's failure.

It is normal for a brand new quartz monitor crystal to display a crystal life
anywhere from 0 to 5% due to process variations in producing the crystal.
Naturally, this invites the question: "Is a brand new crystal indicating 5% life
spent inferior to a crystal indicating 1% life spent?"

If a new crystal indicates 5% life spent, it means that either the quartz blank is
slightly thicker than normal (more mechanical robustness), or the gold
electrode is slightly thicker than normal (better thermal and electrical
properties), or both. In either case, its useful life with regard to material
deposition should not be adversely affected. To verify this assertion, laboratory
testing was performed on crystals which covered the crystal life range in
question. Results indicate that a brand new crystal which indicates 3 to 5% life
spent is just as good, if not better than a crystal indicating 0 to 2% life spent.

As a consequence, it is important to consider the change in crystal life (%), not
just the absolute crystal life (%) Indicated.