shott at stanford.edu
Wed Jun 17 20:31:22 PDT 2009
We believe that drytek2 is available for some careful and cautious use.
Elmer has run the both the descum and nitride etch recipes with stable
power and with reflected power of less than 5 watts .... but there are a
couple of tricks to doing that.
Here are his comments:
> The variable inductor has dirty or damaged contacts. Cleaned the contacts and ran the Descum and Nitride etch recipes with a reflected power < 5 W.
> The variable inductor is still dirty but is working much better. Adjust it only in coil position narrow range of 2.5 ± 0.5. If initially the plasma does not strike, sweep the coil position between 2 - 3. Use only the capacitor to improve the tuning once the plasma is on.
In particular note his comments about using the capacitor and not the
coil adjustment to fine tune .... the roller inductor has multiple
sliding contacts and all of the RF current goes through those contacts.
Once you have struck the plasma and are reasonably well tuned, resist
the urge to fine tune with the coil setting .... there is the danger
that even a slight rotation will move the coil to one of the high
resistance spots and the plasma can either collapse or the reflected
power will suddenly jump to 60-80 watts.
Note: the overall health of the coil is significantly better than it was
a day or so ago and we've run some fairly long runs to convince
ourselves that the plasma is stable .... as long as you don't touch the
coil setting once you've got things running.
We are in the process or ordering replacement roller inductors .... but
they are custom manufactured and have a 4-6 week lead time. As the same
time, we've begun discussions with Jim McVittie to see if we can come up
with an appropriate matching network that doesn't need the variable
But with careful use, we expect that you should be able to get some
useful etching done.
More information about the drytek2