CCD-world: Need help - Cleaning of contaminated CCD surface

[Willie Koorts]

The following was posted to CCD-world:


On Wed, 14 Jul 1999, Byeong-Gon Park wrote:

> Of course my pump is auto-vent type, that's why I don't understand
> the situation. The only information I heard from the guy at the place is
> that when he went to the telescope room he found the guage showed
> several hundreds of Torr(!) while the rotary part of the pump was running
> only.
> 
> Do you have any preferred slvent to try to soak the CCD?


Hi Byeong-Gon

One possibility for oil from the pump finding its way into your dewar is
with a power failure, power dip, or worse, a power grey-out, while pumping.
With power grey-outs, strange things can happen to automated electronic
systems.

Automated pumps are normally fitted with a solenoid-operated hi-vacuum
valve at the vacuum connection to the dewar.  When a power failure occurs
while pumping, this valve closes instantly, isolating the dewar from the
pump.  If this valve fails to close, or take too long to do so, or leaks,
the dewar, being the lowest pressure in the system, can now suck in the
oil from the backing pump into the dewar.  Some turbo pumps also have
their bottom bearing, oil lubricated - the design of the pump is such that
this is OK under normal operation but who knows what happens when the
dewar starts sucking in reverse on a stationary pump?

You are probably also aware of the fact that an oil based pump (without a
cold trap or the turbo pump running) will cause lots of oil back-streaming
when soly evacuating the dewar and the pressure drops down to the
molecular flow region.

Assuming that the guage that showed "several hundreds of Torr" was reading
the dewar pressure, suggests that the dewar must have lost its pressure,
probably sucking through the pump(s) at some stage.  When the power came
back, the automatic system probably realized that things are wrong and
shut the valve (and turbo pump?).

If you look back in the archives (http://www.not.iac.es/CCD-world/), a lot
of CCD cleaning discussions appeared on this list at the end of 1995 and
during 1996.  I attached one here below which would be of immediate
interest to you.

We have once tried cleaning a CCD using heated Ethyl (Absolute) alcohol
and more or less the procedure described by Kirk below.  We were also told
that a droplet hanging from a bond wire can break it and that the CCD
should be lifted out VERY SLOWLY from the solvent to prevent the formation
of a drop.  We therefore devised a "hoist" to do the lifting.  Kirk's mail
below seems to suggest that this is not so critical.

We also investigated the compatibility of TEK/SITe AR-coating to different
substances and it seemed that water is about the most destructive.  We
found when experimenting with a dead chip, that blowing on the chip while
wet with solvent, causes ice (followed by water when it melts) to form on
the surface because of rapid evaporation of the solvent and came to the
conclusion that blowing is not a good idea.  Obviously, full static
precautions and shorting the pins together must be applied throughout
while handling the chip.

Regards
Willie

                       Mr. W.P. Koorts ( wpk at saao.ac.za )    
                     South African Astronomical Observatory 
                   PO Box 9   Observatory  7935  South Africa
                 Tel.(27) (21) 447 0025  Fax.(27) (21) 447 3639
               World   Wide   Web  ( Work ) http://www.saao.ac.za
             World Wide Web ( Personal ) http://www.saao.ac.za/~wpk/


Date: Wed, 12 Jun 1996 10:46:54 -0700
From: Kirk Gilmore <gilmore at ucolick.org>
To: ccd-world at cfht.hawaii.edu
Subject: Cleaning CCD's

In response to the "cleaning CCD's" e-mail, I have been
cleaning CCD's routinely for a number of years with no
adverse effects.  We coat our thick devices with a laser
dye solution to improve the UV response and I have found
that cleaning the device before applying the solution is
essential in eliminating contaminants. Immersing the CCD
in a 250ml beaker of warm (~50C) acetone for about five 
minutes and then lifting slowly out at about a 45 degree
angle cleans the device well and leaves no residue marks.
I short out all pins to ground during the wash.
As far as the bonding wire problem goes, I have cleaned
and phosphor coated many devices without losing a bond wire.
The deposition process calls for the CCD to be spun at 1400
RPM for several minutes and I can't recall a bond wire
coming off.

Kirk Gilmore

> From: walker at ctiow3.ctio.noao.edu (a.walker x305)
> Message-Id: <9606121440.AA13202 at ctiow3>
> To: ccd-world at cfht.hawaii.edu
> Subject: cleaning CCDs
> Status: R


> This may be of interest?  Now only if they could come up with a way
> to prevent the bond wires from detaching while cleaning...

> Alistair

*****************************

>From david at photosol.com Mon Jun 10 08:03:37 1996
> Date: Mon, 10 Jun 1996 07:59:24 -0400
> From: "David M. Stone" <david at photosol.com>
> Organization: PHOTOGRAPHIC SOLTUTIONS, INC.
>
> My company has developed a cleaning product for wiping over a CCD array 
> without scratches or residue. It is recommended by a leading 
> manufacturer of high end digital cameras and is to be released in 
> September. How can I contact someone at your company to see if this 
> might be an item of interest? I can send a sample mid-summer.
> Thanx.
- -- CCD-world -- --
CCD-world is fully moderated. Send posts to CCD-world at astro.ku.dk
Standard replies will go to the list; address personal replies manually.
For more information, please go to:  http://www.not.iac.es/CCD-world/