CCD-world: Pleas help TASS

[Tom Droege]

The following was posted to CCD-world:

Harry,

Thank you very much for all your useful comments.  I am slowly moving
toward a gas bottle and just pumping dry Argon through the cameras at a
small positive pressure.  Of course first a bake out under vacuum.  This
would be an OK solution, but a pain for my amateur associates that may not
have a gas bottle in the dark room, as I do.  (Left over from cold fusion
experiments - I ended up being a debunker showing where many experiments
were in error.)

I do not think that I have a connector problem.  But I could have a bad
lead here and there.  They were made out of DB-25 solder cup connectors.
Leads were soldered in with long stripped sections.  Thus there was a lot
of soldered wire for the epoxy to seal.  

The useful trick I learned was how to keep the epoxy from dripping out the
other side.  After trying wax,  Play Dough, and several stages of gluing, I
finally found liquid latex in a hobby store.  I jsst paint the back side of
the connector with the latex, and this seals it well enought that the epoxy
does not drip out.  I can then get a 1/4" plus epoxy thickness surronding
the bare wires.  Once the epoxy has set, all the latex can easily be
stripped out of the connector. 

Thanks again for your help.

Tom 

At 06:47 PM 5/5/00 -0700, you wrote:
>The following was posted to CCD-world:
>
>Tom,
>We've used 3M 2216 (grey) epoxy as a vacuum sealant quite
>successfully--epoxying connectors into a vacuum head.  The difficulty is
>that if you epoxy pins going through the 'vacuum' wall, any movement of the
>pin after the epoxy sets will create a leak.  Soldering the pins into
>protoboard and THEN epoxying the board in place with full epoxy coverage
>works well.  Note that you cannot use insulated wires (especially stranded!)
>through an epoxy interface because you have a continuous leak between the
>wire and its insulation.
>
>A lot of your problem may be in water contamination trapped on the inside
>surfaces of your head (or in PCB material, et al).  Baking for a long time
>(hot as you can stand it, at least 70C for 3 days) usually helps this out
>(in a dry environment--best to have vacuum, then purged with dry N2 or At at
>regular intervals then back to vacuum to drive the water off the surfaces).
>This approach will also drive manufacturer-provided water out of your
>windowed CCDs!  Note that 85C is "TEC kit making" temperature!
>
>Filling your container with dry gas will only be a temporary solution.
>Water will eventually migrate into the head.  I've used Zeolite as a
>molecular sieve, but your leak rate may be too great to be much good.
>
>    Harry H. Marsh
>    PixelVision, Inc.
>
>
>----- Original Message -----
>From: Tom Droege <droege at wwa.com>
>Subject: CCD-world: Pleas help TASS
>
>
>The following was posted to CCD-world:
>
>Hello experts,
>
>Some time ago I started "The Amateur Sky Survey".  The general goal is to
>build a bunch of 2k x 2k cameras and survey all the sky all the time.  I am
>well on the way to the goal, and have solved an immense number of problems.
>  There are 20 dual cameras systems in the pipeline.  Since I am fairly
>old, I made the decision early on to take a "Bob Wilson" (The builder of
>Fermilab) approach.  The general idea is that you just start the production
>line building things as cheaply as you can.  The over designed things give
>no problem.  The under designed things stick out like sore thumbs, where
>you can give them all your attention.  The idea is that it is cheaper to
>throw away the production and redesign a few things than to build
>everything with large safety factors.  This, Bob demonstrated, is a cost
>effective way of doing things.  (By building Fermilab at twice the design
>energy at less than half the proposed cost.)  Since I am paying for this
>out of my own pocket, it seemed appropriate to try Bob's approach.  It
>involves a lot of failure, so it is not something that you can easily get
>by a funding agency.  But I have a tolerant funding agency, me.  There is
>also the problem that at 70 I do not have time to build a succession of
>prototypes.
>
>OK, the problem of the day.  It is getting close to the last problem, as I
>am taking pretty good pictures with the prototype system.
>
>I am buying CCD 442a's from Lockeed at bargain prices.  Dealing with them
>is a pain as they do not seem to care if they sell them to me at $1700 ea.
>(2k x 2k grade 3 device) or not.  So I get little help from them and feel
>lucky to get any devices.  This includes accurate data sheets.  I had to
>learn the pin out by experimentation at the cost of several devices.  (The
>diagram said I had a single stage output amplifier when it was two stage,
>and the pins were labeled some like the old design and some like the new.)
>
>I had planned on operating the 442 with the lid sealed on.  Thus I could
>cool it and only have to worry about condensation on the cover glass.  It
>turns out that Lockheed seals moisture inside the chip.  So if you cool it
>down, one gets ice crystals.  Lockheed was not helpful about this.  I then
>switched to chips with taped on covers.  The idea is to leave the taped on
>cover in place so that this gives moisture a chance to get out.  I do not
>have the clean room facilities to keep dirt off the chips, or the
>microscopes, etc., to pick off dirt that get on.
>
>To save costs, I did not design an enclosure to hold vacuum.  I made my own
>connectors with epoxy.  Not recommended high vacuum practice.  It is pretty
>tight, some have a time constant of order 20 minutes.
>
>The plan was to use a desiccant.  The material I have is called "Drierite"
>which is mostly CaS04 with 3% CoCl2.  The design allows a large volume of
>desiccant, coupled to the head by a short 1/4" id tube.  The catalog says
>"dew point of -70 F.  This I hoped would allow me to go as cold as I could
>with a two stage TEC.
>
>The situation:
>
>I start getting ice crystals at -10 C or so.  I added a thermoelectric
>cooler to act as a getter which runs colder than the CCD.  This gets me to
>-15 C or so.  The cooling system will get to -30 C which is about as cold
>as I need to go.  In fact, -15 C is almost good enough for the amateur
>locations.  The sky brightness for me near Chicago with the big survey
>pixels (7") means that there is always more light from the sky than dark
>current.  As you all know there is structure in the dark current above -20
>or so.  It would be nice to run there.  I have found a couple of
>professional locations.  These would benefit if I could get to -30.  It has
>been a struggle to get to -30, so I hate to give it up.
>
>Note that I have tried various cool down procedures, some of which help,
>but the above is the best result.
>
>The questions:
>
>Have any of you run using a desiccant?  What did you use?  Some have
>suggested that I fill with dry Nitrogen.  I have a bottle of Argon that I
>could use.  I do not have much hope that it will stay in my not very tight
>enclosure.  Any suggestion as to what I might do?   Sorry, the camera head
>enclosure was not designed to hold pressure, so pressurizing is out of the
>question without a complete redesign.
>
>Good advice would be appreciated.  Good advice might be "give up on this
>plan and design for vacuum."  Sigh!  I hope it is not.  That would set me
>back a year.
>
>Any help would be appreciated.  I am very close to getting a number of
>survey cameras into the field.  This is close to the last problem.  (Sure!)
>
>Tom Droege
>
>
>
>- -- 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/
>
- -- 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/