CCD-world: Flex Cable for inside dewars

[Craig D Mackay]

The following was posted to CCD-world:

Dear CCD world

I have seen a number of communications about the use of flexible printed
circuits as a way of providing electrical connection between a detector
inside a vacuum dewar and the outside world.  I thought you might be
interested to hear of the experiences that we had some years ago at
AstroCam.

We were looking for a method that would allow us to manufacture vacuum
dewars for CCD use quickly and effectively in order to minimise
manufacturing time but also to improve long-term reliability.  We used
flexible printed circuits which are made on Kapton.  The flexible
circuit was passed through the wall of the dewar and epoxied (using
vacuum compatible epoxy) on to a metal plate which was in turn sealed
with a O-ring against the wall of the dewar.

We had no problems with the materials themselves.  The printed circuit
material was checked for out gassing by Oxford Instruments and found to
be very good indeed.  The problems that we had work twofold:

1.	The copper that is used on the printed circuits is dramatically
better at conducting heat them the more traditional dewar wiring
materials such as beryllium copper.  As a result there is a necessity to
make it is thin as possible.  This did not seem to cause problems
initially, and repeated flexing of the PCBs at room temperature seem to
have no harmful effect on electoral continuity. However we did have
significant failures particularly at around the solder pads with
cracking of the relatively fine PCB tracks.  We believe that this was
caused by the stresses induced by repeated thermal cycling.

2.	The biggest problem, however, was that the flexible PCB is formed as
a sandwich of two layers.  We found that it was virtually impossible to
get this sandwich not to include paths by which air leakage occurred
down the length of the PCB.  This gradually degraded the vacuum which
was a big problem for us since many of the laboratories using liquid
nitrogen dewars do not want to pump them as frequently as astronomers
seem to be entirely happy to do.

The net effect is that now we prefer to wire up our dewars in order to
give them the longest time between vacuum pumping.  Looking back on our
experiences of flexible printed circuits I would say that it was an
expensive, time-consuming and ultimately unsuccessful piece of work and
unless you have good reasons for believing you can overcome these
problems I personally would not touch them for this application.

Best wishes

Craig D. Mackay
- -- 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/