From: email@example.com (Glen Walpert)
Subject: Re: Side effects of potting compound !
X-Newsreader: News Xpress 2.01
Date: Sat, 02 Nov 2002 14:44:17 GMT
NNTP-Posting-Date: Sat, 02 Nov 2002 09:44:17 EST
Organization: Net Access (netaxs.com)
In article <firstname.lastname@example.org>, email@example.com (Gibbo) wrote:
>For anyone interested here is what I have found following various combinations
>of the potting experiments regarding the multiturn trimmer (circular track with
>Potted in epoxy - affected trimmer position during curing.
>Potted in polyeurethane - affected the trimmer position during curing.
>Preset entirely covered in silicon then potted in polyeurethane. No effect on
>Conformal coated then potted in epoxy - affected the trimmer position.
>Conformal coated then potted in polyeurethane - affected the trimmer position.
>Grub screw superglued then potted in either epoxy or polyeurethane - affected
>the trimmer position.
>Go figure !
Looks like the method that worked is the one that prevented potting compound
from entering the trimmer, and wicking in between the wiper and the
resistance element. The conformal coating might have been expected to do this
also, but many conformal coatings have too low a viscosity to be very
effective at sealing gaps.
BTW, *nothing* is *completely* incompressible, not even diamond, and the
compressibility of all elastomers is significant in many situations, and high
enough to be easily measured. The material manufacturer will normally provide
the compressibility (bulk modulus) if asked. Typically differences in thermal
expansion are more important, however - under normal use the changes in volume
due to thermal expansion are much larger than changes in volume due to
pressure alone. Clearly where an elastomer is restrained by being potted or
otherwise, any differential thermal expansion between the restraining
structure and the elastomer must result in both compression of the elastomer
and deformation of the restraining structure, since no material has an
infinite modulus of elasticity, bulk or otherwise. Detailed calculation of
these deformations are straight-forward, I did them regularly when I was a
seal designer many years ago.
When I want high reliability in any assembly, especially one which will be
potted, I never use trimmers, which are among the most unreliable electronic
components made. I use select-in-test resistors, usually a pair in series to
allow for fine adjustment. Obviously this is expensive, perhaps more
expensive than many applications can justify, but SIT resistors (or electronic
pots if future adjustment or automated adjustment is required) are the only
way to avoid long term reliability problms with trimmers.