Re: [volt-nuts] Low-cost voltage reference questions
Dave wrote:
Your table mentions at the bottom the LTZ1000, but you don't have that
device listed in the table. I assume that is an oversight, but perhaps it
was intensional.
The table I posted was a condensation of the larger table that I use,
with entries removed that were not pertinent to the on-list
discussion. But I didn't remove the notes regarding shunt regulators.
As regards humidity, I wonder if an reasonable attempt at sealing a package
combined with silica gel inside would give an internal humidity that keeps
fairly stable.
That is exactly what I do. I typically package references in small,
cast aluminum boxes, and include as large a dessicant packet as will
fit in the unused space. Without any special precautions as to
sealing, I have never seen one of the dessicant packs more than
slightly used, even after a decade or more and even if there was a
hole in the box wall to allow for a potentiometer adjustment. With
precautions (using a gasket or sealing the seam with RTV after final
testing, and sealing any adjustment holes), I think you'd be good for
a lifetime.
Best regards,
Charles
On 29 November 2015 at 10:53, Charles Steinmetz csteinmetz@yandex.com
wrote:
Dave wrote:
As regards humidity, I wonder if an reasonable attempt at sealing a package
combined with silica gel inside would give an internal humidity that keeps
fairly stable.
That is exactly what I do. I typically package references in small, cast
aluminum boxes, and include as large a dessicant packet as will fit in the
unused space. Without any special precautions as to sealing, I have never
seen one of the dessicant packs more than slightly used, even after a
decade or more and even if there was a hole in the box wall to allow for a
potentiometer adjustment. With precautions (using a gasket or sealing the
seam with RTV after final testing, and sealing any adjustment holes), I
think you'd be good for a lifetime.
Best regards,
Charles
Are you, or anyone else, aware of any reasons silica gel should not be used
in electronic enclosures?
A friend and I were thinking of developing antennas, which would be used
outside, but enclosed in a fibreglass or similar tube. I suggested
flushing the antenna with nitrogen, then putting silica gel inside.
He was of the opinion that silica gel was not a good idea. Instead he
reckoned on pressuring the inside with nitrogen above ambient pressure,
and having gas-tight seals would be better. I'm personally of the opinion
that's a bit over the top, as then you need valves to get the nitrogen in.
His idea stuck me as a bit of a sledgehammer to crack a walnut, but he has
better mechanical engineering skills than me.
The difference with the antennas compared to a voltage standard is antennas
would be used in a more hostile environment than a laboratory. So perhaps
whats reasonable in a lab, is not so sensible for an antenna exposed to
wind, rain, ice etc.
Dave
On Sun, 29 Nov 2015 11:49:14 +0000
"Dr. David Kirkby (Kirkby Microwave Ltd)" drkirkby@kirkbymicrowave.co.uk wrote:
A friend and I were thinking of developing antennas, which would be used
outside, but enclosed in a fibreglass or similar tube. I suggested
flushing the antenna with nitrogen, then putting silica gel inside.
I know I am a bit late, but I'd like to comment anyways.
At my former job, we have built devices that have been installed
in high alpine environments and left there unattended for years.
The "technique" we used was standard aluminium casings with rubber seals
(similar to Hammond 1550W series), with a gore-tex breather hole and
filling all unsued space with silica gel.
First thing you need to know is, that it is impossible to keep an
housing completely airtight or rather watertight, as long as you
cannot control ambient temperature and pressure. The air pressure inside
will change and there will be a very slight breathing. If you don't give
it a safe way to breath, it will breath through the rubber seal. But this
breathing will be nasty: Breathing out the air from inside and sucking
in the (condensated) water droplets from the outside. Ie it will slowly
accumulate water. That's why we had do add a breather hole with a gore-tex
seal. You can change this to a long and narrow hole pointing downwards,
that should be equally good. If you make the hole volume large enough to
be larger than the worst case breathing volume, then you will never suck
in outside air. Depending on the environmental conditions you might still
need silca gel to keep any water vapor that diffused through the hole
from condensating at unsuspecting electronics (and I assume that you
will need a lot of silca gel in the UK weather).
I am pretty sure, that the idea with the overpressure nitrogen would work,
but not with a fiberglas tube. These are not pressure tight enough. They
will leak out all overpressure within a couple of months (if not weeks).
For this to work, you'd at least need aluminium or stainless steel and
build the whole thing in the way how high vacuum chambers are built.
Nothing for the faint of heart.
If you reuse silica gel from bought electronics, you should bake it out
for a couple of hours, somewhere between 100°C and 150°C. (electrical
oven, not gas oven)
Attila Kinali
--
It is upon moral qualities that a society is ultimately founded. All
the prosperity and technological sophistication in the world is of no
use without that foundation.
-- Miss Matheson, The Diamond Age, Neil Stephenson
In message 20160122221053.8362277eb59f0f8cabca9816@kinali.ch, Attila Kinali w
rites:
[...] and filling all unsued space with silica gel.
Just don't forget that silica gel is a relatively excellent material
for thermal insulation.
Silica Gel is a popular dessicant mainly because it doesn't matter
if you flip it over to turn it upside down.
In stationary applications there are a large selection of salts,
including almost any chloride you have heard about, which are very
volume efficient compared to silicagel provided the beaker/glass/bucket
you put them in doesn't get tipped over.
--
Poul-Henning Kamp | UNIX since Zilog Zeus 3.20
phk@FreeBSD.ORG | TCP/IP since RFC 956
FreeBSD committer | BSD since 4.3-tahoe
Never attribute to malice what can adequately be explained by incompetence.