1 Feb
2014
1 Feb
'14
9:45 a.m.
On 02/01/2014 11:02 AM, Jules Richardson wrote:
On 01/27/2014 01:03 PM, dwight wrote:
There is cold and there is COLD. For the moment COLD is near freezing
(0C) and cold is 50F.
Most will tolerate cold but many not work reliably. That is no harm
will likely occur to electronics.
the reason I be vague is a large machine like say and older PDP-8 with
core may like the cold but
core might be on the edge of margin and not work reliably, and the power
supplies may drift off
the set points make it more unreliable. None of that is permanent
failure.
HOWEVER,
Magnetic media will be potentially damaged! Why? The electronics might
go unreliable and write trash
or read trash, or worse a positioner may be off and put stuff in a servo
area on media.
Also lubricants will even is fresh increase in viscosity when cold so
things may not move well
until they are warmer.
The best example of the above was a TTY (asr33) that would not work at
below 60 as it needs more pads
(CR end dashpot was new and tight fit.) so it took longer to return the
carriage. The punch was fussy
as well when cold.
In the end its a system level call but some parts may not like the cold
and other may thrive!
Condensation is prohibited as old and likely dusty boards will have some
containments that are
conductive when wet and will upset bias, signal paths or regulation
(power supplies) all can be
minimally unreliable or potentially destructive. Also condensation is
problematic with older epoxy
packaged part as they may absorb moisture over time and when warm
convert that to die level
corrosion.
One last area, temperature cycling. That is going from cold to hot or
reverse with a sharp
change in temperature over time. This can crack solder joints, bonds
to dies, or mess with
mechanical tolerances in thing slike tapes and disks. If the machine is
cold and you power it
the self heating may be great enough to induce the sharp temperature
change and stress
the construction or connections.
A lot of things change with temperature in electronic
circuits.
Silicon is no exception. Shifting thresholds can cause
just as much problems, cold or hot.
ICs have to be made for specific temperature spans.
Too hot or too cold and they don't do what they are intended
to.
OK, so on that note - is there the possibility of lasting damage from
running electronics when they're too cold? Or is it more likely that
they'll just be glitchy or simply won't work when cold, but will be
fine again when warmed up?
I'm in much the same situation as Ethan, where one
of the rooms I've
got vintage equipment in gets down to around 50F until the heaters
have brought things up to a more comfortable level, so it's useful to
know if there's a risk in trying to operate machines when it's that cold.
In doing so it should be gradual without condensation (dry heat). 50F
is cold but generally inside the
proscribed range is the air is DRY. I'd suggest that warming to around
60-65 is a good thing. Most of
the larger machines make enough heat that keeping the room warm is
rarely an issue.
It sounds like magnetic media should be OK (unless
there's an age
factor which makes them more susceptible to cold than when new), and
that was originally my biggest concern - but maybe I should be
worrying about the electronics, too...
Most magnetic media is plastic binder and has a greater physical change
than metals over temperature.
Think of ferric oxide on aluminum platters with heads at microinches.
Take care with mechanical things.
Allison