6 Feb
2006
6 Feb
'06
8:26 a.m.
Pete asked:
I have here a uVAX II in a ba23 box, most likely not
powered up for 2 or 3
years.
The innards are clean, no nasty smells, no other trace of prior problems.
Even so, to minimise the risk of damage I'm going to disconnect all the 5v &
12v leads, then attach some suitable dummy load(s), switch on and wait for
smoke.
Does this seem an appropriate plan? Can anyone point me to pinouts for the
various supply leads from an H7864?
Does anyone have any other recommendations/experience of going though this
process, and is there anything I've missed that I should have included
above?
Just plug it in and turn it on.
Unless the economics have drastically changed in the past 5 or 10 years,
a uVax II in a BA23 has a street value of $50 tops. If there's any irreplacable
components/assemblies/drives you may indeed want to remove them during
initial testing (in any event as you reconfigure the system you certainly want to
start with the simplest "has to work" configuration and then add on all the
unknowns.)
Ignore anyone who tells you to bring this up on a Variac. You will be increasing
the chances of damage by doing so. Variac's are great for linear supplies in
reforming electrolytic caps, but they only make life worse in a switching
power supply because you're gonna cause all sorts of out-of-usual
conditions. See, a switching power supply has negative impedance as seen
from the "in" connector: if you decrease
input voltage,
it will increase its current draw to keep power out constant. Outside
of a
reasonable range (say +/- 30%) of input voltage the flyback pulse widths etc. will
be seriously out of whack too, also probably triggering a shutdown.
My personal preference with linear supplies is to just plug it in and let any
bad electrolytics blow up :-).
The BA23's power supply (like most switching supplies) has more than adequate
shutdown-on-overvoltage-out circuitry.
The one thing to check: look at the 0.156" spacing harness from the PS
to the backplane. Wiggle it on and off, look for charring and overheating. Some
signs of past heat are likely, but if any of the contacts are burnt up you
might want to look for or fabricate a different jumper. Worst case will be that
you burn up the jumper, and you won't be the first to have it happen!
After you get it up and going, you will probably find that the NiCads that
hold the CPU configuration and run the TOY clock don't hold a
charge. It's just 3 AAA's in series. It's not really necessary if you
don't
mind setting the clock and selecting the boot device manually
at every startup. NiMH's are more available today and
probably a better technical choice for replacement if you decide you need to
TOY clock and config RAM.
Tim.
6 Feb
6 Feb
10:24 a.m.
My personal preference with linear supplies is to just
plug it in and let any
bad electrolytics blow up :-).
This is not a bad idea, and reasonable for "modern" electrolytics (the
quality of components, especially electrolytics, increased greatly in the
1950s, thanks to the military). If something has not been turned on for 15
or 20 years - OK, maybe it is a worthwhile precaution, but nothing to go
nuts over. Otherwise, just a power up, perhaps with the supplies not
feeding anything, will almost always work. If a cap blows, chances are
great that it was failing due to another reason (moisture getting thru the
seals is a culprit), and would have blown at some point anyway.
All the years I have been doing this, I have never had an electrolytic
drastically fail, and have only personally seen it once on someone else's
equipment. The failure did not damage anything else.
Electrolytics from the 30s and 40s are a different story.
William Donzelli
aw288 at osfn.org
11:19 a.m.
New subject: Affects of moisture (or lack thereof) on components (was: Powering up sleeping uVAX II)
In article <Pine.SUN.4.20.0602061013310.26496-100000 at osfn.org>,
William Donzelli <aw288 at osfn.org> writes:
[...] If a cap blows, chances are
great that it was failing due to another reason (moisture getting thru the
seals is a culprit), and would have blown at some point anyway.
I live in the desert. I've seen comments that excessive moisture can
be damaging for components (such as above) and I've seen photos of
monitors that have some funky spotting on them from being stored in
very humid environments (like a SF bay area garage).
Does anything bad happen to electronics that are stored in very dry
conditions for long periods of time? Cities like this would be anything
in Utah, Arizona, Nevada and Idaho (Salt Lake City, Las Vegas, Phoenix,
Boise, Reno, etc.).
--
"The Direct3D Graphics Pipeline"-- code samples, sample chapter, FAQ:
<http://www.xmission.com/~legalize/book/>
Pilgrimage: Utah's annual demoparty
<http://pilgrimage.scene.org>
12:56 p.m.
New subject: Affects of moisture (or lack thereof) on components (was: Powering up sleeping uVAX II)
Does anything bad happen to electronics that are
stored in very dry
conditions for long periods of time? Cities like this would be anything
in Utah, Arizona, Nevada and Idaho (Salt Lake City, Las Vegas, Phoenix,
Boise, Reno, etc.).
I would not think so, as long as the items are protcted from the sun and
static discharg (higher in the dry heat, obviously).
William Donzelli
aw288 at osfn.org
3:05 p.m.
New subject: Affects of moisture (or lack thereof) on components (was: Powering up sleeping uVAX II)
On Monday 06 February 2006 08:19, Richard wrote:
Does anything bad happen to electronics that are
stored in very dry
conditions for long periods of time? Cities like this would be anything
in Utah, Arizona, Nevada and Idaho (Salt Lake City, Las Vegas, Phoenix,
Boise, Reno, etc.).
I pickup a lot of vintage gear the southern S.F. Bay Area (mostly Silicon
Valley) - which has what is called a "Mediterranean" climate - generally
"mild" temperatures and a typical humidity in the 25%-35% range. (Not quite
desert, but practically no rain from May to October).
Other than the usual DEC foam problems and HP rubber/capstan problems, which
seem to be unrelated to climate, equipment stored under a roof here is
generally in excellent condition.
Equipment I've received from humid environments often has rust, mildew,
damaged capacitors, etc.
Based on years of collecting experience, I'm a firm believer that climate
plays a very important role in the condition of vintage gear that hasn't
"lived" in a controlled environment.
BTW: Slightly "off-topic" - vintage cars are also in abundance here for the
same reason. I have a 1973 Mustang Convertable, and it has practically no
rust - and it's rubber parts are almost entirely original and in good shape.
Cheers,
Lyle
--
Lyle Bickley
Bickley Consulting West Inc.
Mountain View, CA
http://bickleywest.com
"Black holes are where God is dividing by zero"
3:19 p.m.
Unless the economics have drastically changed in the
past 5 or 10
years, a uVax II in a BA23 has a street value of $50 tops.
I'd resist this characterization. I witnessed uVII's go through the $50
point 3 or 4 years ago. When that happens to a price of a thing of this
size usually it will first disappear, then, (bounce) increase in value for
those still looking for one. Also, VAXes have completely disappeared
from "Processor", the resellers magazine for
one.
John A.
7:11 p.m.
Pete asked:
So? There are not an unlimited supply of MVIIs (it may seem like they're
common now, but they won't always be). IMHO there is never an excuse not
to same sensible precaustions like testing the PSU first. It doesn't take
long, and it's good practice...
As an aside at this point, I have read seveal books on clock and camera
repair (2 of my other interests). Many of them make the same sort of
point. Although you must practice on junk devices, you must not let
yourself thing 'What the heck, this one is ruined anyway so I'll cut
corners'. You must not learn bad habits.
I have here a uVAX II in a ba23 box, most likely
not powered up for 2 or 3
years.
The innards are clean, no nasty smells, no other trace of prior problems.
Even so, to minimise the risk of damage I'm going to disconnect all the 5v &
12v leads, then attach some suitable dummy load(s), switch on and wait for
smoke.
Does this seem an appropriate plan? Can anyone point me to pinouts for the
various supply leads from an H7864?
Does anyone have any other recommendations/experience of going though this
process, and is there anything I've missed that I should have included
above?
Just plug it in and turn it on.
Unless the economics have drastically changed in the past 5 or 10 years,
a uVax II in a BA23 has a street value of $50 tops. If there's any irreplacable
components/assemblies/drives you may indeed want to
remove them during
initial testing (in any event as you reconfigure the system you certainly want to
start with the simplest "has to work" configuration and then add on all the
unknowns.)
Ignore anyone who tells you to bring this up on a Variac. You will be increasing
the chances of damage by doing so. Variac's are great for linear supplies in
reforming electrolytic caps, but they only make life worse in a switching
power supply because you're gonna cause all sorts of out-of-usual
conditions. See, a switching power supply has negative impedance as seen
from the "in" connector: if you decrease input voltage,
it will increase its current draw to keep power out constant. Outside of a
reasonable range (say +/- 30%) of input voltage the flyback pulse widths etc. will
be seriously out of whack too, also probably triggering a shutdown.
Sure.
My personal preference with linear supplies is to just plug it in and let any
bad electrolytics blow up :-).
My preference on any computer power supply (linear or switcher) is to
inspect it for obvious problems (physically leaking electrolytics, burnt
components, etc), and it if looks OK, give it a dummy load and turn it
on.
The BA23's power supply (like most switching supplies) has more than adequate
shutdown-on-overvoltage-out circuitry.
Which I would rather not depend on.
-tony
6745
days inactive
6746
days old
6 comments
6 participants
participants (6)
-
allain＠panix.com -
ard＠p850ug1.demon.co.uk -
aw288＠osfn.org -
lbickley＠bickleywest.com -
legalize＠xmission.com -
tshoppa＠wmata.com