13 Mar
2013
13 Mar
'13
7:01 p.m.
Hey all --
This'll probably rank as a "duh" moment for most of you but I thought
I'd
share (mostly because it amused me):
I was trying to figure out why the unregulated 8VDC line of my SWTPC 6800's
power supply kept dipping so suddenly under moderate load (and why it would
keep falling and falling as long as I left it running that way...)
The overall behavior was:
- Power up with minimal load (just the motherboard w/it's onboard
regulator): 8V line was fine at about 8.7V.
- Add up to two RAM cards and the 8V was still at around 8.3 or so.
- Add a third card and 8V drops to 6.5V and continues to drop steadily over
time.
This had me scratching my head. Then I happened to notice that the fuse
(F1 on the 12V supply board) was glowing a dull orange and getting brighter.
Well, turns out that someone (me!) had installed the wrong fuse -- it was a
2.5A fuse where a 10A was required (F2 takes the 2.5A, btw). Apparently
when fuses start to give out they draw a lot of current -- lesson learned
:).
Fuse replaced and all is happy. So, the lesson is: check your fuses
carefully.
- Josh
13 Mar
13 Mar
7:46 p.m.
On Mar 13, 2013, at 7:01 PM, Josh Dersch <derschjo at gmail.com> wrote:
Well, turns out that someone (me!) had installed the
wrong fuse -- it was a
2.5A fuse where a 10A was required (F2 takes the 2.5A, btw). Apparently
when fuses start to give out they draw a lot of current -- lesson learned
:).
Good thing to look out for! If I were to nitpick, though, I'd probably
point out that it's not that the fuse is starting to draw a lot of
current, but rather that it's starting to drop a lot of voltage as the
I in the IV term of Ohm's law goes way above its intended range. The
glowing is a tangential result (P = I^2*R).
- Dave
9:49 p.m.
On 3/13/2013 4:46 PM, David Riley wrote:
On Mar 13, 2013, at 7:01 PM, Josh Dersch <derschjo
at gmail.com> wrote:
True enough, thanks for the correction :).
- Josh
Well, turns out that someone (me!) had installed
the wrong fuse -- it was a
2.5A fuse where a 10A was required (F2 takes the 2.5A, btw). Apparently
when fuses start to give out they draw a lot of current -- lesson learned
:).
Good thing to look out for! If I were to nitpick, though, I'd probably
point out that it's not that the fuse is starting to draw a lot of
current, but rather that it's starting to drop a lot of voltage as the
I in the IV term of Ohm's law goes way above its intended range. The
glowing is a tangential result (P = I^2*R).
- Dave
14 Mar
14 Mar
10:08 a.m.
On Mar 13, 2013, at 9:49 PM, Josh Dersch <derschjo at gmail.com> wrote:
On 3/13/2013 4:46 PM, David Riley wrote:
And of course if I were to nitpick again, I'd point out that I said IV
instead of IR, which is what I meant. If nothing else, I try to be
consistent in my pedantry. :-)
- Dave
On Mar 13, 2013, at 7:01 PM, Josh Dersch
<derschjo at gmail.com> wrote:
True enough, thanks for the correction :). Well, turns out that someone (me!) had installed
the wrong fuse -- it was a
2.5A fuse where a 10A was required (F2 takes the 2.5A, btw). Apparently
when fuses start to give out they draw a lot of current -- lesson learned
:).
Good thing to look out for! If I were to nitpick, though, I'd probably
point out that it's not that the fuse is starting to draw a lot of
current, but rather that it's starting to drop a lot of voltage as the
I in the IV term of Ohm's law goes way above its intended range. The
glowing is a tangential result (P = I^2*R). 
7:50 p.m.
At 06:46 PM 3/13/2013, you wrote:
On Mar 13, 2013, at 7:01 PM, Josh Dersch <derschjo
at gmail.com> wrote:
Good point. At least he didn't have any LERs on
the power supply besides. Encountering an SED is also AVBT.
(LER: Light Emitting Resistor. Ohm's law has been broken.)
(SED: Smoke Emitting Diode. Don't break Ohm's law. It will get its revenge.)
(AVBT: A Very Bad Thing, vs ABT: A Bad Thing.
Marth a Stewart was not consulted.)
878 . [Government]I don't make jokes. I just
watch the government and report the facts. --Will Rogers
a50mhzham at gmail.com ? N9QQB (amateur radio)
"HEY YOU" (loud shouting) ? Second Tops (Set Dancing) ? FIND ME ON FACEBOOK
43? 7' 17.2" N by 88? 6' 28.9" W ? Elevation 815' ? Grid Square
EN53wc
LAN/Telecom Analyst ? Open-source Dude ? Musician
? Registered Linux User 385531
Well, turns out that someone (me!) had installed
the wrong fuse -- it was a
2.5A fuse where a 10A was required (F2 takes the 2.5A, btw). Apparently
when fuses start to give out they draw a lot of current -- lesson learned
:).
Good thing to look out for! If I were to nitpick, though, I'd probably
point out that it's not that the fuse is starting to draw a lot of
current, but rather that it's starting to drop a lot of voltage as the
I in the IV term of Ohm's law goes way above its intended range. The
glowing is a tangential result (P = I^2*R). 
8:14 p.m.
On 14 March 2013 23:50, Tom <a50mhzham at gmail.com> wrote:
Good point. At least he didn't have any LERs on the power supply besides.
Encountering an SED is also AVBT.
(LER: Light Emitting Resistor. Ohm's law has been broken.)
(SED: Smoke Emitting Diode. Don't break Ohm's law. It will get its revenge.)
(AVBT: A Very Bad Thing, vs ABT: A Bad Thing. Marth a Stewart was not
consulted.)
Excellent. Thanks! Promptly nicked for Twitter. :?)
--
Liam Proven ? Profile: http://lproven.livejournal.com/profile
Email: lproven at cix.co.uk ? GMail/G+/Twitter/Flickr/Facebook: lproven
MSN: lproven at hotmail.com ? Skype/AIM/Yahoo/LinkedIn: liamproven
Tel: +44 20-8685-0498 ? Cell: +44 7939-087884
13 Mar
13 Mar
8:03 p.m.
On 2013 Mar 13, at 4:01 PM, Josh Dersch wrote:
Hey all --
This'll probably rank as a "duh" moment for most of you but I
thought I'd
share (mostly because it amused me):
I was trying to figure out why the unregulated 8VDC line of my
SWTPC 6800's
power supply kept dipping so suddenly under moderate load (and why
it would
keep falling and falling as long as I left it running that way...)
The overall behavior was:
- Power up with minimal load (just the motherboard w/it's onboard
regulator): 8V line was fine at about 8.7V.
- Add up to two RAM cards and the 8V was still at around 8.3 or so.
- Add a third card and 8V drops to 6.5V and continues to drop
steadily over
time.
This had me scratching my head. Then I happened to notice that the
fuse
(F1 on the 12V supply board) was glowing a dull orange and getting
brighter.
Well, turns out that someone (me!) had installed the wrong fuse --
it was a
2.5A fuse where a 10A was required (F2 takes the 2.5A, btw).
Apparently
when fuses start to give out they draw a lot of current -- lesson
learned
:).
Fuse replaced and all is happy. So, the lesson is: check your fuses
carefully.
Interesting fault mode, but I wouldn't express that as the fuse
"drawing a lot of current". That fuse should be the one in the +8V
line and it looks like what was happening is your cards were drawing
current just short of blowing the fuse. As the fuse filament heated
up it's resistance increased, so more of the +8 supply voltage was
dropping across the fuse, reducing the +8V output level. (Overall
current should actually be going down).
Yay, another SWTPC 6800 (I have one too).
14 Mar
14 Mar
7:18 a.m.
On Wednesday (03/13/2013 at 05:03PM -0700), Brent Hilpert wrote:
Also, specific to the SWTPC 6800, you might want to closely inspect the
white Molex connector(s) at the power supply PCB. The pins in those
connectors were too whimpy for the amount of current typically carried
through them. They will heat up and degrade over time. On one of my
systems the PCB is even brown from the heating.
I mod'd this part of the machine, installing a heavier duty in-line Molex
connector off the board and then soldering wire directly into the holes
in the PCB where the (burnt) connector used to be.
In my case, this was the reason the +8 was too low on the backplane.
Chris
--
Chris Elmquist
Fuse replaced and all is happy. So, the lesson is: check your fuses
carefully.
Interesting fault mode, but I wouldn't express that as the fuse
"drawing a lot of current". That fuse should be the one in the +8V
line and it looks like what was happening is your cards were drawing
current just short of blowing the fuse. As the fuse filament heated
up it's resistance increased, so more of the +8 supply voltage was
dropping across the fuse, reducing the +8V output level. (Overall
current should actually be going down).
Yay, another SWTPC 6800 (I have one too). 
10:22 a.m.
Putting some DC#4 silicone grease on the connector pins willkeep it from over heating
because of bad contact.Dwight
Date: Thu, 14 Mar 2013 06:18:14 -0500
From: chrise at pobox.com
To: cctalk at classiccmp.org
Subject: Re: Fun Power Supply repair tip
On Wednesday (03/13/2013 at 05:03PM -0700), Brent Hilpert wrote:
Also, specific to the SWTPC 6800, you might want to closely inspect the
white Molex connector(s) at the power supply PCB. The pins in those
connectors were too whimpy for the amount of current typically carried
through them. They will heat up and degrade over time. On one of my
systems the PCB is even brown from the heating.
I mod'd this part of the machine, installing a heavier duty in-line Molex
connector off the board and then soldering wire directly into the holes
in the PCB where the (burnt) connector used to be.
In my case, this was the reason the +8 was too low on the backplane.
Chris
--
Chris Elmquist
Fuse replaced and all is happy. So, the lesson is: check your fuses
carefully.
Interesting fault mode, but I wouldn't express that as the fuse
"drawing a lot of current". That fuse should be the one in the +8V
line and it looks like what was happening is your cards were drawing
current just short of blowing the fuse. As the fuse filament heated
up it's resistance increased, so more of the +8 supply voltage was
dropping across the fuse, reducing the +8V output level. (Overall
current should actually be going down).
Yay, another SWTPC 6800 (I have one too). 
13 Mar
13 Mar
8:07 p.m.
On 3/13/2013 4:01 PM, Josh Dersch wrote:
This had me scratching my head. Then I happened to
notice that the fuse
(F1 on the 12V supply board) was glowing a dull orange and getting brighter.
Well, turns out that someone (me!) had installed the wrong fuse -- it was a
2.5A fuse where a 10A was required (F2 takes the 2.5A, btw). Apparently
when fuses start to give out they draw a lot of current -- lesson learned
If you
had that behavior, it sounds like you were using a fusetron time
delay fuse.
The fuses will run at 100% overload indefinitely, handle 10x for a spike
and blow greater than that.
if you had one of the fuses w/o the slow blow, then it would have run
for a while and then blown as well, but you don't get the more precise
overload time delay on that type of fuse. The fusing element may blow
by overload somewhere in the conductor, or with an overload, the solder
at each end of the fuse envelope may fail and it will blow then.
It is interesting that you saw a change in current going thru the fuse
though, sort of indicates the fuse wasn't working so well. Also with
that amount of heat into the phenolic fuse holder, you can probably
expect it to be very fragile and fail at some point due to the spring
load in it. I've seen fuse holders which didn't have glowing fuses, but
overloaded ones fail that way.
Never had one I knew first hand had an incandescent fuse in it though,
that is a bit worrisome.
jim
8:46 p.m.
On Wed, 13 Mar 2013, Josh Dersch wrote:
Hey all --
This'll probably rank as a "duh" moment for most of you but I thought
I'd
share (mostly because it amused me):
I was trying to figure out why the unregulated 8VDC line of my SWTPC 6800's
power supply kept dipping so suddenly under moderate load (and why it would
keep falling and falling as long as I left it running that way...)
The overall behavior was:
- Power up with minimal load (just the motherboard w/it's onboard
regulator): 8V line was fine at about 8.7V.
- Add up to two RAM cards and the 8V was still at around 8.3 or so.
- Add a third card and 8V drops to 6.5V and continues to drop steadily over
time.
This had me scratching my head. Then I happened to notice that the fuse
(F1 on the 12V supply board) was glowing a dull orange and getting brighter.
Well, turns out that someone (me!) had installed the wrong fuse -- it was a
2.5A fuse where a 10A was required (F2 takes the 2.5A, btw). Apparently
when fuses start to give out they draw a lot of current -- lesson learned
:).
Fuse replaced and all is happy. So, the lesson is: check your fuses
carefully.
"Hey, why is there a new light bulb in this thing?"
--
David Griffith
dgriffi at cs.csubak.edu
A: Because it fouls the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?
14 Mar
14 Mar
5:24 p.m.
This had me scratching my head. Then I happened to
notice that the fuse
(F1 on the 12V supply board) was glowing a dull orange and getting brighter.
Well, turns out that someone (me!) had installed the wrong fuse -- it was a
2.5A fuse where a 10A was required (F2 takes the 2.5A, btw). Apparently
when fuses start to give out they draw a lot of current -- lesson learned
:).
Not quite.. Remeember the fuse is in series iwt hthe laod and carries the
load current.
THink of how a fuse works. A fuse melts when the current through it gets
too high. The fuse wire melts, of course,becuase it gets too hot. And
that heat comes from the power -- I^2*R -- disipated in the fuse wire.
So a fuse must have some resistance and disipate power to work. In
general, lower-current fuses have higher resistanec than thigh current
ones. So your 2.5A fuse probably had a higher resistane than the correct
10A one, and thus got hot at hte notmal operating current. Having a
higher resistane, it also dropepd more voltage at that current than the
correct oen would have done.
Fuse replaced and all is happy. So, the lesson is: check your fuses
carefully.
Dre I admit that I once spent a good few minutes attemtpign t ofind out
why a coplece PSU weas giving no outputs at all only for find that some
rotten sod ahd 'borrowed' the fuse from the mains plug (UK BS1363 mains
plugs ahve an inte4rnal cartridge fuse). ARGH!
-tony
4148
days inactive
4150
days old
11 comments
10 participants
participants (10)
-
a50mhzham＠gmail.com -
ard＠p850ug1.demon.co.uk -
chrise＠pobox.com -
derschjo＠gmail.com -
dgriffi＠cs.csubak.edu -
dkelvey＠hotmail.com -
fraveydank＠gmail.com -
hilpert＠cs.ubc.ca -
jws＠jwsss.com -
lproven＠gmail.com