Triple failure. Software? Hardware? Please help
micky
Power Supply?
Last night, when I was running, as I have for at least 6 years, a
home-built computer with 1.5Gig memory and an 800MHaz CPU. All PATA,
no SATA.
I'm ib a another computer now and don't have details about the broken
one, but I can probably find them.
I clicked Back on my Firefox3 and the compuer restarted.
(It got to the point where it urges me to run chkdsk, but I was in
the middle of something and skipped that. The screen went black (or
blue) as normal, but before any Windows screen showed up, it restarted
again.
This time it offerred me safe mode etc. but I chose regular. I let it
run chkdsk and all it reported were two lengthy errors in avg log
files. But when it got to the same place, it restarted again.
This time I chose Safe Mode with Internet and it started fine, and I
even retrieved a couple of things from Usenet.
I wanted to go back to regular mode so I restarted.
This time it only got as far as the memory count-off, and it stopped.
Normally it shows a grid about my hard drives and CD dirves.
When I pushed the restart button, it took me to the CMOS/setup page.
Looking there, I found in the Boot Order, for hard drive it was set to
None, and had only two options, None and Skip. But the CD options
were both of my CD drives. A network was set for the fourth place to
look even though I don't remember ever doing that.
In the first page of setup, I have all four possible PATA drives set
to Auto, so iirc there woudl be no indication there about what drive
it finds.
I Escaped out, checking, Don't save any changes. Usually that is
plenty on other times, once or twice a year, when I get kicked to the
Setup screen.
It only got to the memory count-off again, Reset again put me on the
Setup screeen.
Turning it off completely and restarting showed a black screen. Even
though the red light of one CD drive goes on for a moment (I didn't
see the second one's light)
The screen is black partly because the monitor never wakes out of
standby, and if I turn it off and back on, it displays a little square
in the middle of the screen which says, Standby, Wake with PC, or
something like that.
It was very low humidity last night so I guess it was hotter than I
thought, mauybe 85, and for the first time I had five Firefox Windows
open, with at least 100 tabs. I had turned the fan/temp monitor off
for the winter.
I would think I overheated the CPU but it did start fine in Safe Mode,
and I sort of think some other chip is responsible for the POST
routine.
Waited 6 and 15 hours and works no better than it did.
I have I burned out my CPU, my video card**, the harddrive, my power
supply?
**The video card has acted up 5 times ln the last year, with a red
colored ghost for each letter and lots of funny dots on the screen to
where it's impossible to read. . If I put the computer in standby for
even a couple minutes, it's been all right again for another 6 hours.
I have a back up of the hard drive, but of course it's tailored for
this computer. I'd rather fix this computer then try to copy piece by
piece from the harddrive only.
I can fix this if it's not the CPU, but helpful advice would be much
appreciated.
Thank you.
micky
Paul
[Based on the attempts in the header of one of the posts,
crossposted to alt.comp.hardware . I think that is where you
meant it to go.]
There are a few things you can start with.
1) Visual inspection. Use your senses, to detect something is amiss.
2) Simplify hardware setup and retest.
3) Use the PC "beeper" function for hardware testing, if
video no longer works well enough to get error messages
from the PC.
For visual inspection, you take the side off and look for loose wires.
Or, perhaps a strap on a heatsink broke, and the heatsink is hanging down.
Check, with power off, that fans turn freely. And that heatsinks are not
clogged with dust.
Now, to the hardware specifics. You want to inspect the "capacitors" on
the motherboard, for bulged tops or brown or orange colored deposits on
top, or near the base of the capacitor. The capacitors are aluminum cylinders
with a plastic sleeve, with the component value printed on it. On the top
of the cylinder, there is a "stamped" pattern, perhaps the letter "K",
where the seams in the metal are for pressure relief. If hydrogen gas
builds up inside the capacitor (it is failing), the seams open releasing
the pressure before there is an explosion.
Your computer predates the "capacitor plague" incident. So your capacitors
might not be failing prematurely. Capacitors may last for fifteen years or
more, if the internal chemistry is still good. The capacitor plague ones,
lack stability internally, and chemical breakdown can happen, cold, over
a period of a couple years (I've had some fail like that).
http://en.wikipedia.org/wiki/Capacitor_plague
The top of the caps should be flat, and the pressure relief seams should
still be intact. If the capacitor has failed, a "slight" failure causes
an increase in Vcore ripple. If the capacitor shorts out internally,
you may see smoke, hear "sizzling" at startup and so on. If a cap fails
near the processor socket, it can take out a toroidal coil of wire
or one or more MOSFETs (things with three legs soldered near the CPU
socket). While the Vcore regulator is protected against some fault
types, collateral damage can still occur.
Inside the power supply, you can have similar issues. If your power supply
was a replacement, purchased in the last five to seven years, it could be
it was affected by bad caps as well. I had an Antec power supply (contract
manufactured by ChannelWell CWT), that failed pretty much exactly
like this one. You can see orange deposits on at least four caps here.
http://upload.wikimedia.org/wikipedia/commons/8/84/PSU_Caps.jpg
It's only advisable to inspect the power supply, if the warranty
has expired. Unplug the supply. Remove four screws on the top. Remove the
top plate, noting how any insulation sheet fits into place. Put any
insulation back, exactly as you found it. You may "look but don't touch".
This is an inspection only. If you see the orange deposits, you'll need
a new supply (unless you live in a country with a ready supply of good
quality caps). The reason you "can't touch" in there, is the main cap
is charged to 300VDC or so, and if any juice remains on there, it's
dangerous. You always assume the bleeder resistor has failed, and the cap
is fully charged. Put the cover back on when you're finished. (The reason
you can't do this with a warranty in place, is usually there is a tear-away
sticker over one of the screws, so the manufacturer can detect tampering,
and refuse warranty service if the supply has been tampered with.)
So you're looking for visual clues, loose wires, broken heatsink or fan,
"bad caps", the smell of burning or chemical smells and so on.
The motherboard may have a coin cell to power the CMOS (BIOS settings).
If you use a multimeter, and ground the black lead to a screw on the
chassis (like a screw on an I/O connector), you can probe the top of the
CR2032 coin cell and check the voltage. It should read around 3V. Below 2.3
or 2.4V or so, the battery can fail to maintain BIOS settings when all
power is off on the PC. I didn't see anything in your symptom
description to implicate the battery.
Your startup where disks weren't detected, could have resulted in
the BIOS modifying the boot order. BIOS designs have evolved over
the years, in how they handle the boot order. Time was, the BIOS was
pretty good at keeping a reasonable boot order. I have some PCs here,
which are terrible at it, and the BIOS frequently does whatever the
hell it feels like. And puts the disks in some order I didn't specify.
This is normally triggered by a change in the hard drive configuration.
*******
In the "Simplify" department, you remove stuff according to the symptoms.
If you can no longer get video, you'll have to unplug a fair bit of stuff.
To offload the power supply, you can unplug the hard drives and CDROM drives.
Disconnect both the power cables (Molex 1x4) and data cables, making note
of the orientation and location of the cables for later. Leave the
floppy connected, so you can run memtest86+ later.
If you have no video, you can run "beep tests". These rely on a working
computer case speaker, connected to the SPKR 1x4 header pins on the
motherboard. If you were to unplug an AGP or PCI video card, such that
there was no video, the BIOS would "beep" a three beep pattern, indicating
a video failure. This has the beneficial side effect, of proving the
processor works, it read some BIOS code, it carried out tests. So in
fact, any beeps heard, are a positive sign.
If you power off again, and remove RAM sticks (placing them in an
antistatic bag), you can repeat the beep test again. A different pattern
will be heard. Video and RAM failures use two or three beep patterns,
and the pattern really isn't important, except that each kind of failure
uses a different pattern. Since the "success case" (normal PC startup)
is a single beep, any other beep pattern means the BIOS has discovered
a fault.
If you get zero beeps, then the processor might not be running BIOS
code. Occasionally, on an older computer, the BIOS chip gets "bit rot"
and the code is corrupted. But in your case, this didn't start with
a BIOS level failure - your computer was running at the time, so it
isn't likely to be a bit rot problem.
Your PC description says "800 MMz". That could be 800 MHz Pentium III
or Celeron. That would be powered by the main 20 pin power cable.
It wouldn't likely have the auxiliary power cable ATX12V with 2x2
power connector on the end. The other interpretation of your
description, is you have an Athlon processor. That could be
powered from the main cable as well. I have an AthlonXP motherboard,
and it doesn't use ATX12V, so up to 65W is drawn from the 5V rail of
the main power cable. (Athlon PCs may need a pretty decent +5V
current rating on the power supply. Not all modern power supplies
are a good match for that.)
The Pentium III and Athlon, may have BIOS monitoring of CPU temperature.
The BIOS is capable of a relatively slow response to overheating. If
a heatsink falls off an Athlon, it can overheat so quickly, the
processor crashes before the BIOS code can switch off the power.
Then, the power stays on and the processor is cooked. It was in
later generations, that "hardware overheat protection" in the
form of THERMTRIP was added to PCs, causing the power supply to
shut off, if a monitoring diode on the CPU silicon die is
overheating. But that happened a bit later perhaps, than your
800 MHz CPU.
In terms of probabilities, the power supply is a fairly unreliable
component. If you had no diagnostic equipment, and just played
the part of an equipment swapper, you'd swap that out first.
The CPU is relatively reliable.
RAM, isn't as reliable as they'd like you to believe. I had yet
another failure in that department, a week ago. (Fortunately,
not on one of my good PCs.) The computer vintage was around the
same age as your machine. I had (3) 512MB PC133 SDRAM in the
machine. Two sticks had failed. I had bought a grand total of
eight sticks of that stuff. What is funny, is the five remaining
sticks all tested good. Two of three sticks sitting in the PC
for years, were the ones that failed. It almost suggests some
kind of metallurgy problem while in storage, as that PC doesn't
get to run very often.
As part of your "beep testing", you remove video card and RAM,
and listen for beeps. If you hear two beeps, then maybe the processor
is still running. Then, you add in one stick of RAM (with the
power cord unplugged, no power present). Now, does the beep
pattern change to three beeps (video card missing) ? If so,
then the RAM might be OK. If you heard zero beeps, after
adding back the RAM, it could be that locations below 640K
have gone bad. You'd power down, and try one of the other
RAM sticks from your collection, trying them one at a time
until you find a (relatively) good one.
Say you install one RAM stick, hear the "missing video" beeps,
add in the video card, and now the machine POSTs again. You can
leave the hard drives and CDROM disconnected, and boot a memtest86+
floppy for a test. Now, you can verify memory locations above 640K.
http://www.memtest.org (scroll half way down, for downloads)
Do at least one full pass, with that test program. If you own
multiple RAM sticks, you can test each one individually, to
determine if any of the sticks are bad. Install any RAM that
completes a full pass. Once you've weeded out the duds, you
can do one final test run with all candidate sticks reinstalled.
Always fully power off, before making hardware config changes.
So now we've got RAM, video, floppy, but no hard drives. Now
you can cable those back up, and look for them in the BIOs
detection screen. Are they still missing ? Test your storage
devices one at a time (being careful to properly jumper the
drive for whatever cable config you're using to test). If
two drives disappear, and they share a common cable, you'll
need to test the drives one at a time, either as "Master" or
"Cable Select", depending on the ribbon cable type (80 wire
cable preferred).
Anyway, that's a few ideas to try out.
HTH,
Paul
Bob F
> Triple failures? Hard Drive Data? Hard drive? Video Card? CPU?
> Power Supply?
>
> Last night, when I was running, as I have for at least 6 years, a
> home-built computer with 1.5Gig memory and an 800MHaz CPU. All PATA,
> no SATA.
>
> I'm ib a another computer now and don't have details about the broken
> one, but I can probably find them.
>
> I clicked Back on my Firefox3 and the compuer restarted.
>
> (It got to the point where it urges me to run chkdsk, but I was in
> the middle of something and skipped that.
Don't do that! If you want to screw up a system, just go on and do whatever else
when it tells you to run chkdsk.
micky
wrote:
You had me really scared at first, but then I started thinnking about
the past. When I used win98, I used to have crashes etc. and there
was no reminder to run chkdsk, and I'd go weeks or months without
doing so, but win98 would work fine. It might eventually crash again
in the same circumstances. When I eventually ran chkdsk, I would
save as files all the whatchamacallits that it let me save, then I'd
use the 4DOS List command to look inside all of them, before I deleted
them, and most seemed to be data, not system files. After I looked at
them, if I was missing data, I would try, sometimes successfully, to
rename the files or copy the data from them. Then I would delete them
all. But it never interfered with win98 running.
Are you sure it can make things worse to run windows without running
chkdsk?
micky
>micky wrote:
>> Triple failures? Hard Drive Data? Hard drive? Video Card? CPU?
>> Power Supply?
>>
>> Last night, when I was running, as I have for at least 6 years, a
>> home-built computer with 1.5Gig memory and an 800MHaz CPU. All PATA,
>> no SATA.
>
>[Based on the attempts in the header of one of the posts,
>crossposted to alt.comp.hardware . I think that is where you
>meant it to go.]
Thanks
And thanks a lot for the enormous effort you put into writing this.
My point by point reply is amazingly well written, but you don't
really have to read it . The upshot is so far......
1) The SATA harddrive tests fine USBed to another computer.
2) Can I use any little speaker, even if it's bigger than the usual PC
speaker? (Why do I ask stupid questions like this?)
3) The quickest way out of this may be replacing my CPU, if that is
the problem.
They have them on Ebay from 5 to 11 dollars, shipping included.
Do I need to match more than 800 MHz, Pentium III?????
My current one might be AMD, I can't remember, though I could take the
heat sink off and look.
(One of them includes grease, although I still have some grease and
I'm pretty sure I know where that is.)
4) Other fans good but a 1" x 1" fan on top of a chip has been running
badly for a year. It doesn't start unless I push it, and I almost
never do. My excuse is that I was supposed to migrate to the next
computer. It's above the AMD761 System Controller. Could
that have burned out. If it did, it's not worth repairing, right:?
-----------
>There are a few things you can start with.
>
>1) Visual inspection. Use your senses, to detect something is amiss.
No smell of burning anywhere, esp. power supply. Also CPU but harder
to get my nose close to that.
No bad caps seen.
>2) Simplify hardware setup and retest.
>3) Use the PC "beeper" function for hardware testing, if
> video no longer works well enough to get error messages
> from the PC.
I don't think I put in a beeper when I assembled this, but I ran it a
couple more times with the sound card and amp'ed speakers and no
beeps. I realize that the power the sound card could be bad.
>For visual inspection, you take the side off and look for loose wires.
>Or, perhaps a strap on a heatsink broke, and the heatsink is hanging down.
Heatsink okay, and I checked before I disconnected things, CPU fan
running as normal
A 1" x 1" fan on top of a chip has been running badly for a year. It
doesn't start unless I push it, and I almost never do. My excuse is
that I was supposed to migrate to the next computer.
It's above the AMD761 System Controller. Could that have burned
out. If it did, it's not worth repairing, right:?
>Check, with power off, that fans turn freely. And that heatsinks are not
>clogged with dust.
Fans turn. Heatsinks cleaned about 6 months ago. Still good.
>
>Now, to the hardware specifics. You want to inspect the "capacitors" on
>the motherboard, for bulged tops or brown or orange colored deposits on
>top, or near the base of the capacitor. The capacitors are aluminum cylinders
>with a plastic sleeve, with the component value printed on it. On the top
>of the cylinder, there is a "stamped" pattern, perhaps the letter "K",
>where the seams in the metal are for pressure relief. If hydrogen gas
>builds up inside the capacitor (it is failing), the seams open releasing
>the pressure before there is an explosion.
That's good. I don't want an explosion. No bulges or brown or orange
deposttis, top or bottom. >
>Your computer predates the "capacitor plague" incident. So your capacitors
Yes, I forgot that the mother board was given to me by a friend, who
had used it for years before IO did. So it 's even older.
Asus A7M266, but soon I'll be testing the board. The google search
for the manual says 2001.
>might not be failing prematurely. Capacitors may last for fifteen years or
>more, if the internal chemistry is still good. The capacitor plague ones,
>lack stability internally, and chemical breakdown can happen, cold, over
>a period of a couple years (I've had some fail like that).
>
>http://en.wikipedia.org/wiki/Capacitor_plague
>
>The top of the caps should be flat, and the pressure relief seams should
>still be intact. If the capacitor has failed, a "slight" failure causes
>an increase in Vcore ripple. If the capacitor shorts out internally,
>you may see smoke, hear "sizzling" at startup and so on. If a cap fails
>near the processor socket, it can take out a toroidal coil of wire
>or one or more MOSFETs (things with three legs soldered near the CPU
>socket). While the Vcore regulator is protected against some fault
>types, collateral damage can still occur.
>
>Inside the power supply, you can have similar issues. If your power supply
>was a replacement, purchased in the last five to seven years, it could be
>it was affected by bad caps as well. I had an Antec power supply (contract
>manufactured by ChannelWell CWT), that failed pretty much exactly
>like this one. You can see orange deposits on at least four caps here.
>
>http://upload.wikimedia.org/wikipedia/commons/8/84/PSU_Caps.jpg
>
>It's only advisable to inspect the power supply, if the warranty
>has expired.
I don't rememer how old it is -- it is a replacement --, but I'm never
going to send it back for warranty replacement. .
Still, I would rather disconnect it and test the voltages at the mobo
connector. That would work, right?
>Unplug the supply. Remove four screws on the top. Remove the
>top plate, noting how any insulation sheet fits into place. Put any
>insulation back, exactly as you found it. You may "look but don't touch".
>This is an inspection only. If you see the orange deposits, you'll need
>a new supply (unless you live in a country with a ready supply of good
>quality caps). The reason you "can't touch" in there, is the main cap
>is charged to 300VDC or so, and if any juice remains on there, it's
>dangerous. You always assume the bleeder resistor has failed, and the cap
>is fully charged.
Ooo. Good to know!! I will remember this.
>Put the cover back on when you're finished. (The reason
>you can't do this with a warranty in place, is usually there is a tear-away
>sticker over one of the screws, so the manufacturer can detect tampering,
>and refuse warranty service if the supply has been tampered with.)
>
>So you're looking for visual clues, loose wires, broken heatsink or fan,
>"bad caps", the smell of burning or chemical smells and so on.
>
>The motherboard may have a coin cell to power the CMOS (BIOS settings).
>If you use a multimeter, and ground the black lead to a screw on the
>chassis (like a screw on an I/O connector), you can probe the top of the
>CR2032 coin cell and check the voltage. It should read around 3V. Below 2.3
>or 2.4V or so, the battery can fail to maintain BIOS settings when all
I will do that, but I don't remember seeing wierd things in the BIOS
settings. I think the time was correct, for example. .
>power is off on the PC. I didn't see anything in your symptom
>description to implicate the battery.
>
>Your startup where disks weren't detected, could have resulted in
I remove the hard drive and used a Roseill adapter/USB cable/power
supply to connect it to the computer I'm using. Everything was
there. That's a relief. I have a backup but it was more than a day
old.
>the BIOS modifying the boot order. BIOS designs have evolved over
>the years, in how they handle the boot order. Time was, the BIOS was
>pretty good at keeping a reasonable boot order. I have some PCs here,
>which are terrible at it, and the BIOS frequently does whatever the
>hell it feels like. And puts the disks in some order I didn't specify.
>This is normally triggered by a change in the hard drive configuration.
Okay, so that's fine, although it's not fine that it didn't find my
still working HDD.
>*******
>
>In the "Simplify" department, you remove stuff according to the symptoms.
>
>If you can no longer get video, you'll have to unplug a fair bit of stuff.
>To offload the power supply, you can unplug the hard drives and CDROM drives.
>Disconnect both the power cables (Molex 1x4) and data cables, making note
>of the orientation and location of the cables for later. Leave the
>floppy connected, so you can run memtest86+ later.
I have that on a floppy somewhere, and if I can't find it, I put a
floppy in every desk computer I have. For 10 dollars, less when you
already have one, I don't know why people don'.t
>If you have no video, you can run "beep tests". These rely on a working
>computer case speaker, connected to the SPKR 1x4 header pins on the
The mobo manual is on my HDD, which is disconnected again, but it's
easy to dl again. I have a speaker from some computer I stripped,
although it's getting harder and harder to find things the more things
I have.
>motherboard. If you were to unplug an AGP or PCI video card, such that
>there was no video, the BIOS would "beep" a three beep pattern, indicating
>a video failure. This has the beneficial side effect, of proving the
>processor works, it read some BIOS code, it carried out tests. So in
>fact, any beeps heard, are a positive sign.
I should look for the speaker now.
>If you power off again, and remove RAM sticks (placing them in an
>antistatic bag), you can repeat the beep test again. A different pattern
>will be heard. Video and RAM failures use two or three beep patterns,
>and the pattern really isn't important, except that each kind of failure
>uses a different pattern. Since the "success case" (normal PC startup)
>is a single beep, any other beep pattern means the BIOS has discovered
>a fault.
>
>If you get zero beeps, then the processor might not be running BIOS
>code. Occasionally, on an older computer, the BIOS chip gets "bit rot"
>and the code is corrupted. But in your case, this didn't start with
>a BIOS level failure - your computer was running at the time, so it
>isn't likely to be a bit rot problem.
Very good to know.
>Your PC description says "800 MMz". That could be 800 MHz Pentium III
>or Celeron. That would be powered by the main 20 pin power cable.
>It wouldn't likely have the auxiliary power cable ATX12V with 2x2
>power connector on the end.
Right!! You know my computer better than I do. (I thought I had the
2x2, but I'm thinking of some other computer I fidded with. I think
it's the one I can't find!!! I know it's here somewhere. It was
going to be my next computer until I couldn't find it and a friend
gave me his old one.
> The other interpretation of your
>description, is you have an Athlon processor. That could be
>powered from the main cable as well. I have an AthlonXP motherboard,
>and it doesn't use ATX12V, so up to 65W is drawn from the 5V rail of
>the main power cable. (Athlon PCs may need a pretty decent +5V
>current rating on the power supply. Not all modern power supplies
>are a good match for that.)
>
>The Pentium III and Athlon, may have BIOS monitoring of CPU temperature.
>The BIOS is capable of a relatively slow response to overheating. If
>a heatsink falls off an Athlon, it can overheat so quickly, the
>processor crashes before the BIOS code can switch off the power.
>Then, the power stays on and the processor is cooked. It was in
That may be! I could still buy another CPU, I guess. Then I woudln't
have to filddle with setup .
>later generations, that "hardware overheat protection" in the
>form of THERMTRIP was added to PCs, causing the power supply to
>shut off, if a monitoring diode on the CPU silicon die is
>overheating. But that happened a bit later perhaps, than your
>800 MHz CPU.
>
>In terms of probabilities, the power supply is a fairly unreliable
>component. If you had no diagnostic equipment, and just played
>the part of an equipment swapper, you'd swap that out first.
>
>The CPU is relatively reliable.
>
>RAM, isn't as reliable as they'd like you to believe. I had yet
>another failure in that department, a week ago. (Fortunately,
>not on one of my good PCs.) The computer vintage was around the
>same age as your machine. I had (3) 512MB PC133 SDRAM in the
>machine. Two sticks had failed. I had bought a grand total of
>eight sticks of that stuff. What is funny, is the five remaining
>sticks all tested good. Two of three sticks sitting in the PC
>for years, were the ones that failed. It almost suggests some
>kind of metallurgy problem while in storage, as that PC doesn't
>get to run very often.
I"ve noticed that some tv's often have problems if I don't watch them
for years.
WOW!! That should keep me busy!!
>HTH,
> Paul
GlowingBlueMist
> On Sun, 03 Jul 2011 03:58:50 -0400, Paul<nos...@needed.com> wrote:
>
>> micky wrote:
>>> Triple failures? Hard Drive Data? Hard drive? Video Card? CPU?
>>> Power Supply?
>>>
>>> Last night, when I was running, as I have for at least 6 years, a
>>> home-built computer with 1.5Gig memory and an 800MHaz CPU. All PATA,
>>> no SATA.
>>
>> [Based on the attempts in the header of one of the posts,
>> crossposted to alt.comp.hardware . I think that is where you
>> meant it to go.]
>
> Thanks
>
One other thing worth mentioning is that the life of the motherboard
battery is usually only rated for 5 years from date of manufacture. I
have seen many strange things happen when one finally dies or gets too
low to fully drive the motherboard, including symptoms like you describe.
The battery is relatively cheap. I would replace this one just as a
precaution due to the age of the motherboard, then tell the BIOS to
return to factory settings and go from there. If your BIOS has a
setting that sounds similar to Failsafe rather than factory original try
that one first.
micky
wrote:
>
>3) The quickest way out of this may be replacing my CPU, if that is
>the problem.
>
>They have them on Ebay from 5 to 11 dollars, shipping included.
>
>Do I need to match more than 800 MHz, Pentium III?????
Yes, I think I do.
The mobo manual says
I have Socket 462, socket A
Not every ebay ad includes the socket, but some do.
And that I can use
Athlon CPU's up to 1.2 ghz 1.1g, 1g, 950m, 900m, 850m, 800m
or Duron 750 to 650 but this is slower than what I have now, 800.
Pentium, AMD, Cyrix, IBM
These seem to start at 20 dollars on ebay.
I'm still not sure the problem is the CPU. I have the pinouts for the
PSU and willl be testing. IIUC, I have to keep pushing the on button
every few seconds.
micky
wrote:
>On Mon, 04 Jul 2011 09:42:43 -0400, micky <NONONO...@bigfoot.com>
>wrote:
>
>>
>>3) The quickest way out of this may be replacing my CPU, if that is
>>the problem.
>>
>>They have them on Ebay from 5 to 11 dollars, shipping included.
>>
>>Do I need to match more than 800 MHz, Pentium III?????
>
>Yes, I think I do.
>
>The mobo manual says
>
>I have Socket 462, socket A
> Not every ebay ad includes the socket, but some do.
>
>
>And that I can use
>
>Athlon CPU's up to 1.2 ghz 1.1g, 1g, 950m, 900m, 850m, 800m
And if I'm using 800 Mhz now with no problem** should I be able to
move up to 1.1 or 1.2 gigs??
What if I have a bigger heat sink and fan I can use? Or at least a
new clean one?
**except overheating when the heatsink fins were really dirty, and I
opened too many tabs, And I guess again 2 days ago, if the CPU is
burned out, but I had open 5 windows, with 100 to 150 tabs. It was a
mistake, I never did it before and won't do it again. I've
cleaned the heatsink fins, and I should be able to do a much better
job, or use an new heatsink, when I get the new CPU, and I've had my
chimney cleaned so it won't get nearly as dirty again nearly as fast.
When I first bought the CPU, 800 seemed plenty fast!
dadiOH
> Triple failures? Hard Drive Data? Hard drive? Video Card? CPU?
> Power Supply?
>
> Last night, when I was running, as I have for at least 6 years, a
> home-built computer with 1.5Gig memory and an 800MHaz CPU. All PATA,
> no SATA.
>
> I'm ib a another computer now and don't have details about the broken
> one, but I can probably find them.
>
> I clicked Back on my Firefox3 and the compuer restarted.
>
> (It got to the point where it urges me to run chkdsk, but I was in
> the middle of something and skipped that. The screen went black (or
> blue) as normal, but before any Windows screen showed up, it restarted
> again.
>
> This time it offerred me safe mode etc. but I chose regular. I let it
> run chkdsk and all it reported were two lengthy errors in avg log
> files. But when it got to the same place, it restarted again.
>
> This time I chose Safe Mode with Internet and it started fine, and I
> even retrieved a couple of things from Usenet.
>
> I wanted to go back to regular mode so I restarted.
>
> This time it only got as far as the memory count-off, and it stopped.
> Normally it shows a grid about my hard drives and CD dirves.
>
> When I pushed the restart button, it took me to the CMOS/setup page.
> Looking there, I found in the Boot Order, for hard drive it was set to
> None, and had only two options, None and Skip.
So the BIOS can't find a hard drive. Strange, since you were able - once -
to boot to Safe.
Why don't you try booting from a floppy or bootable CD (maybe the XP install
CD?) and try to read the C: directory? If it can't, either the HD or HD
controller is likely bad, maybe both. It is also possible for the power
supply to screw up reading the HD if the PSU is bad.
--
dadiOH
____________________________
dadiOH's dandies v3.06...
...a help file of info about MP3s, recording from
LP/cassette and tips & tricks on this and that.
Get it at http://mysite.verizon.net/xico
Paul
It could be the power supply. Be careful with "excessive test cycles".
If a power supply is going bad, each "on" cycle tests the protection
circuits on the power supply (overvoltage, overcurrent, overtemp).
I remember one bright person, pushed the power button fifty times
in rapid succession, convinced that with enough tries something
good would happen. The supply popped, because of the abuse :-)
Um, don't do that.
The problem with maintenance on your CPU, is the processor could be
a "bare die" kind. Those are easily chipped on the edges, if you
press the heatsink into place tilted. The processor uses either a
"shim" or rubber bumpers on the four corners of the processor
package, to help guide the heatsink into a parallel fit.
You can see the bumpers on this one. This one would be reasonably
well protected against install accidents.
http://cdn.cpu-world.com/CPUs/K7/S_AMD-AX2000DMT3C.jpg
You can see some details, for applying thermal paste to the
silicon die of the processor, here.
http://www.arcticsilver.com/pdf/appmeth/amd/ss/AMD_app_method_surface_spread_v1.1.pdf
But that doc, doesn't explain how to remove the heatsink from the
processor as currently installed. Sometimes the heatsink "sticks"
to the processor, with such good adhesion, you end up pulling
the CPU right out of the socket. (That doesn't hurt a ZIF socket,
at least in my experience. I've never seen a ZIF get permanently
damaged from such an event.) It helps, if the heatsink
assembly is warm, when you try to remove it, to aid in
separating the heatsink from the top of the CPU. And you'd only
do that (take it apart), if you already had a tube of thermal paste
available for re-installation. Try not to get the paste, on any
tiny components present next to the silicon die of the processor.
Use isopropyl alcohol for cleanup, again, avoiding transport of
thermal paste, all over the place. Apply isopropyl to a cleaning
cloth, then wipe the paste off.
*******
The A7M266 compatible processors are here. The motherboard is limited
to FSB266, and judging by the list, it looks like it may have a
PLL filter to aid in usage of models later than the original Athlon.
If you were buying off Ebay, you'd have to be fairly careful to
verify what you were buying.
http://support.asus.com/Cpusupport/List.aspx?SLanguage=en&m=A7M266&p=1&s=10
There is probably a site around somewhere, that decodes the part number,
like the AX1900DMT3C thing. An Ebay advert, should include such information,
so you can do your checking before purchase.
http://www.cpu-world.com/Cores/Palomino.html
It's possible processors other than the ones in the list will work,
but I'm not going to guess at that.
In terms of the power they use, I used to keep a list, but the list
doesn't have the original Athlons 1000Mhz or lower in it. So this list
doesn't allow you to compare the difference in power usage, between
the existing processor and a new one.
http://groups.google.com/group/alt.comp.hardware.homebuilt/msg/fb649923396c8b65?dmode=source
*******
Things I'd check
1) Connect a 2" speaker to the SPKR terminals. You need to be
able to listen to the PC Beep function. It doesn't necessarily
come out of a sound card. It uses a separate pin header on the
motherboard, and a separate speaker. All my computer cases here,
have a small speaker in the front of the case. You can try an
8 ohm speaker, and see if you get sound from it or not.
Your motherboard has a 2x10 "panel" header in the lower right
corner. The 1x4 SPKR connector goes on the pins on the upper
right of that 2x10. There is room to plug up to six cables
into that header, and you'll have at least one cable for
the front panel POWER button.
If you had one of those $200+ fancy computer cases, which lack
a speaker, you'd buy one of these. But look inside the case,
for a 1x4 pin header connector like the one in this diagram,
so you can connect it up. If you see a 1x4 with "SPKR" printed
on it, that's the one you want.
http://tekgems.com/images/large/CASE-SPEAKER-BULK-unit.jpg
One "beep" means "good POST" and the computer thinks it has
a working video card to display with. Other beep patterns
indicate problems. Since the processor and BIOS must be
running to make beeps, *any* beep pattern is a relatively good
sign. It's not completely dead, if it beeps. If there are no beeps,
and you're sure the speaker is wired up, then the processor isn't
executing the BIOS code (bad processor, bad chipset, bad BIOS chip,
bad motherboard or reset circuit).
2) Test the power supply. If you don't know how to effectively
test a power supply, the next best thing is to swap in a
known good supply. In terms of skill set, this doesn't require
too much skill, except avoiding dropping the power supply while
trying to get it into place.
I use a load box to test supplies here, and it draws a light
load to simulate an idling computer. Then, I leave it for a
couple hours test, and use the multimeter on it to check
voltage levels. And that isn't even the beginnings of a good
test, but it's all I can afford to set up. (A Chroma tester
can do a thorough test.)
You have to take careful note, of where all the cables go. I'm
inside computer cases enough on a daily basis, this all seems
easy to me. But it wasn't easy the first time I did it :-)
I had to make drawings then, noting the red mark on the data
cable for pin one, whether the cable had an alignment pin, and
the like. Same goes for disk power cables. While the connector
has a "keyed" shape to prevent wrong insertion, there are been
at least a few individuals who (somehow) managed to jam one in
rotated 180 degrees (destroying the hard drive). It must take
super-human strength to do that :-)
The last supply I bought, for an older system, was this one.
http://www.newegg.com/Product/Product.aspx?Item=N82E16817103013
Sparkle Power Inc ATX-400PN-B204
+3.3V @ 30A, +5V @ 28A, +12V1 @ 18A, +12V2 @ 18A, -12V @ 0.5A, +5VSB @ 2.0A
<--- 150W max -------> <--- 348W max ---------->
That contains sufficient +5V amperes, to handle an Athlon plus
an ATI 9800Pro video card and a couple disk drives.
That is not a perfect power supply, but it hasn't blown up on me.
It makes my UPS beep when I switch on at the back of the PC,
which is why I "subtract one point" from the rating of it. It
means the unit has excessive inrush current. Not a big deal.
3) If the power supply isn't helping, take note of the new symptoms
and post back. Does the power stay on longer, with a new supply ?
Did the new power supply get damaged by the motherboard you're using ?
4) Regular re-application of thermal paste, is a good idea. Generally,
you note the CPU temperature at a given room temperature, a couple
days after you freshen the paste. Then, a couple years later, you
check the CPU temperature again (while room temperature is the same
as before). If it idles say, 10C warmer than it used to, you may want
to take it apart and reapply paste. For CPUs with a metal top surface,
this is a slam dunk and easy to do. With the Athlon, with bare silicon
die, each disassembly/reassembly cycle is "taking a chance".
With a metal top CPU, the only chance you take, is snapping off a
plastic tab on the CPU socket, while re-installing.
You can "cook out" an Athlon. But your symptoms aren't consistent
with that. Your symptoms at the moment, sound like a power supply
issue. A cooked Athlon will be discolored, and fail to POST. This
seems to happen, even if the heatsink was in good contact. It could
be, that an internal short develops inside the processor, and cooking
is the result. But again, you might get some warning, such as
excessive CPU temperature as measured in the BIOS "hardware monitor"
page. See section 4.5.2 in the user manual, for a sample picture.
A CPU temperature of 61C might be "normal" for those old processors,
simply because the silicon die size is tiny, and it's burning up to
65W. That is a lot of thermal flux to push through a small area,
and even with a primo cooler bolted into place, the CPU can still
run pretty hot. You'd want a heatpipe S462 cooler, to improve the
situation. Back in the day, an aluminum heatsink with copper slug
was more common, and those aren't nearly as good. A heatpipe cooler
could get the temperature down below 61C.
HTH,
Paul
Dominique
news:4g0017tsa98ub39e9...@4ax.com:
> (Sorry for the duplicate)
>
<snip>
Don't be sorry, just don't do it!
Bob F
Chkdsk goes through the disk to look for errors. The Errors it is looking for
can cause more errors. The system thinks there might be errors. If you want to
avoid the system messing up more, run chkdsk when it tells you to. It's silly
not to unless you know exactly why it thinks there is a problem, and you know
otherwise.
micky
<glowing...@truely.invalid> wrote:
>On 7/4/2011 8:42 AM, micky wrote:
>> On Sun, 03 Jul 2011 03:58:50 -0400, Paul<nos...@needed.com> wrote:
>>
>>> micky wrote:
>>>> Triple failures? Hard Drive Data? Hard drive? Video Card? CPU?
>>>> Power Supply?
>>>>
>>>> Last night, when I was running, as I have for at least 6 years, a
>>>> home-built computer with 1.5Gig memory and an 800MHaz CPU. All PATA,
>>>> no SATA.
>>>
>>> [Based on the attempts in the header of one of the posts,
>>> crossposted to alt.comp.hardware . I think that is where you
>>> meant it to go.]
>>
>> Thanks
>>
>
>One other thing worth mentioning is that the life of the motherboard
>battery is usually only rated for 5 years from date of manufacture. I
I think I've replaced it once, but that could be 5 years ago already!
>have seen many strange things happen when one finally dies or gets too
>low to fully drive the motherboard, including symptoms like you describe.
Does it need the mobo battery when the computer is plugged and
running?
larry moe 'n curly
micky wrote:
>
> Triple failures? Hard Drive Data? Hard drive? Video Card? CPU?
> Power Supply?
>
> Last night, when I was running, as I have for at least 6 years, a
> home-built computer with 1.5Gig memory and an 800MHaz CPU. All PATA,
> no SATA.
> 1) The SATA harddrive tests fine USBed to another computer.
If you get hard disk errors under Windows, it's probably best to quit
using the drive, rather than let Windows fix the errors because
Windows could write a lot more errors to the disk and make it harder
to recover your data.
> 2) Can I use any little speaker, even if it's bigger than the usual PC
> speaker? (Why do I ask stupid questions like this?)
I'd try to get a speaker rated for at least 8 ohms impedance,
preferrably at least 32 ohms, to prevent loading down the transistor
or chip that drives the speaker. 8 ohms is the most common value, but
some speakers are just 4 ohms, which I'd avoid.
> 3) The quickest way out of this may be replacing my CPU, if that is the problem.
The CPU isn't the problem because the computer wouldn't run nearly as
well as it does even now. The only CPU-related problems you could be
experiencing would be related to overheating because the heatsink is
really dirty, the fan isn't turning fast enough, the heatsink isn't
fastened down right and is crooked (not making complete contact with
the CPU), or there's way too little or way too much heasink grease
between the CPU and heatsink. BTW if you use silver-based grease (no
reason to prefer it), be careful not to get it on the tiny components
surrounding the center square on the CPU package.
> They have them on Ebay from 5 to 11 dollars, shipping included.
If you have an Asus A7M266 mobo, you need a Socket 462 (AKA Socket A)
CPU -- Athlon or Duron. That means you cannot use an Intel or Cyrix
CPU. Also the motherboard's BIOS has to support the CPU or it may not
boot. Some BIOSes are strict about this, while others boot anyway if
they can't identify the CPU. Here's a list of CPUs supported by the
A7V266:
http://support.asus.com/Cpusupport/List.aspx?SLanguage=en&m=A7M266&p=1&s=10
Apparently you need BIOS ver. 1006 or newer to use hard drives bigger
than 137GB, unless you run such drives from a PCI IDE or SATA
controller card.
> 4) Other fans good but a 1" x 1" fan on top of a chip has been running
> badly for a year. It doesn't start unless I push it, and I almost
> never do. My excuse is that I was supposed to migrate to the next
> computer. It's above the AMD761 System Controller. Could
> that have burned out. If it did, it's not worth repairing, right:?
Why, oh why did you let it run it like that for a year? That tiny
fan is for the north bridge chip, which tends to run very hot and can
burn out if overheated, and unless the heatsink has lots of fins at
least 3/4" long, that chip really needs a fan as well as a heatsink.
If you can't find a heatsink that will fit the motherboard, replace
just the fan, but if you can't find a new fan, take it apart and clean
& lube it. Use light machine oil only, not silicone spray or WD-40.
However those tiny fans often burn out electronically, probably from
overheating.
Does the motherboard have some yellow capacitors with a "K" stamped
into the top of each one, near the CPU? If so, they're probably
Fujitsu Functional Polymer, and the early ones were made wrong and had
high failure rates. But any brand of caps, even the highest quality
ones, can go bad in ten years, and brands not from Hong Kong or
Japanese companies can be a lot worse. But even some Japanese
companies made duds -- Nichicon screwed up their normally excellent HM
and HN models from about 2001-2004, Nippon/United Chemicon models KZG
and KZJ have always been undesirable, and you don't want Toshin Kogyo
(TK). Caps don't always bulge or leak when they fail, especially
those made without water in them, like the yellow Fujitsus. The caps
most likely to fail are those next to the north bridge and in the
voltage regulator circuits (between CPU and back of motherboard, near
AGP or PCI-E slot, next to memory module slots), where you'll usually
find donut coils. My last 3 mobo failures were with caps located
next to the memory slots.
Capacitors can go bad in power supplies, too, and the A7M266
motherboard taxes the +5V rail a lot because that's what it uses to
power the CPU, unlike the +12V used by most newer mobos.
Don't try to fix a power supply if you don't know about electronics
because unlike the motherboard it operates at dangerous high voltage,
as much as 340V (and it IS the voltage that can kill you because high
voltage allows higher current to flow through your body). If you do
try to fix it, don't turn on the power except with the power supply
case completely put back together, with all the screws in place
(protects against shock and explosion), and with a GROUNDED 3-wire AC
power cord. Beware that some houses have 3-wire AC outlets that are
not grounded.
Generally when I suspect a hardware failure and don't immediately find
an obvious problem, I first run self-booting diagnostics, like
MemTest86, to bypass Windows and all the software problems it may
cause. I also measure voltages with a digital meter, but this won't
detect rapid variations in voltages. MemTest86 and MemTest86+ are
both very good programs and are based on the same design, but
sometimes one will find errors while the other won't. Also consider
Gold Memory, which for one person detected an error that the other
diagnostics missed. It's important to test for a long time because in
that person's case, it took Gold Memory 9 hours before detecting the
error again.
A few motherboards won't boot if the battery is low. I've heard of
only two designs like that but have seen it in only one.
micky
I'll try. After I read this I counted the number of hot pins and it
was only 10, so I figured a maximum of 10 presses, with rests after
every 2 or 3.
But in practice, since I had to remove the connector, and connect a
jumper between 14 and ground (17) I only had to turn it on once.
Most of the fans didn't go on, but one I had connected stright to the
power supply did and I could feel the air on my fingers, reminding me
the power was still on!!!
The first time through, a lot of voltages were absent or low, and I
thought I found it, but the third time, they were all there.
I reconnected the PSU, and if I were 10 y.o. merely taking it apart
would be enough to fix it, but that didn't work this time. I had
already replaced the video card with an identical new one.
IS THIS IMPORTANT:
The 12 volts was 10.9 in one place and 11.5 in another. With no
load.
>The problem with maintenance on your CPU, is the processor could be
>a "bare die" kind. Those are easily chipped on the edges, if you
>press the heatsink into place tilted. The processor uses either a
>"shim" or rubber bumpers on the four corners of the processor
>package, to help guide the heatsink into a parallel fit.
Well, I got it in the first time, and mounted the heat sink, with the
thermal paste, so I think I can do it.
But I read all the urls to do it better this time.
Thanks. I have a speaker now. no beeps at all!!
> If you had one of those $200+ fancy computer cases, which lack
> a speaker, you'd buy one of these. But look inside the case,
> for a 1x4 pin header connector like the one in this diagram,
> so you can connect it up. If you see a 1x4 with "SPKR" printed
> on it, that's the one you want.
>
> http://tekgems.com/images/large/CASE-SPEAKER-BULK-unit.jpg
>
> One "beep" means "good POST" and the computer thinks it has
> a working video card to display with. Other beep patterns
> indicate problems. Since the processor and BIOS must be
> running to make beeps, *any* beep pattern is a relatively good
> sign. It's not completely dead, if it beeps. If there are no beeps,
> and you're sure the speaker is wired up, then the processor isn't
> executing the BIOS code (bad processor, bad chipset, bad BIOS chip,
> bad motherboard or reset circuit).
>
>2) Test the power supply. If you don't know how to effectively
> test a power supply, the next best thing is to swap in a
> known good supply. In terms of skill set, this doesn't require
> too much skill, except avoiding dropping the power supply while
> trying to get it into place.
>
> I use a load box to test supplies here, and it draws a light
So even if the voltages are good, it coudl be bad, right?
The 10.9 and 11.5 volts are bad signs?
> load to simulate an idling computer. Then, I leave it for a
> couple hours test, and use the multimeter on it to check
> voltage levels. And that isn't even the beginnings of a good
> test, but it's all I can afford to set up. (A Chroma tester
> can do a thorough test.)
>
> You have to take careful note, of where all the cables go. I'm
> inside computer cases enough on a daily basis, this all seems
> easy to me. But it wasn't easy the first time I did it :-)
> I had to make drawings then, noting the red mark on the data
> cable for pin one, whether the cable had an alignment pin, and
> the like. Same goes for disk power cables. While the connector
> has a "keyed" shape to prevent wrong insertion, there are been
> at least a few individuals who (somehow) managed to jam one in
> rotated 180 degrees (destroying the hard drive). It must take
> super-human strength to do that :-)
>
> The last supply I bought, for an older system, was this one.
>
> http://www.newegg.com/Product/Product.aspx?Item=N82E16817103013
>
> Sparkle Power Inc ATX-400PN-B204
Thanks I'd like one that sparkles. Especially on the fourth of
July.
> +3.3V @ 30A, +5V @ 28A, +12V1 @ 18A, +12V2 @ 18A, -12V @ 0.5A, +5VSB @ 2.0A
> <--- 150W max -------> <--- 348W max ---------->
>
> That contains sufficient +5V amperes, to handle an Athlon plus
> an ATI 9800Pro video card and a couple disk drives.
>
> That is not a perfect power supply, but it hasn't blown up on me.
> It makes my UPS beep when I switch on at the back of the PC,
> which is why I "subtract one point" from the rating of it. It
> means the unit has excessive inrush current. Not a big deal.
>
>3) If the power supply isn't helping, take note of the new symptoms
> and post back. Does the power stay on longer, with a new supply ?
> Did the new power supply get damaged by the motherboard you're using ?
Oops, I hope not. Haven't tried it yet.
>4) Regular re-application of thermal paste, is a good idea. Generally,
> you note the CPU temperature at a given room temperature, a couple
> days after you freshen the paste. Then, a couple years later, you
> check the CPU temperature again (while room temperature is the same
> as before). If it idles say, 10C warmer than it used to, you may want
> to take it apart and reapply paste. For CPUs with a metal top surface,
> this is a slam dunk and easy to do. With the Athlon, with bare silicon
> die, each disassembly/reassembly cycle is "taking a chance".
Wow. I didn't know that.
> With a metal top CPU, the only chance you take, is snapping off a
> plastic tab on the CPU socket, while re-installing.
>
> You can "cook out" an Athlon. But your symptoms aren't consistent
> with that. Your symptoms at the moment, sound like a power supply
> issue. A cooked Athlon will be discolored, and fail to POST. This
But it doesn't POST anymore. The last two times I got video, it did
the memory count-off, accurately, to 1.5 gigs. And then it stopped,
no more displays I pressed the Reset button and it went to the BIOS
display. I exitted out of there making no changes, and again it
stopped after the memory count off. And again Reset took to to the
BIOS screen. Then I turned it off with the On/off button and a few
seconds later turned it On again, and no display.
When it's first turned on, it polls both CD roms, the red light goes
on for a second. When I press Reset it only polls one of them.
(It doesn't poll the two floppies, but I thnk I turned that off since
I rarely use them and they never break afaik.)
And the harddrive light goes on for 10 or 15 seconds, but i have no
harddrive installed since yesterday. So I guess it looks and can't
find.
I forgot until now to try a boot CD. I put Hiren's Boot CD in each
drive (I forget which is set to boot.) No good. The CD lights seem
to behave the same -- at most they were on a little longer, but that's
because I was watching. They didn't go off and on again.
And the other lights the same for sure.
One thing is good. The CPU fan and case fan plugged into the mobo
run when the computer is turned on. And the little fan runs a bit
when pushed and stops in a quarter turn.
So given all this, do yuou think my Athlon is toast?
And what about the Southbridge? Northbridge? Could that cause these
symptoms and is it replaceable,
AND BETTER YET, HOW DO I distinguish a bad Northbridge from a bad CPU?
> seems to happen, even if the heatsink was in good contact. It could
> be, that an internal short develops inside the processor, and cooking
> is the result. But again, you might get some warning, such as
> excessive CPU temperature as measured in the BIOS "hardware monitor"
> page.
No temperature hardware monitoring software running.
> See section 4.5.2 in the user manual, for a sample picture.
> A CPU temperature of 61C might be "normal" for those old processors,
> simply because the silicon die size is tiny, and it's burning up to
> 65W. That is a lot of thermal flux to push through a small area,
> and even with a primo cooler bolted into place, the CPU can still
> run pretty hot. You'd want a heatpipe S462 cooler, to improve the
> situation. Back in the day, an aluminum heatsink with copper slug
> was more common, and those aren't nearly as good. A heatpipe cooler
> could get the temperature down below 61C.
I forget what the numbers were that it told me. I tried to find out
what numbes to set my alarm level at, but didn't get firm answers. So
I just used a few degrees above what it was running.
>HTH,
> Paul
Yes, it helps a lot. Thanks a lot.
I plan to reply to Larry M and C tomorrow.
GlowingBlueMist
to swap the battery than it is to determine if it really needs replacing.
About the only definite proof is if the clock looses or resets the time
when the system is unplugged from the wall, but I have seen batteries
that were too low to allow a good boot while still barely maintaining
the clock.
Paul
Well, say the ATX spec was +/- 5% on voltage tolerance. That would
allow the 12V to drop as low as 11.4V. Some of that tolerance
is there, to allow for "crossloading", where one heavily loaded
rail, causes another voltage rail to shift. (That's because
they share a common transformer, and there is only one feedback
path to control all of them. Only a few supplies, like perhaps
an Antec Truepower, have tight control of all rails.)
Common power supply designs, can have one 12V rail or two 12V rails.
If your power supply is of the same vintage as your A7M266 motherboard,
then there will be one 12V rail. And even power supplies with two rails,
may actually be feeding them from one source, and using current limiters
in each path to emulate dual rail operation. None of that is important,
except to suggest all the cables should have the same voltage on them,
as there would be one voltage source, the cabling has low resistance,
so there isn't a good reason for the cables to have radically
different voltages on them.
The basic form of the equation for this would be: V - (R * I) = Vconnector.
The first V there, is the power supply internal voltage source.
The R is the resistance in the path. The I is the current flowing
in the cable.
To get your 10.9 or 11.5 from the cables, you'd need a large R
or a large I. A large R, comes from using thin thin cables.
Or, from a connector making a bad connection (it happens - I
had a connector burn because of that). As a metric, if you
have a hard drive connected to a lossy power cable, and you
hear the drive "spin-down, spin-up", that means the voltage
has dropped to around 11V. I have experienced a voltage that
low here (on a perfectly healthy power supply), when connecting
several heavy loads to a Molex power cable (the power supply
cable might have had three Molex and a floppy power connector
on it). The solution in that case, was to feed the video card
power connector from an entirely separate cable, and the disk
drives shared the other cable. Once I did that, the
hard drive stopped behaving badly (because the voltage
on the cable went back above 11V again).
Try sampling the voltage on a cable that isn't used to power
stuff. A "no load" reading on that cable, will be more indicative
of how healthy the power supply is. Even allowing for crossloading,
the supply should be able to stay above 11.4V on a "no load"
cable. Using our equation again V - (R * I), the no load cable
has zero current flow, so the voltage drop term (R * I)
contributes no voltage drop, and then V = Vconnector and
a voltage reading gives a "true" value.
Both kinds of readings are valuable. Your low readings, could
indicate that the computer needs better power balancing (moving
some loads to another cable). If you can arrange to measure
a "no load" voltage, that would tell us how stressed the
supply is, inside the PSU casing.
But an A7M266 system, shouldn't be loading the 12V much.
The processor runs from the 5V rail, at least on some
of them. The single 12V wire on the main power cable,
would be barely enough to run Vcore, if they did that.
So +5V is a more likely source, if no ATX12V cable is
present.
<<snip>>
>
> Thanks. I have a speaker now. no beeps at all!!
>
Perhaps you'll hear something when you test another power
supply in there. The supply you swap in, must have a decent
+5V ampere rating.
>
> So even if the voltages are good, it coudl be bad, right?
>
> The 10.9 and 11.5 volts are bad signs?
It could still be bad. But voltage readings are largely a waste of time.
Sometimes you get lucky, and they tell you a story. But
a multimeter can't tell you everything you need to know.
There could be excess ripple on the supply. There could be
poor transient response (causing the computer to crash
every time the CPU loading changes). These are things
that can be captured with a digital storage scope, but
those cost $35000 for a good one (the ones they gave us
at work to use, were like that).
<<snip>>
>>
>> You can "cook out" an Athlon. But your symptoms aren't consistent
>> with that. Your symptoms at the moment, sound like a power supply
>> issue. A cooked Athlon will be discolored, and fail to POST. This
>
> But it doesn't POST anymore. The last two times I got video, it did
> the memory count-off, accurately, to 1.5 gigs. And then it stopped,
> no more displays I pressed the Reset button and it went to the BIOS
> display. I exitted out of there making no changes, and again it
> stopped after the memory count off. And again Reset took to to the
> BIOS screen. Then I turned it off with the On/off button and a few
> seconds later turned it On again, and no display.
If you could get the memory count off, a couple things have been tested.
It means you were running BIOS code. To fetch BIOS code, requires
using a path through both the Northbridge and Southbridge. So right
there, as soon as there is evidence BIOS code is running, you've
proved a percentage of the chipset is working.
The memory countoff, only proves the Northbridge was good for millions
of cycles. Fetching BIOS code, provided better hardware coverage, but
perhaps doesn't run quite as many cycles.
With your memory counting behavior, you'd reach for a "PCI POST card",
which displays "progress codes", and it would tell you what subroutine
is failing after the memory test. It could be that something like
a DMI/ESCD update is failing.
(I can't find a good picture - basically, it's a card with at least
a two digit hex display on it, to display progress codes. Price
varies from $100 at a local computer store, to $10 straight from
Hong Kong.)
Changing the number of sticks of RAM present in the computer, will
trigger a DMI update in the BIOS. The BIOS records what hardware is
present, and writes that information into the BIOS chip (in an
area reserved for that purpose). Sometimes, you get a more interesting
response from the computer, by changing the number of RAM sticks,
from one test run to the next. (Make RAM changes, with all power off,
as in, computer unplugged.)
>
> When it's first turned on, it polls both CD roms, the red light goes
> on for a second. When I press Reset it only polls one of them.
>
> (It doesn't poll the two floppies, but I thnk I turned that off since
> I rarely use them and they never break afaik.)
>
> And the harddrive light goes on for 10 or 15 seconds, but i have no
> harddrive installed since yesterday. So I guess it looks and can't
> find.
>
> I forgot until now to try a boot CD. I put Hiren's Boot CD in each
> drive (I forget which is set to boot.) No good. The CD lights seem
> to behave the same -- at most they were on a little longer, but that's
> because I was watching. They didn't go off and on again.
>
> And the other lights the same for sure.
>
> One thing is good. The CPU fan and case fan plugged into the mobo
> run when the computer is turned on. And the little fan runs a bit
> when pushed and stops in a quarter turn.
If the "little fan" is the Northbridge fan, you need to get that
running.
There is a new 40mm here. You'd want to check whether that's the
same size as the one you've got now.
http://www.newegg.com/Product/Product.aspx?Item=N82E16835185039
On those old chipsets, the power level wasn't that high. The fan
might have been overkill. But the thing is, the fan, if it isn't
turning, may actually get in the way of good cooling. If you were
never to use the fan again, you'd want to replace the cooler with
a better passive cooler.
http://www.ixbt.com/mainboard/images/roundamd760may2k1/a7m-big.jpg
The datasheet for the AMD761 Northbridge, says it uses 3W to 4W of
power. A tall passive cooler would be good enough for that, if you
couldn't find a fan. If the passive cooler had a theta_R of 12 C/W,
then at 3W, that would be 36C above 35C case ambient, for a chip
temperature of 71C. And it's allowed to go up to 100C. If a little
air spills from the CPU cooler, and the air flow hits the
passive cooler, the temperature drops further (to maybe 50-60C).
>
> So given all this, do yuou think my Athlon is toast?
>
> And what about the Southbridge? Northbridge? Could that cause these
> symptoms and is it replaceable,
>
> AND BETTER YET, HOW DO I distinguish a bad Northbridge from a bad CPU?
When the thing is totally broken, no beeps, no video, no nothing,
you really can't tell what is broken. Could be BIOS chip, NB, SB,
Processor, Vcore regulator, you name it.
That is when you have to come up with more test cases, such as
swapping in known good stuff, simplifying the hardware setup
by removing some hardware, trying stuff that will (normally)
cause the computer to beep, and so on.
>
> I forget what the numbers were that it told me. I tried to find out
> what numbes to set my alarm level at, but didn't get firm answers. So
> I just used a few degrees above what it was running.
>
>> HTH,
>> Paul
>
> Yes, it helps a lot. Thanks a lot.
>
> I plan to reply to Larry M and C tomorrow.
A good regular operating temperature, is less than 60C to 65C or so,
as a ballpark figure.
The shutdown temperature, might be set to 85-90C or so. (I used to
have some datasheets with that info, but can't find them now.) Your
algorithm, of dialing it a few degrees above normal max, is
also a good way to do it. That way, it'll alert you if cooling
isn't quite as good as it used to be.
Paul
Ken Blake, MVP
wrote:
> On Mon, 04 Jul 2011 09:19:13 -0500, GlowingBlueMist
> <glowing...@truely.invalid> wrote:
>
> >One other thing worth mentioning is that the life of the motherboard
> >battery is usually only rated for 5 years from date of manufacture. I
>
> I think I've replaced it once, but that could be 5 years ago already!
>
> >have seen many strange things happen when one finally dies or gets too
> >low to fully drive the motherboard, including symptoms like you describe.
>
> Does it need the mobo battery when the computer is plugged and
> running?
No. It not only doesn't need it, it doesn't use it. That's the reason
why I've posted the following many times:
Before anyone whose clock is running slow rushes out to buy a new
battery, he should first take note of whether he is losing time while
the computer is running or while it's powered off. If it's while
powered off, the problem *is* very likely the battery. But if it's
while running, it can *not* be the battery, because the battery isn't
used while the computer is running.
--
Ken Blake, Microsoft MVP (Windows Desktop Experience) since 2003
Please Reply to the Newsgroup
Paul
> On Mon, 04 Jul 2011 09:19:13 -0500, GlowingBlueMist
> <glowing...@truely.invalid> wrote:
>> One other thing worth mentioning is that the life of the motherboard
>> battery is usually only rated for 5 years from date of manufacture. I
>
> I think I've replaced it once, but that could be 5 years ago already!
>
>> have seen many strange things happen when one finally dies or gets too
>> low to fully drive the motherboard, including symptoms like you describe.
>
> Does it need the mobo battery when the computer is plugged and
> running?
When the computer is plugged in, and switched on at the back,
the +5VSB rail will be running. That rail runs, even when the
fans aren't spinning. The +5VSB takes the place of the
battery, as a source of current for the clock (RTC).
Another purpose of +5VSB, is to power the system RAM
sticks, while the system is in Standby Suspend To RAM (sleep
or S3).
The power supply has two halves. This shows how the
CR2032 coin cell battery fits in the picture.
CR2032 coin cell -----+
\___ Clock
/
ATX PSU ----- +5VSB supply ---------+
----- All other rails ----------- Motherboard power
If you unplug the computer from the wall, the CR2032 lasts
for three years. In that case, it is the sole source of
power for the clock.
If the computer is plugged in and switched on at the back, then
the +5VSB source is available and takes the place of the battery.
In that case, the battery can last for up to ten years (or
whatever its shelf life rating is). If +5VSB is present,
then no current flows out of the CR2032. As a result, the
battery lasts much longer (up to "shelf life" years).
The motherboard is not allowed to charge the battery. The current
flows out of the battery, but a diode prevents it from flowing
backwards.
The simplest alternative, if you own a multimeter, is to take
a reading off the top surface of the coin cell battery, with
respect to chassis ground. You can pick up a ground connection,
using a metal screw on an I/O connector in the I/O place area
of the computer. I clip on there with one lead of the
multimeter, and then use the red lead to make a voltage
reading from the top of the coin cell. Slightly above +3V,
is a good battery. Below 2.3V is a bad battery. The "knee"
of the battery is relatively sharp, so if the battery is
"on the decline" and near end of life, it'll be flat after
three or four weeks or so.
If you don't own a multimeter, you can remove the battery
and take it to your local Radio Shack. And they can test it
with a meter. But removing the battery is just a PITA,
and I'd just replace it on the spot. If I was going to
go to the trouble of getting it out of that damn socket,
I'd want to resolve the issue immediately, instead of
wasting the gas on a trip to Radio Shack for a test.
You should buy your replacement battery, from a store
with a high "turnover" rate. It is possible to find stores,
selling weak or flat batteries. And at the Mall of all places!
Paul
Rob
I've even seen bad batteries make the clock run fast. Sounds
unlikely I know, but true (was an Asus P4C800E.)
Cheers,
--
Rob
Bob F
> A low battery can keep a system from booting. Fortunately it is
> easier to swap the battery than it is to determine if it really needs
> replacing.
Not if you have a voltmeter. If it reads 3V, it's fine.
Ken Springer
It's always smart, IMO, to check a battery while it's under a load, not
just sitting out on your desk. :-)
--
Ken
Mac OS X 10.6.8
Firefox 5.0
Thunderbird 3.1.11
LibreOffice 3.3.2
Paul
> On 7/7/11 9:38 AM, Bob F wrote:
>> GlowingBlueMist wrote:
>>> A low battery can keep a system from booting. Fortunately it is
>>> easier to swap the battery than it is to determine if it really needs
>>> replacing.
>>
>> Not if you have a voltmeter. If it reads 3V, it's fine.
>>
>>
>
> It's always smart, IMO, to check a battery while it's under a load, not
> just sitting out on your desk. :-)
>
You can take an open circuit reading, and use the curve here to
estimate lifetime remaining. The upper left curve shows voltage
versus temperature and life. About 2.7V, as measured while
it's still sitting in the PC, tells you there is roughly 10% life
left.
http://www.panasonic.com/industrial/includes/pdf/Panasonic_Lithium_CR2032_CR2330.pdf
Paul
Ken Springer
Chart saved to the hard drive.
That would take a much more accurate volt meter than I have. :-)
But, my suggestion works for any battery. I learned that years ago when
in aircraft mechanics school.
Paul
You'll get more of the lifetime of the battery, if you just
meter it in the PC. A "light bulb" test, might cause you to
chuck the battery, while 30% life remains.
Paul
Bill in Co
> On 7/7/11 9:38 AM, Bob F wrote:
>> GlowingBlueMist wrote:
>>> A low battery can keep a system from booting. Fortunately it is
>>> easier to swap the battery than it is to determine if it really needs
>>> replacing.
>>
>> Not if you have a voltmeter. If it reads 3V, it's fine.
>>
>>
>
> It's always smart, IMO, to check a battery while it's under a load, not
> just sitting out on your desk. :-)
True enough! Sometimes the open circuit voltage can be close to normal,
but it's still bad (when tested under load, due to excessive internal
resistance in the cells).
Don't tell anyone, but I sometimes test these dinky batteries directly (but
briefly!) with an ammeter (i.e., for their short circuit current value).
Caveat: I said dinky battery. Don't even think about trying this on a car
battery (at least not with most meters). :-)
Paul
But the load on the CMOS cell is 10 microamps, meaning
the open circuit voltage is very close to the voltage when
a 10 microamp load is present. A load test is unnecessary.
Just using a multimeter reading, and the curve from the datasheet,
is enough to predict lifespan.
Even with a high internal cell resistance, 10 microamps times
that resistance is nothing.
As a matter of fact, there is a 1K ohm resistor in series
with the CMOS coin cell, as a current limiter. That should give
you some idea how little they care about cell resistance. A
ceramic cap is placed at the end of the circuit, where it
joins to the Southbridge, to provide good transient response,
which happens mainly when the computer is running. There are
next to no transients when the battery is being called on
to deliver the 10 microamps.
Paul
Nobody > (Revisited)
Not always...
If you are using a digital voltmeter, it's input impedance is so high
that a crappy battery can read 3.0 v, but put even a small load (like
the mobo CMOS) and it can drop badly.
(DVMs basically put no load on the battery)
Best way to check a CMOS battery is with one of those "real cheap"
analog meters (one that's rated 2Kohms per volt is perfect for this).
That puts a reasonable load on said battery.
Best bet, replace it anyway with a new one (preferably from a source
that has fairly fresh ones).
I've been shot in the foot by CMOS batts to even worry about checking
voltage.
--
"Shit this is it, all the pieces do fit.
We're like that crazy old man jumping
out of the alleyway with a baseball bat,
saying, "Remember me motherfucker?"
Jim “Dandy” Mangrum
Bill in Co
Well, maybe you're right for testing these CMOS/BIOS batteries. 10 ua is
practically nothing, I agree (and it would take a really high internal
resistance to drop much of anything, as you said).
But for flashlight batteries, and the like, I find the load test often
necessary. :-)
But I stand corrected on the former. Thanks. :-)
micky
wrote:
>GlowingBlueMist wrote:
Yes, my battery does read 3V. Easy to measure, since any part of the
the case is the ground, and the battery face is pretty big.
3V was under the tiny load of the CMOS, but if any of you think I
should, I'll measure again while the computer is on. I thought the
load was no greater then, since at that point, the power supply powers
everything.
A mobo battery doesn 't face the load that many batteries do. In fact
it's not much more than a digital watch to keep the clock ticking and
a little more data saved than in a watch. But less load than a
digital watch because it's not used to display the time, change the
time display, or do the alarm beeping. And the battery is several
times as big as most digital watch batteries.
>
Bill in Co
I'd be surprised if it reads any different in circuit, in that case.
If it did, it could indicate an excessive leakage problem somewhere in the
circuit connected to the battery.
micky
Good to know.
I havent' had any problems with computer batteries, unless you count
twice in 27 years that the battery got used up and had to be replaced
(Actually iirc there was no battery in the PCJr and one had to put the
date and time in every time you logged in.)
But I do have major problems with my DVDR-HDD. It needs to know what
time it is to start recording at the right time, yet the time is wrong
soon after it's set. Is this because there's a bad piece of
quartz in the built in clock and if, when this breaks, I buy another
of the same model, will I likely get a better clock?
It has three settings for "Automatic Time Set"
Off
Automatic
Manual.
Whjen it's on Off, the time runs slow (or maybe fast) and the
recording starts late (or ends early**)
When it's on Auto, I think the time jumps around by a minute or two,
and even when it's just "jumped" its not very likely to be right.
When it's on Manual, I guess that means that's Manual Time Set, not
automatic Time Set like the whole box is called., the next screen lets
me set it. I don't want to have set it every few days. I don't even
wear my watch to the bedroom and no clock in the room has a second
indicator.
Because the time is usually wrong, I set it to record starting a
minute or two early and ending a minute late. But if I want to record
or watch something in time period before or after that, it can't
record both at the same time, so I lose part of one show, and when the
clock is wrong in the other direction, I lose part of the other.
Ugh.
>
micky
>
>Both kinds of readings are valuable. Your low readings, could
>indicate that the computer needs better power balancing (moving
>some loads to another cable). If you can arrange to measure
>a "no load" voltage, that would tell us how stressed the
>supply is, inside the PSU casing.
Those numbers I gave you, 10.9 for one 12 volt pin and 11.5 for the
other** were indeed no load readings. (All the other readings were
very close to nominal, 3 and 5.)
**I didn't note positive or negative. I assumed those were correct)
>But an A7M266 system, shouldn't be loading the 12V much.
>The processor runs from the 5V rail, at least on some
>of them. The single 12V wire on the main power cable,
>would be barely enough to run Vcore, if they did that.
>So +5V is a more likely source, if no ATX12V cable is
>present.
>
>>Thanks. I have a speaker now. no beeps at all!!
>>
>
>Perhaps you'll hear something when you test another power
>supply in there. The supply you swap in, must have a decent
>+5V ampere rating.
Well, a friend gave me a bunch of PSUs, and the first two were 235
watts instead of the 275 Silencer (PC Power and Cooling) that I've
been using, but I figured I should test them before installing them.
First one tested dead and then the other. I don't mind him giving me
his junk, but I'm surprised he let it sit around his own apartment
without marking it as bad.
The next one I tool out of a young friend's junk computer and all the
pin voltages were wrong.** Not dead but worng. One was over 12v,
16v iirc.
Finally I found one from the first friend that was still in the
wrapping bag. Also a Silencer, Brand new, and 275 watts.
I tested it and the votlages were all good except the 12 volt pins
were actually a little lower than the one I removed!!!
Rather than install it, I just plugged in the 20-pin conector and
ignored alll the plugs to the CDs, the floppies, harddrrive,, the
extra fan.
That should be okay, shouldn't it?
Turned it on, green light goes on, red light goes on, and still
displays nothing on the monitor.
Doesn't beep either. (I may have unplugged the USB keyboard, but I
don't remember right now.)
Question about the AGP video card in the AGP slot. At the bottom
corner of the card, the blank circuit board has an L-shaped part that
I thought was supposed to hook onto a tab at the end of the slot.
But it seems to just slide into the slot at the same time all the
contacts and the rest of the board slide into the slot. I couldn't
manage to hook it onto anything. I installed ti right, didn't I?
**I don't remember if it was a Dell, but I guess so. I'm going to
call him to make sure. No indication on the Delta Electronics case
that it was meant for a Dell!! .DPS-200PB-101
With a suffix F it's used in a Compaq
Ken Springer
> 3V was under the tiny load of the CMOS, but if any of you think I
> should, I'll measure again while the computer is on.
Sort of out of the box here, but if you've got a computer that is
working right, switch batteries.
But, unless you're also looking for the knowledge, by now you could have
bought a new battery and known for sure. :-)
micky
>
>The simplest alternative, if you own a multimeter, is to take
>a reading off the top surface of the coin cell battery, with
>respect to chassis ground. You can pick up a ground connection,
>using a metal screw on an I/O connector in the I/O place area
>of the computer. I clip on there with one lead of the
>multimeter, and then use the red lead to make a voltage
>reading from the top of the coin cell. Slightly above +3V,
Mine was exactly 3. IIRC. Two days ago and I'm starting to forget
details. :(
>is a good battery. Below 2.3V is a bad battery. The "knee"
>of the battery is relatively sharp, so if the battery is
>"on the decline" and near end of life, it'll be flat after
>three or four weeks or so.
>
>If you don't own a multimeter, you can remove the battery
>and take it to your local Radio Shack. And they can test it
>with a meter. But removing the battery is just a PITA,
Yes it is. Amazingly, some times the battery learns to fly.
>and I'd just replace it on the spot. If I was going to
>go to the trouble of getting it out of that damn socket,
>I'd want to resolve the issue immediately, instead of
>wasting the gas on a trip to Radio Shack for a test.
Thanks.
Paul
One word of warning, is Dell computers had a period of product years,
where the wiring on the main power connector was non standard.
If you're going to pick through a junk pile, it's a good idea
to verify wire color. As long as the manufacturers of the supplies
stick to the standard color scheme (i.e yellow = 12V), then you can
detect a "Dell specific" supply from an "ATX standard" supply. For
a reference on color codes to expect, you can use an ATX spec.
http://www.formfactors.org/developer/specs/atx/ATX12V_1_3dg.pdf (page 30)
You would at least, compare the wire color of your (previously working)
ATX supply, to any new supply you want to test out.
In terms of power rating, you want to check the label on the supply,
and see if it has adequate +5V amps. On my Nforce2 Athlon system,
that number was around 25 amps, which left a bit of room for
the +5V current my video card was using. If those 250 or 275 watt
supplies have relatively low 5V ratings, they might not be the
best choice for a substitute.
*******
To complicate matters, supplies come in two types. One type is the
"honest" type. The label contains both maximum and minimum current
ratings. The "minimum" value tells you that the supply may not
regulate the rail properly, under "no load" conditions. For
example, if you have a supply with a 12V 1A minimum spec, then
if you disconnect all 12V loads, the output won't read 12V. No
damage should result, but the +/-5% spec in the ATX standard
may be violated. That might be viewed as an admission they
don't meet crossloading specs, and by putting the information
on the label, there should be no surprises.
The other kind of supplies, only have maximum current values
listed. With those, we *assume* the supply stays in regulation
all the way down to zero.
When probing your existing power supply, you want to do as follows.
ATX ------- +5V/+12V ------ disk#1 ----- disk#2 ----- Disk#3
Supply ------- +5V/+12V <--- Cable with no loads connected
You take a reading off both cables. The "cable with no loads connected"
gives a "true" reading of current output voltage. Taking a reading
off the end of the first cable, would show the impacts of R * I
voltage drop.
If you read the second cable, and it reads 12V, then you check the first
cable, and it reads 11V, you might conclude from that, that there
is too much resistance in the cabling to support that amount of
peripheral loading.
If both cables read 10.6, especially the no-load cable, then you might
conclude the power supply is out of spec. If the offered load is
below the "minimum" current specified on the label, then you
have your explanation. 10.6 volts is too low to run a hard drive,
so the hard drive may refuse to spin up, or may spin up and then
spin down, over and over again.
A standard supply works roughly as follows. There is a transformer
in the supply. The primary side has high voltage chopped DC on it.
The output voltage is established by "turns ratio" of the transformer
windings. Rectifiers at the output, convert the AC from the
transformer, back into DC.
Core
||
Primary_winding ||3.3V winding
||
||5V winding
||
||12V winding
The power supply observes the outputs, and tried to compensate for
loading. If we heavily load the +5V, the power supply immediately
notices this, and "turns up" the primary. All output voltages rise.
The 5V rises, until it's close to 5V again. The 12V has now
risen above the 12V value, and might be 12.6V. You can actually
tell how the supply is being loaded, by watching the direction
the output rails are moving in. If I saw the five volt rail at
4.8V, and the twelve volt rail at 12.6V, then I conclude from
that, that the five rail is "heavily loaded".
*******
PC Power and Cooling, did make replacement supplies for Dell computers,
which means some of their products will have the "alternate wire
color" pattern and can't be used on your computer. The documentation
may have been removed from their site, as to which models were
Dell substitutes.
*******
You can use the main power connector only, if you want. As long
as the motherboard places some loading on each rail, it should
stay in spec enough to be usable. Your 12V rail will still be
used to power fans.
*******
The L-shaped AGP "heel" has two potential functions. It
can be used to keep the card in the slot, as in this example.
This shows the "slide lock" style retainer, on the AGP slot.
But there was also the spring-loaded side button heel lock,
which is more annoying to disengage.
http://www.interfacebus.com/agp-expansion-card.png
But in an AGP Pro motherboard, the heel would fit right into
the slot, where some extra power pins are located. The heel
has no electrical contacts on this, so it doesn't matter.
The heel may provide some mechanical support in that case,
but not be as secure as a card with a heel lock. AGP Pro
adds a section before and after the regular slot area.
http://www.motherboards.org/files/techspecs/apro_r11a.pdf (Fig.6 page 12)
If you own an AGP Pro card, and have an AGP Pro motherboard,
then the heel area has electrical contacts on it. Not too
many people own AGP Pro cards.
http://upload.wikimedia.org/wikipedia/commons/8/82/Quadro_ELSA_GLoria_II_Pro.jpg
Paul
micky
>micky wrote:
>> Triple failures? Hard Drive Data? Hard drive? Video Card? CPU?
>> Power Supply?
All these are good except, I think, the CPU.
I took the heat sink off and the thermal compound looks pretty bad.
10 or 20% of surface has no compound, on either the sink or the chip.
The rest is dry although I don't know if that matters.
CPU's of this vintage are pretty cheap now, under 20, some under 10,
shipping included.
I have 800 MHz now but I can get up to 1.1 GHz, which would be faster
I suppose.
The question is, The 800 MHz I had overheated, but does that mean I
shouldn't use a 1.1 GHz???
It will have new Arctic Silver Paste, and I will use a new heat sink
just like the old heat sink but clean.
Thanks.
meerkat
"micky" <NONONO...@bigfoot.com> wrote in message
news:fole17t0li66rndku...@4ax.com...
properly. That might be why the 800 packed up.
micky
wrote:
>On Sun, 03 Jul 2011 03:58:50 -0400, Paul <nos...@needed.com> wrote:
>
>>micky wrote:
>>> Triple failures? Hard Drive Data? Hard drive? Video Card? CPU?
>>> Power Supply?
>
>All these are good except, I think, the CPU.
>
>I took the heat sink off and the thermal compound looks pretty bad.
>10 or 20% of surface has no compound, on either the sink or the chip.
>
>The rest is dry although I don't know if that matters.
>
>CPU's of this vintage are pretty cheap now, under 20, some under 10,
>shipping included.
>
>I have 800 MHz now but I can get up to 1.1 GHz, which would be faster
>I suppose.
>
>
>The question is, The 800 MHz I had overheated, but does that mean I
>shouldn't use a 1.1 GHz???
And is a mobile CPU like
http://cgi.ebay.com/AMD-Mobile-Athlon-4-900MHz-CPU-Processor-AHM0900AVS3B-/310327196745?pt=CPUs&hash=item4840f13c49
different from a desktop CPU? Yes, I'm sure it is, but I can't
figure out how.
Bill in Co
> micky wrote:
>> On Tue, 05 Jul 2011 04:31:13 -0400, Paul <nos...@needed.com> wrote:
>>
>>> Both kinds of readings are valuable. Your low readings, could
>>> indicate that the computer needs better power balancing (moving
>>> some loads to another cable). If you can arrange to measure
>>> a "no load" voltage, that would tell us how stressed the
>>> supply is, inside the PSU casing.
<snip>
> A standard supply works roughly as follows. There is a transformer
> in the supply. The primary side has high voltage chopped DC on it.
> The output voltage is established by "turns ratio" of the transformer
> windings. Rectifiers at the output, convert the AC from the
> transformer, back into DC.
I thought a switching power supply just took the 120 VAC input into the
primary of a power transformer, and the secondary of that transformer
stepped down the AC voltage, and that was rectified, and then THAT was
converted into chopped lower voltage DC (PWM, pulse width modulation, where
the duty cycle is continually adjusted (through feedback circuitry) so as to
maintain a constant average DC voltage after filtering).
Are you saying they dispensed with the a need for an input power
transformer? Interesting, if that's the case (I don't know).
Paul
> Paul wrote:
>
>> A standard supply works roughly as follows. There is a transformer
>> in the supply. The primary side has high voltage chopped DC on it.
>> The output voltage is established by "turns ratio" of the transformer
>> windings. Rectifiers at the output, convert the AC from the
>> transformer, back into DC.
>
> I thought a switching power supply just took the 120 VAC input into the
> primary of a power transformer, and the secondary of that transformer
> stepped down the AC voltage, and that was rectified, and then THAT was
> converted into chopped lower voltage DC (PWM, pulse width modulation, where
> the duty cycle is continually adjusted (through feedback circuitry) so as to
> maintain a constant average DC voltage after filtering).
>
> Are you saying they dispensed with the a need for an input power
> transformer? Interesting, if that's the case (I don't know).
>
There is a great schematic here, of an early supply.
http://www.pavouk.org/hw/en_atxps.html
120AC-- rectifier --- 300VDC --- Q1,Q2 drive the --- secondary side
C5,C6 T3 transformer, rectified and
chopped DC filtered to give +12.+5,-5,-12
The transformer T2 separates the base excitation of Q1 and Q2 high
voltage transistors, from the low voltage chip TL494.
That means there is isolation between AC input and DC output, and
between the low voltage control circuit and the high voltage
primary. The control circuit ground and +5VSB ground, are the same
ground as the +12.+5,-5,-12 outputs use.
Using broad brush strokes, it looks like this -
Multiple
Transformers
||
AC input, HV DC side || DC outputs +12, +5, -5, -12
|| DC output +5VSB
|| Low voltage control circuit
Safety ground <---------------------------------------> Black wire
logic ground
It's two power supplies in one. The +5VSB is a separate power supply from
the others. The +5VSB separate power supply, may also deliver other
voltages needed by the control/supervisor circuits. The main supply
is "soft switch-able", as a function of PS-ON (left, middle of diagram).
When the computer sleeps, the +5VSB supply is still running, but
the fan is turned off. Convection cooling is expected to remove the
5-10W of heat coming from the +5VSB supply. On some supplies,
on initial powerup, if you hold your hand near the supply cooling
vent, you can feel instantaneously, warm air that was coming from
the +5VSB circuit. If you're "charging your iPod" while the computer
sleeps, that means +5VSB has to work harder, without the benefit
of fan cooling. So don't throw a blanket over your PC, if it is
in S3 Suspend To RAM :-) There is still stuff in there that gets
warm.
HTH,
Paul
Paul
I used a Mobile in my Nforce2 motherboard. But that's another time,
another place. There are a few details about using mobile processors.
All it takes is several days of "Googling research", to dig up the
salient details.
http://fab51.com/cpu/barton/athlon-e23.html
http://www.ocinside.de/go_e.html?/html/workshop/pinmod/amd_pinmod.html
Your A7M266 may lack certain features that make a Mobile fun to use,
such as good multiplier control. You can do "socket mods", and I was
fortunate that while my motherboard was missing a necessary logic
signal for overclocking, it wasn't needed. (My motherboard had
four of the five necessary multiplier bits, under BIOS control.)
Quite a few of the necessary control functions, can be faked by
cuts and straps around the processor socket, but if you had to
do everything that way, to use your Mobile, that would be nuts.
You certainly want to get a mobile that is socket compatible. In some
cases, laptop mobiles come with different pinout than desktop ones.
It's possible to get screwed on Ebay, if the item description isn't
accurate.
If I were you, I'd stick to the chart and be happy :-)
http://support.asus.com/Cpusupport/List.aspx?SLanguage=en&m=A7M266&p=1&s=10
If you wanted to buy two processors, like a Mobile you had your
eye on, as well as a Palomino Model 6 2100+, you could play with
both of them. If the Mobile is being a PITA to work with
(stuck at ~600MHZ, and you can't figure out how to change it),
then you'll always have the Palomino to use. If you're short
on funds though, buying the Palomino should be enough to get the
computer working again.
*******
When you allow a portion of a "bare die" silicon chip go
dry, and have no thermal paste on it, that area of the
chip gets hotter than the other, still cooled portions.
The delta_T can be high enough to stress the materials
and crack something.
This is why, on the one hand, I don't want people to
obsess over the thermal paste, by changing it too frequently.
But on the other hand, if you use a poor grade of paste
product, and it is "thermally pumped" out from between
the heatsink and silicon die, your processor is going
to end up experiencing more stress than necessary.
The worst kind of paste, is a product Radio Shack
used to sell. It's a zinc oxide paste in silicone oil
base material I've used that stuff on audio amplifier
power transistors (with the mica insulating washer), and
needed to refresh the paste after about six months.
The better enthusiast materials, purchased at places
other than RadioShack, contains a heavier organic
base material, that stays put better. There have
been some products (which are not too popular),
where the organic base was so stiff, users have
troubles applying it. That stuff undoubtedly doesn't
get pumped out in a couple years, but it's also a
bear to get installed right. Using a middle of the
road product, a little less stiff, is a good choice.
This is a follow on, to the stuff I use. I have a
tube of, I think it's AS3. This is AS5. My tube
has lasted so long, that's why it's no longer for
sale. One large plunger lasts a long time, if you're
careful with it. I've done all my computers, over
the years, with one tube like this one :-)
http://www.newegg.com/Product/Product.aspx?Item=N82E16835100007
There are a number of products with similar performance.
About all you want to avoid in a product, is a paste
with a runny, silicon oil, base material. Many
materials contain the same ceramic micro-particles,
so you're getting comparable performance from all of them
(plus/minus 2 or 3 degrees running temp on the processor).
You can use the MSDS (material safety data sheet), to
get the composition of the materials. This one uses Boron
Nitride, just like a lot of competing products. That
is a ceramic material, hard as rock, but in powder form.
http://www.arcticsilver.com/PDF/thermcom/AS5_MSDS_3.pdf
Paul
Bill in Co
Thanks, Paul.
micky
Actually, I left that part of the story out. I think the fan is
okay and it won't get drity faster than normal again for another 20
years or more, and only if I don't maintain my chimney well enough.
The story is OT. I have oil heat. When I first bought this house
and had annual cleaning, they would come and measure several things,
or at least a couple, including stack (flue) temperature, but as time
went on, or I changed companies for some good reason, they seemed to
do things more and more carelessly. For example, there's a
barometric damper, a T in the chimney pipe with an end door in the
short horizontal pipe, and a pivot and a counterweight, that opens a
little when the furnace turns on or off. One guy just taped it closed
once. When I asked the next guy, he just untaped it without answering
my question, Why did he tape it closed? If the orginal installer went
to the trouble to put it in, it shouldn't have been taped shut.
I think they just relied on the color of the flame, which is important
but at least not precise enough. They also never cleaned the oil
filter which is just before the oil pump. (Just a metal screen. For
some reason there is no oil filter the size of an auto oil filter.
There should be one of those, although dirty oil hasn't actually been
a problem. (The nozzles clog after two years but they are normally
replaced every year.)
But more importantly, none of them ever suggested I clean my chimney.
After I was here a year or two I had hired a sweep to clean my
fireplace chimney. amd even though he was on the roof already and
could have made twice as much money just by reminding me I had two
chimneys, he didnt' say a word. I dind't think of it until weeks
after he was there. (Even when I wasn't depressed, I could be very
stupid.) After that I didn't use the fireplace that much and never
for pine etc. and I forgot, and when I finally had the fireplace
chimeny cleaned again 28 years leater, last year, hardly any dirt came
out.. (Both are round, double-wall, metal tubing, maybe stainless,
with caps on top to keep out the rain and animals.). And the furnace
guys never suggested cleaning the furnace chimney.
Eventually one year the CO detector went off while I was sleeping..
Had to turn off the furnace on a fairly cold niight, after opening the
window for an hour! The guy came the next day, and the furnace pipe, a
12 foot sloping section of 4" single wall pipe that leads to the
chimney, only had a 1" opening for exhaust gas, at least in the couple
feet I could see. The rest was soot. He vacuumed that out, but
even he said nothing about having the whole chimney cleaned. I was
also cliincially depressed at the time, or I might have figured this
and more out without their help. .
And by the end of the winter there was soot on the walls where the
furnace output grills are, and soot on tv screens and where there is
electrostatic charge. And even soot where the screw heads are that
hold the sheetrock in place. Why is that electrically charged?
Anyhow I finally thought, on my own, to clean the chimney, and since
then afaict the furnace is as clean as when I got here.
Even that was troulble. The second guy i called was 40 dollars
cheaper and closer to where I lived But he only ran the brush once
up and once down the fireplace chimney, and at the furnace he looked
around, disonnected iirc the sloping pipe at the chimney, and then
said I needed new pipes for 680 dollars. I said I'd think about it
and he only charged me for one chimney. I called his office to find
out why I couldnt' use the pipes I had. She said "creosote".
I called the other guy, who told me my pipes were getting old (thin,
I think, although I think that only shows at the ends where one pipe
goes into another. There are a couple holes at the ends where the
pipes are screwed together, and the inside pipe hole sometimes gets
bigger.) , but they didnt' need replacement yet. I had read that oil
furnaces that use number 1 or 2 oil don't make creosote, but he said
they did sometimes, BUT I DIDN'T HAVE ANY. He charged me his regualr
price, 160 or 180 or something, and didn't recommend any other work.
Then I told him about the other company. He said that their price
for the pipes (680 for 12 feet of single wall galvanizzed stove pipe,
with 4 angle pieces) was about twice what it should be. He obviously
felt there was no reason he couldn't clean the pipes.
An;yhow, that's mj story. :)
larry moe 'n curly
Nobody > (Revisited) wrote:
>
> Best way to check a CMOS battery is with one of those "real cheap"
> analog meters (one that's rated 2Kohms per volt is perfect for this).
> That puts a reasonable load on said battery.
But a cheap analog meter costs more than a Harbor Freight digital
meter that includes a loaded battery test function, especially when
you use a store coupon and get it free with purchase. That HF meter
is actually very accurate, but I'd be leery of using it with high
voltage,
Harbor Freight used to give out those meters free with coupon, but the
coupons have changed, and now require a purchase anywhere from $0.01
to $19.99. I believe the worst HF coupons are in Motor Trend, Popular
Mechanics, Popular Science, and Auto Week, while Playboy, Road &
Track, Car & Driver, and the weekend Wall Street Journal have the
"free with any purchase" coupons.
larry moe 'n curly
micky wrote:
>
> Finally I found one from the first friend that was still in the
> wrapping bag. Also a Silencer, Brand new, and 275 watts.
>
> I tested it and the votlages were all good except the 12 volt pins
> were actually a little lower than the one I removed!!!
Some PSUs, especially old ones like that and the newer high efficiency
designs, don't regulate their voltages very well at light loads and
may require draws of several amps before they work right. For
example, I have some high quality 12-year-old 300W Delta PSUs that I
can't use with low power computers because the +12V rail won't put out
enough voltage to make hard drives spin, unless the +5V rail is loaded
down with at least 3-4 amps. IOW measure voltages under realistic
loads.
micky
Amazing, at least to me. Thanks.
As to my not connecting anything but the mother board, not the other
outputs too, surely the floppies and the CD drives use almost nothing
when they are not in use. The hard drive would use somet. Do
computers with cheap power supplies have a hard time displaying the
entire POST or the Setup/BIOS screens if there is no harddrive?
Or are the fans enough?
larry moe 'n curly
Bill in Co wrote:
>
> Paul wrote:
>
> > A standard supply works roughly as follows. There is a transformer
> > in the supply. The primary side has high voltage chopped DC on it.
> > The output voltage is established by "turns ratio" of the transformer
> > windings. Rectifiers at the output, convert the AC from the
> > transformer, back into DC.
>
> I thought a switching power supply just took the 120 VAC input into the
> primary of a power transformer, and the secondary of that transformer
> stepped down the AC voltage, and that was rectified, and then THAT was
> converted into chopped lower voltage DC (PWM, pulse width modulation, where
> the duty cycle is continually adjusted (through feedback circuitry) so as to
> maintain a constant average DC voltage after filtering).
With 60Hz AC, wouldn't that require a pretty big and heavy transformer
to get a few hundred watts? Because I have the 35-45W PSU from a pre-
IBM PC computer that uses a 60Hz mains stepdown transformer measuring
about 2.5" across (~60mm), and I don't think even modern 1,200W PC
PSUs have transformers that large because they operate them at
something like 100KHz.
micky
>Your A7M266 may lack certain features that make a Mobile fun to use,
>such as good multiplier control. You can do "socket mods", and I was
I don't think so! No socket mods for me. :-)
I've snipped a lot of your post, and plan to reply to that later.
>fortunate that while my motherboard was missing a necessary logic
>signal for overclocking, it wasn't needed. (My motherboard had
>four of the five necessary multiplier bits, under BIOS control.)
>Quite a few of the necessary control functions, can be faked by
>cuts and straps around the processor socket, but if you had to
>do everything that way, to use your Mobile, that would be nuts.
>
>You certainly want to get a mobile that is socket compatible. In some
>cases, laptop mobiles come with different pinout than desktop ones.
>It's possible to get screwed on Ebay, if the item description isn't
>accurate.
>
>If I were you, I'd stick to the chart and be happy :-)
>
>http://support.asus.com/Cpusupport/List.aspx?SLanguage=en&m=A7M266&p=1&s=10
This chart is great. Thanks a lot. It has a lot of info that is not
in the owners manual. It makes clear I can use 266 bus speed, and
that I can go to 1.2 GHz.
I had my mind made up when I noticed the CPU I was looking at was a
server cpu instead of desktop? Does that make any difference?
Should I avoid it?.
It uses the same socket A / socket 462
http://www.cpu-world.com/CPUs/K7/AMD-Athlon%20MP%201200%20-%20AHX1200AMS3C.html
http://cgi.ebay.com/AMD-Athlon-MP-1-2GHz-CPU-Processor-AHX1200AMS3C-/140564215641?pt=CPUs&hash=item20ba47b759
>*******
>
Paul
That's a hard combination to Google for. If I try to find someone running
an MP in a A7M266, all the answers that show up are for A7M266-D.
The FAB51 site claims that the L5 bridge is more like a Product ID,
than an actual hardware changer. So you should be able to plug the
MP processor into the motherboard. Then, it's a matter of what the
BIOS will do, when it reads "MP".
*******
Get that Google-Fu to work, and see what you can dig up. It looks
like some people on "pcper" have tried a few things on A7M266.
http://forums.pcper.com/showthread.php?threadid=234592&highlight=a7m266+barton
Things that higher end processors need, are filtering for the PLL supply
voltage, and a BIOS with the right ROMSIP stuff in it.
Paul
Bill in Co
Perhaps that's the reason, in addition to the size, bulk, and extra cost.
The one thing nice about any electronic devices using a power transformer is
the complete isolation from the mains, but apparently that potential safety
issue has somehow been resolved.
I can still remember the days of the cheapie, "hot chassis", "All American
Five" tube radios. "hot", because they ran directly off the mains line
voltage, without using a power transformer for any isolation from the power
mains (to save money).
Paul
The ATX power supply, still uses transformers. The difference is,
the circuit pumping an alternating current through the transformer,
operates at a high frequency. The physics of the transformers,
means they can be made smaller when operating at the higher
frequency. That improves the power density (watts per cubic foot).
AC --- DC --- Hi_freq_AC --- transformer ---- rectifier --- low_voltage_DC
switching (smaller due
to high Freq)
The ATX supply has isolation between input and output, and this
is actually tested on each unit. You should see a sticker on
the case of the supply, something like "Hi-pot tested". The
Hi-pot test voltage, is much higher than line voltage. And
the specification, may be something like IEC950 or 60950.
That defines the voltage level to be tested to. (For a number
of these standards, they cost money, which is why i don't have
a copy.)
http://webstore.iec.ch/webstore/webstore.nsf/standards+ed/IEC%2060950-1%20Ed.%202.0?OpenDocument
Paul
Bill in Co
Seems to be true. I looked briefly at the schematic again, and didn't see
any direct connection between either side of the AC line and chassis ground.
But I do remember some grounding issues (i.e., potential "hot chassis"
issues) with the earlier "AC-DC" (transformerless) 120V tube radios of
yesteryear, which still seems to be stuck in my brain. And that you had to
use "floating" (ground isolated) instruments to check them out :-). Of
course, they didn't use switching power supplies and the like. They simply
used a 35W4 (as I recall) tube rectifier, and some RC filtering, to get high
voltage DC (as needed for the tube plate voltages), etc.
Paul
There is one here, claiming to be transformerless. I don't see how
they're powering the filaments though. This design is rectifying
power from the line directly, to get V++. You'd have to be very careful
what you touched in here.
http://www.oddmix.com/tech/r39_telefunken_239u_2.html
Paul
Bill in Co
Those transformerless sets powered the filaments directly, in series, using
higher voltage filaments (not the standard 6.3 VAC or 12.6 VAC tubes).
(Actually they did have an audio output transformer for the audio output
stage to drive the loudspeaker, so they weren't completely transformerless).
From distant memory, for the "All American Five" tube radios of the 1950's
and early 1960's, the miniature tubes used were: 12BE6, 12BA6, 12AV6, 35W4
(the rectifier), and 50C5 (audio output). If you hook up all the filaments
in series, you get 12 + 12+ 12 + 35 + 50 = the line voltage, so no filament
transformer was needed. There were some earlier variations of the All
American Five, and what tubes they used (not miniature), but this was the
last one, as I recall.
So by eliminating the power transformer, they saved cost. But in the
process they made the sets a bit more hazardous, due to the inherent "hot
chassis" design.
In short, most of the older radio sets were generally made better (circa
1940's), before they started cutting corners, as they always so. Just like
is true for homes these days vs in the past, where you could see some actual
care, in the detailing.
I could say something similar about radio, and the death of AM radio, and
really good DJ's, but that's another story. Today everything is more
"canned", "prescripted", and is essentially completely "conglomeratized" -
owned by just a few large corporations. I'd call it the wholesale
"Musak-ing" of radio. :-) Just like the legacy of the individual "mom and
pop" stores, it's dying out, or dead already.
micky
We had several radios of that design, and I still have most of them.
I have an Emerson radio, a classic one with a big dial on the right,
with a dial light in a white piece of plastic at the top of it, and a
smaller metal speaker grill on the left, with lots of little holes,
maybe copper colored.
The black plastic case is chipped and when it sat on a small metal
table in the kitchen, sometimes one leg of the radio would go over the
edge. Then when I touched the top of one of the four metal table legs,
that extended above the top shelf, I got a tingle. I didn't worry
about it, but now iiuc, had we reversed the plug, it might have been
wors???
Sometimes the metal trim around the formica kitchen counter**, the
full length of it maybe L-shaped 10 feet, gave the same tingle.
**I don't think they had formica counter fronts in 1950 or 53 when
this house was built. They had a metal channel that covered the
whole front edge, overlapping the top a quarter inch.
When the radio is on a bigger table, or a non-metal table, the bottom
of the chassis doesn't touch anything it shouldn't.
micky
<surly_cu...@earthlink.net> wrote:
>
>
>In short, most of the older radio sets were generally made better (circa
>1940's), before they started cutting corners, as they always so. Just like
It's also weight. The emerson has a hole in the top that makes a
carrying "handle". It's truly portable, if you plug it in when you
get where you're going. Everyone liked that part, I think.
A lot of things were done to make color tv's portable in that way.
Including getting rid of the high-voltage cage. That still scares me
a little, but as one who has to drag the tv that just broke out of the
basement and drag another CRT tv down there, I'm glad they are lighter
than they were. I have a 21 CRT computer monitor that is so heavy,
I can't make it up one flight of stairs in a single try.
>is true for homes these days vs in the past, where you could see some actual
>care, in the detailing.
>
>I could say something similar about radio, and the death of AM radio, and
>really good DJ's, but that's another story. Today everything is more
>"canned", "prescripted", and is essentially completely "conglomeratized" -
>owned by just a few large corporations. I'd call it the wholesale
>"Musak-ing" of radio. :-) Just like the legacy of the individual "mom and
>pop" stores, it's dying out, or dead already.
Yes, so true.
Nobody > (Revisited)
> I thought a switching power supply just took the 120 VAC input into the
> primary of a power transformer, and the secondary of that transformer
> stepped down the AC voltage, and that was rectified, and then THAT was
> converted into chopped lower voltage DC (PWM, pulse width modulation, where
> the duty cycle is continually adjusted (through feedback circuitry) so as to
> maintain a constant average DC voltage after filtering).
>
> Are you saying they dispensed with the a need for an input power
> transformer? Interesting, if that's the case (I don't know).
There's so many variations on "switchmode power supplies" that there's
no single way to explain them.
Some actually don't have true transformers, just inductors, caps,
conttroller circuits and switching transistors. You "shouldn't" see that
in consumer gear (safety issues), but the "East Asians" have snuck them
into some counterfeit wallwarts and bricks. IIRC, this was the case with
a lot of those laptop fires a few years ago.
Years ago, I was a "pre-prototype" tech on two separate switchmode
supplies intended for a proposed variant of the ADCAP MK48 torpedo.
They took the "raw" 275vdc off the torp's powerplant or battery stack
and put out 1200 A @5vdc on one, 850A @5dvc on the other, with a mix of
various +12, -12, +24 to 28, -24 to 28, and about 14 other variables.
Some were common ground or common hot, others were totally floating.
On top of that, the switcher frequencies were "agile" from (IIRC) 2kHz
to 80kHz so that the vibes of the magnetics wouldn't interfere with the
sonar and other stuff.
Rough guess from us tech and the engineers was that there were over 16
different variations in those two units.
This was '76-'82 era technology, and kinda buggy. If there was a logic
glitch in the wrong place/time, the switching transistors would do what
was called "double turn-on" and dump all the capacitor energy into the
6-phase rectifier blocks (literally "forests" of stud-mount diodes).
The results were pretty much the same as setting off a small grenade
inside the case. One of these "events" even took out the side wall of a
burn-in oven. It would have made Dick Cheney look like Mickey Mouse.
--
"Shit this is it, all the pieces do fit.
We're like that crazy old man jumping
out of the alleyway with a baseball bat,
saying, "Remember me motherfucker?"
Jim “Dandy” Mangrum
Nobody > (Revisited)
> "free with any purchase" coupons.
I do remember seeing those 2K/v cheapies at HF a few years ago, seems
like they were about $4, also seen at auto parts stores.
I haven't been keeping track of voltmeter pricing, specs, or such much
lately as I'm awash in the damned things.
At last guesstimate, here's what I have kicking around.
1 genuine USAF PSM-6 "possumeter", 20K/v, mirrored scale
(rebopped for standard batteries)
1 Fluke 77
1 Fluke T5 "not a clamp"
1 genuine "Wiggy" branded Wiggy (1936?)
1 Ideal "combo wiggy" (does continuity as well)
3 or 4 "Tic Tracers"
1 Greenlee clamp
3 Simpson 260s
2 Tripplett 630s
2 RCA VoltOhmysts
1 B&K bench DVM
3 of the old HP "Triple Nickle" transmission test sets
4 or 10 of those little foldup RadShak DVMs stuffed in various
gloveboxes and drawers.
I even have one of those cute little NLS (NonLinear Systems) "little
blue brick" DVMs
Don't even ask about O-scopes... or tube testers...
(but I am proud of a minty Tek 464 storage scope (maybe it's a 465 as
it's got the DVM piggyback top)
(I need to have a garage sale)
Nobody > (Revisited)
> There is one here, claiming to be transformerless. I don't see how
> they're powering the filaments though. This design is rectifying
> power from the line directly, to get V++. You'd have to be very careful
> what you touched in here.
>
> http://www.oddmix.com/tech/r39_telefunken_239u_2.html
>
> Paul
That "C1" "barreter" is probably like the 50A1 "ballast tube" used in
some of the old Zenith TransOceanics to feed the filaments
Excerpt from
http://antiqueradios.com/forums/viewtopic.php?t=68641&highlight=50a1
"The 50A1 is not a tube. It's a ballast resistor under vacuim. I have
found these to be very stable. Usually, the only reason they would go
bad is from a short in the power supply. Zenith used them in
transoceanics, and, I have found it to be stable. Short? Not likely.
Burn out is more possible, and that, generally, when a short occurs.
Bill Cahill"
Euro-style schematics from back in the tube days were often "funken" to
us yanks, but if you chase the right side line from "C1", you'll see the
"humpies" they used to indicate tube filaments.
Nobody > (Revisited)
> On Sat, 9 Jul 2011 16:59:39 -0600, "Bill in Co"
> <surly_cu...@earthlink.net> wrote:
> A lot of things were done to make color tv's portable in that way.
> Including getting rid of the high-voltage cage. That still scares me
> a little, but as one who has to drag the tv that just broke out of the
> basement and drag another CRT tv down there, I'm glad they are lighter
> than they were. I have a 21 CRT computer monitor that is so heavy,
> I can't make it up one flight of stairs in a single try.
Wasn't color, but an old Motorola 19" B&W tube "portable" shook the hell
out of me and scared the crap out of my Gramma. (It had been passed
around the rest of the family for years). IIRC, it was about 70-80
pounds, it had an actual power transformer!
The "Aquadag" coating on the CRT (HV anode ground return) was grounded
to the chassis by a coil spring sorta "half-wrapped" over the top of the
CRT. With all the bouncing around that this tank of a TV had over the
years, the 'dag wore off. No pic...
I'd been fixing TVs since I was 9, was 12 at the time. The old cheater
method for troubleshooting HV back then was to just ground a HV test
lead (rough guess 100K resistor) to the chassis and probe around to see
where you could draw sparks or corona.
Gramma was doing some knitting, but somewhat watching me.
Things were a little cramped, and I ended up leaning my bare arm over
the top of the jug (on the 'dag). You can guess the rest.
Gramma's eyes were like saucers, don't remember if it was from the foul
language coming from inside the TV or the funny jerking around the rest
of my body was doing.
Fixed it by relocating the spring to catch "good" 'dag, then measured
the anode cap .... 27KV!
micky
BTW, I decided I like the old drity heat sink better than the new one
of the same size and similar but IMO inferior design. The clean one
has 20 flat plates (all attached to the bottom) made by cutting 19
deep grooves in a cube of aluminimum. The old one is the same total
size but has a hundred or more black 2mm posts sticking up, so they
radiate from the circumference of the posts, not the surface of the
plates. I'm not positive which is better, or if black is better than
natural aliuminum".
So I put 3 inches of warm water and some liquid dish soap in a white
bowl and put the heat sink in it. Out rushed a lot of black stuff,
making all the water black. I swished it around for 4 or 5 seconds,
then changed the water and detergent, but nothing more came out. I
meant to use a haair dryer but dinner interfered and it seems to have
drip dried very quickly.
>That's a hard combination to Google for. If I try to find someone running
>an MP in a A7M266, all the answers that show up are for A7M266-D.
I do google before I post almost every time, and I did this time, but
this search didn't even occur to me. Instead I searched on
server vs. desktop processor and got several hits right on
topic, but some seemed to say it would be fine, mostly because any
problmes they raised I didn't think applied to me. Some answers were
amateurish, based on a person's experiience with 2 or 3 computers and
the usual other problems. And so many of the hits thoughout this
have dealt with overclocking, which I have no interest in.
>The FAB51 site claims that the L5 bridge is more like a Product ID,
>than an actual hardware changer. So you should be able to plug the
>MP processor into the motherboard. Then, it's a matter of what the
>BIOS will do, when it reads "MP".
>
>*******
>
>Get that Google-Fu to work, and see what you can dig up. It looks
I will try, but maybe not tonight. I was worn out, after I spent
a lot of time on slower ones, then changed my attention to 1.2 Ghz and
266MHz (and this temporary computer doesn't help, becuse I have no
good place for the keyboard Usually it's on my lap, but I have to keep
one ankle on the other knee, and I get tirred of that. Hmmm. I have a
TV table left over from 1961. I should try that. ).
So since you too think it might work, I just ordeed the one above,
hoping he might ship it today. He probably didn't, siince I got
only the Paypal email, but iIt will still probably be here by
Wednesday. If it doesn't work, it was only 10 dollars and another
several days' wait, but I always believe things are going to work.
The therrmal paste is also coming Wedneday or so.
>like some people on "pcper" have tried a few things on A7M266.
>
>http://vip.asus.com/forum/view.aspx?id=20060131035417383&board_id=1&model=A7M266&page=1&SLanguage=en-us
>
>http://forums.pcper.com/showthread.php?threadid=234592&highlight=a7m266+barton
>
>Things that higher end processors need, are filtering for the PLL supply
>voltage, and a BIOS with the right ROMSIP stuff in iit.
> Paul
micky
That's a lot! Maybe it explains why I can't understand a word you
write.
Just kidding.
Glad it didn't kill you. The most I've gotten is far below that, 2K,
and it started my shoulder dislocating again after almost 10 years of
not doing it.
larry moe 'n curly
Nobody > (Revisited) wrote:
>
> At last guesstimate, here's what I have kicking around.
> 1 genuine USAF PSM-6 "possumeter", 20K/v, mirrored scale
> (rebopped for standard batteries)
>
> 1 Fluke 77
>
> 1 Fluke T5 "not a clamp"
>
> 1 genuine "Wiggy" branded Wiggy (1936?)
>
> 1 Ideal "combo wiggy" (does continuity as well)
>
> 3 or 4 "Tic Tracers"
>
> 1 Greenlee clamp
>
> 3 Simpson 260s
>
> 2 Tripplett 630s
>
> 2 RCA VoltOhmysts
>
> 1 B&K bench DVM
>
> 3 of the old HP "Triple Nickle" transmission test sets
>
> 4 or 10 of those little foldup RadShak DVMs stuffed in various
> gloveboxes and drawers.
>
> I even have one of those cute little NLS (NonLinear Systems) "little
> blue brick" DVMs
>
> Don't even ask about O-scopes... or tube testers...
> (but I am proud of a minty Tek 464 storage scope (maybe it's a 465 as
> it's got the DVM piggyback top)
I envy you. My only good meters are a Fluke 73 (somebody owed me
money) and a Radio Shack 50K/volt analog meter built from a kit.
I remember NLS making tiny scopes that could run off battery, and I
once saw about ten of them monitoring signals in some equipment room.
larry moe 'n curly
Bill in Co wrote:
>
> I can still remember the days of the cheapie, "hot chassis", "All American
> Five" tube radios. "hot", because they ran directly off the mains line
> voltage, without using a power transformer for any isolation from the power
> mains (to save money).
I have a 35-year-old TV built like that -- still works, has needed
only 2 capacitors and its flyback connections soldered (not resoldered
-- originally wire wrapped). Then about 15 years ago I bought a new
TV with direct video and audio inputs and thought it would be
transformer isolated, but instead they used optical isolators on those
direct inputs. 4-5 years ago, I got rid of that TV only because the
case had become so brittle that I thought it would one day suddenly
collapse from the weight of the CRT.
larry moe 'n curly
micky wrote:
>
> A lot of things were done to make color tv's portable in that way.
> Including getting rid of the high-voltage cage. That still scares me
> a little, but as one who has to drag the tv that just broke out of the
> basement and drag another CRT tv down there, I'm glad they are lighter
> than they were.
I remember an office full of CRT monitors sitting next to each other
that would twitch like clockwork because the flybacks caused
interference, and this would happen even when they were 6 feet apart.
OTOH most of our own monitors could sit right next to one another and
not twitch, and almost all of them had high voltage cages or cases
that were copper plated. One of them without much of a cage instead
had an extra winding on the flyback that was connected to nothing but
a 1/4" x 3" circuit board that sat by itself -- an antenna to put out
an out-of-phase signal to cancel interference.
Bill in Co
I wonder if the current generations of TVs even use a power transformer. I
think it's a good idea (for the simplest isolation), but I'm old school.
:-) I suppose if someone had a 1000W TV, the added bulk might be a
consideration.
larry moe 'n curly
micky wrote:
>
> On Sat, 9 Jul 2011 01:48:42 -0700 (PDT), "larry moe 'n curly"
> <larrymo...@my-deja.com> wrote:
>
> > Some PSUs, especially old ones like that and the newer high efficiency
> > designs, don't regulate their voltages very well at light loads and
> > may require draws of several amps before they work right.
>
> As to my not connecting anything but the mother board, not the other
> outputs too, surely the floppies and the CD drives use almost nothing
> when they are not in use. The hard drive would use somet. Do
> computers with cheap power supplies have a hard time displaying the
> entire POST or the Setup/BIOS screens if there is no harddrive?
>
> Or are the fans enough?
Every computer will start with just the motherboard installed because
every time I get a new mobo, I first test it without any drives
installed. A mobo applies a bigger load than a hard drive. BTW a
very few mobos won't boot at all if the clock battery is dead or
missing.
Fans typically draw just half their rated amps, and I think you'd need
5-10 of them for a decent load. Some people build PSU test loads from
automotive bulbs, especially old sealed beam headlamps, or they string
together a bunch of 5-10 watt resistors of equal resistance value in
series or parallel until they get the right amps and watts (BTW,
operate resistors at no more than half their rated wattage or they'll
get hot enough to melt plastic, burn fabric and wood, maybe even start
fires).
Paul
You might want something around 12V @ 1A load, and fans might draw
around 0.5A. Your motherboard doesn't power the processor from +12V
(most likely), so your +12V rail might only have fans on it.
Hard drives are a reasonably light load. The motor uses +12V, and
the current could be between 0.3A to 0.6A or so (idling). A couple hard
drives, with data cable disconnected, might serve as a load. Hard
drives can draw up to 2.5 amps during spinup, for the first ten
seconds, so the loading isn't constant as such. The power supply
has to supply more current, right after you turn on the circuit.
After ten seconds, the spindles are up to speed on the drives,
and the current flow level drops back to 0.3A to 0.6A on
each drive.
When testing power supplies of unknown quality, you don't
use your "good" hard drives :-) That's the problem with
using hard drives, is what happens if the power supply 12V
rail is running at 15V.
If your ATX supply had a 12V @ 15A rating, and you connected
two hard drives, the hard drives will draw up to 5A for the
first ten seconds, settling down to 1.2A or less, once up to
speed. If you use too many hard drives, you could violate
the rating, due to the spinup current draw. In this example,
six hard drives would be "max" (as then, you'd be drawing
15A for the first ten seconds).
Hard drives are convenient as a load, since the connector
is ready to go, and easily mates with the ATX supply. As long
as you're aware of the spinup current, and don't use
too many, it makes a fine source of loading.
*******
I use power resistors from an electronics store, for test loads,
and those can be used for test purposes. This is an example
of what you can find for resistors. These particular ones have
a metal body, and can handle 25W or 50W. The 50W ones are
a bit bigger. And power resistors get good and hot. On my
load box, I use an 80mm fan for cooling, across the resistors.
There is material between the metal body and the wire wound resistor
inside, so the metal should not be electrified.
http://www.galco.com/techdoc/nte/25wm220_cp.pdf
The 25WM012 is a 12 ohm resistor. If connected across the 12V
rail, the current draw is 12V / 12 ohms = 1 amp. The power
dissipated, is 12V * 1A = 12 watts. And 12 watts, in such a
small package, should have some cool air across it. If you
were only dumping a couple watts, it might be sufficient to
convection cool it.
You adjust the ohms of the one you purchase, to get the
loading you desire. If you used a 3 ohm resistor,
12V / 3ohm = 4 amps. And 12V * 4A = 48W, and that 50W resistor
is cooking. The 25W resistor is probably in serious trouble.
You can use combinations of resistors, in series or in
parallel, to help spread out the heating. I don't generally
set up my load box resistors, to get near their name plate
rating in watts. I try to keep the watts within reason,
using a fraction of the watt rating so the surface
temperature won't be too high.
Some ceramic resistors, can operate at 300C, in which
case, you don't need the fan, but don't touch it :-)
The 25WM012 above can take 275C, but I probably wouldn't
push that one that hard. I have some wirewound ceramic
power resistors, and you can tell by looking at them,
they'll take a lot of heat.
Generally speaking, light bulbs are not recommended as test
loads, at least when power is tightly constrained. To
give an example, I bought a DC adapter (wall wart) at the store,
something like 12V @ 2A, connected an automotive light bulb,
perhaps 12V @ 1A, and it causes the DC adapter to shut off.
The bulb would not light. So I can't test that adapter with the bulb
as a load. The reason for this, is when the automotive light
bulb is cold, the initial current draw can be double or more,
the "running" current flow level. When the filament is hot,
that is when it draws 1A. My DC adapter, being a switcher,
detects the overload instantly, and won't allow the cold
filament overload to last for more than a fraction of a second.
More crude power technologies, don't react immediately like that,
which is why they might allow the bulb to warm enough not to continue
to be an overload. Switching supplies can be set up, to shut off
very quickly, before the bulb even gets warm.
So in some ways, the "bad habits" of the disk drive
(the 2.5A spinup current), are mirrored in the operation
of the automotive light bulb (fraction of a second, cold
filament current flow). As long as you're aware of
the electrical characteristics of the initial inrush,
you can figure out a safe number of loads to use.
If the ATX supply had a 12V @ 15A rating, and you connect a
12V @ 1A automotive light bulb, no problem. The initial
cold filament surge current will be handled with ease.
But if the total loading of light bulbs, say eight light bulbs
in parallel, is slightly over half the supply ampere rating,
there is a chance the supply would "turn off on overload",
due to the initial cold filament surge. This is one reason,
I use real power resistors, because while they do have a
temperature coefficient, it isn't nearly as non-linear as a
light bulb.
There are metals with a pretty low temperature coefficient of
electrical resistance, such as manganin and related alloys.
The resistor used inside a multimeter, when making current
flow measurements, will have a low temperature coefficient,
and the resistance value won't change much at all when it
gets warm. By comparison, my load box would be inferior to
one of those. But using the automotive light bulb, as a
current shunt inside the multimeter, while cheap, would
throw any measurements out the window (50% error plus).
You use this stuff, for utmost accuracy, when the resistance
must remain stable. For best results, this stuff has to
be handled with care, annealed and the like.
http://en.wikipedia.org/wiki/Manganin
http://www.o-digital.com/uploads/2179/2192-1/Manganin_Shunt_42.jpg
You wouldn't waste that stuff for making load resistors, as
it would be too expensive.
So those are examples of some loads and resistor types.
Since you have hard drives on hand, those are the
closest thing within reach, to draw a bit of current.
Paul