Overhauling a Dell Dimension 8400

  • Thread starter Thread starter Searcher7
  • Start date Start date
and t could run around three applications before it was out of RAM. Maybe 256MB will be enough to just boot it. Some Googling finds GX110 using Intel 810 chipset. And that might take two low-density 256MB DIMMs max. The DIMM would have sixteen chips on it, to be low-density. In the past, there were a lot more high density DIMMs for sale on places like Ebay. I got mine from Crucial (I have an older machine I bought four of those for it), but Crucial no longer sells that stuff. This is an example of what you don't want. This is an 8 chip module. http://www.valueram.com/datasheets/KVR133X64C3_256.pdf I'd just give it a try, without changing the RAM. And since the GX110 is a Dell, you could use your Dell WinXP CD and it should activate immediately. If it boots OK, then you have an incentive to buy some more RAM. ******* The listed minimums for RAM here, are 64MB and 128MB. The 128MB is the "recommended" amount. So you never know, you might be able to run one application on it :-)
http://en.wikipedia.org/wiki/Windows_xp Paul
I guess I should just try re-installing XP on the drive in the GX110. The drive already has a clean install of XP on it, but it was done in a different machine. So perhaps I can avoid the blue screen-restart cycle by having the XP disk in the DVD drive when I attempt to boot up.

On the subject of ram. After doing some reading about how the HP 7955 can be finicky even with the recommended ram I did some more swapping and managed to get one stick of 256mb to put an end to the 1-3-3-1- code. But I then got a 1-2-2 code which says DMA failure. So I started adding the compnents. It wasn't until I added the video card/monitor that I got past that code also.

But like the DELL GX 110, at thje moment I can only use 256mb of ram. And now, even though I can get into the HP 7955 BIOS, when I attempt to boot up I can only get as far as a black screen with the blinking "underscore". (And the only way to turn the HP 7955 on and off is by plugging and unplugging the AC cord).

Thanks.

Darren Harris
Staten Island, New York.

On the HP 7955, then you need to find some means to boot, where
you won't run into the storage device problem. Does the 7955 boot from USB ?
You could even be stuck at the flashing cursor, if there's a problem with
the contents of the drive (as far as the computer is concerned).

As an example of an alternate boot device, I have Ubuntu
installed on a USB flash, and I use that one flash stick, to boot
different computers here. It's booted my main computer and my laptop,
even though they have different hardware. So that's an example of a
universal test boot. What that consists of, is a read-only portion
containing the Linux LiveCD files, plus a file which serves as
a storage area when the OS is running.

Where that won't work, is on the oldest computer, because it won't
boot from USB under any circumstances. It only boots from regular
storage devices, like hard drives or CD/DVD drives.

Does the HP 7955 have a floppy drive ? If so, does a copy of memtest86+
boot and run ? That's another form of crude boot test.

(Various download files, half-way down this page...)

http://www.memtest.org/

If the machine is ignoring the front power switch, the power switch
itself could be defective. Or, there could be a stress on the
Southbridge, which causes the PS_ON# signal to be ON all the time.
I had that problem here, on my P2B-S when the IDE cable was half out
of its socket. That caused enough stress on the Southbridge, to
immediately cause the power to come on. It's even possible for
a computer ignoring the front power button, to have a PSU problem.
(A defect in the ability to detect the logic level of PS_ON#.)

Using my "simplify" strategy, I'd disconnect storage cabling, and
try booting with the USB flash. Or, run with just the floppy
cabled up and try the memtest86+ floppy. Only try the hard drive,
after you get something else working.

Also, for the hard drive aspect, you can enter the BIOS and see
if the hard drive is properly identified. The Southbridge storage
ports, the connected storage devices should be listed. That proves
that an attempt to get identity information, worked OK. It helps
prove the cabling is working, the drive is powered, and so on.

If the computer has a USB card reader, instead of a floppy drive
(in the 3.5" bay), try disconnecting the card reader. Again,
this is part of the simplify strategy. Only have as much hardware
connected, as is needed for your current experimental attempt.
Once you get booted and running, you can slowly add a piece of
hardware at a time, and see if it stops booting or otherwise
complains.

Paul
 
and t could run around three applications before it was out of RAM. Maybe256MB will be enough to just boot it. Some Googling finds GX110 using Intel 810 chipset. And that might take two low-density 256MB DIMMs max. The DIMM would have sixteen chips on it, to be low-density. In the past, there were a lot more high density DIMMs for sale on places like Ebay. I got mine from Crucial (I have an older machine I bought four of those for it), but Crucial no longer sells that stuff. This is an example of what you don't want.This is an 8 chip module. http://www.valueram.com/datasheets/KVR133X64C3_256.pdf I'd just give it a try, without changing the RAM. And since the GX110 is a Dell, you could use your Dell WinXP CD and it should activate immediately. If it boots OK, then you have an incentive to buy some more RAM. ******* The listed minimums for RAM here, are 64MB and 128MB. The 128MB is the"recommended" amount. So you never know, you might be able to run one application on it :-)

http://en.wikipedia.org/wiki/Windows_xp Paul






On the HP 7955, then you need to find some means to boot, where

you won't run into the storage device problem. Does the 7955 boot from USB ?

You could even be stuck at the flashing cursor, if there's a problem with

the contents of the drive (as far as the computer is concerned).



As an example of an alternate boot device, I have Ubuntu

installed on a USB flash, and I use that one flash stick, to boot

different computers here. It's booted my main computer and my laptop,

even though they have different hardware. So that's an example of a

universal test boot. What that consists of, is a read-only portion

containing the Linux LiveCD files, plus a file which serves as

a storage area when the OS is running.



Where that won't work, is on the oldest computer, because it won't

boot from USB under any circumstances. It only boots from regular

storage devices, like hard drives or CD/DVD drives.



Does the HP 7955 have a floppy drive ? If so, does a copy of memtest86+

boot and run ? That's another form of crude boot test.



(Various download files, half-way down this page...)



http://www.memtest.org/



If the machine is ignoring the front power switch, the power switch

itself could be defective. Or, there could be a stress on the

Southbridge, which causes the PS_ON# signal to be ON all the time.

I had that problem here, on my P2B-S when the IDE cable was half out

of its socket. That caused enough stress on the Southbridge, to

immediately cause the power to come on. It's even possible for

a computer ignoring the front power button, to have a PSU problem.

(A defect in the ability to detect the logic level of PS_ON#.)



Using my "simplify" strategy, I'd disconnect storage cabling, and

try booting with the USB flash. Or, run with just the floppy

cabled up and try the memtest86+ floppy. Only try the hard drive,

after you get something else working.



Also, for the hard drive aspect, you can enter the BIOS and see

if the hard drive is properly identified. The Southbridge storage

ports, the connected storage devices should be listed. That proves

that an attempt to get identity information, worked OK. It helps

prove the cabling is working, the drive is powered, and so on.



If the computer has a USB card reader, instead of a floppy drive

(in the 3.5" bay), try disconnecting the card reader. Again,

this is part of the simplify strategy. Only have as much hardware

connected, as is needed for your current experimental attempt.

Once you get booted and running, you can slowly add a piece of

hardware at a time, and see if it stops booting or otherwise

complains.



Paul

I noticed a message after the HP splash screen during boot up that says that the system CMOS checksum is bad and as a result the default configurationis used. (So the boot order goes back to default).

The CMOS battery holder is loose and definitely needs to be re-soldered tight. But I wouldn't think that would affect things after I powered up.

I also tried putting in my XP disk, but I'll have to change out the original "CD Writer" that came with the HP 7955 for one of my DVD drives.

I haven't yet figured out how to install Ubuntu on my flash drive yet, but will try again when I get home. I downloaded that "Pre-Compiled Bootable ISO (.zip)" and also the "Auto-installer for USB Key (Win 9x/2k/xp/7)". If I'm lucky enough to be able to install Ubuntu to my flash drive I'll also tryusing it with the DELL GX110 that resets after the blue screen is reached.

Thanks.

Darren Harris
Staten Island,New York.
 
I noticed a message after the HP splash screen during boot up that says that
the system CMOS checksum is bad and as a result the default configuration is used.
(So the boot order goes back to default).

The CMOS battery holder is loose and definitely needs to be re-soldered tight.
But I wouldn't think that would affect things after I powered up.

With a flaky battery holder, if you turn off mains power, then you could lose
the CMOS settings. If the computer remains switched on at the back, the
CMOS should still be powered up, and the settings preserved. (The CMOS runs
from the battery, or the CMOS runs from +5VSB. ORing diodes select a source
of power, whatever is available. If the battery is pulled, then +5VSB can
take its place. +5VSB is regulated down to 3V, otherwise known as 3VSB
on an Intel schematic.)

If the CMOS settings are corrupt, you can load setup defaults, enter the BIOS
and correct the boot order, save and exit. The new CMOS values should be valid
as long as the mains power stays on at the back of the computer (you can do warm
restarts, push the reset button, etc.). Only if there is a fault in the 3VSB that
feeds the CMOS circuit, would you have continual complaints about bad CMOS contents.

For example, if you "clear CMOS" and leave the AC power on by accident, it burns
the ORing diodes in 3VSB, and then you can have a permanent problem with the
CMOS receiving power. Whether the battery holder is soldered or not. One poster
I conversed with, he managed to fix his diodes by soldering in some 1N914/1N4148
type diodes as replacements, and was satisfied with the results. So I've worked
with at least one poster, to repair that kind of damage. I have three Asus motherboards
here, that all use the same three pin, dual diode device, and it's a pretty tricky
soldering job to fit regular diodes to take its place. So that poster must have
been pretty skilled with a soldering iron. The pads on that device, are pretty tiny.

What boggles the mind, is why so many motherboards are booby trapped with
that diode burning "feature". Even on chipsets that don't need to handle
CMOS clearing the old-fashioned way, they've still been set up with the
burn-able CMOS setup. Sad really. It means I have to continually remind
people to "turn off all power" before clearing CMOS. Some manuals neglect
to mention it, and the motherboards are booby trapped. A recipe for
unnecessary repair jobs and damaged goods.

The CMOS battery cannot burn the diodes, because there is a 1K ohm
resistor in series with the CMOS battery. The source of power that
burns the diodes, is +5VSB from the ATX power supply. As long as you turn
off the power supply, then, it's safe to clear CMOS.
I also tried putting in my XP disk, but I'll have to change out the original
"CD Writer" that came with the HP 7955 for one of my DVD drives.

I haven't yet figured out how to install Ubuntu on my flash drive yet, but will
try again when I get home. I downloaded that "Pre-Compiled Bootable ISO (.zip)"
and also the "Auto-installer for USB Key (Win 9x/2k/xp/7)". If I'm lucky enough
to be able to install Ubuntu to my flash drive I'll also try using it with the
DELL GX110 that resets after the blue screen is reached.

Thanks.

Darren Harris
Staten Island,New York.

I'm quite confident you'll get that working :-) You have the skills.

Paul
 
(e-mail address removed)2.com wrote:










With a flaky battery holder, if you turn off mains power, then you could lose

the CMOS settings. If the computer remains switched on at the back, the

CMOS should still be powered up, and the settings preserved. (The CMOS runs

from the battery, or the CMOS runs from +5VSB. ORing diodes select a source

of power, whatever is available. If the battery is pulled, then +5VSB can

take its place. +5VSB is regulated down to 3V, otherwise known as 3VSB

on an Intel schematic.)



If the CMOS settings are corrupt, you can load setup defaults, enter the BIOS

and correct the boot order, save and exit. The new CMOS values should be valid

as long as the mains power stays on at the back of the computer (you can do warm

restarts, push the reset button, etc.). Only if there is a fault in the 3VSB that

feeds the CMOS circuit, would you have continual complaints about bad CMOS contents.



For example, if you "clear CMOS" and leave the AC power on by accident, it burns

the ORing diodes in 3VSB, and then you can have a permanent problem with the

CMOS receiving power. Whether the battery holder is soldered or not. One poster

I conversed with, he managed to fix his diodes by soldering in some 1N914/1N4148

type diodes as replacements, and was satisfied with the results. So I've worked

with at least one poster, to repair that kind of damage. I have three Asus motherboards

here, that all use the same three pin, dual diode device, and it's a pretty tricky

soldering job to fit regular diodes to take its place. So that poster must have

been pretty skilled with a soldering iron. The pads on that device, are pretty tiny.



What boggles the mind, is why so many motherboards are booby trapped with

that diode burning "feature". Even on chipsets that don't need to handle

CMOS clearing the old-fashioned way, they've still been set up with the

burn-able CMOS setup. Sad really. It means I have to continually remind

people to "turn off all power" before clearing CMOS. Some manuals neglect

to mention it, and the motherboards are booby trapped. A recipe for

unnecessary repair jobs and damaged goods.



The CMOS battery cannot burn the diodes, because there is a 1K ohm

resistor in series with the CMOS battery. The source of power that

burns the diodes, is +5VSB from the ATX power supply. As long as you turn

off the power supply, then, it's safe to clear CMOS.













I'm quite confident you'll get that working :-) You have the skills.



Paul

The library computers are very limited in what they allow you to do. So I don't think that will happen.

Anyway, before I head back home and try to get past the black screen with the blinking underscore on my HP 7955, this is where I'm at with two other machines.

DEL GX110: I did manage to be able to get a total of 512mb to be seen by the DELL GX110 PC. I attempted to install XP three times using a regular XP disk and also a DELL OEM disk. The machine would consistently freeze at the end of the install upon reaching the desktop. On all subsequent re-boots the pc would freeze sometimes just before reaching the desktop, but usually within several seconds of getting there. If not immediately, then when I move the mouse.

ASUS CUV4X: I did manage to also get a total of 512mb to be seen by the PC with the ASUS CUV4X motherboard. (That is using the same two sticks). Afteran XP installation the best I was able to get was a media player to open but not produce much in the way of audio along with off color video:

So after a second XP install opening up a media player would either freeze the system or just not work with the video files at all. (Depending on whatmedia player I tried).

So I'm wondering if the problem could be the ram, which I used in both systems or perhaps the flash drive, which I've been plugging and unplugging between systems to install the media players and transfer video files. Could it have some sort of virus that corrupts anything it touches?

It just seems extremely unlikely that I can't get a media player to work (completely) with these two systems as well as my Mini-ITX.

Thanks.

Darren Harris
Staten Island, New York.
 
The library computers are very limited in what they allow you to do. So I don't think that will happen.

Anyway, before I head back home and try to get past the black screen with the blinking underscore on my HP 7955, this is where I'm at with two other machines.

DEL GX110: I did manage to be able to get a total of 512mb to be seen by the DELL GX110 PC. I attempted to install XP three times using a regular XP disk and also a DELL OEM disk. The machine would consistently freeze at the end of the install upon reaching the desktop. On all subsequent re-boots the pc would freeze sometimes just before reaching the desktop, but usually within several seconds of getting there. If not immediately, then when I move the mouse.

ASUS CUV4X: I did manage to also get a total of 512mb to be seen by the PC with the ASUS CUV4X motherboard. (That is using the same two sticks). After an XP installation the best I was able to get was a media player to open but not produce much in the way of audio along with off color video:

So after a second XP install opening up a media player would either freeze the system or just not work with the video files at all. (Depending on what media player I tried).

So I'm wondering if the problem could be the ram, which I used in both systems or perhaps the flash drive, which I've been plugging and unplugging between systems to install the media players and transfer video files. Could it have some sort of virus that corrupts anything it touches?

It just seems extremely unlikely that I can't get a media player to work (completely) with these two systems as well as my Mini-ITX.

Thanks.

Darren Harris
Staten Island, New York.

There's an "offline scanner" available here. 237MB download.

http://support.kaspersky.com/8092

http://rescuedisk.kaspersky-labs.com/rescuedisk/updatable/kav_rescue_10.iso

You don't have to use the rescue2usb thing. I just burn a CD with the
ISO9660 file, then boot that. I've also tried it from a USB flash, and
I think that worked OK.

That tool works best, if you update the virus definitions, but
if your networking isn't working, you can still run it with
the definitions on the CD. The disc will be "freshest", if
you use it immediately after downloading the image.

If you think it's a RAM problem, you can always give memtest86+ a try.
The downloads are half-way down this page. Memtest86+ is a program
you boot the computer with, so you don't need Windows or Linux to be
able to use it. I keep a floppy with this on it, to test the RAM
on my current Core2 system.

http://www.memtest.org/

Paul
 
There's an "offline scanner" available here. 237MB download.



http://support.kaspersky.com/8092



http://rescuedisk.kaspersky-labs.com/rescuedisk/updatable/kav_rescue_10.iso



You don't have to use the rescue2usb thing. I just burn a CD with the

ISO9660 file, then boot that. I've also tried it from a USB flash, and

I think that worked OK.



That tool works best, if you update the virus definitions, but

if your networking isn't working, you can still run it with

the definitions on the CD. The disc will be "freshest", if

you use it immediately after downloading the image.



If you think it's a RAM problem, you can always give memtest86+ a try.

The downloads are half-way down this page. Memtest86+ is a program

you boot the computer with, so you don't need Windows or Linux to be

able to use it. I keep a floppy with this on it, to test the RAM

on my current Core2 system.



http://www.memtest.org/



Paul

I can't burn anything, so I'd have to go the USB route.

As for memtest86+, I tried installing that last week to no avail. (I couldn't boot from the USB or floppy drives).

Yesterday after wading through apps that required an internet connection toinstall or payment I came across USBantivirus which found something in thememory path called SYSINFO2. (But it would fix it unless I upgraded for $15). I couldn't find any reference to in the search of "My Computer".

Thanks.

Darren Harris
Staten Island, New York.
 
Back
Top