CDROM listed as "unknown" in BIOS but Windows detects correct name

[kev833]

I discovered an interesting problem. Maybe someone can help me figure it out. My Dell Optiplex GX520 Desktop running Windosws XP SP3 had a bad CDROM drive which I replaced with a dusted off Sony CD-RW/DVD-ROM. This is what happens when I played with the master/slave/cable-select jumper pin.

1. When master pin is jumpered: bios detects correct name of cdrom, windows doesn’t boot to desktop

2. When slave pin is jumpered: bios detects cdrom name as unknown, windows boots to desktop and detects correct name of cdrom

3. When cable-select pin is jumpered: bios detects correct name of cdrom, windows doesn’t boot to desktop

4. When no pin is jumpered: bios detects cdrom name as unknown, window boots to desktop and detects correct name of cdrom.

The model of the cdrom is Sony CD-RW CRX320EE.

 

[Paul]

I discovered an interesting problem. Maybe someone can help me figure it out. My Dell Optiplex GX520 Desktop running Windosws XP SP3 had a bad CDROM drive which I replaced with a dusted off Sony CD-RW/DVD-ROM. This is what happens when I played with the master/slave/cable-select jumper pin.

1. When master pin is jumpered: bios detects correct name of cdrom, windows doesn’t boot to desktop

2. When slave pin is jumpered: bios detects cdrom name as unknown, windows boots to desktop and detects correct name of cdrom

3. When cable-select pin is jumpered: bios detects correct name of cdrom, windows doesn’t boot to desktop

4. When no pin is jumpered: bios detects cdrom name as unknown, window boots to desktop and detects correct name of cdrom.

The model of the cdrom is Sony CD-RW CRX320EE.

Is the drive all by itself, or is there a second
storage device on the ribbon cable ?

IDE cables, you fill them from the end. If only
one storage device is present, it goes on the end
connector, not the middle one. The middle one is
only filled, if the end spot is occupied.

Other than that suggestion, I haven't a clue how the
detections would be different, between BIOS and OS.

Try the drive by itself on the cable, before giving up.

Paul

 

[kev833]

Is the drive all by itself, or is there a second

storage device on the ribbon cable ?



IDE cables, you fill them from the end. If only

one storage device is present, it goes on the end

connector, not the middle one. The middle one is

only filled, if the end spot is occupied.



Other than that suggestion, I haven't a clue how the

detections would be different, between BIOS and OS.



Try the drive by itself on the cable, before giving up.



Paul

The drive is connected by a single cable (by itself) to the PATA controller.. End to end. And there is only one IDE controller on the motherboard.

 

[philo]

I discovered an interesting problem. Maybe someone can help me figure it out. My Dell Optiplex GX520 Desktop running Windosws XP SP3 had a bad CDROM drive which I replaced with a dusted off Sony CD-RW/DVD-ROM. This is what happens when I played with the master/slave/cable-select jumper pin.

1. When master pin is jumpered: bios detects correct name of cdrom, windows doesn’t boot to desktop

2. When slave pin is jumpered: bios detects cdrom name as unknown, windows boots to desktop and detects correct name of cdrom

3. When cable-select pin is jumpered: bios detects correct name of cdrom, windows doesn’t boot to desktop

4. When no pin is jumpered: bios detects cdrom name as unknown, window boots to desktop and detects correct name of cdrom.

The model of the cdrom is Sony CD-RW CRX320EE.

If the OS recognized the drive OK, I would not worry about how the bios
sees it. The bios really only has to properly recognize the boot drive
as the OS does not require a BIOS call for auxiliary drives.

 

[kev833]

From: "kev833" <[email protected]>

















This is an ol platform. What is the system's BIOS version ?



--

Dave

Multi-AV Scanning Tool - http://multi-av.thespykiller.co.uk

http://www.pctipp.ch/downloads/dl/35905.asp

Yes the system is pretty old. Bought in an offfice auction.
BIOS vendor: Dell
version: A07
firmware: 108.108
date: 03/31/2006

I searched the net for an image of the board but could not find one.

 

[kev833]

If the OS recognized the drive OK, I would not worry about how the bios

sees it. The bios really only has to properly recognize the boot drive

as the OS does not require a BIOS call for auxiliary drives.

Correct. Just a strange thing to see the drive named "unknown" but Windows detects the right name. Oh well.

 

[kev833]

Correction. There are two SATA IDE controllers plus the one PATA IDE controller I've already mentioned.

 

[philo]

Correct. Just a strange thing to see the drive named "unknown" but Windows detects the right name. Oh well.

Whenever I troubleshoot, all I usually worry about is getting it going.
No two machines have ever been the same
and so many times, I experience something that makes no sense...but
somehow works.

If I was going to try and figure out the reason, I'd have a backlog of
100 machines on my repair bench.


That said, I have seen my share of Dell Optiplexes that sure seemed to
have problems with two devices on the same IDE cable. I don't recall
seeing that issue anywhere else.

 

[David H. Lipman]

Yes the system is pretty old. Bought in an offfice auction.
BIOS vendor: Dell
version: A07
firmware: 108.108
date: 03/31/2006

I searched the net for an image of the board but could not find one.

The BIOS is at A11.

Since it is the BIOS that interacts witrh the hardware connecting to the motherboard and
the OS, I suggest going to http://support.dell.com
and obtaining the BIOS A11 for the Optiplex GX520.

 

[J. P. Gilliver (John)]

If the OS recognized the drive OK, I would not worry about how the bios
sees it. The bios really only has to properly recognize the boot drive
as the OS does not require a BIOS call for auxiliary drives.

Until the OS won't boot one day and you have to use a recovery CD (-:.
But the OP knows what link to change (though it may be worth checking
that it will actually boot from a CD, not just "recognise the drive").

 

[Paul]

Whenever I troubleshoot, all I usually worry about is getting it going.
No two machines have ever been the same
and so many times, I experience something that makes no sense...but
somehow works.

If I was going to try and figure out the reason, I'd have a backlog of
100 machines on my repair bench.


That said, I have seen my share of Dell Optiplexes that sure seemed to
have problems with two devices on the same IDE cable. I don't recall
seeing that issue anywhere else.

Is the GX520 one of the "bad cap" machines ? I can find a reference
here to one. No details though, as to which caps.

http://hardforum.com/archive/index.php/t-1599530.html

This article has a picture.

http://ben-engel.blogspot.ca/2010/0...010/04/replacing-electrolytic-capacitors.html

One has a split "K" on top, another is completely discolored on top.
They look to be part of the chipset power source.

http://4.bp.blogspot.com/_uEiDRkLxhMo/S9EwaJgY_yI/AAAAAAAAD3U/K-FnNLq8X9o/s1600/IMG_0378.JPG

*******

The BIOS should be able to get the identity check right, as it
exchanges very little data across the cable to do that. It's
strange that Windows does a better job. Almost like something
isn't ready, or stable, at the instant the BIOS does its check.

As another test case, turn off the computer power at the back.
Wait a minute. Now, turn it on and enter the BIOS. Does the
BIOS properly detect it now ? Push the reset button. Does
the BIOS properly detect it after reset ? I'm curious about
the timing. And what pre-conditions make it "unknown".

Paul

 

[kev833]

[snip]

Update. This seems to be a Windows XP SP3 problem after all. Though I haven't been able to pin point what area this problem exist in Windows. I confirmed it this way. I placed the jumper on the master pins of the CDROM drive.The proper cable connections were already made. Boot to BIOS to make sure the BIOS detected the drive. Then I booted the PC from a Fedora Linux USB stick. Confirmed that the CDROM was detected by the Linux XFCE desktop and even played some DVDs. But from Windows, the PC freezes while it is trying to load the desktop. So it's not the drive, the controller, the BIOS, the jumper, or the cable. Windows won't even boot to safe mode.

 

[kev833]

From: "kev833" <[email protected]>

[snip]
Update. This seems to be a Windows XP SP3 problem after all. Though I

haven't been able to

pin point what area this problem exist in Windows. I confirmed it this

way. I placed the

jumper on the master pins of the CDROM drive. The proper cable connections

were already

made. Boot to BIOS to make sure the BIOS detected the drive. Then I booted

the PC from a

Fedora Linux USB stick. Confirmed that the CDROM was detected by the Linux

XFCE desktop

and even played some DVDs. But from Windows, the PC freezes while it is

trying to load the

desktop. So it's not the drive, the controller, the BIOS, the jumper, or

the cable.

Windows won't even boot to safe mode.

No matter what, I still suggest upgrading the BIOS to A11 especially since

there are fixes dealing with CD/DVD drives.



--

Dave

Multi-AV Scanning Tool - http://multi-av.thespykiller.co.uk

http://www.pctipp.ch/downloads/dl/35905.asp

Hmmm, I'll consider it. Thanks.

 

[philo]

Is the GX520 one of the "bad cap" machines ? I can find a reference
here to one. No details though, as to which caps.

http://hardforum.com/archive/index.php/t-1599530.html

This article has a picture.

http://ben-engel.blogspot.ca/2010/0...010/04/replacing-electrolytic-capacitors.html


One has a split "K" on top, another is completely discolored on top.
They look to be part of the chipset power source.

http://4.bp.blogspot.com/_uEiDRkLxhMo/S9EwaJgY_yI/AAAAAAAAD3U/K-FnNLq8X9o/s1600/IMG_0378.JPG


*******

The BIOS should be able to get the identity check right, as it
exchanges very little data across the cable to do that. It's
strange that Windows does a better job. Almost like something
isn't ready, or stable, at the instant the BIOS does its check.

As another test case, turn off the computer power at the back.
Wait a minute. Now, turn it on and enter the BIOS. Does the
BIOS properly detect it now ? Push the reset button. Does
the BIOS properly detect it after reset ? I'm curious about
the timing. And what pre-conditions make it "unknown".

Paul

I've also gotten in a lot of Dells with bad caps.
Once I see that I scrap out the machine.

 

[J. P. Gilliver (John)]

In message <[email protected]>, philo  <[email protected]>
writes:
[]

I've also gotten in a lot of Dells with bad caps.
Once I see that I scrap out the machine.

As opposed to scrapping it in?
--
J. P. Gilliver. UMRA: 1960/<1985 MB++G()AL-IS-Ch++(p)Ar@T+H+Sh0!:`)DNAf

One death from beef on the bone might be expected every 20 years at the current
rate, but 40,000 people will die falling downstairs in that time. Should we make
bungalows compulsory? ("Equinox" on Risk, April 1999, paraphrased by Polly
Toynbee in Radio Times)

 

[Paul]

I've also gotten in a lot of Dells with bad caps.
Once I see that I scrap out the machine.

As opposed to scrapping it in?[/QUOTE]

Getting it "out the door", is the right direction

Who wants to spend hours soldering caps ? I've done it,
and even with a $3000 rework station, it still takes
a couple hours. It's not really all that cost
effective, unless the product has sentimental value
to you.

Caps are easy to remove and replace, if they use
big holes in the motherboard. If the legs of the
caps are jammed into the holes (interference fit),
then the job is messy. And you don't want to
tear them out with brute force, because you
can rip out the plated holes. (Been there, done that.)

Paul

 

[philo]

In message <[email protected]>, philo <[email protected]>
writes:
[]

I've also gotten in a lot of Dells with bad caps.
Once I see that I scrap out the machine.

As opposed to scrapping it in?

Getting it "out the door", is the right direction

Who wants to spend hours soldering caps ? I've done it,
and even with a $3000 rework station, it still takes
a couple hours. It's not really all that cost
effective, unless the product has sentimental value
to you.

Caps are easy to remove and replace, if they use
big holes in the motherboard. If the legs of the
caps are jammed into the holes (interference fit),
then the job is messy. And you don't want to
tear them out with brute force, because you
can rip out the plated holes. (Been there, done that.)

Paul

I sometimes get in 20 machines at a time and no way is it worth it to
replace caps on an obsolete machine. Since I get the machines free, I
don't want to spend money and time trying to repair mobos. Last batch I
got in, I ended up with enough parts to get about half of them going.

I only gave it a try once, and the mobo still did not work right as
evidently it had other bad caps that just hadn't visibly leaked yet.

I did pick up a few spare Optiplex boards on eBay that cost less than a
set of caps...just for the heck of it.

 

[Paul]

In message <[email protected]>, philo <[email protected]>
writes:
[]
I've also gotten in a lot of Dells with bad caps.
Once I see that I scrap out the machine.

As opposed to scrapping it in?

Getting it "out the door", is the right direction

Who wants to spend hours soldering caps ? I've done it,
and even with a $3000 rework station, it still takes
a couple hours. It's not really all that cost
effective, unless the product has sentimental value
to you.

Caps are easy to remove and replace, if they use
big holes in the motherboard. If the legs of the
caps are jammed into the holes (interference fit),
then the job is messy. And you don't want to
tear them out with brute force, because you
can rip out the plated holes. (Been there, done that.)

Paul

I sometimes get in 20 machines at a time and no way is it worth it to
replace caps on an obsolete machine. Since I get the machines free, I
don't want to spend money and time trying to repair mobos. Last batch I
got in, I ended up with enough parts to get about half of them going.

I only gave it a try once, and the mobo still did not work right as
evidently it had other bad caps that just hadn't visibly leaked yet.

I did pick up a few spare Optiplex boards on eBay that cost less than a
set of caps...just for the heck of it.

If the Vcore has seven caps on output and five caps on input
(parallel arrays), you replace every cap in the array. If one
leaks in the array, you replace all of them. And its for
the reason you state - the others are going bad as well,
and might not be visibly damaged yet.

On a seven cap array, if one fails, the other six
carry more ripple current than they used to. This tends
to accelerate failure of the others, even if they
don't leak while doing so. That helps the failures to
correlate.

Some cap failures can be near to being perfectly harmless.
If a bulk bypass cap fails (chemical breakdown, not stress),
you might not miss it in operation. These would be caps
near the PCI slots. The liquid electrolyte though, could
corrode other things.

Some day, you'll be getting used gear with 100% Polymer caps,
and this nightmare will be over

Paul

 

[philo]

I've also gotten in a lot of Dells with bad caps.
Once I see that I scrap out the machine.

As opposed to scrapping it in?[/QUOTE]

 

[philo]

On 06/10/2013 06:21 PM, Paul wrote:
X

Some cap failures can be near to being perfectly harmless.
If a bulk bypass cap fails (chemical breakdown, not stress),
you might not miss it in operation. These would be caps
near the PCI slots. The liquid electrolyte though, could
corrode other things.

Some day, you'll be getting used gear with 100% Polymer caps,
and this nightmare will be over

Paul

I guess what amazes me is that the machines I am getting with bad caps
are of varying ages spanning quite a few years.

OTOH: Why should I be surprised?