Help! 5.25" drive must be alone to work!

[leguerri]

Hello,
I have an IBM P166MMX system, model PC300GL IIRC. The problem is that I
can't use a 5.25" drive properly if it's not alone.
That means a 1.2M drive works perfectly as A:, as long as there is no
B: drive physically connected.
But when I connect the B: drive, which is a 3.5" 1.44M, I can no longer
format or write to the A: (5.25"), but B: seems to work fine. The sectors
on A: drive seem unrecognizable. If I try to format a brand new disk, it
seems to work, but after FORMAT says "Unable to write BOOT" message.

Ultimately, I tried swapping A: and B:, so now the 3.5" is A: and the
5.25" is B:. All the write problems swapped as well to the B: drive, and
A: works fine.

I thought the drive was the problem, but that 1.2M drive works perfectly
alone, or on another system.

Of course, I always set up the correct BIOS diskette types for A: and B:
whenever I change them physically.

Does anyone have any advise, or experience to share?

Thanks in advance

Louis-Luc

 

[Michael Hawes]

Hello,
I have an IBM P166MMX system, model PC300GL IIRC. The problem is that I
can't use a 5.25" drive properly if it's not alone.
That means a 1.2M drive works perfectly as A:, as long as there is no
B: drive physically connected.
But when I connect the B: drive, which is a 3.5" 1.44M, I can no longer
format or write to the A: (5.25"), but B: seems to work fine. The sectors
on A: drive seem unrecognizable. If I try to format a brand new disk, it
seems to work, but after FORMAT says "Unable to write BOOT" message.

Ultimately, I tried swapping A: and B:, so now the 3.5" is A: and the
5.25" is B:. All the write problems swapped as well to the B: drive, and
A: works fine.

I thought the drive was the problem, but that 1.2M drive works perfectly
alone, or on another system.

Of course, I always set up the correct BIOS diskette types for A: and B:
whenever I change them physically.

Does anyone have any advise, or experience to share?

Thanks in advance

Louis-Luc

See if the 5.25" drive has an address setting, if so, change it. Both
drives should have same address and end drive on cable is A. Does the 3.5
work ok on its own? Have you checked Bios settings? Drives must be set
correctly in BIOS to format, when reading they identify format from track 0
info.

Mike.

 

[daytripper]

Hello,
I have an IBM P166MMX system, model PC300GL IIRC. The problem is that I
can't use a 5.25" drive properly if it's not alone.
That means a 1.2M drive works perfectly as A:, as long as there is no
B: drive physically connected.
But when I connect the B: drive, which is a 3.5" 1.44M, I can no longer
format or write to the A: (5.25"), but B: seems to work fine. The sectors
on A: drive seem unrecognizable. If I try to format a brand new disk, it
seems to work, but after FORMAT says "Unable to write BOOT" message.

Ultimately, I tried swapping A: and B:, so now the 3.5" is A: and the
5.25" is B:. All the write problems swapped as well to the B: drive, and
A: works fine.

I thought the drive was the problem, but that 1.2M drive works perfectly
alone, or on another system.

Of course, I always set up the correct BIOS diskette types for A: and B:
whenever I change them physically.

Does anyone have any advise, or experience to share?

Tried a different cable yet?

 

[Rod Speed]

Normally due to a bad cable if it used to work properly.

Can be a bad floppy controller if a new cable doesnt fix it.

 

[Arno Wagner]

Hello,
I have an IBM P166MMX system, model PC300GL IIRC. The problem is that I
can't use a 5.25" drive properly if it's not alone.
That means a 1.2M drive works perfectly as A:, as long as there is no
B: drive physically connected.
But when I connect the B: drive, which is a 3.5" 1.44M, I can no longer
format or write to the A: (5.25"), but B: seems to work fine. The sectors
on A: drive seem unrecognizable. If I try to format a brand new disk, it
seems to work, but after FORMAT says "Unable to write BOOT" message.
Ultimately, I tried swapping A: and B:, so now the 3.5" is A: and the
5.25" is B:. All the write problems swapped as well to the B: drive, and
A: works fine.
I thought the drive was the problem, but that 1.2M drive works perfectly
alone, or on another system.
Of course, I always set up the correct BIOS diskette types for A: and B:
whenever I change them physically.
Does anyone have any advise, or experience to share?
Thanks in advance
Louis-Luc

One higly speculative idea: Some (all?) 5.25" drives have a low-value
pullup resistor on the write-data line. I have ssen 220R. This
amounths to about 14 TTL loads. I have not checked any 3.5" drives
lately, but years ago I was surprised to find that one had a 1500R
pullup resistor, amounthing to 2 TTL loads. Now comes the speculation:
If the driver for the write-data line is borderline, it could still
function well enough with the 14 TTL loads, but not with the 16 TTL
loads of both floppies (there is only one write-data line).
If, in addition, the 3.5" floppy has better input signal tolerances
(quite possibly, since with its larger pullup resistor it will
get a worse signal), it could still get a good enough signal for
it to function.

The clean solution is to use a better controller, today usually
meaning mainboard.

Arno

 

[Donald G. Davis]

I have an IBM P166MMX system, model PC300GL IIRC. The problem is that I
can't use a 5.25" drive properly if it's not alone.
That means a 1.2M drive works perfectly as A:, as long as there is no
B: drive physically connected.
But when I connect the B: drive, which is a 3.5" 1.44M, I can no longer
format or write to the A: (5.25"), but B: seems to work fine. The sectors
on A: drive seem unrecognizable. If I try to format a brand new disk, it
seems to work, but after FORMAT says "Unable to write BOOT" message.

Ultimately, I tried swapping A: and B:, so now the 3.5" is A: and the
5.25" is B:. All the write problems swapped as well to the B: drive, and
A: works fine.

I thought the drive was the problem, but that 1.2M drive works perfectly
alone, or on another system.

Of course, I always set up the correct BIOS diskette types for A: and B:
whenever I change them physically.

Does anyone have any advise, or experience to share?

To work correctly, some (all?) floppy drives need to have jumpers
set to match their positions on the cable (first, second, or single). If
you're not sure about yours, do a Google search for each drive's brand and
model number and "jumper," to see if yours are set correctly for the
position you want each in.

 

[DaveinOlyWa]

the drive has a master single jumper that has to be set.

this is only on old stuff but it sounds like that is what you have.

on mine the jumpers are labeled vaguely but basically 0 is for
master, 1 is single and 2 is slave.

these jumpers im sure just control connects to a terminate resistor.
so the signal is being terminated on the drive before it has a chance
to get to the a drive.

have you tried switching cable positions ?

 

[Rod Speed]

To work correctly, some (all?) floppy drives need to have jumpers
set to match their positions on the cable (first, second, or single).

Nope. Thats the reason for the twist in the cable between the
two floppy connectors, so the drives dont need to be rejumpered.

If you're not sure about yours, do a Google search for each
drive's brand and model number and "jumper," to see if yours
are set correctly for the position you want each in.

Pointless with a normal PC.

 

[Folkert Rienstra]

Nope. Thats the reason for the twist in the cable between the
two floppy connectors, so the drives dont need to be rejumpered.

Not all drives are PC drives, oh mighty clueless one. That is why he said
'some'. Universal drives have jumpers or even a switch which allows you
to set them as device select 0 (or A), 1 (B) , 2 (C) or 3 (D).

IIRC they should be jumpered 1 (or B) for use in a PC to get the proper
drive A and drive B detection depending to where they are on the cable.

Pointless with a normal PC.

Nope, not with universal drives that can be jumpered to respond to
drive select A, B, C or D.

 

[Rod Speed]

Some pathetic excuse for a troll claiming to be
message just the pathetic excuse for a troll thats all it can ever manage.

Try harder, child. You might actually manage to fool someone, sometime.

 

[Rod Speed]

the drive has a master single jumper that has to be set.

this is only on old stuff but it sounds like that is what you have.

on mine the jumpers are labeled vaguely but
basically 0 is for master, 1 is single and 2 is slave.

There is no master or slave with floppy drives.

You can in theory specify the drive ID using that jumper
but for a very long time indeed now the tradition has been
to have both drives jumpered identically and its the twist in
the ribbon cable between the floppy drives that swaps the ID.

That approach was used very early on with the IBM PC.

these jumpers im sure just control connects to a terminate resistor.

Nope. Both sets of terminators are used together.

Different system to SCSI terminators.

so the signal is being terminated on the drive
before it has a chance to get to the a drive.

Thats not what termination is about either.

 

[Arno Wagner]

[...]

Pointless with a normal PC.

Nope, not with universal drives that can be jumpered to respond to
drive select A, B, C or D.

Actually I have one of these drives. Mightily old (720kB 3.5").
And I have seen a newer one as well somewhere. Not all 3.5" floppies
are designed for PC use only. The Shuggart-bus definitely supports
4 floppies. I think the design with the twisted cable was just to
make it idiot-proof. So no surprise Rod does not know about this ;-)

Arno

 

[leguerri]

If you're not sure about yours, do a Google search for each

Actually I have one of these drives. Mightily old (720kB 3.5").
And I have seen a newer one as well somewhere. Not all 3.5" floppies
are designed for PC use only. The Shuggart-bus definitely supports
4 floppies. I think the design with the twisted cable was just to
make it idiot-proof. So no surprise Rod does not know about this ;-)

Arno

Absolutely, the twist is just to avoid changing jumpers. It should work
the same if both drives are on a straight non-twisted cable, as long as
one is jumpered DS0 and the other DS1. That's how I found an old XT back in
'96. It was next to the garbage bin. It was not booting, and both floppies
were connected to a STRAIGHT cable. I had the choice to either put a twisted
cable, or change the jumper of one drive to DS0.

Louis-Luc

 

[J. Clarke]

[...]

If you're not sure about yours, do a Google search for each
drive's brand and model number and "jumper," to see if yours
are set correctly for the position you want each in.

Pointless with a normal PC.

Nope, not with universal drives that can be jumpered to respond to
drive select A, B, C or D.

Actually I have one of these drives. Mightily old (720kB 3.5").
And I have seen a newer one as well somewhere. Not all 3.5" floppies
are designed for PC use only. The Shuggart-bus definitely supports
4 floppies. I think the design with the twisted cable was just to
make it idiot-proof. So no surprise Rod does not know about this ;-)

Not so much "idiot proof" as easier to assemble. With jumpers you either
have to have two bins of drives or you have to have the guy on the line set
the jumpers each time. Either of these will increase manufacturing cost
due to rework--with the jumpers he'll set them incorrectly from time no
matter how diligent he is, to time, with the preconfigured drives they'll
get binned incorrectly from time to time. With the twisted cable you only
have to maintain one bin of identically jumpered drives, eliminating that
source of manufacturing errors, at the cost of a slightly more expensive
cable that can be easily inspected.

 

[Jim Jones]

[...]

If you're not sure about yours, do a Google search for each
drive's brand and model number and "jumper," to see if yours
are set correctly for the position you want each in.

Pointless with a normal PC.

Nope, not with universal drives that can be jumpered to respond to
drive select A, B, C or D.

Actually I have one of these drives. Mightily old (720kB 3.5").
And I have seen a newer one as well somewhere. Not all 3.5" floppies
are designed for PC use only. The Shuggart-bus definitely supports
4 floppies. I think the design with the twisted cable was just to
make it idiot-proof. So no surprise Rod does not know about this ;-)

How odd that I've actually commented on it in the past and
even a pathetic excuse for a troll could find it in groups.google.

Not so much "idiot proof" as easier to assemble. With jumpers you either
have to have two bins of drives or you have to have the guy on the line set
the jumpers each time. Either of these will increase manufacturing cost
due to rework--with the jumpers he'll set them incorrectly from time no
matter how diligent he is, to time, with the preconfigured drives they'll
get binned incorrectly from time to time. With the twisted cable you only
have to maintain one bin of identically jumpered drives, eliminating that
source of manufacturing errors, at the cost of a slightly more expensive
cable that can be easily inspected.

Pity there have been bugger all PCs shipped with
more than one floppy drive for years and years
now, and even less with two identical floppy drives.

Nice theory. Pity about the reality tho.

 

[J. Clarke]

In comp.sys.ibm.pc.hardware.misc Folkert Rienstra


[...]

If you're not sure about yours, do a Google search for each
drive's brand and model number and "jumper," to see if yours
are set correctly for the position you want each in.

Pointless with a normal PC.

Nope, not with universal drives that can be jumpered to respond to
drive select A, B, C or D.

Actually I have one of these drives. Mightily old (720kB 3.5").
And I have seen a newer one as well somewhere. Not all 3.5" floppies
are designed for PC use only. The Shuggart-bus definitely supports
4 floppies. I think the design with the twisted cable was just to
make it idiot-proof. So no surprise Rod does not know about this ;-)

How odd that I've actually commented on it in the past and
even a pathetic excuse for a troll could find it in groups.google.

Not so much "idiot proof" as easier to assemble. With jumpers you either
have to have two bins of drives or you have to have the guy on the line
set
the jumpers each time. Either of these will increase manufacturing cost
due to rework--with the jumpers he'll set them incorrectly from time no
matter how diligent he is, to time, with the preconfigured drives they'll
get binned incorrectly from time to time. With the twisted cable you
only have to maintain one bin of identically jumpered drives, eliminating
that source of manufacturing errors, at the cost of a slightly more
expensive cable that can be easily inspected.

Pity there have been bugger all PCs shipped with
more than one floppy drive for years and years
now, and even less with two identical floppy drives.

Nice theory. Pity about the reality tho.

The reality is that the twisted cable was use for both diskette and hard
drives and it was put into use in the era in which dual diskette drives
were the norm. Once it was in place as a standard there was no reason to
change it when single-drive machines became the norm. With IDE they
formalized cable select so that a twist was no longer required.

 

[Jim Jones]

Arno Wagner wrote:

In comp.sys.ibm.pc.hardware.misc Folkert Rienstra


[...]

If you're not sure about yours, do a Google search for each
drive's brand and model number and "jumper," to see if yours
are set correctly for the position you want each in.

Pointless with a normal PC.

Nope, not with universal drives that can be jumpered to respond to
drive select A, B, C or D.

Actually I have one of these drives. Mightily old (720kB 3.5").
And I have seen a newer one as well somewhere. Not all 3.5" floppies
are designed for PC use only. The Shuggart-bus definitely supports
4 floppies. I think the design with the twisted cable was just to
make it idiot-proof. So no surprise Rod does not know about this ;-)

How odd that I've actually commented on it in the past and
even a pathetic excuse for a troll could find it in groups.google.

Not so much "idiot proof" as easier to assemble. With jumpers you either
have to have two bins of drives or you have to have the guy on the line
set
the jumpers each time. Either of these will increase manufacturing cost
due to rework--with the jumpers he'll set them incorrectly from time no
matter how diligent he is, to time, with the preconfigured drives they'll
get binned incorrectly from time to time. With the twisted cable you
only have to maintain one bin of identically jumpered drives, eliminating
that source of manufacturing errors, at the cost of a slightly more
expensive cable that can be easily inspected.

Pity there have been bugger all PCs shipped with
more than one floppy drive for years and years
now, and even less with two identical floppy drives.
Nice theory. Pity about the reality tho.

The reality is that the twisted cable was use for both diskette and hard drives
Duh.

and it was put into use in the era in which
dual diskette drives were the norm.
Nope.

Once it was in place as a standard there was no reason
to change it when single-drive machines became the norm.

The reason he claimed clearly wasnt the reason.

With IDE they formalized cable select
so that a twist was no longer required.

That came well after the first IDE implementations.

 

[Folkert Rienstra]

"Rod Speed" (e-mail address removed)> wrote in message
[...]

If you're not sure about yours, do a Google search for each
drive's brand and model number and "jumper," to see if yours
are set correctly for the position you want each in.

Pointless with a normal PC.

Nope, not with universal drives that can be jumpered to respond to
drive select A, B, C or D.

Actually I have one of these drives. Mightily old (720kB 3.5").
And I have seen a newer one as well somewhere. Not all 3.5" floppies
are designed for PC use only. The Shuggart-bus definitely supports
4 floppies. I think the design with the twisted cable was just to
make it idiot-proof. So no surprise Rod does not know about this ;-)

Not so much "idiot proof" as easier to assemble. With jumpers you either
have to have two bins of drives or you have to have the guy on the line set
the jumpers each time. Either of these will increase manufacturing cost
due to rework--with the jumpers he'll set them incorrectly from time no
matter how diligent he is, to time, with the preconfigured drives they'll
get binned incorrectly from time to time.

With the twisted cable you only have to main-
tain one bin of identically jumpered drives,

To use jumperless PC-type drives that have drive select fixed at pin 12.

 

[Folkert Rienstra]

"J. Clarke" (e-mail address removed)> wrote in message news:[email protected]...

Arno Wagner wrote:
"Rod Speed" (e-mail address removed)> wrote in message
[...]

If you're not sure about yours, do a Google search for each
drive's brand and model number and "jumper," to see if yours
are set correctly for the position you want each in.

Pointless with a normal PC.

Nope, not with universal drives that can be jumpered to respond to
drive select A, B, C or D.

Actually I have one of these drives. Mightily old (720kB 3.5").
And I have seen a newer one as well somewhere. Not all 3.5" floppies
are designed for PC use only. The Shuggart-bus definitely supports
4 floppies. I think the design with the twisted cable was just to
make it idiot-proof. So no surprise Rod does not know about this ;-)

How odd that I've actually commented on it in the past and
even a pathetic excuse for a troll could find it in groups.google.

Not so much "idiot proof" as easier to assemble. With jumpers you either
have to have two bins of drives or you have to have the guy on the line
set
the jumpers each time. Either of these will increase manufacturing cost
due to rework--with the jumpers he'll set them incorrectly from time no
matter how diligent he is, to time, with the preconfigured drives they'll
get binned incorrectly from time to time. With the twisted cable you
only have to maintain one bin of identically jumpered drives, eliminating
that source of manufacturing errors, at the cost of a slightly more
expensive cable that can be easily inspected.

Pity there have been bugger all PCs shipped with
more than one floppy drive for years and years
now, and even less with two identical floppy drives.

Nice theory. Pity about the reality tho.

The reality is that the twisted cable was use for both diskette and hard
drives

It was?

and it was put into use in the era in which dual diskette drives
were the norm. Once it was in place as a standard there was no
reason to change it when single-drive machines became the norm.

With IDE they formalized cable select so that a twist was no longer required.

IDE never needed a twist. Cable select has nothing to do with a twist.
IDE doesn't even have a specific device select line. IDE devices react
to a register delivered ID (0/1). And even a twist to avoid nipping
a wire doesn't work as pin 27 (2 wire twist) is a used signal and both
26 and 30 (3 wire twist) are ground, similar to CSEL (pin28)
The floppy wire twist involves 7 wires, not just 2 (or 3).

 

[J. Clarke]

Arno Wagner wrote:

In comp.sys.ibm.pc.hardware.misc Folkert Rienstra


[...]

If you're not sure about yours, do a Google search for each
drive's brand and model number and "jumper," to see if yours
are set correctly for the position you want each in.

Pointless with a normal PC.

Nope, not with universal drives that can be jumpered to respond to
drive select A, B, C or D.

Actually I have one of these drives. Mightily old (720kB 3.5").
And I have seen a newer one as well somewhere. Not all 3.5" floppies
are designed for PC use only. The Shuggart-bus definitely supports
4 floppies. I think the design with the twisted cable was just to
make it idiot-proof. So no surprise Rod does not know about this ;-)

How odd that I've actually commented on it in the past and
even a pathetic excuse for a troll could find it in groups.google.

Not so much "idiot proof" as easier to assemble. With jumpers you
either have to have two bins of drives or you have to have the guy on
the line set
the jumpers each time. Either of these will increase manufacturing
cost due to rework--with the jumpers he'll set them incorrectly from
time no matter how diligent he is, to time, with the preconfigured
drives they'll
get binned incorrectly from time to time. With the twisted cable you
only have to maintain one bin of identically jumpered drives,
eliminating that source of manufacturing errors, at the cost of a
slightly more expensive cable that can be easily inspected.

Pity there have been bugger all PCs shipped with
more than one floppy drive for years and years
now, and even less with two identical floppy drives.
Nice theory. Pity about the reality tho.

The reality is that the twisted cable was use for both diskette and hard
drives
Duh.

and it was put into use in the era in which
dual diskette drives were the norm.

Nope.

Nope?

I see. So would you be kind enough to provide the date on which machines
with dual diskette drives ceased to be commonplace and the date on which
IBM introduced the twisted diskette cable?

The reason he claimed clearly wasnt the reason.

So what was the reason?

That came well after the first IDE implementations.

The first IDE implementations preceded the existence of any standard.