PCs that support only one floppy drive in hardware

[Scott Stevens]

On Sat, 8 Oct 2005 14:25:41 -0700
Tim Mann <tim at tim-mann.org> wrote:

> On Sat, 8 Oct 2005 03:25:49 -0500 (CDT), cctech-request at classiccmp.org
> wrote:
> > Message: 22
> > Date: Fri, 7 Oct 2005 21:51:09 -0700
> > From: jim stephens <james.w.stephens at gmail.com>
> > Subject: Re: TEAC FD-55GFR = Quad Density?
> > To: "General Discussion: On-Topic and Off-Topic Posts"
> > 	<cctalk at classiccmp.org>
> > Message-ID:
> > 	<ae0bc2000510072151o541c0ad3u8dbc114f37216beb at mail.gmail.com>
> > Content-Type: text/plain; charset=ISO-8859-1
> > 
> > <snip>
> > >
> > >
> > > <snip>
> > > Actually, I am still slowly getting over my long painful struggle
> > > to get two floppy drives enabled on a Dell Optiplex system.
> > > There's support for a second floppy in the BIOS but it appears the
> > > hardware support is entirely missing. I had to just do away with
> > > the 3-1/2" disk to get a 5-1/4" disk installed in one of my (many,
> > > since I get them for < $5 all kitted out with Pentium III
> > > processors) beloved (!!??!!) Optiplex boxes. Was there THAT much
> > > savings in not including hardware support for two floppies, DELL?
> > > Why not patch settings for the second drive out of the BIOS so we
> > > don't pound our heads against the wall trying??
> > >
> > > I'm curious about the comment about second drive support missing.
> > > I
> > am curious why the addition of a proper multiheaded cable doesnt
> > fix this?
> > 
> > they actually dont drive the drive select in the cable?
> 
> I built a new machine around an EPoX 9NPA nForce 4 motherboard and
> found out to my surprise that it was only capable of running one
> floppy drive. In my case the BIOS provides no way to enable a second
> drive, so at least it's consistent.  Having a BIOS that lets you put
> in settings for a second drive when the hardware doesn't support it
> would really be a screwup.
> 
> EPoX makes schematics available, so I downloaded one and studied it to
> try to confirm whether there really was no way to get a second drive
> to work.  It turns out the super-IO chip they use to run the floppies
> and several other system functions (fan control, temperature sensors,
> serial and parallel ports, etc.) has only a limited number of pins. 
> Several have multiple functions, and the motherboard designer has to
> select which function he wants to use for each pin and do without the
> others. EPoX chose to use the pins that could have driven the second
> floppy's motor-on and drive select for other functions (I forget what
> offhand). So those pins are no-connects on the floppy cable.
> 
> As a small compensation, it's possible to reconfigure the parallel
> port as an external floppy connector.  If you do that, you can put two
> drives on it.  (Whoopee.)
> 
> This all is not a huge deal for me since I can pop a Catweasel card
> into the machine if I really want more floppies in it, but it's a bit
> annoying.
> 

One thing I am wondering is if an 8-bit floppy controller card (the ones
from the PC-XT generation, which didn't have floppy IO on the
motherboard) could be modified (if needed) and contrived to work in a
more modern system that still has the ISA bus.  If the particular I/O
locations are being used, simple cuts and adds could re-direct the I/O
ports.  This would give the enterprising programmer a 765 controller
with relevant hardware to plug additional drives into.  The original
PC-PC/XT controller is fully documented in the TechRef, and even has all
the cabling in place to support four floppies.

> -- 
> Tim Mann  tim at tim-mann.org  http://tim-mann.org/