[Coco] Disk drive questions?

[Gene Heskett]

On Thursday 04 December 2003 14:44, jdaggett at gate.net wrote:
>Jim
>
>If I remember correctly yes you can have two disk controllers in the
> MPI. Drive 0 must be in the first slot. Also, if my memory serves
> me corrrectly, you can have four single sided drives or two double
> sided drives or two single sided and one double sided drive. So yes
> three physical drives are the max allowable if one is the double
> sided otherwise four.

Actually, since the drive selection has nothing to do with the floppy 
disk controller chip, one *could* decode those 3 lines used for drive 
select into 8 real enable signals.  So one could run as high as 7 or 
8 double sided drives by using os9.  3 was enough for me, so I never 
explored that little tributary to increased storage.

>The main controllers used by the Coco are the 1773 and the 1793
> chips. The 1773 and 1793 will handle both single and double
> density. These controllers only work with single sided drives only!

Oh?  You should tell that to the 5 or 61 controllers I have that all 
run double sided disks just fine.  AFAIK, when the side select line 
goes true for side 1 as opposed to side 0, the sector count (written 
to the sector register in these chips, is expected to continue to 
increment, for os9, sector 19 is the other side of the disk IF the 
Logical Sector Number arrived at by the seek commands math has a 
sector number modulo of 19-36.  So the chip has no concept of "side", 
only the track and sector numbers writtren to those corresponding 
registers.

> Therefore the capcity for 35 tracks is 160K and 320K respectively.
> These two can be replaced with a 1777 or a 1797 respectively and
> access double sided disk drives. Warning the 1777 conroller is a
> rare as hen's teeth. The 1797 is available.
>
>RSDOS treats double sided drives as two single sided drives. This is
> because the 1773/1793 controller chips do not have side select
> capabilities.

"Side Select" is not a function of any of these chips, its a seperate 
latching register on the controller board, and is driven by the disk 
driver software.  The chip in use doesn't enter into the picture 
other than searching for the sector number written to its sector 
address register by the driver.  As thats an 8 bit value, and the 
chips don't check for any overflows that my experiments found, one 
could theoreticly have a single track that was 255 sectors long.  
That disk would be several feet in circumference though. :)

> Also the 177x series of controllers, FD501 and FD502,
> can only read and write a maximum of 360K per side. The 1793, FD500
> will read and write 720K per side.

Untrue.  The only limitation in these chips is that the 1773 family 
cannot do a data rate of 500khz, only the std (for Tandy stuffs) of 
250khz, thereby limiting an 80 track double sided disk to 720k.  
Halve that for a 48 tpi drive.  If the drive itself will step on in, 
then each additional track is another 4604 bytes per side/track 
gained IIRC.  I usually ran drives like that at an effective capacity 
of 765k.

>Also the 1793 controller will work on the COco3 if and only if it is
> plugged into the MPI.

Only because the mpi furnishes the +12 volts that particular Radio 
Shack contoller needs for its offchip data decoder's input offset 
bias.  Coco 1 and 2's supplied that at the cartridge port, the coco3 
doesn't due to running out of power in the builtin psu, so it only 
works in the mpi which does furnish that voltage.

Hard Drives International made one that didn't need the +12 volts.  
Made in a black metal box.  However, you are correct in that those 
were apparently made in limited quantities.  I did have one, but NDI 
where it went by 2003.  Its been years since I stumbled over it.  
Actually, IIRC, that controller had a fujitsu MB8877 in it, a 1793 
clone in hcmos that could be run as high as 1 megabaud for its data 
rate.  Somewhere, I have a data sheet on that puppy from fujitsu if 
it hasn't moldered away with age.

>james
>
>On 2 Dec 2003 at 23:53, jimcox at miba51.com wrote:
>> Hi All!
>>
>> Spent some time re-reading the manuals and realize I
>> should have done a better job of reading them the first
>> time through, cause I think I would have stuck with the
>> CoCo longer instead of abandoning it for the Atari ST I
>> got long ago.
>>
>> One thing I noticed while going through the exercises in
>> the manual, was once I wound down, and started powering
>> off everything, I noticed that the back of the drive,
>> where the power supply is, was warmer than what I would
>> have expected it to be.  Then again, this IS 20 year old
>> technology that hasn't been powered up for some time.  Any
>> thoughts?
>>
>> I'd also like to ask again, how easy is it to drop a 3.5"
>> drive into a system.  I know I have the answer some where
>> in the thousands of emails I have saved, but I thought I'd
>> save time and ask here.  Any pointers?
>>
>> I know that someone has said that the maximum number of
>> drives in a system is 3, though in the manual, it says the
>> maximum drive number is 3, so if you start with 0, my
>> logic says you can have 4 drives total.  Is it possible to
>> have two different drive controllers in an MPI and toggle
>> between them?
>>
>> BTW, after all the typing, I now realize how good we got
>> it now on current systems.  Boy, do I sure miss the "up
>> arrow" key.  Cheers!
>>
>> Jim
>>
>> PS: Anyone heard from Carl Boll lately?
>>
>> --
>> Coco mailing list
>> Coco at maltedmedia.com
>> http://five.pairlist.net/mailman/listinfo/coco

-- 
Cheers, Gene
AMD K6-III at 500mhz 320M
Athlon1600XP at 1400mhz  512M
99.27% setiathome rank, not too shabby for a WV hillbilly
Yahoo.com attornies please note, additions to this message
by Gene Heskett are:
Copyright 2003 by Maurice Eugene Heskett, all rights reserved.