Percom Disk Drives ?

Percom Disk Drives ?

Post by Bob Klaa » Mon, 03 May 1999 04:00:00



I need some assistance, those of you that are fimiliar with the Percom
Drives have surfed the net
and have places I may have missed..  From the conseption Percom produced
single density disk drives, then double density drives.  They went on to
produce double sided single and double sided double density drives..

Does anyone have a clue to the Internet where information about any
Percom drive produced can be obtained?  Your assistance in this area
will not only enhance my knowledge but benifit yourself and anyone using
a Percom.

I have several, modles to play with but would like if avalable a list of
what they actually produced.

Just for those that didn't know, picked out one to test and it has a
Printer Ribbon cable attached, so if you have one that has a 34pin
(actual pins) sticking out the back it will interface a printer to yourr
computer..The card edge if sticking out the back will access 3 slave
disk drives, provided the SLAVES are turned on first.   More on this
great Improvment for the Atari 8  Bit in the next couple weeks.

Yea you got it the Best Disk Drive for the Atari 8-bit is the Percom
delending upon the modle you own

Bob

 
 
 

Percom Disk Drives ?

Post by LWhite64 » Mon, 03 May 1999 04:00:00


Quote:>From the conseption Percom produced
>single density disk drives, then double density drives.  They went on to
>produce double sided single and double sided double density drives..

Hi Bob-
I don't know of any place on the net that has
Percom info, but here's what I have found
going through Compute!, Antic, Analog, and
Softside.  Softside was easy - NOTHING!

By far the greatest number of ads were in
Compute!   My Compute!s start in May of '81.
The first Percom Data Company ad appeared
in 10-81.  Unfortunately, it was an MFD series
for AIM, KIM, and SYM computers - $599.
They advertised almost monthly through early
1982, then stopped.  

The next ad was June, 1982,
and this is the first Atari Percom ad - the RFD
series - $799 (!) and slave drives were $399.
The RFD series were evidently all D/D.
The ads talk about the ability to dynamically
control 4 drives, querrying and adjusting -
the Percom Block.  In August of 1982, the
price of the first add-on drive was $459.

In December of 1982, the first ad for the
AT 88 series appeared - $488.  But single
density only.   In November of 1983, the
first Compute! ad appeared for the
AT88 S1PD - $599.  Now, the model
designation makes sense.  Printer port and
double density.    I find no ads after 11-83 in
Compute! for Percom Data Company, Inc.

Antic.  
My whole issues of Antic start in April 1983.
I decided to save space years ago, and cut the
early issues up.  Sigh!)  The AT88 was
advertised monthly from 4-83 thru 8-83.
Then in 9-83, the AT88 S1PD is advertised.
That was the last ad that I found in Antic
for Percom.

Analog:
Just a couple of ads in Analog.  I found an
ad in #9 (11/12-1982) and #10 (1/2-1983).
These were both for the AT88 model.
There was no ad in #6, and I do not have
#7 or #8.  

Summarizing the info.
About June of 1982, Percom starts advertising the RFD drive for $799.   About
December,
they started promoting the AT single density drive at a much lower $488.  In
Sept, 1983,
the the AT88 S1PD double density model
appears, still for $200 less than the original
RFD series.  In Nov, 1983, the last ad that I
could find appears for Percom.

In March and September of 1984, Analog
and Antic (respectively) carried disk drive
reviews.  Percom got a brief mention in the
Analog article, but was not carried in the
reviews in either.  So judging by the
advertising, Percom's moment in the sun
was pretty brief.

Larry

 
 
 

Percom Disk Drives ?

Post by LWhite64 » Tue, 04 May 1999 04:00:00


Oops!  The Antic "Disk Survey" was in
August, 1983.  Don't know why I typed
"September".  Old age?  Nah!  (I hope!)
Larry
 
 
 

Percom Disk Drives ?

Post by Michael Curre » Fri, 07 May 1999 04:00:00



Quote:>I have several, modles to play with but would like if avalable a list of
>what they actually produced.

I just happen to be trying to build such a list!

Based on articles and ads in Antic, Analog, and Atari Classics, I believe
the following Percom drives for the Atari were actually produced:

RFD40-S1   SS/DD, single drive, 40-track (180K/disk)

RFD40-S2   SS/DD, dual drives, 40-track, 180K/disk

RFD44-S1   DS/DD, single drive, 40-track, 360K/disk

RFD44-S2   DS/DD, dual drives, 40-track, 360K/disk

AT88-S1  SS/SD, single drive, 40-track, 90K/disk

AT88-S2  SS/SD, dual drives, 40-track, 90K/disk

AT88-S1PD  SS/DD, single drive, 40-track, 180K/disk (+printer port)

This doesn't take into account the slave drives they kept talking about.  I
wonder whether the slave drives are non-Atari-specific?

My other question concerns two more models referred-to in an Antic review,
but which aren't mentioned anywhere else.  They would be:

RFD80-S1  SS/DD, single drive, 80-track, 360K/disk

RFD84-S1  DS/DD, single drive, 80-track, 720K/disk

These names I made up, but the specs are straight from the Antic review.

Do these exist??
--

8-bit Atari FAQ and Vendor Lists, http://www.faqs.org/faqs/atari-8-bit/
Cleveland Free-Net Atari SIG, telnet://freenet-in-c.cwru.edu (go atari)
St. Paul Atari Computer Enthusiasts, http://www.library.carleton.edu/space/

 
 
 

Percom Disk Drives ?

Post by LWhite64 » Fri, 07 May 1999 04:00:00


Hi Michael-

Quote:>These names I made up, but the specs are straight from the Antic review.

Is the Antic Review from the "New Products"
in volume 1-2 (June, 1982) the review that
you are referring to here?  Or is there
another review later on?  In all of the ads
that I looked at, I never saw an ad for a
drive with more than one unit built in - only
add-on "slave drives" at $399-459 each.
(Doesn't mean they didn't sell them, or
enterprising dealers didn't, but I just did not
find any ads that said they did).   Also doesn't
mean I didn't overlook it.  :-)   Likewise
in the brief ads for the AT88 series, never
saw an ad for an "S2".  Did you locate one?
Unfortunately, I didn't have access to any of
their sales literature that one could send in for.
There may have been more info in that, if
someone would still have it.  Logically, it
seems more likely that they would have
pursued the "slave drive" strategy because
of the high power consumption of the early
drives, especially the full-height models
used in the RFD's. (?)
Regards,
Larry
 
 
 

Percom Disk Drives ?

Post by BobKlaa » Fri, 07 May 1999 04:00:00


Ok, not sure on the latter drives as I shipped one and can't rember the
modle number but it was not the last drive you posted..Larry White will
have it in a few days.

The best disk Drive ever built in its year for the Atari 8-bit would
have to be the Percom.  If it will do a double sided double density disk
then it will format out a 720k 3 1/4 in disk drive adapted to the
cable.  It will also, do a new 1.44mg 3 1/4 inch drive with a 720k disk
into it, I could not get it to do a 1.44mg disk with the sensor hole
covered, received errors.  Perhaps somone else can persue the project
and come up with a way to enhanse the rom?  But at this time the 2
double density Percoms I had here would do what was needed to arrive at
extra storage.

On A 5 1/2 inch disk you could format it out as a double sided double
density disk at 40 tracks, and 3 as the step rate, those below would
also work.  Any other information would be appreciated..The last two
modles they produced had a Printer interface built in, Larry White has
the only cable I had in existance.

Now on any modle to run slaves, you had to attach them with a 34pin
Ribbon cable, have the 3 inline and turned on first then booted up the
master.

On the newer 1.44mg drives I could not get it to boot as a master but it
would read as a second drive.  The jumper must have to be moved on the
1.44mg drive, requires unsoldering..the 720k drives work fine as there
is a jumper you can move on the drive..perhaps a modified cable like the
IBM has to see the first drive could correct this but have not gotten
into that area at this date.

bob

This post is about 3 weeks earlier than I antisipated.



> >I have several, modles to play with but would like if avalable a list of
> >what they actually produced.

> I just happen to be trying to build such a list!

> Based on articles and ads in Antic, Analog, and Atari Classics, I believe
> the following Percom drives for the Atari were actually produced:

> RFD40-S1   SS/DD, single drive, 40-track (180K/disk)

> RFD40-S2   SS/DD, dual drives, 40-track, 180K/disk

> RFD44-S1   DS/DD, single drive, 40-track, 360K/disk

> RFD44-S2   DS/DD, dual drives, 40-track, 360K/disk

> AT88-S1  SS/SD, single drive, 40-track, 90K/disk

> AT88-S2  SS/SD, dual drives, 40-track, 90K/disk

> AT88-S1PD  SS/DD, single drive, 40-track, 180K/disk (+printer port)

> This doesn't take into account the slave drives they kept talking about.  I
> wonder whether the slave drives are non-Atari-specific?

> My other question concerns two more models referred-to in an Antic review,
> but which aren't mentioned anywhere else.  They would be:

> RFD80-S1  SS/DD, single drive, 80-track, 360K/disk

> RFD84-S1  DS/DD, single drive, 80-track, 720K/disk

> These names I made up, but the specs are straight from the Antic review.

> Do these exist??
> --

> 8-bit Atari FAQ and Vendor Lists, http://www.faqs.org/faqs/atari-8-bit/
> Cleveland Free-Net Atari SIG, telnet://freenet-in-c.cwru.edu (go atari)
> St. Paul Atari Computer Enthusiasts, http://www.library.carleton.edu/space/

 
 
 

Percom Disk Drives ?

Post by Technoid Mutan » Tue, 11 May 1999 04:00:00


The Percom controller gave us the control block and the method of configuration
which still rules today.  Only the 815 disk drive does not follow this Percom
standard.  As a reliable drive it sucks hard.  I have had several ds/dd  models
and they all were unreliable.  In 1986 or 1985, I tracked down the last vestige
of the Percom company for floppy drives.  I ordered and installed a newer rom to
correct some problems.  Here are some problems I am aware of and some info on the
percom drive controller for the at88d2?

The bare controller board is capable of four drives ds/sd

The addition of a daughter card plugged into the controller's CPU? socket or
drive controller socket yeilded density dumb double density mfm.  This inability
to autodetect the density of a disk let to special progams and utilities released
for DOS XL and for Spartados.  MYDOS/Ataridos users could run a keyed in
basic/machine language program I call DriveZap in order to edit the control block
for density.  The Indus GT and Rana1000 disk drives made attempts at density
autoswitching and did ok but might need fiddling such as from thier keypads.

The Percom controller in most of it's incarnations formats disks 'wierd' by Atari
standards. It writes '1A' hexidecimal to each blank sector on format as opposed
to '00' which leads sector copier programs to read, buffer and write the whole
disk to the destination.  This might be only an annoyance in time lost, but disks
formatted with  an other than '00' blank sector data may not work with
copy-protected copies of programs (say make a bad sector by pulling on a piece of
tape attached to a diskette during a disrete write from a basic language bad
sector maker and then writing all data but that sector).  The program may still
freak at an unbad, expected empty, sector filled with "1a".  This led to a
program for all of us called Pervert which would re-high level write a disk to
read '00' for each unoccupied sector.  No drive or dos for the Atari is freaked
by this as the sectors are not allocated in the bitmap and so logically contain
no data by inference.  It is programs which make direct sector access expecting a
'clean' '00' that freak.

Even though you can configure a percom for enhanced density by editing the
control block, it will not be able to read or write to this format.

The Percom drive uses a 6809 processor - same as Radio Shack COCO computers and a
Western Digital controller - both 40 pin packages.

Percom drives are soft sectored and make no use of the sector timing hole on
soft-sectored disks so use of hard sector disks will not freak the drive out.

The Percom controller will read, write and format a 1.44 mb 3.5inch drive as 1mb
(not 720k) in any Atari dos capable of a 1mb format such as Sparta or Mydos.  The
drive must be jumper configurable to emulate a 77track,26 sector per track,
high-density 8.5inch floppy drive.  The mechs are not all that tough to find.
Look for a chunky jumper block at the rear versus none or just one three pinner
for 0/1.

'AT' style 'flipped' cables just don't work.  Don't even try.  Just use a
straight cable, id and terminate.

Percom drives will not pass a rigorous real-world test.  Your percom drives will
not reliably read and write thier own disks much less disks from your 1050 even
if you tune your standard mechs to 288rpm standard and stick with that even
though there is a minor performance penalty.  Nothing will fix a percom but Bob
Wooley and his soldering iron.  I know a lot about them I swear.

The most common failure for a percom is the legs of the power transistor and
voltage regulators.  These components are mounted to the circuit board, but the
heat-sinking bodies are screwed to the back of the 'L-shaped' baseplate.  This
causes strain on these legs when the sio cables are plugged and unplugged.
Eventually the legs will crack and you will need to install jumpers or replace
the components (10.00 at radio shack for all three).

Having owned several Percom drives and even gone to the extent of purchasing an
upgrade rom in 1986 or so, the drive never was reliable.  I used it for mass
storage above 1050 capacity, but used a real 1050 for archiving disks as I knew
they would allways read. I still have that 400plus disk collection and they STILL
read.

The ATR8000 is a much more interesting controller.  It has a LOT of ram even in
it's default configuration of 16kb, has it's own o/s and CPU and up to 64kb of
ram in most configurations.  I often wondered why more attention wasn't paid to
this capability.  I have documentation and a program which uploads configuration
information.  It is possible to write a 64k software floppy controller using this
upload method. There were a couple of program loaders which used disk images were
created to use the extra space on larger floppies to say store 8 games on a disk,
but no program for Ultraspeed which I allways suspected was possible given the
ram available for buffering.  No Super Archiver in software for ATR8000 which
would have been relatively easy compared to the hardware approach in modifying
other drives.

The ATR8000 is relatively density smart but benefits from DriveZAp or DOSXL's
program too.

The ATR8000 will not ever read or write to Enh density disks though it too will
accept a DriveZap order for enh mode.

The ATR8000 will read, write and format disks up to 1350+/-kb in cp/m mode if
your drive supports high density emulation of 77track/26sectors/double density 8
1/2 inch drives.  Usually these drives will have an unusual number of jumper
blocks for a 1.44 3.5 inch drive.  These drives are limited to 1mb in any Atari
implementation but the mechs still must be the same as for CP/M mode in that they
have the jumpers to do the emulation.

The ATR8000 uses the sector timing hole in soft sectored disks as an indicator as
to the nature of the drives connected. It polles this hole for a couple rotations
for each drive at power up.  This tells the controller's firmware what geometry
drives are connected.  8inch, 5 1/4 inch high-density drives, and 3.5inch drives
rotate at 360 rpm and 5.25inch low density drives such as 92,180,360, and 720k
drives rotate at 300rpm unless you are running them for atari drive
interchangeability at 288rpm which does not affect the controller.

The above means that the ATR8000 will not read or write the back side of  a
'flippy' disk without modifying the floppy media it'self.  There is a trick which
involves powering the atr with the front side of the disk inserted and then
flipping the disk, but I never really got it down. A hardware mod to the drive
mech would fix this but would involve precise drilling of the mech's frame and
installation of another optical sensor/sender pair to make work.  Pertec and
Aerocomp used to produce 'flippy-ready' mechs but that was a LONG time ago and I
have never seen a mech which had both pairs of sensors.

The ATR8000 uses a z80a cpu and a Western Digital controller like most Atari
drive controllers it's reliablity for storage sucks as bad as Rana or Indus but
at least the controller never dies.  The electronics of a percom or atr will
almost allways seem to work properly but never do.

BobKlaas wrote:
> Ok, not sure on the latter drives as I shipped one and can't rember the
> modle number but it was not the last drive you posted..Larry White will
> have it in a few days.

> The best disk Drive ever built in its year for the Atari 8-bit would
> have to be the Percom.  If it will do a double sided double density disk
> then it will format out a 720k 3 1/4 in disk drive adapted to the
> cable.  It will also, do a new 1.44mg 3 1/4 inch drive with a 720k disk
> into it, I could not get it to do a 1.44mg disk with the sensor hole
> covered, received errors.  Perhaps somone else can persue the project
> and come up with a way to enhanse the rom?  But at this time the 2
> double density Percoms I had here would do what was needed to arrive at
> extra storage.

> On A 5 1/2 inch disk you could format it out as a double sided double
> density disk at 40 tracks, and 3 as the step rate, those below would
> also work.  Any other information would be appreciated..The last two
> modles they produced had a Printer interface built in, Larry White has
> the only cable I had in existance.

> Now on any modle to run slaves, you had to attach them with a 34pin
> Ribbon cable, have the 3 inline and turned on first then booted up the
> master.

> On the newer 1.44mg drives I could not get it to boot as a master but it
> would read as a second drive.  The jumper must have to be moved on the
> 1.44mg drive, requires unsoldering..the 720k drives work fine as there
> is a jumper you can move on the drive..perhaps a modified cable like the
> IBM has to see the first drive could correct this but have not gotten
> into that area at this date.

> bob

> This post is about 3 weeks earlier than I antisipated.

> Michael Current wrote:

> > In a previous article, bobkl...@sisna.com (Bob Klaas) says:

> > >I have several, modles to play with but would like if avalable a list of
> > >what they actually produced.

> > I just happen to be trying to build such a list!

> > Based on articles and ads in Antic, Analog, and Atari Classics, I believe
> > the following Percom drives for the Atari were actually produced:

> > RFD40-S1   SS/DD, single drive, 40-track (180K/disk)

> > RFD40-S2   SS/DD, dual drives, 40-track, 180K/disk

> > RFD44-S1   DS/DD, single drive, 40-track, 360K/disk

> > RFD44-S2   DS/DD, dual drives, 40-track, 360K/disk

> > AT88-S1  SS/SD, single drive, 40-track, 90K/disk

> > AT88-S2  SS/SD, dual drives, 40-track, 90K/disk

> > AT88-S1PD  SS/DD, single drive, 40-track, 180K/disk (+printer port)

> > This doesn't take into account the slave drives they kept talking about.  I
> > wonder whether the slave drives are non-Atari-specific?

> > My other question concerns two more models referred-to in an Antic review,

...

read more »

 
 
 

Percom Disk Drives ?

Post by LWhite64 » Sat, 15 May 1999 04:00:00


Well, I now have two Percom's to compare.
They are both AT88 series, but one is the
AT88-S1 (SS,SD only) and the other is
the AT88 SPD (purchased from Bob Klaas -
thanks, Bob).  They are very different
internally as regards the controller boards.
I have not dug into the SPD much yet, except
to look briefly at it's innards, but I will be
doing further investigation, and will post the
results.  BTW, an interesting feature that
I found in the single density version is that
there are several traces scratched off the
main controller board in the area where the
daughter board is installed (with the
1771 FDC.)   My take on this is that Percom
planned at one time to have one controller
board, and then could alter the drive
easily/cheaply by using different FDC
daughter cards.  Of course there are plenty
of alternative possibilities.  In one post it was
mentioned that the Percom was not very
reliable as regards R/W.  I must admit, that
my tests of the single density drive bear that
out.  I picked out 5 S/D Dos disks (no copy
protection, of course), and the Percom
choked on one of them.  My 1050's ate them
up.  Doesn't really prove much, but I
do plan on putting a brand new mech in it
to test to see if it is just slightly mis-aligned.
BTW, the champion Atari drive as regards
reading reliably is the XF551.  No contest.
Anybody else out there use a Percom on a
regular basis?  Reliable?  
Larry
 
 
 

Percom Disk Drives ?

Post by Technoid Mutan » Sun, 16 May 1999 04:00:00


I recall in a previous message I am too lazy to dig up, that one person
had some trouble with the Jumper settings on floppy drive mechanisms.

At risk of offending anyone: Here is what needs to be done to a drive to
connect it to the ATR8000:

My example is an older and very common Teac FD-55B 5.25" double-sided,
40 track drive mech.

First you need to set the drive id.  As there are only four possible
drives in the SA400 spec, you have four choices.  My drive has these
choices labeled:

DS0  (DS for DRIVE SELECT)
DS1
DS2
DS3

DS0 would correspond to D1: and DS3 to D4:

Some drives are labeled differently but this is the jist of it.  No two
drives can have the same id.

A controller such as the Percom and ATR8000 are terminated.  They belong
at one end of the cable.

New PC-type controllers have done away with the requirement for
termination and id settings by flipping a portion of the cable to
autoselect the drive by position on the cable.  Atari 8-bit controllers
can't take these type cables.  A straight cable must be used.

All drive mechs which will work with Atari8 controllers must have a
socketed 'terminator' resistor network chip.  Usually 'RN1' on the
drive's mech.  All drives but the drive connected to the endmost
connector on the cable must have this chip REMOVED.  THe end drive MUST
have the chip installed in the socket.

The terminator acts as a baffle, absorbing signals that would otherwise
reflect off the end of the cable and generate 'noise' which would
interfere with operation.

|----------------|-----------------|----------------|---------------|

^
^
Controller (Terminated
permanantly)                                              Last drive
(Terminate)

Drive Select setting has NOTHING TO DO with termination.  Drives can be
out-of-order on the cable such as 3, 2, 1, 0 or 3,1,0,2, or 3,0,2,1  or
0,1,2,3.  By 'first in the chain' I mean the drive connected to the
connector after the controller's connector.  The terminated drive for
each of the four examples previous would be 1) Drive 0,  2) Drive 2,  3)
Drive 1  4) Drive 3 as they happen to have ended up as the  last drive
on the cable but not neccessarily the last drive id in the chain.

SCSI, "MFM" (ST506/412) compatible drives and controllers (bridge
controllers like OMTI, ADAPTEC, XEBEC  etc) use EXACTLY method of
configuration as the above.  This 34 pin cable was flipped for the PC
clones a long time ago also.  Atari uses a straight cable just like
floppies above. Same rules and procedures for termination etc.  Only
difference is that these 'mfm' mechs only have two available id's
instead of four as in SA400 floppys.

SCSI Drives follow the exact same rules for termination and id except
that the host adapter takes an id (usually 6 or 7) and there are eight
available id's (seven available for controllers).

If you are following then you will realize that the controllers or host
adapters in the case of scsi have thier own terminator.  Often the
Terminators are socketed and can be removed.  A "V" configuration with
the controller unterminated in the middle of the chain with two drives
terminated (the two at the end of each leg of the "V") is perfectly
legal in all the above configurations and interfaces.

Thus, if you have a CSS Black Box, you will look in the corner of the
board nearest the scsi connector and note a 16pin DIP chip which may be
a wierd color such as red, yellow, blue, white, or shiny black.  This is
your terminator pack.  If you pull this chip, and then plug your BB into
the middle connector on your scsi cable you then Must have a drive
connected to the end -connectors on the  scsi cable and those two drives
must be terminated.

Some new scsi drives have a jumper-enabled termination settings.  No
more terminating resistor packs to remove, lose and then later need
again.  Now you can just lose the jumper instead.  This jumper might be
labled "TE" for Terminate enable.

SCSI drives for the 8-bit on the Black Box are required to power the
termination.  I actually ran a wire from +5volts to pin 20 on the scsi
connector of a CDC drive I bought because it had no jumper for powered
termination.  This allowed me to use it as drive 0 without having to
have another drive that did have powered termination at the end of the
cable just to make the drive useable.

No scsi/sasi host adapter for the 8-bit I am aware of will tolerate a
drive set to use a Parity Bit.  This is usually a jumper on the drive
labeled "P" or "PE" for parity enable disable.  This jumper should be
installed or removed as appropriate to provide a Parity OFF condition or
the drive will simply not exist to the Atari's host adapter.

All SCSI host adapters for the 8-bit I am aware of use id 7.  A drive
set to id 7 will not exist on the chain and the host adapter may not
work.

SA400 - Shugart Associates specification for controlling floppy drives
which has been and is still universal.

SASI - Shugart Associates Systems Interface.  The precursure of SCSI.
Hard disks only.  No tapes etc.  Drives are dumb and don't map bad
sectors without help from the computer operating system.

SCSI - Small Computer Systems Interface.   A superset of SASI developed
by a consortium of drive and controller manufacturers.  Development of
SCSI adds Tape drives and just about any other device to run on a SCSI
host adapter.  Atari 8's use small word scsi and don't provide all of
SCSI's features but those related to storeage.

ST506/412 - A 'dumb' hard disk interface developed by Seagate
Technologies.  The hard drives had no controller onboard, just the bare
minimum.  Things such as format and encoding are decided by a controller
such as  an OMTI or Xebec or Adaptec which would be cabled to the drive
by the installer.  That the controller is a separate device and that the
controller decides the encoding (RLL, MFM are most popular) proves that
useing the description "MFM" for this type of hard disk drive is
inaccurate.  Mechs made to this interface spec are encoding neutral.
"RLL Certified" drives are a little faster and have better media.
Faster because of 26 physically shorter sectors per track and better
media to satisfy the linear density required by 40% more sectors in the
same amount of space.  You might call many older IDE drives "RLL" drives
as they use the RLL encoding method by a different interface scheme.

All of the disks, interfaces and encoding schemes above assume that
there will be a fixed number of sectors in a track.  A 17 sector per
track format means that the innermost track (the shortest track on the
platter) can store 17 sectors.  The outer tracks have more area to
record on but fixed-sectors per track makes for less complicated
electronics.

This brings me to an example encoding method called Zone Bit Recording
or ZBR.

Lots of ide and scsi disks now use this method.  With cheaper and faster
electronics, we now have the luxury of electronics smart enough to fully
utilize the disk.  This is accomplished by dividing the disk into
logical "Zones"  The outermost zone might store 80 sectors per track,
the next inner zone only 64 and so on down to 17 for the innermost zone
for example.  The controller onboard the drive determins which zone it
will write in and configures it'self accordingly.  The encoding method
is often a modification of RLL.

The above means that your hard disk 'lies' to your IBM PC compatible and
lets the machine think that there are a fixed number of sectors per
track when there are not.  Useing an older hard disk formatter intended
on a ZBR encoded drive that has bad sectors may result in the wrong
sectors being marked out as these formatters mark according to head and
cylinder and relative sector number as in 1, 620, 17 for the last sector
on an ST225 disk drive versus a linear sector number (21080).  In this
case of a 620,17,2heads drive, a zbr drive would be confused because to
it, you have marked an arbitrary sector leaving the bad one at the end
of the disk free for use.

I don't know what got into me with this message.  I hope it is useful
and doesn't bug anybody with the lecture....

OH, One BIG thing to remember:

Say you only have one drive, that drive must be terminated and on the
last connector on the cable.  There allways allways must be a device
which is terminated and connected to each end for anything to happen.
You can't leave an unconnected "tail" longer than 1/2 inch under any
circumstances or the system may not work.

 
 
 

Percom Disk Drives ?

Post by Technoid Mutan » Sun, 16 May 1999 04:00:00


I have tried new mechs lots of times with the Percom and ATR8000. As a
computer tech I have access to all the parts I need for testing and
still can't make them much more reliable.  Back in 1985 or 1986 when I
got my Percom in a  trade, it was a double density model and was
somewhat unreliable.  It used a belt-driven full-height drive mech.  I
went out and bought two 360k ds/dd Newtronics gear-driven mechs and
installed them.  I have experimented with dialing them down to 288rm etc
to no avail.  There is something wrong with the controller that makes
writes unreliable if verify is off.  Reliability improves with verify
but who wants that?  I actually kinda 'got into' the percom for a while
as it is a four-disk controller and was perfect for a bbs back then.  I
even painted mine Haze Grey once then later sanded the cover to bare
metal and clearcoated it.  I guess I Loved that drive..... Ok, but you
all know what I mean about a treasured peice of gear you lovingly clean
an tune up regularly.   My percom was something of a trophy back then.
Since then I have had three.  One was brand new in a box in 1987 or 88.
It had all the same problems.  I think Bob Wooley or Bob Puff could fix
whatever is wrong with the controller's design.

I hate that it sounds as though I am flaming the Percom drive, but I am
only trying to share my experience with them and how exhaustively I
tried and failed to make the damn thing perfect.


>  I must admit, that
> my tests of the single density drive bear that
> out.  I picked out 5 S/D Dos disks (no copy
> protection, of course), and the Percom
> choked on one of them.  My 1050's ate them
> up.  Doesn't really prove much, but I
> do plan on putting a brand new mech in it
> to test to see if it is just slightly mis-aligned.
> BTW, the champion Atari drive as regards
> reading reliably is the XF551.  No contest.
> Anybody else out there use a Percom on a
> regular basis?  Reliable?
> Larry

 
 
 

Percom Disk Drives ?

Post by sup8p.. » Sun, 16 May 1999 04:00:00




Quote:

> New PC-type controllers have done away with the requirement for
> termination and id settings by flipping a portion of the cable to
> autoselect the drive by position on the cable.  Atari 8-bit controllers
> can't take these type cables.  A straight cable must be used.

Another thing about pc floppies  is that only TWO can be connected.
I have found it is possiable with SOME pc twister cable drives to
connect them to ATR, BBFloppy etc by changing around the jumpers. A
bit of trial and error here as well as a contuinty tester.

Quote:> SCSI, "MFM" (ST506/412) compatible drives and controllers (bridge
> controllers like OMTI, ADAPTEC, XEBEC  etc) use EXACTLY method of
> configuration as the above.  This 34 pin cable was flipped for the PC
> clones a long time ago also.  Atari uses a straight cable just like
> floppies above. Same rules and procedures for termination etc.  Only
> difference is that these 'mfm' mechs only have two available id's
> instead of four as in SA400 floppys.

I have never seen such a cable and all MFM/RLL  mechs I have here can
be set to one of four id's. Finding a controller card that can run
four mechs is a bit of a problem however. Don't forget that a sperate
20way cable for each drive comes from the controller to the mech. ie 2
drives, two 20 way cables plus the 36 way cable etc

Quote:

> SCSI Drives follow the exact same rules for termination and id except
> that the host adapter takes an id (usually 6 or 7) and there are eight
> available id's (seven available for controllers).

The MIO and BB Don't have an id, so it is possiable to have eight scsi
devices.

Quote:

> SCSI drives for the 8-bit on the Black Box are required to power the
> termination.  I actually ran a wire from +5volts to pin 20 on the scsi
> connector of a CDC drive I bought because it had no jumper for powered
> termination.  This allowed me to use it as drive 0 without having to
> have another drive that did have powered termination at the end of the
> cable just to make the drive useable.

Whatout for the MIO. The TPWR pin is grounded.  This usally pops a
fuse in the scsi device that is between +5v and TPWR. It usally cannot
br replaced because it is soldered in and very small.

Quote:

> No scsi/sasi host adapter for the 8-bit I am aware of will tolerate a
> drive set to use a Parity Bit.  This is usually a jumper on the drive
> labeled "P" or "PE" for parity enable disable.  This jumper should be
> installed or removed as appropriate to provide a Parity OFF condition or
> the drive will simply not exist to the Atari's host adapter.

I have heard of people getting around this by using a PAL ot whatever
to generate Parity.
I don't know if it checks Parity, but I would say it woulden't

Quote:

> All SCSI host adapters for the 8-bit I am aware of use id 7.  A drive
> set to id 7 will not exist on the chain and the host adapter may not
> work.

No it doesn't. See above.

Quote:

> SA400 - Shugart Associates specification for controlling floppy drives
> which has been and is still universal.

> SASI - Shugart Associates Systems Interface.  The precursure of SCSI.
> Hard disks only.  No tapes etc.  Drives are dumb and don't map bad
> sectors without help from the computer operating system.

> SCSI - Small Computer Systems Interface.   A superset of SASI developed
> by a consortium of drive and controller manufacturers.  Development of
> SCSI adds Tape drives and just about any other device to run on a SCSI
> host adapter.  Atari 8's use small word scsi and don't provide all of
> SCSI's features but those related to storeage.

SCSI is being updated all the time. Scsi 1 could only support
massstorage devices (from memory) Scsi II could support scanners and
cd roms as well as a speed increase.
Scsi III is faster again with more decive support.
The Harddrive in my pc is a 9.1G Ultra2 scsi LVD.

James.

 
 
 

Percom Disk Drives ?

Post by Jon Mel » Mon, 17 May 1999 04:00:00





>> New PC-type controllers have done away with the requirement for
>> termination and id settings by flipping a portion of the cable to
>> autoselect the drive by position on the cable.  Atari 8-bit controllers
>> can't take these type cables.  A straight cable must be used.

This is inaccurate.  (at least I'm pretty sure <grin>)  The twist
merely changes around the ID lines.  When you use twist style floppy
cables in PC (in general practice) you only use two drives max on a
cable.  You set the device ID jumpers on BOTH drives to drive two
(drive one when counting from zero) and the twist in the cable changes
it back to drive one (or zero depending on the counting scale).  The
twist DOES NOTHING in relation to termination.

What modern PC floppies use (and IDE devices as well I think) is a
form of termination where there are no resister packs to install or
remove and no jumpers to change to set termination.  Termination is
automatic.  The method used is called DISTRIBUTED TERMINATION.  It
gets a little bit technical and to be honest I would have to look it
up in order to explain exactly how it works.  There is a really great
book out wich covers PC hardware called "Upgrading And Repairing PCs"
by Scott Mueller and is published by Que.  It explains it.  If anyone
is interested I can re-read that portion of the book to try and
explain it better.

Quote:>Another thing about pc floppies  is that only TWO can be connected.
>I have found it is possiable with SOME pc twister cable drives to
>connect them to ATR, BBFloppy etc by changing around the jumpers. A
>bit of trial and error here as well as a contuinty tester.

There is no mystical properties here.  Did you know that PCs can have
4 drives as well?  In the old days of PCs (especially the days when a
lot of systems were floppy only) it was not unheard of to have a PC
with 4 floppy drives on it.  You can use non twisted cables perfectly
well on a PC if but you then must take care to set the device ID
jumpers correctly.  All the drives you find in PC style systems have
jumpers to set the ID to 0 through 3 (or 1 through 4) somewhere on
them.  The real big problem that I find with such drives is that you
NEVER seem to get documentation on the floppy drives UNLESS you buy a
retail packaged floppy drive on it's own.  Even then you may not get
the all the jumpers documented.  It goes without saying that OEM drive
bundles or drives that come pre installed in whole system are never
documented.  In the "old" days it was unheard of not to get such
information.  How times have changed.  Fortunately a lot of times you
can find the jumpers for device ID by looking at the silkscreaned info
on the PCB of the drive or other clues.  Then it is a simple matter to
set the drive ID.  You can't always find such clues though, and
sometimes there are other jumpers whose function is somewhat of a
mystery.

Quote:>> SCSI, "MFM" (ST506/412) compatible drives and controllers (bridge
>> controllers like OMTI, ADAPTEC, XEBEC  etc) use EXACTLY method of
>> configuration as the above.  This 34 pin cable was flipped for the PC
>> clones a long time ago also.  Atari uses a straight cable just like
>> floppies above. Same rules and procedures for termination etc.  Only
>> difference is that these 'mfm' mechs only have two available id's
>> instead of four as in SA400 floppys.

I also want to point out, that when termination is required to be
manualy set, keep in mind it is usually if not always independent of
device ID.  Termination is determined only by whether a drive is at
the end of the cable or not.  In other words, the last device in the
chain gets the termination.  You can go change around all the device
IDs to a different order all you like but the termination must be on
the end or distributive across all, depending on the implementation.
Quote:>I have never seen such a cable and all MFM/RLL  mechs I have here can
>be set to one of four id's. Finding a controller card that can run
>four mechs is a bit of a problem however. Don't forget that a sperate
>20way cable for each drive comes from the controller to the mech. ie 2
>drives, two 20 way cables plus the 36 way cable etc
>> SCSI Drives follow the exact same rules for termination and id except
>> that the host adapter takes an id (usually 6 or 7) and there are eight
>> available id's (seven available for controllers).
>The MIO and BB Don't have an id, so it is possiable to have eight scsi
>devices.

>> SCSI drives for the 8-bit on the Black Box are required to power the
>> termination.  I actually ran a wire from +5volts to pin 20 on the scsi
>> connector of a CDC drive I bought because it had no jumper for powered
>> termination.  This allowed me to use it as drive 0 without having to
>> have another drive that did have powered termination at the end of the
>> cable just to make the drive useable.

>Whatout for the MIO. The TPWR pin is grounded.  This usally pops a
>fuse in the scsi device that is between +5v and TPWR. It usally cannot
>br replaced because it is soldered in and very small.

>> No scsi/sasi host adapter for the 8-bit I am aware of will tolerate a
>> drive set to use a Parity Bit.  This is usually a jumper on the drive
>> labeled "P" or "PE" for parity enable disable.  This jumper should be
>> installed or removed as appropriate to provide a Parity OFF condition or
>> the drive will simply not exist to the Atari's host adapter.

>I have heard of people getting around this by using a PAL ot whatever
>to generate Parity.
>I don't know if it checks Parity, but I would say it woulden't

>> All SCSI host adapters for the 8-bit I am aware of use id 7.  A drive
>> set to id 7 will not exist on the chain and the host adapter may not
>> work.

>No it doesn't. See above.

>> SA400 - Shugart Associates specification for controlling floppy drives
>> which has been and is still universal.

>> SASI - Shugart Associates Systems Interface.  The precursure of SCSI.
>> Hard disks only.  No tapes etc.  Drives are dumb and don't map bad
>> sectors without help from the computer operating system.

>> SCSI - Small Computer Systems Interface.   A superset of SASI developed
>> by a consortium of drive and controller manufacturers.  Development of
>> SCSI adds Tape drives and just about any other device to run on a SCSI
>> host adapter.  Atari 8's use small word scsi and don't provide all of
>> SCSI's features but those related to storeage.

>SCSI is being updated all the time. Scsi 1 could only support
>massstorage devices (from memory) Scsi II could support scanners and
>cd roms as well as a speed increase.
>Scsi III is faster again with more decive support.
>The Harddrive in my pc is a 9.1G Ultra2 scsi LVD.

>James.

 
 
 

Percom Disk Drives ?

Post by Sysop Fox » Wed, 19 May 1999 04:00:00


Original Mail Date: May 18, 1999   Time: 12:17 am
Quote:>SCSI, termination.  No Atari 8-bit HD-controller with parity I'm aware

of.

The CSS Black-Box cannot handle SCSI harddisk which require parity.
But...

CSS made an upgrade-PAL to add the parity-signal (the 9th bit) to the
SCSI-bus of the Black-Box.  This PAL had to be mounted on top of the HCT
line-driver/buffer nearby the SCSI resitor-pack and SCSI-bus.
It's a bit expensive (12,- USD or something) but it works.

I made my own upgrade, for about 1.- USD which works exactly the same, but
only requires a few more lines to "redirect" to that same HCT i.c...  :-)

Sysop Fox-1  (fox-1 AT aq DOT nl)

Gateway: THUNDERDOME, Atari 8-bit BBS. Running on ATARI 130XE, Pro-node 471
---
 t

 
 
 

Percom Disk Drives ?

Post by BobKlaa » Sat, 22 May 1999 04:00:00


Ok the expitement continues, I have the XF551 3 1/2 in drive, and
several 720k disks.  The Drive Larry has will read the XF551 720k disks
and execute files placed on it with a 551.  This is done by changing the
drive mech to a 3 1/2 in 720k mech.  Unless you have the file on a 3 1/2
inch disk to boot the Percom you need a file to change the drive on the
fly a P/D program uploaded to my old BBS  a few years ago.  Have it
somewhere but can't remember the name of it, was written by Kenneth
Burgess.

bob

 
 
 

Percom Disk Drives ?

Post by LWhite64 » Sun, 23 May 1999 04:00:00


Quote:>Unless you have the file on a 3 1/2
>inch disk to boot the Percom you need a file to change the drive on the
>fly a P/D program uploaded to my old BBS  a few years ago.

Hi Bob-
I'm sorry, but I'm not following you here.
Are you saying that you need a configuration
changing program, since the Percom don't
automatically change density?  (I.e. send
a new "Percom Block" to the drive?)
Larry
 
 
 

1. Percom Disk Drive

I found a PercomData Disk drive at a bargain store and it has some odd
interface on the bck of it.  Somewhere I read that's it's supposed to be
a parallel port.  It's Model # is AT88-S1PD if that helps at all.  It
looks as if it's the same kind of parallel port that's on the back of an
800xl, but it also has a 34 pin connector, that almost looks like I
could hook a 3.5" floppy up to it.  Any info on this drive would be
greatly appreciated, thanks..


2. problem with special characters

3. Percom Disk Drives, other stuff.

4. AS400 Melrose Park

5. Percom Disk Drive - Drawings?

6. Using DTD With Namespace In Xerces - Possible?

7. PERCOM DISK DRIVES

8. need help about cross-realm

9. My Percom Disk Drive Troubles [Info-Atari8 Digest V87 #114, etc.]

10. Percom disk drive

11. F/S Percom Disk Drive

12. PERCOM Disk Drives