Re: CDC 9429 Floppy maintenance manual

[Chuck Guzis]

On 05/08/2016 02:33 PM, Philip Pemberton wrote:

> My awards list more or less goes as follows:
> 
> Hardest to align: Nintendo Famicom Disk System. For bonus points,
> when you replace the drive belt, you have to realign the drive hub,
> which sets the "start of track" position. There must be a jig or
> procedure to do this, but I've never seen it. Homebrew procedure is
> to loosen the hub and rotate it a few degrees until things align and 
> the drive works... The hub alignment, incidentally, is critical
> because the discs are written as a continuous spiral track, not a
> series of concentric tracks.

Used on the Smith-Corona PWPs as well.  Sad part is that the old belt
turns to goo and is hard to clean and get running again.  There is/was a
seller on eBay who was offering  polyurethane replacements.  I've got a
pile of those sitting in my freezer.

> Nicest half-height 5.25in: Teac FD-550 series I love these drives to
> bits. There are a bunch of variants (40/80 track, 1.2Meg and 360K)
> but they're pretty solid performers. Fairly good at reading crusty
> old disks. Keep a few Bemcot wipes and some isopropyl around to clean
> the heads.

I think you mean FD-55 series.  Shame that they never made any 100 tpi
varieties, but Teac did rebadge someone else's FH 100 tpi drive.

> Weirdest drive interface: the NEC 8-inch drive Uses something called
> a "VFO" interface (I think I remembered that right?), which is a
> Japanese standard. Also needs to be rejumpered to provide raw data
> output. This is jolly good fun, because the jumpers (if memory
> serves) have quite odd labels...

You can find old PC98-era Japanese stuff (along with some CNC gear) that
requires these.  Getting the gear going with a commodity legacy drive is
a real chore--there used to be a manufacturer of external PLL data
separator boards to accomplish this.

> The "What were they thinking?!" award: Amstrad 3-incher, made by
> Panasonic. PC style power connector pinout. With the 5V and 12V
> swapped. You can bet every one of these you'll find that's been
> "tested working, motor spins when powered but that's normal" will
> have a fried ASIC. Again, has a drive belt, but at least you can
> replace this without cocking up the alignment.

I've done in a 3.5" drive on a Joyce by thinking that nobody would be so
stupid as to swap the +5 and +12 on the same connector.  Yes, it fried
the drive.   Also, the interface cable requires some serious 'reweaving'
to interface to a traditional (Shugart-style) interface.

> For 3.5in PC drives, I quite like the Sony drives. From experience
> with DiscFerret, they're pretty good at pulling a clean signal off
> discs some other drives won't even read. Some Panasonic drives are
> better built, though. Apples and oranges.

I've lately taken up with Samsung SFD-321B drives for general use.
Well-made and very flexible.  I occasionally provide drives to the CNC
people where pin 34 is READY/ and pin 2 is DISK CHANGED/ with DS0 and
1.6MB (360 RPM) mode.  All easily done with the Sammies.

-
On a related note of "weird and wonderful", does anyone have media for
the Western Digital "Take Ten" cartridge drive?

--Chuck

Re: CDC 9429 Floppy maintenance manual

[Philip Pemberton]

On 08/05/16 21:49, Chuck Guzis wrote:
> On 05/08/2016 01:30 PM, tony duell wrote:
> 
>> They are considerably better than the Shugart drives with the plastic
>> disk with a spiral groove for the head positioner
> 
> Ah yes, the SA-400.  When I was evaluating one, I wondered if Shugart
> was really serious about the things.  Doubtless some engineer at Shugart
> was quite proud of himself for designing a ball-bearing follower for the
> spiral groove.  Marks for being cheap.
> 
> Oddly, my own troubles with the SA-400 (believe it or not, this was used
> as the original IBM offering for the 5150 drive) were with the tach
> circuit.  Mine blew a small inductor.
> 
> Among all of the 5.25" FH drives offered at the time, my vote still goes
> to Micropolis for designing the Sherman tank of drives.  Leadscrew
> positioner with multiple steps per track.  Probably among the most
> expensive 5.25" drives also.
> 
> To their credit, Shugart did have the best hub-clamp setup.
> 
> --Chuck
> 
> 


My awards list more or less goes as follows:

Hardest to align: Nintendo Famicom Disk System.
  For bonus points, when you replace the drive belt, you have to realign
the drive hub, which sets the "start of track" position. There must be a
jig or procedure to do this, but I've never seen it. Homebrew procedure
is to loosen the hub and rotate it a few degrees until things align and
the drive works...
  The hub alignment, incidentally, is critical because the discs are
written as a continuous spiral track, not a series of concentric tracks.


Nicest half-height 5.25in: Teac FD-550 series
  I love these drives to bits. There are a bunch of variants (40/80
track, 1.2Meg and 360K) but they're pretty solid performers. Fairly good
at reading crusty old disks. Keep a few Bemcot wipes and some isopropyl
around to clean the heads.


Weirdest drive interface: the NEC 8-inch drive
  Uses something called a "VFO" interface (I think I remembered that
right?), which is a Japanese standard. Also needs to be rejumpered to
provide raw data output. This is jolly good fun, because the jumpers (if
memory serves) have quite odd labels...


The "What were they thinking?!" award: Amstrad 3-incher, made by Panasonic.
  PC style power connector pinout. With the 5V and 12V swapped. You can
bet every one of these you'll find that's been "tested working, motor
spins when powered but that's normal" will have a fried ASIC.
  Again, has a drive belt, but at least you can replace this without
cocking up the alignment.


For 3.5in PC drives, I quite like the Sony drives. From experience with
DiscFerret, they're pretty good at pulling a clean signal off discs some
other drives won't even read. Some Panasonic drives are better built,
though. Apples and oranges.


Cheers,
-- 
Phil.
classic...@philpem.me.uk
http://www.philpem.me.uk/

Re: CDC 9429 Floppy maintenance manual

[Fred Cisin]

It would have at least had to be the SA400-L, since the SA400 was 35 track, 
and PC-DOS, from the get-go, used 40 tracks.


Or [trivially] patched.  Either done right, or kludged by putting a 
non-existent file in the directory occupying tracks 35-39.



There were some strange patches in the early days!

PC-DOS 1.00 was single sided only, but the FDC supported double sided, and 
double sided drives were very available.  There was a patch being passed 
around that would permit installing a double sided drive, and the second 
side would be treated as though it were an additional drive.  I never got 
around to trying that with 4 drives installed, . . .

Does anybody know what the actual origin was for that patch?

Re: CDC 9429 Floppy maintenance manual

[Chuck Guzis]

On 05/08/2016 02:02 PM, Paul Berger wrote:

> Are you sure about the SA-400 being used in 5150s?  All the ones I
> ever saw where Tandon TM-100s which look very much like them and I
> saw lots as a support person in an IBM lab in the early 80s.  They
> have the same sort of motor setup to turn the diskette, but use a
> different head positioner.   The TM-100s used a stepper with a taut
> band positioner.

Quite.  I purchased my 64K 5150 from Computerland in San Jose (I
probably still have the receipt) and included an MDA, a floppy
controller, the SA-400, as well as PCDOS 1.1, MASM 1.0 and the techref.
 I didn't buy a monitor because I already had an OEM kit monitor that
was probably better than the 5151.  I later added a Quadram Quadboard
with 256KB of memory--and then moved to PCDOS 2.1 and worked out a
simple interface to a WD1000 controller with a Shugart SA-1000 hard
drive, giving me 4MB of real hard disk.

When I was on a floppy-only system, I added a Micropolis 1115-VI second
drive.  An odd beast with buffered seek.

So, yes, I'm quite sure.

--Chuck

Re: CDC 9429 Floppy maintenance manual

[Fred Cisin]

Oddly, my own troubles with the SA-400 (believe it or not, this was used
as the original IBM offering for the 5150 drive) were with the tach
circuit.  Mine blew a small inductor.


On Sun, 8 May 2016, Paul Berger wrote:
Are you sure about the SA-400 being used in 5150s?  All the ones I ever saw 
where Tandon TM-100s which look very much like them and I saw lots as a 
support person in an IBM lab in the early 80s.  They have the same sort of 
motor setup to turn the diskette, but use a different head positioner.   The 
TM-100s used a stepper with a taut band positioner.


The first 5150s that I had, or saw, had Tandon TM100-1 (single sided).
Followed (along with PC-DOS 1.10) with the TM100-2 (double sided).  Later, 
I started to see some other brands of drives used by IBM in 5150s, but 
NEVER Shugart SA400.
It would have at least had to be the SA400-L, since the SA400 was 35 
track, and PC-DOS, from the get-go, used 40 tracks.


The SA400 was, however, the original offering for the TRS80.  Followed 
later by TM100-1 and a few others.


And the SA390 (same mechanism, but without the circuit board) was used as 
original offering for Apple.  There was an unconfirmed rumor that, for a 
while, when they ran short, that Apple bought SA400s and removed the 
board.



I agree about Micropolis; it was the most reliable drive that I had, 
although kinda slow stepping.



The LEAST reliable drives that I tried were the BASF 2/3 height, followed 
by the Qumetrak 142 (early 1/2 height)



Worst problem that we faced with the TM100s was that students in the 
computer labs would let the door slam when opening, breaking the hinge.

Re: CDC 9429 Floppy maintenance manual

[Paul Berger]

On 2016-05-08 5:49 PM, Chuck Guzis wrote:

On 05/08/2016 01:30 PM, tony duell wrote:


They are considerably better than the Shugart drives with the plastic
disk with a spiral groove for the head positioner

Ah yes, the SA-400.  When I was evaluating one, I wondered if Shugart
was really serious about the things.  Doubtless some engineer at Shugart
was quite proud of himself for designing a ball-bearing follower for the
spiral groove.  Marks for being cheap.

Oddly, my own troubles with the SA-400 (believe it or not, this was used
as the original IBM offering for the 5150 drive) were with the tach
circuit.  Mine blew a small inductor.

Among all of the 5.25" FH drives offered at the time, my vote still goes
to Micropolis for designing the Sherman tank of drives.  Leadscrew
positioner with multiple steps per track.  Probably among the most
expensive 5.25" drives also.

To their credit, Shugart did have the best hub-clamp setup.

--Chuck
Are you sure about the SA-400 being used in 5150s?  All the ones I ever 
saw where Tandon TM-100s which look very much like them and I saw lots 
as a support person in an IBM lab in the early 80s.  They have the same 
sort of motor setup to turn the diskette, but use a different head 
positioner.   The TM-100s used a stepper with a taut band positioner.


Paul.

Re: CDC 9429 Floppy maintenance manual

[Fred Cisin]

> > Ahhh... of course, I should have thought of that.
> > I am even more cautious about using the 9428 manual for 9429
> > service, then.

Actually, many lines of drives, such as the Tandon TM100 have the
same circuitry for both the 48tpi and 96tpi variants. (and the
TM100-4M at 100tpi)


On Sun, 8 May 2016, Chuck Guzis wrote:

One of the things that endears to me those pieces of garbage is that
they apparently changed PCB designs according to what parts they had on
hand.  I've got at least three different versions of the TM-100 PCBs,
depending on the head stepper used (4 wire/6 wire).  Some were better
than others.


Which would be an arument to be cautious about using the 
TM100-2  manual for the TM100-2, not just for the TM100-4.

Re: CDC 9429 Floppy maintenance manual

[Chuck Guzis]

On 05/08/2016 01:30 PM, tony duell wrote:

> They are considerably better than the Shugart drives with the plastic
> disk with a spiral groove for the head positioner

Ah yes, the SA-400.  When I was evaluating one, I wondered if Shugart
was really serious about the things.  Doubtless some engineer at Shugart
was quite proud of himself for designing a ball-bearing follower for the
spiral groove.  Marks for being cheap.

Oddly, my own troubles with the SA-400 (believe it or not, this was used
as the original IBM offering for the 5150 drive) were with the tach
circuit.  Mine blew a small inductor.

Among all of the 5.25" FH drives offered at the time, my vote still goes
to Micropolis for designing the Sherman tank of drives.  Leadscrew
positioner with multiple steps per track.  Probably among the most
expensive 5.25" drives also.

To their credit, Shugart did have the best hub-clamp setup.

--Chuck

RE: CDC 9429 Floppy maintenance manual

[tony duell]

> > Actually, many lines of drives, such as the Tandon TM100 have the
> > same circuitry for both the 48tpi and 96tpi variants. (and the
> > TM100-4M at 100tpi)
> 
> One of the things that endears to me those pieces of garbage is that

They are considerably better than the Shugart drives with the plastic disk 
with a spiral groove for the head positioner

> they apparently changed PCB designs according to what parts they had on
> hand.  I've got at least three different versions of the TM-100 PCBs,
> depending on the head stepper used (4 wire/6 wire).  Some were better
> than others.

HP used the TM100 in some of their machines. Oddly, the PCB is electrically
identical to the normal Tandon one (even down to having jumper options that
are never used in HP machines AFAIK) and has the same layout but is built like
an HP board (entirely gold plated) and many of the ICs have 1820- numbers.

-tony

Re: CDC 9429 Floppy maintenance manual

[Chuck Guzis]

On 05/08/2016 01:09 PM, Fred Cisin wrote:

> Actually, many lines of drives, such as the Tandon TM100 have the
> same circuitry for both the 48tpi and 96tpi variants. (and the
> TM100-4M at 100tpi)

One of the things that endears to me those pieces of garbage is that
they apparently changed PCB designs according to what parts they had on
hand.  I've got at least three different versions of the TM-100 PCBs,
depending on the head stepper used (4 wire/6 wire).  Some were better
than others.

Jugi was not known for quality production, just low cost  Its amazing
that so many survive--must be due the the numbers deployed.

JTS hard drives, anyone?

--Chuck

Re: CDC 9429 Floppy maintenance manual

[Fred Cisin]

I have reason to believe that the CDC 9428 and 9429 are identical
except that the 9429 is jumpered for 80 tracks and the 9428 is
jumpered for 40 tracks... but I'm not 100% sure.

* On Sat, May 07, 2016 at 11:17:50PM -0600, Eric Smith  
wrote:

They would have to have different heads, if the 40 track model was
standards-compliant and industry-compatible.


On Sun, 8 May 2016, Seth Morabito wrote:

Ahhh... of course, I should have thought of that.
I am even more cautious about using the 9428 manual for 9429
service, then.


Actually, many lines of drives, such as the Tandon TM100 have the same 
circuitry for both the 48tpi and 96tpi variants. (and the TM100-4M at 
100tpi)


A specific exception, that I don't think is relevant here, is that a 1.2M 
drive will have a few circuitry changes V a 48tpi/"360K" model, such as 
RWC (reduced write current), and sometimes a 300RPM V 360RPM control.


But, a "360K" (CDC 9428) V "720K" (CDC 9429) are likely to be virtually 
identical other than step distance and head width.

Re: CDC 9429 Floppy maintenance manual

[Seth Morabito]

* On Sat, May 07, 2016 at 11:17:50PM -0600, Eric Smith  
wrote:
> On Sat, May 7, 2016 at 1:19 PM, Seth Morabito  wrote:
> > I have reason to believe that the CDC 9428 and 9429 are identical
> > except that the 9429 is jumpered for 80 tracks and the 9428 is
> > jumpered for 40 tracks... but I'm not 100% sure.
> 
> They would have to have different heads, if the 40 track model was
> standards-compliant and industry-compatible. 

Ahhh... of course, I should have thought of that.

I am even more cautious about using the 9428 manual for 9429 
service, then.

-Seth
-- 
Seth Morabito
s...@loomcom.com

CDC 9429 Floppy maintenance manual

[Seth Morabito]

I don't suppose anyone has a copy of the CDC 9429 floppy drive
maintenance manual they'd be willing to scan for me?

I have reason to believe that the CDC 9428 and 9429 are identical
except that the 9429 is jumpered for 80 tracks and the 9428 is
jumpered for 40 tracks... but I'm not 100% sure. I do have a copy of
the 9428 maintenance manual, thanks to Bitsavers, but having the 9429
manual would put my mind at ease lest I follow the alignment
procedures from the 9428 manual and screw something up.

-Seth
-- 
Seth Morabito
s...@loomcom.com