Topics

Integral PC Floppy Drive


Mark Bielman
 

Hi all! Generally speaking, this topic has been beaten (nearly) to death (9122, etc) but here I am again with a faulty Sony floppy drive, this time from one of my Integrals.

It's a Sony OA-D32W-10. (I hate these things!)

Originally, I could not insert a disk. Figured it was gummed up so pulled it for a clean and lube. Easy to get out, really.
Did not see any goo but gave it a clean anyway. "Almost" working but then noticed the arm that opens the door on the disk was not retracting to 'home' position.
Aha! The spring came off the arm. (unusual) Put the spring back on and working normally, so back in the unit it goes.

But now, insert a disk, light ON, spins up, but the system does not even try to read it.

Head looks OK, homes properly, etc. Used my floppy head cleaning disk but no help there.

Stuck now. Any thoughts?

I guess I could try another drive to verify. (maybe pull the good one from my other IPC)

Mark


Bruce
 

Could you possibly have reversed the cable - this is symptomatic of what happens if you do that with a PC


Quoting Mark Bielman <mbielman@hotmail.com>:

Hi all! Generally speaking, this topic has been beaten (nearly) to death (9122, etc) but here I am again with a faulty Sony floppy drive, this time from one of my Integrals.

It's a Sony OA-D32W-10. (I hate these things!)

Originally, I could not insert a disk. Figured it was gummed up so pulled it for a clean and lube. Easy to get out, really.
Did not see any goo but gave it a clean anyway. "Almost" working but then noticed the arm that opens the door on the disk was not retracting to 'home' position.
Aha! The spring came off the arm. (unusual) Put the spring back on and working normally, so back in the unit it goes.

But now, insert a disk, light ON, spins up, but the system does not even try to read it.

Head looks OK, homes properly, etc. Used my floppy head cleaning disk but no help there.

Stuck now. Any thoughts?

I guess I could try another drive to verify. (maybe pull the good one from my other IPC)

Mark



Mark Bielman
 

That's possible Bruce. I will open my other unit tomorrow and will check that and perhaps swap the drives. Pretty easy to do.

Mark

BTW - such a cool machine! I have one with the SW Engineering ROM Module AND the BASIC ROM installed. My HP 8904A signal generator came today, hooked it up to my Integral and did some basic HP-IB commands and it's SO EASY! Everything works great.


Martin Hepperle
 

The optical track 0 sensor in the SONY drive in my last 9133 box was partially clogged with "pocket fluff"
The drive worked only sporadically, until I very thoroughly cleaned the gap and optical path of the sensor.

Martin


Mark Bielman
 

At one point in the cleaning process, the head stepper would move all the way in (away from track 0) at power up.
Blew air into the hall sensor and that fixed that, but it still would not try to read a disk. :-(

Mark


Mark Bielman
 

Well this is interesting...

Put a known good drive in the IPC and it does the exact same thing!
Maybe a bad cable?

Mark


Mark Bielman
 

OK. Well, tried another drive and another cable. No change. It must be a problem on the Logic B board. :-(
Have in "fold apart" mode now for troubleshooting. The 2MHz clock, VCO and pulse widths (per the service manual) are all in spec.

Poking around the controller revealed next to nothing, although some of the pins seem like they could be incorrect...
READY signal is never true, same with HEAD LOAD, TRACK0, some others. Basically see no activity on these kinds of pins when I insert a disk or not.
Inverters and other gates look OK.

Could I have a faulty controller IC?

Attached a schematic and a data sheet.

Appreciate suggestions!

Mark


 

Mark,

 

Thanks for the update.

 

I’m afraid I don’t have many suggestions, I’m not a hardware/electronics guy, so can’t contribute much along the lines of schematics, traces, scopes, etc. I do PC service and repair, which these days involves ZERO schematics, voltmeters, scopes, etc. it’s all software and peripheral replacement.

 

For floppy issues, I have seen cases in the past when corrosion where the ribbon cable connects to the board causes flaky issues, but I’ve no idea what kind of connection that is in the IPC. If I understand correctly, both drives you have work OK in one of your IPC’s and both don’t work in the other, and the problem remains even with multiple ribbon cable swaps, is that right? It turns out I also have a dual-sided FDD (from a 9114B) that I could send if that might help you to isolate the issue.

 

You mentioned the S/W Dev board – do you have one of those available? And I also see that there is a C Compiler available on Disk (floppies) – Is the Board just a ROMed version of the disk software? I also see that Fortran, Forth, Pascal, etc. are available in addition to Technical BASIC, so this is quite a nice little developers box.

 

Looking in the 86-88 HP Catalogs, I see the IPC was considered a part of the 9000 Series – which surprised me, I had never heard that before.

 

--bob

 

From: VintHPcom@groups.io [mailto:VintHPcom@groups.io] On Behalf Of Mark Bielman
Sent: Sunday, January 03, 2021 7:07 PM
To: VintHPcom@groups.io
Subject: Re: [VintHPcom] Integral PC Floppy Drive

 

OK. Well, tried another drive and another cable. No change. It must be a problem on the Logic B board. :-(
Have in "fold apart" mode now for troubleshooting. The 2MHz clock, VCO and pulse widths (per the service manual) are all in spec.

Poking around the controller revealed next to nothing, although some of the pins seem like they could be incorrect...
READY signal is never true, same with HEAD LOAD, TRACK0, some others. Basically see no activity on these kinds of pins when I insert a disk or not.
Inverters and other gates look OK.

Could I have a faulty controller IC?

Attached a schematic and a data sheet.

Appreciate suggestions!

Mark


Mattis Lind
 

I have almost non experience with the Integral hardware. My own unit seems to work, but I haven't tested the floppy. Until this date I have only studied the schematics, that's all.

Since the chips is under software control I guess it would do nothing unless it detects proper input signals.

Do you have MOTOR ON active (low) ? Drive Select 1 ? I would guess that that is the first signal that the CPU activates. Then it probably waits for the RDY signal to come active (low).

Is there anything happening on U24 outputs?

I am not fully aware of how the floppy operates in the integral, but potentially it is monitoring the DSKCHG signal and only turns on the motor when a new disk has been inserted.

Does the CPU access the U24 or U18 at anytime? Check pin 9 and 3 respectively. Is it reading the DSKCHG status? U25 pin 13.

In the DIAG ROM there are tests for the floppy. Do you have the DIAG ROM? 

/Mattis


Mark Bielman
 

OK. Managed to spend some time on this. (work getting in the way!)

U18 (controller) pins like CS (pin 3) look good. A lot of activity at power on (configuration) and a periodic pulse after that (checking status).
Other pins (RE, WE, data) are all OK too.

Outputs from latch U24 look OK, but do not change. (clock input also has a periodic pulse) Outputs are:
pin 2 (motor on) = H (gets inverted so L to drive)
pin 7 (SS0) = L (also inverted)
pin 10 (DSKRST) = H (also inverted) <--- seems wrong!
pin 15 (Drive Select 1) = L (also inverted) <--- trace cut and signal to drive tied to Vcc

Assume that if both Drive Select signals are H, that selects Drive 0. (these are active low)

That DSKRST line does not seem right. An active low on the floppy side, and always active!
But the floppy manual has this pin as CHG RST and no real explanation as what it does. (something to do with a disk change)

Some other findings are:
DSKCHG (from drive, active low) is always high
WRPT (from drive, active low) is always high (tried changing on the disk)


Data inputs to U24 look good.

Mark


Mark Bielman
 

Also, there is a periodic pulse on U25 enable pins.

I do NOT have a (working) diag rom


Mark Bielman
 

No ideas from the group it seems. Next up... remove the Logic B board and examine w/microscope. THEN, swap in another board from a working IPC. (was hoping to avoid that)
I could also use 2 scope probes to verify that latch U24 is working. (I think I can do that, trigger on the CLK signal)

Mark


Arnie Berger
 

A couple of suggestions:

  1. Might be worth investing in an inexpensive logic probe to examine pin states on ICs. A nice feature is they blink on clock pulses, which is sometimes easier than a scope to observe.
  2. Another option is to invest in a low-cost, 16-channel logic analyzer. That should be enough channels to do some signal tracing. My old HP scope has a 2 analog and 16 digital channels. This has been very handy for debugging these kinds of problems. I’ve seen them on Amazon for under $80, but I have no idea how good they are. We use LogicPort logic analyzers in out undergraduate labs. They work very well, have 34 channels, but cost just under $400. Use the PC as your front panel and connect via USB. Also, you can buy a used HP LA ( my old division) on ebay for under $200.
  3. If you could locate one of HP’s original Signature Analyzers ( HP1600 ?) that would be another way. You’ll need a working system to do that, or at least a working board. Each node on the board can generate a unique signature. Then substitute the faulty board and start tracing. As soon as you see variations in the signature, then start working backwards until you isolated the problem. I’ve never used it but I’ve read about it. Of course, without a schematic, it could be really hard to fix.

 

Arnie

 

 

 

From: VintHPcom@groups.io <VintHPcom@groups.io> On Behalf Of Mark Bielman
Sent: Wednesday, January 6, 2021 5:29 PM
To: VintHPcom@groups.io
Subject: Re: [VintHPcom] Integral PC Floppy Drive

 

No ideas from the group it seems. Next up... remove the Logic B board and examine w/microscope. THEN, swap in another board from a working IPC. (was hoping to avoid that)
I could also use 2 scope probes to verify that latch U24 is working. (I think I can do that, trigger on the CLK signal)

Mark


Mattis Lind
 

What I find puzzling is that the drive is not reporting READY. I don't think that controller will do anything unless the drive is ready.

The drive manual says:

4·3·2-4. Ready (READY) - .f. This line is Logical "0'" when a) disk is inserted and b) motor is rotating and c) drive unit is selected. At all other times this line is Logical "1".

The Disk change reset signal should affect the READY status as far as I understand.

The drive is set as drive 1, right? Maybe it can be useful to measure on the drive to verify that both drive selects are HIGH.
There is a minimum time between MOTOR ON and SELECT that is 400 ms which probably is that it has to see at least a number of index holes pulses to go ready. BTW. Are there any index pulses coming from the drive?

Is is something wrong with the floppy disk itself? You have tested a few others, right? Can you verify that the floppy is rotating correctly?

/Mattis


Mattis Lind
 

I mean to write "The Disk change reset signal should NOT affect the READY status as far as I understand.".


Mark Moulding
 

In many projects over the years, I've never gotten used to the idea of reaching for a logic analyzer to debug a problem, but whenever I did, it worked like no other tool could have.  That said, used logic analyzers that are suitable for equipment of this vintage are readily available on eBay, for not too much money.  A quick look shows this one for $75 looking like a pretty good buy, which also includes the pods (an absolute requirement; old swap meet joke: "The logic analyzer?  Only $10.  The pods are an additional $500...")

I'd go for an old HP analyzer.  HP was the inventor and pretty much king of logic analyzers, and made a zillion different models.  Even the oldest (1600A) is useful, but try to choose a listing that demonstrates it working, at least to the extent of 1s and 0s on the screen.  And make sure you get those pods and the gripper leads that go with them!
~~

Mark Moulding


Anders
 

Yes, the HP163X series is a bit old, loud and clunky, but I have a HP1631D where I replaced the fan for a quieter one. It also has very nice HPIB decodes. See: http://www.dalton.ax/hp1631D/

Shipping is a bummer as it is heavy, but I carried my home in my luggage :)


 

I recently got a used HP1670D with all the cables and clips etc for $210 including shipping and it was a great deal.
It's heavy (it has a CRT), the screen update is slow if there are a lot of transitions, the UI is clunky (better with keyboard and mouse) and file transfer by ethernet is a pain, but it does what I need and more. I strongly recommend them and agree that it's important that it work and have as many of the connectors as possible.
I looked at the 1630 in your link and it looks good - I don't know much about the feature differences myself but the sources I consulted about the 1670 suggested that a 1670 or higher was a better buy than a 1630.
All the best
Brian


Mark Bielman
 

All very interesting but not too helpful at the moment.

Mark


Arnie Berger
 

I second Mark’s opinion. Of course, I come from that Division, so I’m a bit biased in my opinion.

 

Arnie

 

From: VintHPcom@groups.io <VintHPcom@groups.io> On Behalf Of Mark Moulding
Sent: Thursday, January 7, 2021 9:14 PM
To: VintHPcom@groups.io
Subject: Re: [VintHPcom] Integral PC Floppy Drive

 

In many projects over the years, I've never gotten used to the idea of reaching for a logic analyzer to debug a problem, but whenever I did, it worked like no other tool could have.  That said, used logic analyzers that are suitable for equipment of this vintage are readily available on eBay, for not too much money.  A quick look shows this one for $75 looking like a pretty good buy, which also includes the pods (an absolute requirement; old swap meet joke: "The logic analyzer?  Only $10.  The pods are an additional $500...")

I'd go for an old HP analyzer.  HP was the inventor and pretty much king of logic analyzers, and made a zillion different models.  Even the oldest (1600A) is useful, but try to choose a listing that demonstrates it working, at least to the extent of 1s and 0s on the screen.  And make sure you get those pods and the gripper leads that go with them!
~~

Mark Moulding