General Category > Lisa Troubleshooting and Repair

Will the Lisa return an IO error without the "Lite" Interface Card

(1/7) > >>

friedboard:
Trying to isolate a problem on a potentially fried IO floppy controller, does anyone know if the Lisa will return error 57 without the "Lite" Interface Card plugged in?
Mine is returning the error without the interface card, or with the interface card + floppyemu. But no error with original floppy drive (the drive doesn't work).

Trying to see if this is something on the IO board or lite interface card.
Looking at the error code in service mode in address fcc017 shows
10 ERROR: invalid format parameter

sigma7:

--- Quote from: friedboard on August 06, 2025, 11:43:35 pm ---a potentially fried IO floppy controller, does anyone know if the Lisa will return error 57 without the "Lite" Interface Card plugged in?

--- End quote ---

My experience is that error 57 generally indicates the FDC isn't responding to the 68k. There is a CPU ROM assembly listing that can be studied to see the specific conditions, but I think you won't get more troubleshooting information beyond "FDC not working"... ie. you won't find out any details as to why it isn't working.

"Invalid format parameter" is an error returned by the command validation check ("VALIDATE" in the I/O ROM) when the 68k tries to command the FDC to format/initialize a floppy, but hasn't setup the correct settings for formatting. It is set at $170E in the I/O ROM rev. A8 assembly listing (couldn't find that, but this 2/10 listing is similar: https://bitsavers.org/pdf/apple/lisa/firmware/IO_ROM_88_Listing_Dec83.pdf.

Although I'm sure they can fail, I believe Lite Adapters are not a common failure point. More likely is a poor contact between the I/O EPROM or 6504 and their socket, or the I/O or CPU Board and the motherboard.

Perhaps your 6504 isn't running (or not running valid code). Some suggestions:

- Test that you can successfully write and read back to the SRAM shared with the FDC, eg. can you change the byte at FCC017  ? (Note that the 68000 can't access a word at an odd address - if you try, you will get an error bouncing you out of service mode. So you may find it easier to write a 16 bit value to FCC016)
- If the 6504 is doing it's normal "waiting for command" loop, the "ImAlive" byte should be changing at FCC051
- Is the 8T97 U2F on the I/O Board hot; I think that can indicate damage due to an incorrectly plugged in drive?

I suppose it is also appropriate to check you have a self-consistent hardware configuration... ie. if you are trying to use a Lite Adapter, you should have a 2/5 I/O Board (two, not 3 large chips at the top left), a 2/5 Motherboard (with a rear panel parallel port), and 2/5 chassis wiring (two 26 pin ribbon cables in the drive cage area).

friedboard:
Nice, thanks for the suggestions!

- I can write and read back from FCC017 just fine, no errors even if I read or write directly from FCC017 and the value updates.
- The value at FCC051 is always different every time I try to read it, is that a good sign?
- The chip at 8T97 feels normal and is not HOT.

Some background: I plugged in an ESProfile emulator into the 26 pin port right next to the lite adapter board which fried a floppyemu that is connected to the adapter and the Lisa no longer boots through another working floppyemu.

AlexTheCat123:

--- Quote from: friedboard on August 07, 2025, 03:45:36 am ---Some background: I plugged in an ESProfile emulator into the 26 pin port right next to the lite adapter board which fried a floppyemu that is connected to the adapter and the Lisa no longer boots through another working floppyemu.

--- End quote ---

Yep, that's almost certainly what's causing your issues right there. On the Lisa 2/5, the parallel port for the ProFile is going to be on the back of the machine, not in the drive bay like on the 2/10. The extra 26-pin cable next to the Lite Adapter is the cable for the second Twiggy drive, and it's not compatible with hard drives (or hard drive emulators). So plugging an ESProFile into that port fried the floppy stuff and could've fried your ESProFile too.

It's unfortunate that both the ProFile cable and the Twiggy cable use 26-pin connectors; Apple really should've designed it so that you can't accidentally plug one into another!

sigma7:

--- Quote from: friedboard on August 07, 2025, 03:45:36 am ---- The value at FCC051 is always different every time I try to read it, is that a good sign?

--- End quote ---

Yes and no... it probably means the 6504 is running and reading the EPROM correctly, which is good, but means you don't have a simple problem to solve.


--- Quote ---plugged in an ESProfile emulator into the 26 pin port right next to the lite adapter board
--- End quote ---

That's an easy mistake to make. Back in the 80's we didn't have the plethora of inexpensive connectors available today, so it was more difficult to make all connectors unique (eg. the serial and parallel back panel connectors both using DB-25). However, having been made in those 'simpler' times, the Lisa is also much more repairable than modern computers.

The pinout of the Twiggy drive cable can be observed from the Twiggy Digital Board schematic that it once plugged into.
https://lisalist2.com/index.php/topic,524.msg3712.html#msg3712

The pinout of the parallel port connector (that the ESProFile expects) is nicely presented in Patrick's IDEfile documents
http://john.ccac.rwth-aachen.de:8000/patrick/idefile.htm

Since TTL is generally tolerant of misconnections between 0 and 5V, I'd guess the problem on the Lisa side was caused by the ground pins of the ESProFile shorting +12V to something. (You might find more issues beyond this.)

On the parallel port 26 pin header as used by a Widget, these pins are connected together for ground: 2,3,7,12,14,17,19,22,26 I don't know if all of them are used for ground by the ESProFile but for some other device they are likely to be, so we will consider them all.

On the Twiggy Digital Board Schematic, +12 is connected to pins 1,3,5,7.

In addition we have -5V on pin 25.

So plugging those devices together is likely to put +12V on what would normally be Widget ground pins (but are not ground for the floppy connector).

The pins suffering this potential injustice are:
 2 - Phi A
12 - /WREQ
14 - RData
17 - -12V
19 - (n/c)
22 - MClk
26 - (n/c)

-12V is weak supply, but the circuit to the PSU may have been damaged, so first check that:
a) all the voltages from the PSU are within tolerance (measure at points on the I/O Board), and
b) +12 and -5 still make it to the end of the Twiggy connector ribbon cable

Tracing those signals to the I/O Board:

Phi A comes from the 8T97 U2F, which is used by the FDC to interrogate the floppy drive for disk-in-place, write protect, move the head etc. I imagine that if this were damaged, the FDC might hang trying to access the floppy, resulting in error 57.

/WREQ comes from U3F-10, but is only used for writing to the floppy, so maybe U3F is damaged; you may not find out until you try to write to a floppy.

RData goes to U5B-13. Again you may not find out if U5B is damaged until you try to read data from a floppy.

MClk is called MT0 or MT1 (depending on upper or lower drive cable) on the Motherboard and I/O Board schematics. These are driven by U4F. If the motor spindle doesn't run when the FDC tries to read or write to the disk, then U4F would be the suspect. It could be that only the upper drive output is damaged, in which case you may never know until you plug in a set of Twiggies.

On the Lite Adapter, Phi A and MClk go into U6F, so it may be damaged.

(There is the possibility that the connection was made with a connector backwards / 180 degrees around, so the notes above are assuming the connection was made in the normal orientation, not backwards. If the ESProFile header is polarized, that might ensure this.)

And there is the possibility that a high current opened a trace somewhere (like blowing a fuse); I think this is unlikely for this situation, but you could look around for a scorched trace and check the continuity of any suspects.

Starting with those possibilities, you could replace them all or if you have some test equipment, do some troubleshooting (eg. watch each signal with a voltmeter or oscilloscope to see if there is any activity) to determine which ones are still ok.

Navigation

[0] Message Index

[#] Next page