Computer forgets how to read disk?

[Jim Haug]

Hi all. I have finally gotten my ancient S100 system running. It is a North Star Horizon frame with a Compupro CPU-Z Z80 board, a Compupro RAM17 64k board, and a Morrow DJDMA disk controller running two 5.25" Panasonic JU455 drives.

After booting is seems to read the drives okay, run some programs, etc:

https://drive.google.com/file/d/1NTzQx4Gk3YuxUE202T-LJYSiw2z-mx-l/view?usp=sharing

Eventually, it does stuff like this, and I have to reboot:

https://drive.google.com/file/d/10EIYnkQEbtyQCmzcMbfuadAzq67-iWxi/view?usp=sharing

Has anybody seen something similar?

Jim

[Jim Haug]

BTW I replaced the cable and swapped out some disk drives, all with similar results

[Randy McLaughlin]

I would start by cleaning and properly lubricating the drives.

Check all caps, look for leakage on any electrolytic caps.

Look for cold solder joints everywhere.

Look for tarnished IC's.


Randy

[Douglas Miller]

That looks like something may have trashed the DPB in the BIOS (i.e. corrupted memory), such that the BDOS no longer has the right track number for the start of the directory.

[Jim Haug]

Good point. I have four of the Panasonic drives. They’re all in good working order. I previously used them in another system with no difficulties

[Jim Haug]

That’s a good suggestion. I’ll have to track down which chip that’s likely to be on and check it out

[ldkr...@gmail.com]

Jim,
Don't forget that anytime you are at CP/M's prompt, and you change out a floppy in any drive,
to do a CNTL C to read in the directory information. Sometimes a CP/M program will request
a different floppy be inserted, and that's the ONLY time you do not do a CNTL C to refresh
the directory information. It's OK to insert any floppy in any drive as long as it's the same
version of CP/M and the DPB information is the same. Just do a CNTL C and you are good
to do a directory listing or execute some program that access's any floppy.

Of course you can also do a RESET with the Computer's RESET Button vs a CNTL C.

Larry

[Douglas Miller]

*IF* the problem is a corrupted DPB, there may not be a chip at fault. You might try to see if you can find any pattern to when this occurs, such as running a particular program or command. It is possible that a corrupted program on disk could trash the DPB, and once it is trashed it will require a cold boot (RESET or power cycle) to recover. If the corruption is due to bad hardware, it may take some time to track down. Perhaps something to check is the S100 bus connectors, at least those used to have a bad reputation back in the day.

[ldkr...@gmail.com]

Jim,
Michael Minn has an interesting site at:
https://michaelminn.com/linux/mmcpm/

I've used his software and made a few modifications which were tossed by him.

If you are interested I have "mmcpm417.c"

MMCPM has three modes:

With only an image file name parameter, MMCPM lists the directory contents with file
sizes and a list of file blocks:

./mmcpm disk01.img

With the "copy" option, MMCPM copies the contents of the image file into separate
files in the current directory:

./mmcpm copy disk01.img

With the "examine" option, MMCPM performs any specified complimenting or de-interleaving
and prints a hex dump of the image to stdout. This can be redirected to a text file or
piped to "less" for additional examination. This option can be helpful when trying to
figure out the format of unfamiliar disks:

./mmcpm examine disk01.img
./mmcpm examine disk01.img | less

Larry

[Randy McLaughlin]

Your symptoms make me believe the CPU, RAM, and console I/O may be OK.

That said it is still an older system being brought back up after sitting.

You need to go back to basics:

Power first - voltages OK, filter caps OK?

Drives clean and properly lubricated.

Cards pulled and inspected. Inspections includes checking caps, check for cold solder joinds, darkened parts, tarnished IC's, etc. All IC's pushed into sockets. card edge connectors clean? Bus sockets look ok?

With your symptons I would check for thermal problems on disk controller (add heat/cold to chips).

Can you come up into DDT and write simple test code while in it.

Does it fail quickly or is there a delay after it warms up?


Randy

[Jim Haug]

Back to the drawing board. I decided to check my cpu board and realized that it was failing to jump to an onboard EPROM chip on it. A few weeks ago it was working okay, or mostly okay, and now it's not. I'll try to track down what's going on with that. It might be the explanation as to why the system is having trouble reading disks