Stay-at-home SC126 outcomes...

[Jim McGinnis]

Some time ago I started to imagine a system that utilizes the SC126 (Thanks Steve!) and also satisfies some "packaging ideas" of interest to me.

I went about creating "mezzanine cards" for the SC126 system. The cards plug into the first and second 80 pin headers as well as various connector headers on the SC126.

An early prototype looks like the following when all three cards are assembled:

Card 2 links to the SC126 Serial Ports (A & B) as well as both SD Card SPI interfaces. The card creates locations for the FTDI and SD modules to lay flat INSIDE the assembly. I have added SIO and CTC support to the second card. The third and fourth serial ports are brought out as a DE-9 and another "lay-flat" FTDI like adapter.  The Card will accept SIO/2 or SIO/0 bonded 40 pin chips (3 x 2-pin jumpers must be set), MAX232 converts COM2 to RS232.

Card 3 has a SAMTEC mezz card connector to continue the stack above Card 2.  Card 3 contains a WD37C65A floppy controller a-la Dr Scott's implementation (and others) as well as an 82C55 PPIDE interface.

The white structural items are PETG 3d printed supports to keep the stack in place.

The entire assembly is installed into a 3D printed case...

This is the version with the FDD included. Another version of the top exists that only contains the SSD drive option and it is about 26mm shorter.

Up front is the logo ("SC126 STACKED") along with the power switch, a 10 segment LED Bar Graph where the power (top segment) and the disk drive status leds (bottom 8) and one unused below the power led are displayed.  The uSD card access slot is up front as well.

The rear appears as follows:

The rear has a reset button brought out, COM0 through COM3 as well as the second SPI SD Card connector. I didn't break out the I2C interface. DE-9 is in the way right now. It could be "extended" on card2 to the front face.

The left side breaks out the expansion connector as shown below:

The expansion breakout will allow experiments to be conducted using the 80 bin expansion cards as well as the 80 to 50 pin breakouts.

For scale, the right side of the enclosure shows the scale of the enclosure when the FDD is installed:

If anyone is interested in the 3D printer files I can email them to you.

The original 3D design was performed using Fusion360 and sliced using Simplify3D. The parts were printed using Prusa Mk 2.5 (modified frame) at 0.15mm layer height using PETG and PLA fibers.

The 2nd and 3rd card can be found on EasyEDA by searching for "trawlergeek." (a person who admires traveling trawler boats like Nordhaven)

The cards are well supported by RomWBW 3.1pre5 and later.   So many thanks to Wayne for his generous and patient support!

The 3D print time for these designs total over 30 hours. The longest print is about 13 hours. Yes, long prints!

Hope you are all well in these difficult times.

Cheers

Jim

[Alex French]

That looks fantastic. I love seeing complete systems like this.

What software are you running on it?

[Jim McGinnis]

Thank you Alex.

I am running RomWBW + CPM3.

I have been developing C and ASM programs prior to getting this box running. I have an Oki ML321 Turbo dot matix printer on COM2.

I am using the HITECH C compiler because I really can't get accustomed (again) to K&R C without prototypes. I am using ASM, MAC, and M80 to compile the assembler stuff.

Trying to get SYSLIB and related large libraries in place for more ASM/MAC work. I use VEDIT as a full screen editor. It can be easily configured to respond to all the function keys and arrows. My muscle memory for vi-like editors is long gone. LOL.

I have a 24" LCD with a Raspberry Pi 3B+ attached to the rear via the VISA bolt holes. I 3D printed the pi case with the VISA mounting holes.

I run C-Kermit on the pi in Kermit Sever mode - it is very convenient for moving files to and from the Z180/Z80 systems.

I run a customized version of minicom on the pi as a VT102 terminal server. It is modified to default the character transmit delay to 15ms so that it doesn't overrun the other system I have - RC2014 12 SLOT setup.  In that hardware environment, multi-key combinations (function keys, arrows, etc) will overrun the com port buffer with the back-to-back characters generated by the function keys, etc.

I am hoping to use the 2x40 pin expansion connector for some custom hardware experiments and interfacing to displays, LEDS, EF displays, etc.

So, the SC126 is mostly destined to be my "development" box. I have my sights on Modula-2, Pascal, and possibly PL/I.

How about you?

What are your interests?

Cheers

Jim

[Richard Deane]

Looks a really great system. I have an SC126 but working on SC131 currently as it is a tidy package with case.

I found when tinkering with languages over the last few years that PL/I is great under CP/M with good CP/M access, good file handling and good type conversion. I found that Modula 2 was too limited in it's ability (i.e. won't) to convert between text and string. I never learnt 'c' properly and my first major work after college was on PL/I for IBM, but had to switch to Pascal on CP/M becasue it was before DRI released PL/I-80. I have been tinkering with a hexdump program under PL/I having ported it from and between Pascals, attempted and failed under Modula2, and now settled on PL/I - happy to provide as a learning sample if you wish, as it uses some poorly documented features e.g. ON ERROR handling. Would need to add a few more comments, and finish debugging some current feature changes. I have a simple pi.c program working under various c compilers and woud like to port to PL/I to compare performance in fixed and float  numbers with c.

Richard

[Jim McGinnis]

Richard,

Thanks for the interesting experience data using PL/I. Yes, I am interested in your example source for the hex dump program. I am very much starting over with PL/I since it has been at least 35 years since I last used it. I have been working in a limited fashion with PL/M at work (legacy avionics software) and was always remembering the advent of PL/I on DEC minis.  I missed it. Since I am now retired from Collins Aerospace, I have time to pick and choose my focus.

I think that you are correct. Modula (even the second iteration) may be too strongly typed for convenient use. I will tread lightly when I peek at the available compilers and libraries.

I have been involved with "c" development for decades - it is the natural go-to language for me. And I did a lot of assembly language work many years ago - PDP/11, 8080, 6502, HC11, Z8, IBM 360/370 SPASM, and others. Still do it for modern work when that level of abstraction (or lack thereof) or raw performance is required - mostly AD and TI DSP parts for radar signal processing.

The SC131 is really a neat package. I was excited when it came out. I have a small RC2014 5 Slot project and Sergey's EZZ80 board to package, yet. The SC131 will likely make it into the "stable of boards and systems" soon.

Thanks for the thoughtful reply.

Very best regards,

Jim

[Jim McGinnis]

The "SC126 Stacked" 3D design is now on Thingiverse...

https://www.thingiverse.com/thing:4333850

Cheers

Jim