Close

Side quest and redesign progress

A project log for Uno 14500

An MC14500B based single board computer with an old-school front panel for interactive programming

jnp-3rJNP-3R 05/08/2024 at 07:060 Comments

Sorry it's been forever since the last update, inspiration comes in spurts and all, but having gotten myself stuck I did what anyone with sense would do, start another project!

Well there's actually been a few, but the one that's relevant to this project is this little guy:

Heathkit digital trainer console with logic switches, lights, clock generator and breadboard
She looks nice, but everything's a little wobbly. Let's see if we can fix that.

A Heathkit ETW-3700 Digital Logic Trainer. I hate dealing with sprawling wires from power supplies and peripherals, so this fella's a good option to prototype with these older chips without the mess. Unfortunately, it needs work.

The PSU and clock generator work, but the breadboard and terminal strips have seen better days, and logic switch B is super wobbly. Let's take a look:

Underside of the PCB from the digital trainer. Large solder blobs and lifted pads are visible
Oh, ohh.... We're not the first ones in here.


Looks like someone broke the switch off and figured a bigger glob would fix the ripped pads. Thankfully some epoxy to hold the switch and the proper amount of solder has us going.

The terminal strips were also pretty wallowed out and loose, but we found some replacements from Electronix Express that'll work well enough.

With those fixes she's all good to go, and running an ICU on the breadboard:

Heathkit ETW-3700 Trainer all fixed up with an MC14500B on the breadboard
She works! Just loaded a one from data (L4) to RR (L2)!

Now that we can play with some testing we've realized our clock circuitry may not be correct. The datasheet for the 14500B only shows X2 (the internal clock out) and shows it going *low* during the execution cycle, but the unit on the breadboard responds to X1 going *high*. Other builds we see (that don't use the internal clock also use a clock that's high during the instruction, so... We may need to do some redesign.

We like Nicola Cimmino's clock and program counter design for the PLC14500 (and already used his LED suggestion), but will have to make some changes for the parallel preloadable counters we're using. Maybe a flip-flop to delay the clock and implement the 74HC590's "output is one phase behind clock" feature? More tinkering is required!

Discussions