The Permanent Form

8085 Circuit image showing Memory IC

Now Arriving: Solder.

The goal all along has been to build something I can haul around and use. A circuit on a logic board's not gonna do it.

Time to Bend Metal

The PCB

The PCB for the permanent version is going to be one or two Vector Plugboards with integral power buses. Based on the number of tie points I'm using already on the logic lab, I'm expecting to need a linear stretch of about 15-20 inches space on the board for the chips in the plan:

IC         length (pins)
XTAL Osc   7
7400       7
8085       20
74373      10
2x74138    16
74C923     10
4x74HCT574 40
Some extra 16            
Total      126

That's already over 13 inches when giving room to space the chips out a bit, and give room for extras like resistors and all (caps will go on the bottom of the board, I expect.)

One short board (Vector 3677-2) gives about 12 inches of space. If I was willing to give up some of the I/O chips (the 74HCT574s) I could probably cram the whole thing onto one of these. But I want some I/O. For LEDs and DIP switches if nothing else. I gotta have something to play with in my programs, right?

A long board (Vector 3677) is roomier, about 20 inches of space. But it's 9.5" long overall, a bit awkward for a hand-held. However, this is probably the way I'm going to go. I don't want to deal with the problems of splitting unbuffered signals off the board, or with adding buffers for the bus for a multi-board design. The old low part count principle again. If I have two boards, that adds buffers, extra power conditioning, and that much more to go wrong or get damaged while flopping around at the ends of interconnects while I'm trouble shooting it. One board it is.

The keypad and display will be off-board. Their signals are already buffered by the I/O ports anyway. The same will probably go for any other I/O devices. I will probably be able to fit headers into the stray space at the ends of the rows.

The Enclosure

It's hard to find a decent hand-held enclosure for a board that's so long. If I cut off the plug end of the board (long enough to still be used in other projects) I can reduce the length enough to stick it in an old modem's case. I'm also looking at building up a case.

One idea I've had is to split a section of dowel into two half-cylinders. I'd put a pair of saw kerfs in the flat side of each, one to hold the PCB, the other to hold the board for the keypad and display. I'd put some flats on before splitting the dowel to give me a place to mount a face panel and back panel, then cut a couple of endcaps and screw them into the dowel ends. Another idea is to knock together a form and slump some acrylic into it to make a housing. I'd put in relief cuts for the end covers ahead of time, so that they'd already be there once it's been slumped into shape.

Close, Sooo Close

When I was looking for possible enclosures online, I came across Box Enclosures. They've got some really nice hand-held cases. None that quite meet my needs, unfortunately, because the prices are quite reasonable, and the cases are very nice for being generic-type enclosures. I'll keep these in mind for future projects. The colored grips and included battery compartments are really nice. The faces would be really easy to customize with a printout. If I do a PCB set (a core board and a UI board) for the MAG-85, I may do it up to fit one of these. A real PCB would take less space than a protoyping board, so I could almost certainly fit it.


Update: 11 Nov 2009

I've started construction of the permanent hardware now. I'm using the Vector 3677 board, as dsecribed above. So far, I've built and tested the power supply, and added the clock/oscillator circuit. After having to do some rework on the power supply on the PCB, I regret not prototyping every part of the MAG-85 circuit on the solderless breadboard before beginning construction. A simple problem I ran into caused by a databook with a typo could have been avoided.

I had been planning on adding the full RESET and single-step circuits to the permanent version without bothering to prototype them on solderless breadboards first. I've decided that I should do them on solderless boards first.

Update: 24 Nov 2009

testing the circuit prior to a free run test

I have successfully added the CPU to the PCB and tested it with a free-run test using components on a solderless breadboard. See my blog entry on this here:

8085 Permanent Hardware Free Run Test

Next will be adding the address latch and memory. I'll also be breadboarding the single-step circuit on the prototype.

<< Back to MAG-85 Home