Introduction: DEC H-500 Reckoner Research lab Reproduction

Umpteen people reading this will be acquainted with the Digital Equipment Pot (DEC) lines of PDP machines. I would guess though that far fewer have encountered the H-500 Computer Lab. Designed past John Hughes at DEC Canada, and launched in the late 60's, the H-500 was part of a Computing machine LAB curriculum to introduce students and engineers to digital electronics. It's non unexpected that DEC would guarantee this since more than half of it's PDP machines at the time were installed in educational institutions.

The machine itself shipped with a extraordinary Computer Lab Workbook, written by the designer, that contained a staring course in digital electronics. Put together the COMPUTER Science lab package was intended to accompany courses in binary pure mathematics, Mathematician algebra, digital logical system or computer technology. While not a true computer, the H-500 could be "bugged" to perform many of the underlying operations of a true computer using a point-to-point patch cord mechanism.

Motive

So why would I require to undertake an H-500 Electronic computer Laboratory replica? Because most of the projects that I consume been working on lately follow the same model:

  • They are reproductions of cool computer toys and devices from the 60's.
  • Have tremendous instructive value (when first discharged and even now IMHO). Thereto end they usually shipped with well backhand and instructive manuals.
  • They feature film unique and noteworthy designs.
  • Due to their age they have become rare and thus expensive and hard to get.
  • Perhaps most significantly I reckon that the devices themselves and their designers deserve to exist remembered and honored.
  • As a bonus for this particular project, I'm Canadian and the H-500s were manufactured in Canada.

And so the H-500 is a very good fit for my interests and skill set. Plus I really need one ;-)

Design Goals

With this reproduction I Leslie Townes Hope to provide the best H-500 Computer Lab experience possible. What answer I mean by that? Good...

  • Visually my reproduction will be clearly recognizable as an H-500. I bequeath strive to provide an outwardly precise appearing of the original to the best of my power and within the constraints of the implementing technologies. Take over in nou that I can't afford one of these so I will be functioning from images and the kindness of actual owners to provide me with details.
  • Operationally my H-500 volition perform like the original. A exploiter will beryllium able to work through the experiments in the Lab Workbook without change or issues.
  • I would like for this stick out to be elementary to well, reproduce. To this end I will only be using 3D printed parts and readily available components and materials.
  • I will try to keep the total project cost to a minimum.
  • A complete set of operating instructions, parts list, and STL files leave be posted to (this) Instructable at the completion of this project so that others could make one should they choose to.

So hopefully my H-500 reproductive memory bequeath look and work like an original. Lets amaze started.

Supplies

In addition to the printed parts you will pauperism:

  • 1 - Arduino Mega 2560 - Used to go through the clock and switch funtionality.
  • 11 -NO Magnetic Reed Switches - Digi-Cay part number 2010-1087-Neodymium.
  • 1 -Panel Mount Power Jack - Center Incontrovertible 5.5 mm x 2.1 mm.
  • 20 - M3 x 12 Bolts - For the trade meeting place.
  • 6 - M3 Nuts - For the switch assembly.
  • 24 - 2-flag Male Headers - Amazon - Gikfun 40 Pin Strip Breakable Pin Header Male 2.54 mm for Aduino.
  • 48 - 1-pin Female Headers - Amazon - PTSolns 40 Pin 2.54 mm Single Row Female Cope Atomic number 79 Plated.
  • 350 - Rolled Flange Point Eyelets - Spaenaur part enumerate E-412.
  • 1 - 1/8" Brand Rod- About 180 mm.
  • 8 - 3 mm White LEDs - Amazon - Gikfun LED Lamps 3 mm White Light First-rate Bright for Arduino.
  • 1 - Potentiometer with On/Off Change - Amazon - 10K Ohm Single Linear Taper Dimmer Potentiometer with On/Hit.
  • 150+ feet - 22 AWG Hookup Telegraph With Silicon Insulation - Amazon - 22 Approximate Electric Telegraph,Tinned Copper color Wire Kit 22 AWG Flexible Silicone Wire(6 Different Colored 26 Feet spools) 600V Electronic Cabbage raised Wire
  • 100+ feet - 22 AWG Good Core Hooknp Wire With Plastic Insulation - Normal hookup conducting wire.
  • 8 - 3.9K Resistors - 1/4 Watt
  • 1 - Power Provision - Amazon - Haitronic 9V 2A Ac/DC Power Adapter palisade charger for Arduino.
  • 6 feet - 1/2" x 3 1/4" board - For the frame.

Step 1: Print the Parts

I printed the parts with the undermentioned settings (unless early wise specified):

Publish Resolution: .2 millimeter

Infill: 20%

Filament:AMZ3D PLA

Notes: No supports. Print the parts in their nonremittal predilection.

To make a DEC H-500 Computer Lab publish the tailing parts:

  • 8 - Frame Construction Corner Brace (optional) - For aligning fame pieces while gluing.
  • 1 - Frame Profundity Jigs (optional) - Used to measure the depths to install the frame supports.
  • 1 - Frame up Mega 2560 Bearer - Mount the Arduino Mega to the frame.
  • 1 - Human body Power Chew Bracket - Mount the power jack to the frame.
  • 1 - Frame Serial Enumerate - Serial number plaque.
  • 1 - Frame Support Igniter Panel - Frame mounted supports for the Light Panel.
  • 1 - Bod Back Speckle Panel - Set up affixed supports for the Patch Impanel. Print A 4 separate parts.
  • 1 - Inning Support Switch Panel - Frame mounted supports for the Switch Panel. Print As Little Jo separate parts.
  • 1 - Light Jury Brace - Wont to join the Light Panel Left and Right.
  • 1 - Fatless Empanel Fine Tuning Knob - Attach to the potentiometer.
  • 1 - Palish Panel Front - The downward front objet d'art for the Light Panel.
  • 8 - Light Panel LED Fake Bulb - Diffuser.
  • 8 - Light Panel LED Holder With Threads - The rest of the bulb assembly.
  • 1 - Light Panel Left - Left persona of the Light Panel.
  • 1 - Lighter-than-air Jury Properly - Ripe part of the Light Panel.
  • 1 - Fat-free Panel (optional) - Use this rather of the Left and Right parts if your mark screw is big enough.
  • 2 - Patch Panel Bound Support - Accustomed reinforce the Patch Control board.
  • 1 - Patch Control board Four Pieces - Main Piece Dialog box that pot be printed as four pieces.
  • 1 - Patch Panel Independent Support - Used to reinforce the Plot of ground Panel.
  • 12 - Patch Panel PCB Holder - Supports the PCBs in an upright stead.
  • 170 - Patch Panel Electrify Holder - Attach the wires from the PCBs to a pair of rivets.
  • 1 - Patch Panel (optional) - Apply this instead of the Patch Panel Little Jo Pieces if your print bed is thumping enough.
  • 1 - Switch Assembly Base Clip Clip Holder - Holds two Home Clip Holders put together,
  • 2 - Switch Fabrication Base Clip Holder - Prevents the Base Clips from popping out.
  • 1 - Switch Assembly Base Clip - Holds Switch Toggle Bases to the Eight and Three bases.
  • 1 - Switch Assembly Eighter Base - Holds the ON-On Toggle Switches.
  • 3 - Switch Assembly Mounting Block - Victimized to attach the Change Bases to the Change Panel.
  • 11 - Swap Assembly Spacer - Gaskets out between Switches. Ideally 3 white and 8 brown.
  • 1 - Switching Assembly Three Base - Holds the momentary Toggle Switches.
  • 1 - Switch Control panel Digital Logo - The digital logo to go with the front of the Switch Panel.
  • 1 - Switching Control board Joiner - Reinforcement to articulation the Left and Right Switch pieces.
  • 1 - Switch Panel Left - Left over part of the Switch Panel.
  • 1 - witch Instrument panel Right - Reactionary part of the Switch Panel.
  • 1 - Switch Empanel - (optional) Use this or else of the Left and Right part if your print bed is heavy enough.
  • 11 - Exchange Toggle Station Conclusion Chapiter - Closes off the Change over Base retention in the magnets and reed switches.
  • 11 - Alternate On/off switch Base - Holds the magnets and reed switches for the Toggles.
  • 11 - Switch Toggle switch Bottom - Bottom persona of the Toggle switch Switch. Make 3 white and 8 brown.
  • 3 - Switch Toggle switch Attraction Bearer Momentary - Counter weights for the momentary pulse switches.
  • 8 - Shift Toggle Magnet Holder Toggle - Rejoinder weights for the ON-ON rocker switches.
  • 1 Switch Toggle (optional) - Usage this rather of the Top and Bottom pieces if you don't mind cleaning ahead supports. Make 3 white and 8 Brown.
  • 11 - Switch Toggle Top - Top split of the Toggle Switch. Make 3 white and 8 brown.

Stride 2: Urinate the Trade Panel

Overview

The Substitution Panel, the bottom most split up of the H-500, holds three momentary "spring" render typecast pulse switches (white), and 8 Happening-ON variant input switches (brown).

Making a Switch

My protrusive point was an STL version of the PDP 8/I rocker switch that I found at Vince Slyngstad's PDP-8 Overeat page. I modified it slightly to throw a hole at the pivot compass point instead of a shaft.

Then I sculptural a "replication exercising weight" to hold the switch magnets and a commensurate base for the "reciprocating" magnets and the reed switches. You will point out that there are two kinds of counter weights. That's because there are two kinds of cradle switches obligatory, a momentary "spring" render for the three H-500 pulsing switches, and an ON-ON variant for the eight input switches.

The base has slots and telegram guides for each of two magnetic Reed switches. The wires are soldered to the tips of the through hole pins.

Assembly of each switch is pretty untwisted forward. Insert the reed switches and magnets (making sure that the magnets in the rocker and the base are oriented to attract) then snap on the base remnant collection plate.

Note that the momentary contact and the ON-ON switches for this externalise only need one charismatic reed switch.

Putting Jointly the Interchange Assembly

Prepare the eleven rocker switches required, tierce momentary "spring" return types for the H-500 pulse switches, and eight ON-Along variants for the input switches.

I recovered some "buoyant John Brown" AMZ3D filum that I though might glucinium a good fit for the 8 input switches. Before printing the switches I split the STL file into 2 parts to avoid having to function supports.

I'd rather spend a some seconds gluing, than 10 minutes removing support material. You finish up with a small seam (where it volition never been seen) but with a much cleaner print.

The above pictures appear way too lily-livered and come not behave justice to the actual color which has a nice caramel/butterscotch look for to it. The picture beneath is a better indicant. It's certainly in the approximate range based along pictures of H-500s that I have seen online.

I took care and ready-made indisputable that the magnets were all oriented the same way and used my 3D pen to secure them in place.

Next I made the eleven rocker switch bases. They are all the same as the one below. You will notice that there is only a one-man reed switch. I had already decided that I would use an Arduino to grapple the clock and pulse signals (reading the clock order, generating the clock signals, debouncing the pulse switches, etc., more on this later in the project) indeed I opinion I would choose vantage and generate the input switch signals based along a single reed per switch knock about $20 off the BOM.

I designed two "cradles" to hold the switch bases.

The bases slide in snugly.

I ready-made some simple clips to hold the base pieces securely in place.

Same for the octonary permutation cradle. I'm running a bit Low on filament so mixing colors a tur.

Next I printed some mounting plates. They have slots for M3 nuts to ensure a reinforced bond.

Connected the cradles to the climb plates with M3 x 10 bolts.

I cut a 1/8 edge in steel rod to span the length of the combined switch meeting place and written whatsoever small spacers to help align the switches. (NOTE: In the final version of the assembly I moved the wires on the 8 toggle switches to the other pull to have installation a little cleaner.)

Starting from unmatchable end I began slowly sliding the gat through the pivot holes attaching the rocker switches with their counter weights as I went. Between switches I inserted one of the gaskets.

When done the switch assembly should look like this.

I printed the face plate for the panel (in two pieces, need a larger loudness printer). I paused and changed the filament coloring from black to light to print the text. The "digital" logo was printed separately (with appropriate color changes from white to downcast past back to white) so glued to the face plate. All that's left-wing is to attach the switch assemblage to the face collection plate. (Or is it the other way around, the switch assembly is quite a heavy.)

That's information technology. The switch panel is cooked. It is tweaked a trifle from the original. I had to make the characters bigger past about 20% to nonplus them "publish" right when extruded. The switches when collective over up being a itty-bitty wider than the originals. Since I had already written the switches, I ready-made the control panel holes a puny larger to lodge in them. When I get the chance I'll get some black steel bolts and so the heads are less strong. All-in-all though I'm highly halcyon with the upshot.

Step 3: Make the Light Panel

Overview

This piece separates the Switch Panel from the Bandage Panel, holds the eight output lamps and the fine tuning knob for the time focal ratio, and also transitions from the higher Switch level to the lower Patch level within the wooden frame.

Fashioning a Lamp

It's impossible to know from just pictures what kinda bulbs they used in the original, but if I had to guess based on the estimated size (about 4 millimeter) I would say grain of wheat bulbs. I could have ordered some but given the incumbent lead times for delivery of not-essential items I decided to cooccur with what I had on hand, LEDs. I modified my Panel Affixed LED Socket by adding a 4 mm printed diffusor printed with transparent filament which ends looking unintegrated with a slight blue tinge.

If you are superficial for a more bona fide candent look you could print the diffuser with icteric or light brown filament. As an try out I tried the brown filament that I used for the rocker switches and it looked rattling good. You might stimulate to reduce the recommended limiting resistance to get the luminescence correct. If I hadn't already installed and wired the Light Panel I would have switched to the brown diffuser.

When lay out collectively IT looks like this.

Assemble the Light Panel

Make eight lamps. These bulb assemblies were slotted into holes on the back of the Light Panel. The panel itself was written in two pieces and glued together with the assistance of a reinforcing undress. There is also a front piece that is affixed on once the two pieces give birth cured. You can see the clock fine adjustment pot mounted on the bet on as well.

From the front here is the fine tuning thickening that I shapely...

... based on the original.

The "bulbs" finished dormy looking pretty adept. I found that I bequeath had to tote up a 4K limiting resister to each LED to stupefy the brightness down to candent levels.

The Light Panel is ready to go under.

Step 4: Make the Patch Panel

Overview

The Maculatio Control board is where the user creates logic circuits by connecting the "components" displayed on the opencut with wire mend cables of various lengths. Patch cables are terminated with 2 mm banana tree plugs that connect to the Patch Panel's components via cheek rivets.

Assemble the Patch Panel

The Patch Control panel was printed in multiple parts to accommodate my print bed, and the text edition is somewhat bigger than original to work at 3D printing machine resolutions, but a pretty good resultant I conceive.

My big "discovery" for these panels was the "David Hilbert Curve" option for the top layer.

I was unaware of this PrusaSlicer feature until now, and would sure have put-upon information technology in some of my other projects had I known about information technology (Minivac 601 and Digi-Comp II for sure). It produces a beautiful even matte finish unlike the slightly "streaky" result I see with the default Rectilinear option. The downside is that Hilbert is quite bit slower. It added an extra forty minutes or so to each of the Little Jo panels I printed. All worth it in my books.

I glued the four panel pieces put together. Since the Patch Panel itself needs to be bad rigid to support the insertion and removal of the patch cables, I configured and printed some support beams and glued them to the underside.

Then it's just a simple matter of installing the rivets onto this panel. Aside my count, there are 335 "terminals" that patch cords can be plugged into. These rivets are Rolled Flange Head up Eyelets sourced from a local supplier.

I had finished a number of test to try to father the perfect tense hole sizing so that the rivets could be just pushed in with a nice friction fit. Unfortunately halfway direct printing the panels I successful a small adjustment to my first layer tallness. So in the end the two right panels worked as expected, but the ones on the left were a little loose fitting. Not to interest this will comprise fixed in the adjacent abuse.

Step 5: Wire the Patch Panel Logical system Elements

Overview

Circuits represented by logic symbols on the top of the Patch Panel are enforced with a 12 7400 series ICs underneath. Away look the logic symbols and using pictures of the wiring from the original H-500 I was able to determine the chips used:

  • SN7400 - 2 Quadruple 2-Input Formal-NAND William Henry Gates
  • SN7410 - 2 Multiple 3-Input Positive-NAND Gates
  • SN7420 - 2 Dual 4-Input Confirming-NAND Gates
  • SN7450 - 2 Dual AND-OR-Turn back Gates
  • SN7473 - 4 Dual J-K Flip-Flops with Realize

As can be seen above, for the original H-500 wiring was accomplished with a large PCB. The ICs were "come out mounted" directly to the PCB (pretty cool actually) and the rivets soldered directly to the board as well. So much a bombastic PCB would be prohibitively valuable so I went in a different direction.

My Implementation

Part way through the wiring this is what my Patch Panel looked like:

There's a a couple of things to note here. First of all I decided to make my life a pocketable easier by creating some breakout boards for the 74XX chips. My first though was to just "dead-bug" the chips and solder the leads directly to the pins. It would have worked but the solicitation of atomic number 75-arranging the pin-outs to be more consistent with the layout of the panel, plus having a label on each pin won out. Hither's the PCB (you will need 4 of these):

written as one piece to save money and veer into individual pieces with a Dremel. I created a small base for the PCBs so that I could stand them upright since there was no way to set back them out.

My plan was to solder the wires coming from the wear-out boards straight to the plaque rivets. It turned out that this was a lot harder than I thought information technology would be. I really should experience tested this much in the first place in the cognitive process. I assume't know if the governance was coated with something but the bonding iron had to constitute practical to the rivets direction too long to get a good join. As a result the panel holes melted and got too large to keep out the rivet firmly in place. I needed a plan B.

They say the necessity is the mother of invention, and I think the root I came up with is actually better than if the bonding had worked. I successful whatsoever mechanical "fittings" to securely attach the wires to the rivets:

Barely introduce the cable into the level hole and skid the fitting over a pair of rivets representing a circuit lead. These work nifty, are quick and pleasing to install, and in reality reinforce the rivet's attachment to the panel. An additive benefit is that a whole PCB "circuit" can be removed from the panel should the need arise to troubleshoot. Here's a closer aspect.

So with the bugs ironed unsuccessful, I finished the Dapple Panel wiring.

And here you go. Entirely the logic elements are now wired exclude for power which I'll running game when I put all of the pieces together. Future I'll belt together a frame to mount everything in.

Footprint 6: Figure the Frame

If I had accession to my localised maker-space (kwartzlab) I probably would suffer ripped something down pat to the required 1/2" x 3 1/4" board size of it that I needed, but I was "lucky" enough to find close to "craft board" at Lowes in the needed size.

The other maker-space thing that I uncomprehensible was access to their fine selection of corner clamps to put the frame together, thus I had to improvise. I concluded up printing some corner guides that worked quite well.

I mitered the corners of the frame and victimised the guides to align the boards while I glued and popped in few brads. The indoors dimensions of the frame are 292 millimeter (width) x 419 mm (height).

Once the glue was set I tarnished the boards then applied a few applications of clear coat. I was a little foiled in the stain as I was expecting it to be a bit ignitor and redder like the original's "faux teakwood". But donated how long information technology took to incur the stain I went ahead and used it.

I printed some supports to hold the Patch, Light, and Switch Panels in place at their appropriate depths (1/2" for the Switch Panel and 1 3/4" for the Patch Empanel) within the frame in and attached them to the wrong faces with 1/2" wood screws.

I affianced the panels onto the frame supports with some E6000 Gum-like.

Finally I printed a mount for the Arduino Mega...

...and installed it in the corner of the frame underneath the potentiometer.

Step 7: Conducting wire the Lower Incomplete

Overview

Entirely of the active controls for the H-500 are located in the lower part of the device. These included (taken directly from the Computer Lab Workbook):

  • Rocker Switches - Rocker switches tail be used to provide either a HI or a LO logical system level. If the top side of the rocker switch is depressed, the two corresponding switching output terminals (directly before of the switch) are taken to a HI level. If the lower side of the switch is concave, the two corresponding output terminals are taken to a LO level.
  • Pulser Switches - The outputs of the pulser switches are normally LO. When a pulser is indented, the related two final terminals attend a HI level. When the pulser is released its output terminals restoration to the LO consideration. Internal circuits related to to the pulsers make sure that when a pulser is depressed or released, electrical disturbance generated in the switch is non transmitted to the pulser production. This unscheduled circuitry makes the pulsers reusable in applications requiring racket-free transitions from one layer to another. Rocker switches do not have this feature.
  • Clock - The clock provides a endless train of How-do-you-do pulses. Clock pulses are 50 nanoseconds wide. The absolute frequency of time pulses give the sack be continuously varied from less than one pulse per back to over 10 million pulses per second. The slowest range of the time is obtained past connecting the common time uncouth period to the left-most speed-selecting closing. The repetition rate of the clock increases with each terminal to the honourable. The fastest repetition rate is obtained aside leaving the clock range selector disconnected. Repetition rates within to each one coarse range can cost varied using the time fine see to it. (Fully counterclockwise gives the slowest repetition rate; fully dextral provides the fastest rate.) The clock wander coarse terminals are to cost used only when for selection of time repetition rate. The clock production is obtained from the two terminals tagged CLOCK OUTPUT.
  • Lamp Indicators - The operation of experiments constructed on the COMPUTER LAB patchpanel is monitored aside the lamp indicators. A lamp will atomic number 4 ON if its corresponding input is at a HI logical system even out. A lamp will be OFF if its corresponding input is at a LO logic degree. If nary connection is provided to a lamp, the lamp will be OFF. Lamps will respond to sustained logic levels and pulses of sufficient duration to activate the lamp filament.

So how were these functions implemented back in the dead 60's. Like this:

With discrete components As depicted in this schematic! Do it to say that the complexity of implementing this circuit goes counter to two of my primary design goals with this project: take it easy to reproduce and non too expensive.

Luckily I do know what the electrical circuit is supposed to do. So what would be an easy and cheap permutation, understood by today's residential district of makers, well an Arduino of course. This is what I came ahead with:

  • Rocker and Pulser Switches - The magnetic reeds of the rocker and pulse switches are monitored by eleven inputs that are set with their pull-ups enabled. When the normally open John Reed switch is "activated" away a magnet its input line will equal pulled down. The Arduino sketch monitors changes to the reed switches and will only variety the state of it's corresponding output signal line once the switching has been suitably debounced (troika consecutive reads 20 milliseconds asunder with the same state). In this effectuation every last switches are debounced, non just the pulsers. We are already at 22 I/O lines and so it's a good matter that the Arduino I had lying around was a MEGA 2560.
  • Clock - Sextet more inputs with pull-ups enabled were sacred to monitoring the time range coarse jumpers. The clock frequency ranges that each enables fanny be seen on the plot above. Ranges are chequered from lowest to highest, fillet at the initiative one found to be enabled with a jumper. Also the potentiometer is read via an analog in line of products and converted to ten steps (1-10) and applied equally a multiplier to the lowest value in the selected graze. So if the time range go under is 10-100, as the potentiometer is turned clockwise, the clock frequencies will exist changed to 10, 20, 30, ..., 100 alfilaria per second gear. If nobelium coarse stray jump shot is set, I leave the clock absolute frequency fixed at unitary clock per endorse. Clock pulses are emitted via a whole number output transmission line.
  • Lamp Indicators - Enclosed in this diagram for completeness, the lamp inputs are non computerized by the Arduino in whatsoever way. Just apply a Hi signal and they will plow on.
  • Power - A switched potentiometer is used to turn the device on and dispatch as with the underivative.

Using an Arduino is non without some compromises. First off I'm using the superior TimerOne library to implement interrupt driven time pulses. Information technology tops out at about 1,000,000 interrupts per second, indeed that's my maximum frequency as well, 1,000,000 clocks per second. On a similar note the clock pulse itself is generated using the digitalWriteFast program library. Within the clock interrupt the cipher is simply: 

digitalWriteFast(2, HIGH);  digitakWriteFast(2, LOW);

According to the documentation the digitalWriteFast() "squall" takes all but 125 nanoseconds to execute. Thusly that means that the clock length will be somewhere between 125 and 250 nanoseconds. (I'll recognise more exactly when I can go back into my maker space (kwartzlab) and access a background.) A little more than the 50 nanosecond duration the fresh could produce, merely I think it will glucinium adequate for my purposes.

I know that I will probably get some feedback for using a "microprocessor" in this projection. If were trying for a more authentic replica I would totally agree. However in this case, where I and am shooting for more of an H-500 Computer Lab working facts of life, the Arduino allows me to simply replace a caboodle of discrete components with a cheaper, faster (at least for me), easier, better implied solution.

Make out the Wiring

The first matter that I noticed when I started the wiring was that the replacement assembly wires (purple) and the light control board LEDs (grey hexes) were overlapping, making my life harder than it had to be.

So I removed the switch assembly and rebuilt it, moving the eighter from Decatur brown rocker switch wires to the bottom of the assembly (the heart rate switches were OK). There is a nice channel along that incline to run the wires between the frame and the assembly itself. Following wiring diagram to a higher place information technology was pretty readable sailplaning and here is the result:

For reference the wire colors are:

  • empurple - switch inputs to Arduinogrey - switch outputs from Arduino
  • white - punch out and clock delay restraint from and to Arduino
  • yellow - pot input to Arduino
  • Marxist - power +, 9V to Arduino, 5V from Arduino
  • black - establish

The LEDs are wired directly to the lamp inputs with a 4K limiting resistor (large enough to keep the brightness polish to incandescent levels). While wiring the LEDs I noticed a pretty big goof on my part. The label for the lamp terminals should deliver study LAMP INPUTS. Doh!

Buckeye State well nothing to do about IT now. I take over updated the Patch Venire STL file but will live with my error along this build.

Step 8: Wire the Upper Half

Running power to the prison-breaking boards was comparatively straight forward. Each PCB has a 2-tholepin header for both VCC and GND.

Starting from the 5V and earth outputs on the Arduino, I made cables "to length" with single female header ends, and victimized these to "daisy chain" the power and prime from one PCB to the next until all twelve were connected. Occasionally a power or ground cable was tapped off of these to power the HIGH and GND terminals.

Wiring Dead!

Step 9: Make Any Spot Cables

I establish some relatively inexpensive 2 millimetre banana tree plugs at about 56 cents CDN each. The wire insulation is silicon based and very flexible. The cables are colored coded away length and are based on the avant-garde H-500's except blackamoor was substituted for brown and there are no orange cables. Here is what the original cables looked like (from hypertext transfer protocol://WWW.so-much-stuff.com/pdp8/computerlab/computerlab.php):

And Here are the plugs that were misused.

As I stated before the 2 millimetre banana plugs I'm exploitation work great with the rivets so I'm fine with them for now.

Starting with 50 cables.

Stone's throw 10: Give Credit Where Credit Is Collectable, Eh

For ME this is a selfsame important part for the chassis. On the game of each Computer Lab sold was a plaque with the model and serial number arsenic seen above. What's prodigious to me Eastern Samoa a proud Canadian is that so far as I can determine all H-500s were Ready-made In Canada. So I had to make a point that my reproduction had such a plaque.

I consume decided not to put a spinal column connected this project, thus I adorned the brass happening the inside of the frame.

Ready to a-ok!

Step 11: Testing

I fired functioning the H-500 with a 9V 2A power cater that I got from Amazon. When I saw that there were no sparks, weed, or flames I successfully tested each of the basic logical system circuits in work using the cradle switches and "lamps". Then I tried and true unmatchable of the "Experiments" from the book, a "Decimal to Binary Encoder".

Information technology worked peachy.

Finally I used every cardinal flip-flops to create a positional representation system counter. Here is a video of the result in action. It demonstrates the use of the time knob to mogul on the H-500 and ascertain the clock rate from active 1 to 10 clocks/second. A Pulsing switch is used at nonpareil point to reset the count.

Step 12: Final Thoughts

It feel like this build took a really long time, but in fact it has only been three months since I started this project. In that location were discouraging separate delivery delays attributable Covid-19, and not having access to my local Almighty space slowed me down a bit and forced ME to get creative, only whol things considered I'm OK with how it went. It was nice to have something engrossing to work on during this implemented "quiet time".

If you would like a slim more in-depth look at the "fashioning" of my DEC H-500 Computer Lab Reproduction sound out my web log connected Hackaday. There you will also find the Gerber files for the 7400 Serial publication PCBs that I used since I couldn't seem to add them to this Instructable.

Cost the First to Share

Recommendations