Loading
1944GPW

DEC PDP-11 keyring keyfob

by 1944GPW Sep 20, 2018
Download All Files

Thing Apps Enabled

Please Login to Comment

I live in Ottawa, Canada where DEC had a major presence for years.

There's an old Compaq in my home; it's a cabinet only, 1997 or so, the original motherboard long ago failed due to bad capacitors.

And it's clearly a re-branded DEC. For one thing, this one machine's case is heavier than about a dozen HPs. For another, there are DEC logos everywhere on its components. DEC Q/A, DEC part numbers, the real deal.

This machine will never be original or museum-grade. What it is, is a great and well-engineered and massively overbuilt case which has roots in the greatest computers during the genesis of Unix/Linux.

It currently houses my multimedia PC, an ASUS motherboard, runs Linux, and occupies a place of honour in my home.

As a tribute, I'd l like it to wear the digital logo that it was born to wear.

I'm pretty good with my 3D printer but I have no artistic skills. You're obviously a fellow DEC fan, if you can make a digital logo that would look good on a PDP-11 or a VT-100, you'd do a real service to those of us who have touched the same kinds of hardware as Dennis Ritchie used.

G'day Lawrence,
Thanks for your comment, and interesting story. I am working on repopping some more DEC items and whilst they're not perfect, they follow the 'six foot rule' (ie. they look great from six feet away) and should at least suffice until an original can be found :). In the works are a repro H960 masthead, the 11/05 console logo (I have already done this as a decal for the late version, and am doing the raised cast lettering for the early version).
I spent some time on doing the dec font (partial, not complete alphabet) in CAD, SVGA and DXF so ought to be able to do VT-100 badges and the like. I need to do the 11/03 bezel anyway, and its strip badge as my 11/03 doesn't have either. So, there's a good chance they'll eventuate but not sure exactly when!

Nice!

Take your time!

My interest in 3D printing originated in being able to reproduce knobs for antique radios and other long-discontinued replacement parts to keep our technological history alive.

I'm still learning how to make consistently high-quality prints, it seems to be like baking: everything must be perfect. But that's okay. I've made more than a few failed prints (haven't we all?) but I try to learn from each of them.

I love music but I'm not a musician - the best I can do is mix your sound and fix your guitar amp. My technical skill is all about supporting artists. I could never be a painter, but I would build you a damned fine canvas.

To that end, yeah, I really do need to get some CAD skills under my belt so I can make at least a few simple things myself.

But in the meantime, your art is respected, and I thank you for it.

Most of my failed prints result from lifting off the bed, especially ABS. It's a never quite consistent effort for me to achieve a good print, and that's why I think 3D printers are really not a foolproof consumer appliance just yet, even though they've been around a few years now.

I can only speak for myself but having a programming background I find OpenSCAD to really suit my modelling ideas. Essentially you are programming solid geometry. It's free and open source as well. If you don't already have one, get a digital caliper and take some measurements of say a radio knob and draw a rough 3-view on paper. You then "draw" by creating the primitives, adding and subtracting, intersecting. The great thing is, you aren't doing the CAD work by dragging and dropping shapes then stretching and prodding, instead you do it parametrically. ie.

facets=100;
$fn = facets;
KNOB_BASE_RAD=9; KNOB_TOP_RAD=8; KNOB_HEIGHT=19.4; cylinder(r1=KNOB_BASE_RAD, r2=KNOB_TOP_RAD, h=KNOB_HEIGHT);

Paste that into OpenSCAD directly then hit F5 and F6 and you will have a simple tapered knob, it's that easy!. You can then subtract the shaft hole, put in a side hole for the grubscrew, apply a knurl etc. There are good beginner totorials on YouTube and you're welcome to look at the scripts for my models I've uploaded here. Have fun!

Most of my failed prints result from lifting off the bed, especially ABS. It's a never quite consistent effort for me to
achieve a good print, and that's why I think 3D printers are really not a foolproof consumer appliance just yet, even
though they've been around a few years now.

Yeah, we're not quite there yet. Let's look at the VCR as an analogy. I think we're still in the days of the open-reel videotape of the 1960s. Someday, someone invents the videocassette to make it easier to use. Mass-adoption begins. A format war ensues, a victor emerges, and it lives supreme until it is disrupted by something better (like DVD).

We're going to see this happen very soon in the world of 3D printing. Very simple standards around machine and filament capability will be established. G-code will be the medium of choice for home printing in the general public, avoiding the complexity of slicing. You'll download your model (maybe pay, maybe open-source, maybe totally free) and you'll enter your filament type (maybe just by flashing your printer's bed camera with a barcode on the spool) and have a foolproof print. Oh, and the bed camera(s) won't just be providing a web video feed, the printer will also be optically measuring your print and adjusting parameters on-the-fly to make it as close to the model as possible, maybe even automatically repairing, replacing, and restarting failed prints with a bed scraper/clearer mechanism.

At this point, if I were running a dollar store (selling nothing but plastic objects from China), I'd be getting worried the way movie theater owners were scared in the 1970s as the first practical VCRs came out.

The supply chain for replacement automotive parts is about to be massively disrupted. Car dealerships are used to spending tens of thousands of dollars on tools like dynamometers and lifts and paint booths. What if you can sell them a machine that makes manufacturer-approved replacement parts for the stupid crap like the window crank that will take two weeks to arrive? How about ten years from now, "We're printing you a new aluminum cylinder head right now, so your car should be ready to go in about three hours." (Yeah, I know we'll be electric by then, but I was hoping for an analogy for the complexity of stuff that dealership-level metal printers might be able to do in ten years.)

Remember what a Hewlett-Packard LaserJet I laser printer looked like? I do. Back-breakingly heavy. 2400 baud serial connection to my computer, and cartridges for different fonts. The $99 Brother at Staples blows the LaserJet I out of the water in every single way. What will 3D printers look like in a few years?

I haven't perfected PLA printing, I'll do that before I move up to ABS. I intend to install my printer in a fireproof box (old bar fridge should be okay) and allow it to perform double-duty as my temperature-controlled enclosure. Honeywell programmable household thermostat, set for AC mode, turning on and off a small 12V exhaust fan at the top of the cabinet, should be all that is required to maintain a consistent 45C or whatever, just using cheap-to-free off-the-shelf stuff. Done right, it is quality.

Paste that into OpenSCAD directly then hit F5 and F6 and you will have a simple tapered knob, it's that easy!
You can then subtract the shaft hole, put in a side hole for the grubscrew, apply a knurl etc. There are good beginner
totorials on YouTube and you're welcome to look at the scripts for my models I've uploaded here. Have fun!

Thank you! I will try the knob, and yeah, I will absolutely keep this in mind as I get my feet wet in CAD.

I took Engineering in University, and I remember using IntelliCAD on a (then) super-fast Pentium II in my first year. After that, because I was in the Electrical stream, we moved on to SPICE models and stuff like that.

Back in Uni, it was hard for me to get good at the 3D CAD because I didn't really didn't think of 3D model space as explicitly as my (newbie) 3D printer and slicer experience is forcing me to. They're really teaching me to use the origin as a tactile object; I can touch the front left of my build plate and it gives the models on my computer screen a real point of reference in my mind.

I have looking into OpenSCAD, it's already installed on my machine (openSUSE 15.0 Leap) and I have played with it a bit. I love your snippet of code, because I know it will give me immediate visual feedback as I dick with the controls. Thank you.

Your suggestion of doing a basic sketch and measurements onto paper is a really good one. Okay, I will produce, to paper, a well-measured sketch, scaled as well on paper as I can, of a real physical machine part. Once I have that, I will attempt to model the paper blueprint in OpenSCAD, and when I'm happy with that, I'll 3D print it. I need a crash course in 3D modeling and making a real and practical object seems like a good motivation.