Thanks, this great info will be added to my options list when the time comes to decide on a design.
I do have a small 24" CNC machine here, so adding a dremel to drill might just work.
When I did my dic calcs, I used 32 DPI as a max, but with 80 DPI as you have shown, it will work.

Brad

Looking again at the Bezel puzzle, I may even try to create an inverse plug from wood and then vac form my own.
I need two of these...



My original thought was hand carving a solid 3/4" oak plank and then adding epoxy, roughing it a bit with a stippled brush to simulate the look of the PET bezel.

I see that pre-textured ABS sheets are readily available...



And DIY vac forming is really quite easy...



I am not concerned about weight or amount of effort, so will look at all options.

Brad

Completed the metalwork for the keyboard lid tonight.
Just have to edge weld the main cabinet panels and add a hinge system.
The cabinet was easy so far, a lot more so than wiring the massive board!



So, what's up with the small flat bar section welded to the longer one you ask?
It's because I work with scrap I have laying around and didn't have enough!

It doesn't matter though, once ground clean you can't even tell.
I will still run a thin coat of body filler as well before painting to get everything looking showroom quality.

Brad

Wow, winter came to visit without warning this year! Went from decent to below zero overnight and brought snow.
I decided to work out in the frozen North and finish the hinges, which are the most complex mechanical parts of the chassis.
It was a bit cold out there in the snow, but the temperature let up to a balmy minus 5, so not too bad!

Time was not on my side, so I didn't make a 3D model, but instead resorted to old-school NCAD...



The hinge is 1.5" flatbar with a 3/16" thickness. Good and robust since it will be connected to a motorized linear actuator.
The hard part was making the hinge hidden inside the case. Yeah... it was way more complex that way, but it avoids ugly outside hinges or bolts and makes the entire car hood style keyboard almost hidden until it is activated. Much more of a cool reveal that way!

I hacked up the pieces needed to make two hinges, and welded them...



So far the hinge construction went well, but I still had no idea of the idea would work, since it was only guesstimated on paper.
The actual hinge pin will be a set of bolts that screw into these nuts that are welded to more pieces of flatbar.
This is done to keep all hinge parts inside the cabinet. I will not have to drill any holes in the chassis and give away the secret of the hinge...



With a set of temporary bolts installed, the hinge system worked perfectly!
Welding is not fully completed yet, but everything worked with proper clearance when lifted up to 180 degrees...



There is still a bit more work to do on the base chassis, such as some edge reinforcing and the edge to edge welding.
It was nice to get to this point so far, and if the weather stays decent, I might just continue the metalworking next Sunday.
Either way, the Vulcan-74 shell can move indoors soon and become a base for sizing up the massive hand wired board.



Soon I will be able to answer that question we all ask when designing a two headed computer using only 1970's logic parts...
..."Will a 3 foot by 2 foot board be large enough to hold all of the ICs"?

I think it might, although the 4 channel sound system is also going to eat up a good chunk of real estate.

That's all for now, got to go and shovel the snow.

Later,
Radical Brad of the Frozen North.

Forgot to show the "hood" closed.
Added the hinge to the model as well - kind of a reverse way to do prototyping, but it works!



No visible signs of a hinge anywhere.
I also found an actuator that might work to create the robotic open.

Brad

Actually, it seems that paper tape readers are not so hard to build, I've seen plenty of projects around for these, even conference badges that read paper tape.

This one makes some clever use of protoboard and is at about the right level of technology if you replace the Arduino with a real CPU like a 6502. (Or you could even dump directly into RAM via some clever use of 7400 parts, I bet.)

All the ones discussed so far are hand drawn, but I'd imagine an old cassette tape deck and a bit of cleverness could motorise the reader.

The punch, of course, is a bigger problem. I'm not really sure what to do about that. It's possible actually to print tapes (on clear arcrylic or perhaps thin paper), but part of the whole appeal for me is that with paper tape your physical media is quite sturdy, so long as it's stored reasonably. I don't trust toner to stay on paper nearly as much as I trust holes to stay in paper. And of course you can punch plastic or other more sturdy media if you really want to give your programs longevity.

At the very least you can hand-edit paper tape without too much difficulty, albeit it's tedious. There's a reason, by the way, that ASCII NUL is all zeros (no punches) and ASCII DEL is all ones (all holes punched)!

If I were going to take this route myself, I'm not very keen on the idea of taping bits of paper together, and I don't have a good source off-hand for long strips of it. I might consider instead using index cards, which are still commonly available, and figuring out some way to read across the full height of those without going overboard on LEDs and sensors.

_________________
Curt J. Sampson - github.com/0cjs

_________________
http://WilsonMinesCo.com/ lots of 6502 resources
The "second front page" is http://wilsonminesco.com/links.html .
What's an additional VIA among friends, anyhow?

Oh, duh, of course. It's still available in giant rolls (300 m—well over 100 KB for holes at the standard 2.54 mm pitch), though the common ~58 mm width is more than twice that of traditional punch tape (24.4 mm). Perhaps one could just go with 16 bits across and double your storage capacity and read rate.

I am guessing that even it being thermal paper would not be an issue, unless that somehow causes the paper part itself to degrade faster, since one would just be punching holes in it, anyway.

But there's still the question of how to punch the stuff. Perhaps building a punch is not so hard for someone more mechanically inclined than me.

I also just noticed that there are still several sellers on Ebay selling unpunched paper tape rolls. I woudn't really want to rely on that as a source of supply, though.

_________________
Curt J. Sampson - github.com/0cjs

Attached is a photo of some tests I did towards making a paper tape punch. The main idea was to have small pins all driven by a the same geared motor via a comb-shaped piece. Solenoids or small motors would engage each pin or not with said comb if a hole was to be punched or not.

Pins were made of brass, and their ends filed into a V shape sort of like office paper punches. I only built some test pieces. The blue ones are 3d printed. I think it could be made to work, of course, making the pins out of steel and the "cage" that holds them as well would be way better.

And yes, I did forget to add a way to put the pins inside the "cage". I just cut the cage in half and soldered it together again. That's what prototypes are for.

I abandoned the project as I didn't really see much use for it. An UART or an SD card interface seemed something I could get more use out of.

Just in case it inspires someone out there.
Juan

My job has my travelling a lot more lately, sometimes for days at a time into remote communities.
Having said that, it does give me a LOT of time to work out retro project ideas!

Recently, I found a sad PET 2001N system that was sitting in a pile of basement junk. It had no monitor or keyboard and all kinds of critters were calling the motherboard home. Of course, I took it home and promptly plugged it in, making a crude sync inverter to see if anything would come up. Of course, being a very robust machine, it booted perfectly and said hello with Basic 4.0 and 32K of ram!

Because this little dude was left alone for so long, I decided to give him a new life using some of the ideas that I have been working out in the Vulcan-74 graphics engine. This kinda works out since I have so little time, and this project can be done on my workbench.

I will blog about this in much more detail (schematics and photos), but the short of it is that I am going to create an add-on board that will give the PET a 384x240 pixel display with 8192 colors. Yes, that's 8192... twice as many as Amiga HAM and without any limitations. Can't be done with NTSC ya say?, well the PET sitting next to me thinks otherwise!

Of course, I am using ONLY parts available in 1980, which means 74 logic and simple SRAM. No other parts shall be used, so ignoring cost, this project would have been possible when the PET was still produced.

The new board will rest on top of the original PET board (like the SuperPET board does), and the new monitor head (for the 1702) will be formed of metal to look like it belongs on a PET. I also have a 1702 with a great picture and only half a case, so this is a perfect fusion of sad parts!

Anyhow, I wrangled up a basic test circuit on one of my monsterboards to test my theory that I can generate 8192 NTSC colors by using an 8 bit Luma and 5 bit Chroma signal. The result is 256 shades of 32 possible colors, and it looks amazing!

The circuit rebuilds the PET sync pulses to look more like proper NTSC and then adds the original PET video signal to the new Luma signal. This makes the PET 100% compatible, and displays its original video in a 320x200 window, centred on the screen.

The circuit then adds a color burst and overlays a new 8 bit Luma signal with 256 shades of gray, plus a Chroma signal with 5 bits of color data. The Chroma is generated by selecting 1 of 32 outputs from a series of 74HC245 gates that delay the color burst by up to 240 nanoseconds.

I am also simulating the IEEE bus, so I can load a new IDE and OS that will take advantage of the new board or just load up some original PET classics!

Ok, that's it for now. I did not intend to type so much, just wanted to say 'ello and warn everyone that this thread is about to grow to a few more hundred photos over the course of this winter. I have a lot more done on this project already, but need to sort it all out into a story.

Oh yeah, almost forgot to post the teaser vid...

https://www.youtube.com/watch?v=irBI9fVE9dY

https://www.youtube.com/watch?v=tEI9nl3IJtE



More to come!
Cheers,
Radical Brad

Deleted this - image files lost.

Hey, RB -- nice to see you're still out there, and (of course) still dreaming up radical schemes! So glad those deserving orphans have found a loving home! I'm looking forward to hearing more from you and your new-found family.

-- Jeff

_________________
In 1988 my 65C02 got six new registers and 44 new full-speed instructions!
https://laughtonelectronics.com/Arcana/ ... mmary.html

Thanks.
Yep, still kickin, but with much less time in my nerd cave. 90% of my work is travel now.
Brad

I received a few emails regarding this thread and the missing photos. I see that I accidentally used uppercase when I moved to my new server. Most of the images should work again now. Oh, and this project (and myself) are not dead, just altered somewhat!

I have been quietly working on the most ambitions project of my entire life for the last 3 years and everything preceding this post is just one small part of it - mainly some of the video circuitry. I didn't want to even mention anything as I had no idea that this insanity would even work until recently when I finally had some time to test out some of the core circuitry.

I can't stay long (so much work on the farm), but will say that this is still a 6502 powered project using only 1980 era logic components, but instead of a game system it is now a massive 16 channel synthesizer and track sequencer that will rival the Fairlight CMI of the same era.

There are currently 17 x 6502 CPUs on the board and almost 800 supporting logic ICs, all 74HC types. There are six main circuit boards, each measuring 10 by 20 inches, so a total of 1200 square inches of hand wired PCB space. I have tested most of the core system on my two massive breadboards, but they seem so small now and can only handle 2 of the 16 channels and the video circuit!

Here I go rambling on.... seriously, need to get back to work pal, keep it brief for now!

So the project is now called "The Lucid Science Syntharion Computer Music Station".

Built in 1980 using basic logic circuitry and multiple 6502 processors, this complete music workstation aims to go head to head with the Fairlight CMI, bringing more power and affordability to the aspiring musician. A partnership with Commodore has helped to keep costs down as the Syntharion now boasts a 640x224 display using CRT from the PET, a fast loading 5.25 disk drive using core components from the 1541, and a responsive keyboard from a VIC-20. An 88 key piano keyboard was custom made by Sequential Circuits. All other hardware is being developed in the secret Lucid Science lab.

Here are some very crude renders showing the hardware layout. Although dimensionally accurate, all of the user controls and final touches are not shown.
Notice the massive circuit boards displayed behind the glass windows. Backlighting and many channel status LEDs will bring everything to life.



There will be mod wheels, full analog filters on all 16 channels, a spring reverb section, built in amplifier (possibly tube), and a very capable music tracker that focuses on live performance capture rather than boring keyboard editing of song data. A 6502 will power the OS and each channel will also have a 6502 to handle all of the 44Khz sample loading, playback and track data. Each channel is basically a fully functional synth on its own.

Those six 10x20 PCBs are perforated proto boards, and ICs will be hand wired point-to-point on the backside.



The video generator will be a derivative of the previous work shown here, using the very fast dual buffer system and blitting capabilities. The difference here is that the display is basic monochrome because this is required to display a razor sharp 80 columns of text and to draw the high resolution waveforms and envelopes.



Currently, I have enough of the OS also done so that I can jam out with 4 channels. No recording or tracking, but I can load samples (up to 256K per channel) from floppy disk. It is super exciting to think about what this synth will be able to do considering how great it already sounds!

Here is the video circuit on one of my huge breadboards. This is the smallest circuit in the project. The sound channels are 32 times this size for IC count.



For now, I am going to ghost myself again for the rest of summer as I am busy with wood cutting and redoing my chopzone.com website. I will be ordering all of the steel to make this frame soon though and plan to do the welding before the snow flies around here. When I am at that point, I will once again overload this joint with too many photos and long winded text.

Almost forgot to mention... I actually have ALL of the parts for this on my workbench, and that is a LOT of parts!
Well, the two glass panels and cherry wood parts have yet to be cut, but I will do those after the frame is welded in case sizes change slightly.
I have built it to be 70 inches wide as that is all the space I have in my office where it is going to live.
There is also a rolling table that it sits on with storage for the many 5.25 disks that will hold all the wavetable and sample data.

Cheers out to all you retro heads!
I really enjoy tuning into the amazing projects shown here when I have the chance.

Radical Brad

Thanks for recovering the images!


Not quite lost - the Wayback Machine has them here.