Thank you. Yes there is some non-negligible manual labor, although I have limited some of it by doing CAD designs and PCBs. I have been thinking of this and I am definitely interested in making a limited run, to see how much time it takes me, how much I enjoy doing the work and if I can manage to make any money at all at a reasonable price point (although I don't mind not making any or even being under a little bit for said test run).
As for the art, in this context, I don't think I'd be able to get a licence unless I were to do some bigger scale production so I don't think this run can be with the Gottlieb IP.
Stay tuned, there's another chapter coming where I tweak the prototype to bring it closer to that.

Version 1 and version 2, for scale comparison.
Note that Version 1 had an accident last year - it's a topper to my Slap Shot machine and fell down to the ground so that I had to reglue the cabinet. It still works but it seems like some solder got busted and the minutes behave a little weird - didn't bother to fix it.

IMG_3607 (resized).JPG

IMG_3609 (resized).JPG

Hot dog! Thank you all for the kind words.
I can definitely make a run happen.
I can make blanks, just the reels + electronics with no cabinet and I'll see if I can have an idea for an original design.

Although I'm happy with the Atlantis build (I call it Version 2 iteration 1), I'm a bit of a perfectionist and there are still a few things I wanted to improve on before I'd be comfortable sending it out to other people.

The main things are:

-the Arduino Mega does not have a real time clock - this means it needs to be calibrated and then there's no guarantee it won't stray after a few weeks. Although it's not the end of the world (I have this Seiko watch which I really like but goes 5 minutes off every 3 days ), and adjusting the time is part of the fun with this clock, I'll need to update it with a Real Time Clock.

pasted_image (resized).png

-the motors run off 12v so that they're more reliable and always run smoothly. The LED strip runs off 5V, and the Arduino's voltage regulator does the conversion. Although the LED's draw is well within reasonable specs, it does get a little hot to the touch. I don't want anything hot so I will need a beefier regulator.

-although I have made the clock smaller, I still find it a little big and I still have room to work with - I'll make it a little bit smaller still

-I will improve on the mechanical design a bit

-the reels themselves need to be calibrated in the code so that they're perfectly aligned for each digit - to that effect I use a calibration matrix which is hardcoded. I'd rather have a desktop app so that if an owner ever needs to calibrate it they can do it on their end.

-the reel is currently 3D printed with PLA. Although the reel mechanicals are very low impact (an initialization sequence where everything rolls, then everything is mostly still all the time), it is my understanding that the material can deteriorate over time. I will thus try other materials.

Some other stuff too - I've already worked through some of that, keep following the thread to see the progress, and when I feel I'm ready to ship some I'll let you guys know.

Not in this run, but it certainly would be feasible. I want to first focus on making the base product rock solid. It is fun to adjust the time with buttons, but it's also cool to control it remotely so something I might put in the backlog.

Mechanically as it is, it can only go forward - I think doing a stopwatch would only be a matter of coding the functionality, so probably no big deal. In the backlog!

==Part IV - V2-I2 - Improving upon the prototype==

Happy Easter y'all!
Ok so let's keep going and getting into how the Atlantis prototype (Version 2 - Iteration 1, or V2-I1) can be improved upon.

So, first thing, once again the Arduino Mega uses a crystal oscillator as its internal clock. Although it's stable enough for most applications, it did not keep perfect time, and needed to be calibrated (checking to see how many seconds / minutes it wins or loses in a day, then changing the code to make up for the slide) - I'm also not certain that once adjusted it would not deviate again.

The Arduino Mega has pins to be able to have a plug and play Real Time Clock module, so I decided to add that and fix the code to match so that it would keep perfect time.

IMG_3618 (resized).JPGIMG_3619 (resized).JPG

I may try to embed the electronics directly on the PCB at some point, but for now I went with a plug in module.

For sure, El Dorado would be beautiful.
For the first run, labor will not be a concern and I'd like to get a good idea of the time to build one unit, so whether it's a kit or assembled won't make a big difference money-wise.
I could provide a translite template so after that people can customize them as they want and share the results.

Wow that looks pretty cool - yes, I have been testing the RTC prototype for some time now and it seems like the one I have loses about 1.5 minutes a month - definitely within a reasonable error margin IMO and I'm ok with having to press the minutes button (right hand side) every month to keep it right.
Eventually in a further iteration I could look into Wi-Fi capabilities to adjust it to an SNTP server, but I'm looking at making a cool looking clock that you can use in your office, gameroom or some other place at your home, not something that you'll synchronize the next SpaceX launch with, so not that high up on my list.

Alright so we're back from the great Pinside blackout of 2021, which I'll tell my grandkids about.

"Alright kids, sit down, grandpa's gonna tell you about the 2020-2021 pandemic"
"NO! We want to hear about the Pinside blackout again!!"

Now next up on the list of improvements, I was previously using the the Arduino's barrel jack to plug in a 9V power supply so that I get good smooth stepper motor operation.
That means that the Arduino's voltage regulator needs to knock it down to 5V for its own operation, the switches and the LED strip.

pasted_image (resized).png

It's not meant to handle a very big workload, so it would get hot to the touch - I don't want anything hot inside the box so I used this much beefier regulator which is way overkill just to be sure.

pasted_image (resized).png

This also means that I can't use the Arduino's barrel jack anymore, but that's not a problem. It means I can also have a barrel jack on a wire extend outside of the box, which I wanted anyway.

pasted_image (resized).png

Next improvement: an update on the cab.
I still found the clock a little too large and I still have room to work with. Also, as I make it smaller, I think having the reels centered will look better.

So I'm going from this (10" x 7" x 5.5"):

pasted_image (resized).png

To this (9" x 6.5" x 5.5"):

pasted_image (resized).png

For the Atlantis clock, I had a friend of mine who owns a CNC router cut the parts for me. I don't very much like to ask people for help for stuff if I can help it, and most shops will charge a fortune for a one-off, so I decided to get one of those engraving machines on Amazon for the next prototype.
At the same time I was able to teach myself a little bit how gcode works and how to use the manufacturing functionalities of Fusion 360.

20210116_153642 (resized).jpg

It's really not made to cut through 0.5" MDF and it took hours to cut each piece.

Since it takes all this time to cut through a piece and we only have one "home" laptop (as opposed to our work ones), I decided to move the operation from the shed to the basement next to the water heater so that we could at least use the computer if needed.

pasted_image (resized).png

That means my poor little girl had to play Roblox next to a running CNC, next to the water heater - something she can bring up in therapy in a few years. Special thanks to her for putting up with all this

Screenshot 2021-04-17 092925 (resized).png

Eventually, after a week's worth of cutting I got all my parts. When I produce a run the idea is not to produce it with this machine, I just wanted to prototype and understand the output of Fusion 360 so that my CAD/CAM stuff would be legit.

pasted_image (resized).png

Looking through my pictures, I found a mockup I made of what I thought the smaller version could look like (although not to scale).

Original
IMG_3329 (resized).JPG

Smaller
9_5x6_5 (resized).png

Some better pics of the cuts for iteration #2.

IMG_3519 (resized).JPG

With all the time I took cutting these, I didn't even bother to check that I had cut the right version of the side panels
Too wide, so I'll have to recut them.
IMG_3520 (resized).JPG
IMG_3521 (resized).JPG

Here,
that's better - after redoing the sides, then assembly and sanding.
pasted_image (resized).png

Painted white - V2.2 and V2.1 side by side to see the size difference.

IMG_3563 (resized).JPGIMG_3561 (resized).JPG

Thank you for the feedback, very appreciated!

Here is the V2.2.2* version of the electronics:

IMG_3414 (resized).JPG

Again, this is a shield that goes on top of the Arduino Mega.

*the nomenclature being:
-Major version 2, the first one being the one with the actual Gottlieb Score Reels
-Minor version 2, the first one being the Atlantis model
-Iteration 2 - I messed up the first PCB because of touching traces - this one works right

Although V2.2 of the clock will not be Atlantis themed, I have decided to keep this "code name" for now.

Mechanical parts V2.2.2.

I tried to do a little bit of 3-D printing optimizations, to use a little less material on the brackets and to decouple the reels and the little washer to offset it from the motor. When it was all one piece, I had a bunch of support material I needed to remove on each one. Now they're just printed flat and I print the washer separately.
The bracket now also has a clip to hold the stepper wires.
pasted_image (resized).png

These are all the parts that go in one score reel assembly.

Reel assembly V2.2.2

pasted_image (resized).png

V2.2.2 Reels plugged into V2.2.2 electronics

pasted_image (resized).png

I'm trying to make this iteration look good even without graphics, so I have made the inside panel with cedar. All that's left is to put it all together and take a few pics, then we've got a complete V2.2 clock!

pasted_image (resized).png

Actually another thing I had mentioned I wanted to improve is that I have a calibration matrix for the adjustment of each reel that was hardcoded.
Although it worked for my individual purposes, that would mean that if I sell units and that I ever come out with a code update (which can easily be done through the Arduino USB port), the code update would wipe out the calibration matrix unless I kept track of each individual one.
What I did is that I decoupled this matrix from the code itself by writing it to the eeprom of the Arduino.
I then coded a Windows application that connects through the USB port and writes to that eeprom.

Thus, you can update the code without touching the calibration matrix and vice-versa.
Here in this super out of focus video, you can see the first baby steps of this application as I send the first Hours++ command.

And, here is V2.2.1 of the clock!
In my opinion it looks good on its own without graphics and puts more emphasis on the reels themselves - but of course it's still always possible to theme it to your favorite EM.

IMG_3579 (resized).JPG
IMG_3587 (resized).JPG
IMG_3597 (resized).JPG
IMG_3594 (resized).JPG
IMG_3585 (resized).JPG

One thing I'm not entirely happy with yet is the way the reels themselves look, black with a white sticker - I think it is adequate when there is a translite but I'd still like to do a little better, which means we'll have a clock V2.2.2.
More on this shortly.

So here is what I'm still unhappy with at this point - when there is a translite, the sticker overlap can mostly be hidden out of sight. However, without a translite, we get this.

IMG_3595 (resized).JPG

In the best of worlds I would like to have the numbers screen printed on the reels, and I do have an idea of how this might be feasible, but I have another easier way to go about it at first which we'll be looking at.