Over the last few months; I've been working on a new prototype for the Williams Star Trek: The Next Generation pinball machine. Tonight I finally got the Arduino code to a point where it works as I envisioned; for the most part.

First; a problem in search of a solution... I've always thought the back of the STNG was way to dark - there just isn't enough GI lighting back there in a stock machine. The problem was exacerbated by me having the borg ship and cannons painted simularily by Matt at Back Alley Creations. The mostly black paint job may make the problem worse. Here's what my machine looked like after installing my STNG Popcaps with the EVO purple mods:
A dark back

Ofcourse; I could have just put a Comet LED strip back there... in fact; I did over the popcaps; but I wasn't happy with that solution. Late last year; I had ordered some APA102-based digitally addressable LED strips from Alibaba Express for a home walkway project I've been working on... but after experimenting with them for a while; I decided I wanted to have a strip which interacted with gameplay - a plan was born.

The current prototype is a customized PCB which implements an application-specific Arduino Micro which piggybacks on "top" of the 8-driver board in STNG to simplify the wiring and minimize realestate impacts in the machine. Piggybacking on the 8-driver board allows me to easily tie into some power and the flash matrix for the Romulan and Beta ramps. I also allows me to add a ribbon cable interposer to implement Inkochnito 's bullet-proofing mod which adds the pulldown resistors to the 74ALS576 without actually soldering to the board. There also some connections to the GI_circuit; to enable strip dimming during Shuttle simulations and Borg multiball. There is also a small connection to the lamp matrix to detect Borg Multiball (ship lighting) and the Millions insert.

The Prototype features a USB connection to allow for serial debug and more importantly; code updates via the Arduino environment. It features a jumper setting to allow you to select "blue" borg - per stock configurations or "green" borg which many people have updated with the appropriate window kits.

The Digital strip is stuck to the underside of the upper glass channel on the machine for positioning over the borg and romulan ships; giving the following result:
Strip under Glass Channel
while allowing it to be hidden from the player.

Here are some pictures of the prototype driving the strip:
Lit; no feature lamps
With no feature lamps; the LEDs over the Borg ship and to the left are all "borg color" selected by the jumper. In my case; blue. The area over the popcap areas are purple to match my EVO mod.

Romulan Ramp Flash
When the Romulan ramp flashes; the area over the Romulan ship flashes green... and then slowly returns to the base borg color.

Beta Ramp Flash
Similarly; when the beta ramp flashes red; the LEDs over it instantaneously turn red and then fade back to purple over time.

I don't have a picture of the millions implementation; but when the millions insert is lit - 4 LEDs in the center of the purple section lite White.

Here's a short video featuring the GI dimming feature:

The main intent of this video is to show you the GI dimming feature which will dim the strip based upon the game play. Current prototype has some issues at brightness of 1 and 2 (pulsing) but the dimming feature is rock solid at 3-8 settings as you can see in the video.

Right now; this is just a prototype which I had planned on selling at Pinball-Mods.com... but I'm not sure if there is any interest in the product. The code is functional as is the prototype... and I'm sure with a little more work would be ready for sale with a month to design the final product, procure proper length strips, and custom wiring harnesses. But; At this point I'm not pursuing a bom list or driving for "production" - again because I'm unsure if there is enough interest. Additionally, I'd have to get an alpha unit into the hands of a tester who can verify proper operation of the board with stock ROMs as mine has the non-ghost patch applied because I have LEDs in my controlled inserts.

If you are interested in this product; there would have to be about 50 interested people to make the production price reasonable. Sound off here if you are interested.

Here's a gameplay video... I've edited out the "boring bits" were the rear GI was static.

Thanks for the kind comments.
I really need to see a lot more interested people before I'd consider moving forward with production.

Does anyone have any ideas for improvements?

Over the past couple of days; I've been working on eliminating the flicker for GI settings 1,2, and 3.
c_mario gave me the push in the right direction by giving me an idea to use the zero-crossing of the GI signal.
I couldn't do this in my design; but I did have a FWB-rectified signal of the GI string... so I used INT6 of the Atmel 32U4 to detect the pulse width of the GI string during the lower GI settings and effectively "hard code" the brightness setting to prevent "popcorn" flicker when the pulse width was in the correct range for a given brightness.

This change was easy in implement on prototype PCB because I brought unused pins to a 12mil via. I just ran a wire to the OPTO's photo-transistor collector and turned on the uC's internal pull-up. Fab B will implement this change.

I ordered Fab B of the design.

Fab A was completely functional; but I wanted to add some series resistors to measure current draw and add a "Z"-Connector to tap into the 12VP rail in STNG. I also wanted to convert a 0.1 header to a 0.156 header for the strip +5V power. Finally, I made the "bullet proofing" section into it's own island so it can be easily snapped off for actual use in the game.

Here's a picture of the hand-assembled Fab B prototype; remembering that this board piggy backs on the 8 driver board:
STNG RearGI - Fab B

Sadly; this is probably where this project will stop. There simply hasn't been anywhere near the interest level I had hoped there would be. I'm not sure why; maybe there are too many similar, but generic products available from various vendors. We needed near 50 interested people before it'd be financially feasible to move into production. There has only been a few people show there interest here in the forum - so it doesn't make sense to invest more time in the project.

For those attending TPF 2018 in a few weeks; look for my STNG next to the rest of my Star Trek machines. My machine will have the prototype installed and functional for you to check out in person.

If the interest level changes; I'll (of course) reconsider the production topic.

I don't have a GI OCD to test with; so not 100% sure.
Since this would have attached to the 8 driver board; I'd think it'd be mostly immune to the OCD. The only issue I'm aware of would be the GI connection via that 2pin 0.1 header next to USB port at the bottom of the board pictured above. As long as you connected the other end of that cable directly to the GI matrix (and not the OCD) it should work fine.