Pinball Project Update
After a pretty frustrating week, I’m pleased to report that the pinball project is at a really good point. Last weekend I started with high hopes, assembling the power supply components and mounting a flipper assembly to a cardboard box. My plan was to test the power supply by firing the flipper coil. It seemed easy, if a bit scary, since I’d be running 120VAC into the transformer and bringing out a still-beefy 43VDC at the other end.
The power supply appeared to be working well until I attached it to the flipper coil, at which point things went downhill quickly. Several fuses were blown, and eventually it was no longer outputting not 43 but only 2 volts, which I attributed to having ruined the capacitor.
I had also been working with the switch input to PS/2 interface. While I was successful at generating output from it, it quickly became apparent that I could easily design my own switch input reader circuit using a set of parallel-to-serial chips, which allow you to read in a number of inputs using as little as one data line, a strobe/load, and a clock line. Additionally, reading keyboard data off of PS/2 and turning it into simple “switch open” and “switch closed” events is a pain in the ass. I’d also have to build a table to handle converting the non-sequential scan codes back into sequential numbers. Not fun, and expensive on the microcontroller, memory-wise. Plus, it’s a whole lot more satisfying to know that you designed part of a circuit. I ordered the chips and a new capacitor.
With the new capacitor my power supply was back in business, but the solenoid was still causing the fuse to blow. After exhausting every resource known to me (web pages, a book, several co-workers, a newsgroup, a mailing list, and Brad O.), I finally found the problem in the most obvious place: the solenoid’s diode had shorted. Sigh. Fortunately I was able to use that stalled time to implement the new switch circuit and the Arduino software to read it.
This morning I nearly exhausted my stock of alligator and IC clips getting all of the components wired together:
- 120VAC to 43VDC power supply
- DC-DC converter to generate 12VDC to power the Arduino board from the 43VDC (giving me a common ground)
- Arduino I/O board
- Parallel-to-serial chip network (2 chips for this test yielding 16 possible inputs)
- Test switch and flipper assembly
- PinMame-HW driver board
Tonight, after a little more work, I finally reached my goal without blowing any fuses or smoking resistors (I got a little experimental while trying to fix the coil problem): simulating a flipper switch that fires a flipper coil. Hooray! This is a real milestone for the project, a kind of proof of concept point. There will probably be some stumbling blocks in the future, but at least I know that I can build a machine even if with one flipper.
I’m still not sure how I want to do the score display. Late last week I had some success in controlling a dot matrix display (used on games from the 1990’s-present), but there are some challenges in keeping the data moving for animations that may make it impractical. (See YouTube video)
Next steps:
- Build up some connectors and prototype boards. You can only do so much with a precarious rat’s nest of jumper wires.
- Perhaps a whitewood with some flippers, or maybe a mini pop bumper playfield for some stress testing.