First the down side…My chance of making Ottawa with a white wood project is rapidly approaching zero. The layout has been taking me longer than expected, learning visual pinball has been taking a little bit extra time (although it is going much better now), and I think that I would need to either quit my job or quit talking to my family to complete it in time. That added to the fact that I have no idea how tough it would be to get a pinball machine across the Canadian border just makes it impossible.
The current Disaster design requires a lot of ramps and habitrails. I need to make a jig to create these wires and it is going to require a lot of tweaking. The tornado may require gimbals to work in all the different situations. That in itself adds a lot of complexity. All of these things make a working prototype much more time consuming.
The visual pinball layout is starting to progress nicely. I think that I have finally figured out what is important, and what can be dropped. Right now I don’t care about the graphics on the board, so I just color everything absurd colors to make it easy to see in visual pinball. All of the rubbers on the board are magically hovering in mid-air without posts. That works fantastically well in visual pinball because the ball is bouncing off the rubbers, and not touching the actual posts. Basically, I only need to model the things that the ball bounces off. Everything else is lipstick on the pig. The angles for the shots are now set, so the play of the table has improved immensely in the last few days.
Summer has been extremely busy. The wife and kids are away visiting grandma for the next few days, so I should have a good amount of time to work on pinball. The time is quickly being filled up with other obligations. I’m hoping to get the first cut of the visual pinball layout working, but the detail on the tornado may suffer.
Last week I ended up having enough time to cut out the driver and controller boards. I now just need the time to be able to solder the parts in and then that testing can start. I ordered the stencil to populate the controller since it is a relatively small pitched surface mount part. That should reduce the amount of time necessary for populating the board significantly. (Previously I was using a syringe to put the solder paste on each individual pad. It works, but it is excessively time consumer). At work we have been populating more surface mount boards by hand, and using a stencil should be a giant step forward.
Looking at the I/O of the Raspberry Pi, it now seems like I may be able to drop the main controller out of the design completely. That would simplify the build and reduce the cost of the BOM. The LED drivers use a SPI bus to change the LEDs, and an RS232 style bus to communicate with each of the solenoid drivers. Both of the buses are available on the header from the Raspberry Pi. There would need to be a little interface board, but that could be a much simpler design.