Have you ever heard the expression, "A poor carpenter blames his tools."? Well, that's a load of crap because a good carpenter wouldn't use a really ratty tablesaw. How does this apply to me? It doesn't because I use a really ratty tablesaw.

With the production of the second control panel in the works, I needed to start thinking about the control panel frame. I designed some plans in Adobe Illustrator over the course of a few hours. Building a wood control in the same shape and with the same angles as the original metal panel is extremely complex. Each support panel has three unique angles in order to support the upward angle of the button panel.



Once I finally worked out the complicated angles and drew out the plans, I piled up some lumber I had in my storage closet and headed to the workshop. This step caused me to face a problem I had known about for some time and had been avoiding. The problem? My tablesaw is a disaster. I purchased a portable Ryobi tablesaw this past summer knowing I needed a tablesaw for light woodworking. My attempt to saw a piece of pine with the included 24 tooth saw blade was met with the same results as sawing through an iron ship keel. Before I could proceed with any serious construction on my control panel I would have to roll up my sleeves and make some serious adjustments.



First step was to replace the blade. My dad and I installed a 10" DeWalt 80 tooth carbide blade which cut through my test lumber like butter. Next was to square the blade angle. My test cuts initially made 90 degree cuts that were approximately 87 degrees. With a few turns of the screwdriver we were finally making true cuts.

With the saw calibrated correctly, I could finally start working on building the panel. To support the weight of the controls I am making the frame from oak, which is a very strong wood. Oak, however, is outrageously expensive. Because of the complicated angles of the wood panels, I am making numerous pine copies before proceeding to the final oak version to ensure the craftsmanship is top-notch.

Here's some good news for our fighting boys overseas. The wonderful folks at the Sega Corporation have developed a thin plastic material that can't be destroyed by conventional weapons. They were also good enough to coat both sides of my arcade with this substance just incase it was hit by some stray artillery shell.

With the wiring in a good place, I wanted to focus on the cabinet. I spent the first part of this weekend designing a wood frame for the arcade control panel. Next it was time to sand the body. After some initial tests with various types of sandpaper I wasn't making a dent in the plastic laminate. Time to break out the big guns...



I busted out my angle grinder and attached a circular steel wire brush head. After four hours I was able to remove all the laminate from one side and annoy about fifteen neighbors. I hadn't anticipated this step taking as long as it had and my plan of having the second control panel completed this weekend had to be scrapped. Hopefully I can make the most of the upcoming long weekend and get the control panel set.

I walked into my apartment after work and tripped over a small box. That could only mean one thing... My controls have arrived!



I took that as a sign from the zerg overmind that it was time to start constructing the control panel. Having made my paper template, I knew the button positioning I laid out was very good. I picked up a large sheet of .5 inch MDF (medium density fiberboard) at Home Depot and began plotting the holes. The difficult part of this task was fitting the joystick. The joystick can only be set behind a maximum of .5 inches of material. Because I am going to cover the MDF with a layer of .25 inch plexiglass, I needed to use a router to create a .25 inch recess so the overall material thickness between the joystick baseplate and the handle remains .5 inches.



This joystick recess also creates a new problem. With the recess in the MDF, there is now only .25 inches of drillable material to screw the joystick in. Faced with this, I decided to make a second panel and use this first control panel as a test run for various screws of different widths and threadings. Once I know the appropriate hardware, I can then make the second panel correctly and take into consideration a few of the small things I learned from making the first one.

Look out you dirty commie rebels. Your endless fleet of soviet tanks and missile launchers are no match for my seemingly bulletproof helicopter gunship. I have been sent on a ludicrous mission to single handedly destroy nine thousand miles of enemy territory.



After mapping out the key circuits and attaching some of the original arcade buttons to the screw terminal, I fired up my PC and plugged in the keyboard encoder. It was time to test the wiring. SUCCESS. Using my wiring matrix, I was able to test every single arcade control for two players.

With the wiring up and running it was now time to start planning the control panel. I am refurbishing an arcade that is designed to be played by two players. With a loose idea of the games I want to load onto the system, I anticipate needing two eight-way joysticks, six action buttons for each player, two start buttons, an insert coin button, a quit button, a miscellaneous button, and a trackball for games like Centipede and Golden Tee.



I have created a preliminary template of my control layout in Adobe Illustrator. The next step is printing the template and laying it over the original control panel. With the panel positioned at the actual standing height I can get an idea of how the layout will feel to the user and make any necessary adjustments.

The past few days have been devoted almost exclusively to designing and planning the arcade controls. An arcade button is nothing more than a small switch. When the button is pressed, a connection is made and a circuit is created. The onboard computer detects this circuit and performs an action.

Unfortunately, the onboard computer for my arcade is a worthless relic and is laying in pieces next to our bikes in the back hall.

Because I am using a traditional personal computer as the brains for my M.A.M.E. Arcade, there are no open points to connect to the button terminals. There are several parts available online that can solve this problem, but I opted to go for a makeshift (free) solution to improve my technical skills. I decided to attempt what is commonly known as a keyboard hack.

A keyboard hack involves using the keyboard encoder circuit board that is built into any keyboard and wiring it in a way that tricks the circuit into interpreting arcade button clicks as keystrokes. There are two very labor intensive steps in modifying a keyboard encoder for this purpose. The first is mapping the key circuitry and the second is soldering the arcade button circuit wires directly onto the board.



A keyboard uses two sets of inputs to determine a keystroke. This is because having a single input for every key would take up massive amounts of space. The keyboard I am using has 116 keys. Having that many inputs would cause the keyboard encoder to be at least triple the size it is now. Instead, my keyboard has one set of inputs that has 18 terminals (set X) and another set of 8 (set Y). Each keystroke is assigned an input for each set. This allows the keyboard to decipher 144 possible keystrokes using only 26 inputs.

Inside the keyboard are two layers of flimsy (thin plastic) with circuitry paths printed on them. Under each key there are circuitry points on each layer of flimsy that come in contact with eachother when a key is depressed. This contact creates a closed circuit traveling through specific inputs from set X and set Y. I know; this crap is baffling. See the following diagram for some clarification.



As you can see, the diagram shows the circuit path of the keystroke F10 on both layers of the flimsy. Both circuit paths arrive at a specific input on the keyboard encoder. From this diagram you can decipher the specific circuit input information for each key. After I logged all the possible key combinations I would need for my arcade controls, I began soldering wire to each solder point on the back of the encoder. These solder points allow you to wire your own circuits onto the board using your arcade controls.



Once wire has been soldered to each solder point on the encoder, I connected the loose end of each wire to a screw block terminal using spade terminal connectors. By using a screw block terminal, you do not have to touch the solder connections on the board to rewire your controls. Notice one block has 18 terminals to connect with set X and the other has 8 to connect to set Y. I now have the capability to wire arcade buttons to match any key on a keyboard. Where's the w00t button?

Here's a fun fact. A CRT monitor can store anywhere from 20,000 - 30,000 volts of electricity when it's unplugged. There goes my plan of hacking into that shit with a sawz-all. I consulted my new arcade holy bible for some details on how the monitor operates and where the various high-voltage points were. The high voltage is created on the circuit board by a transformer and is stored in the anode on the tube exterior. There are several other points where exposed high voltage is present.

If a repair is needed on the monitor almost all repairs must be conducted when the anode is drained of it's stored electrical charge. This procedure is called 'discharging'. This is a difficult task because draining the stored charge requires the user to stick tools into high voltage places. A mistake can literally cost you your life, and possibly your eyebrows. For those of you playing along at home, consult the following graph before breaking open your Zenith.



When I get the appropriate computer hardware to connect my arcade monitor to my computer, I can test to see how bad or good the picture is. If it's decent, I can adjust the dials on the circuit board and smooth the picture out. If it's a disaster we may have to perform some Back to the Future action. 1.21 Gigawatts dude.

My arcade project just received a massive power up from my good friend/nemesis Nat Tarbox. Nat found this gem at the MIT Bookstore in Cambridge, MA. It's 476 pages of arcade based greatness covering every single component of a traditional arcade unit including the potentially dangerous monitor. It's offically ON.

Recently, I did some research on old arcade units. I've always been fascinated with arcades and classic video games. Since classic arcade and console games are experiencing a renaissance I thought it would be a cool idea to buy an old unit on the cheap, clean it up and put it in the apartment.

That's when I found out about M.A.M.E.

MAME, or Multiple Arcade Machine Emulator, is public software which allows a personal computer to play the raw game files that were installed on the original arcade hardware. Essentially you can program your computer, with some special hardware modifications, to play just about every classic arcade game ever made on the original arcade CGA monitors. The software is available for Windows, OSX and Linux. While I'm normally an OSX loyalist and regard windows as the inadequate piece of crap that it is, I opted to go with a Windows XP based system because the file and hardware support resources are much better.

I have broken this projects into two parts. The first part is the rebuilding of an existing unit. The second part is creating an entirely new system with a customized body and assembly. I purchased a half-working 'Major League Baseball' system by Sega for 25 bucks. After reviewing the condition of the system and cleaning out the evil army of arachnids that had established a basecamp, I began to carefully remove the internal circuit boards and power source while making sure to clearly label each wiring group so I could analyze the assembly.



My friend Dan gave me his old heap of a PC that was chock full of spy/ad/poison-ware. After completely wiping the harddrive I downloaded and installed the MAME software and have begun prepping the system to work with the existing arcade monitor.



The difficult part with this first system is the monitor. A special videocard (AVGA) is required to communicate the video information to the CGA monitor with a specific signal. I'm hoping the existing monitor is still in working order as I don't want to spend money on two brand new arcade displays which can cost anywhere from three to seven hundred dollars. This project is proving itself to be a formidable one and I am really enjoying its complexity.

This is the first post of the Matt Collins project blog. This blog is intended to be a continuously evolving bulletin of my latest industrial and graphic creations. While there is absolutely nothing to see here right now, I have every intention of developing this crappy blog into something actually worth reading.

Labels: first