Back when I was in High School, my father and I started a business operating pinball machines and video games. I still have a few relics from that era in storage. One of the things that I’ve had in storage for about 20 years is an original Space Invader upright machine.
This is one of the original black and white Space Invader machines, dating back to approximately 1978. The face of the monochrome CRT is mounted horizontally, and projects up on to a mirrored surface arranged at 45 degrees. This results in the black and white display from the CRT being overlaid against a colour cardboard backdrop.
I don’t think we ever operated this particular machine. All machines that we had on site were keyed alike. This machine has non-master key locks on it. Fortunately I managed to find the keys for it. I vaguely recall that my father purchased this machine and it ended up going straight into storage.
Anyhow, I have now extracted it from storage and after a couple of minor repairs (as detailed below) it is working again.
Below is a log of the steps I have taken to get it running.
The fault turned out to be fairly simple: a shorted 22uF 25V tantalum across the +12V rail. The faulty capacitor was located immediately below the banks of D2107C 4Kx1 dynamic RAM chips, and is shown in the picture here. The picture shows the capacitor with one leg removed from the PCB. I removed one leg to isolate the capacitor from the PCB to confirm that it was shorted, and that there were no other shorts on the board.
I don’t know what the component reference number of the capacitor is, as there is no silk screen overlay on the PCB.
Before I replaced the capacitor, I had noticed that the +12V rail was not operating (there was only +0.28V showing at the output of the 7812 regulator). The input voltage at the regulator was fine. So I assumed the 7812 was faulty and accordingly I replaced it. However, in hindsight it may not have been faulty at all – it may simply has shut off its output because of the short circuit caused by the faulty tantalum capacitor. I thought I had isolated the +12V from its load when I was testing the +12V output, but it turned out that I had disconnected the wrong connector from the game board (at that stage I did not have any schematics or wiring diagrams).
31-DEC-2016: Removed machine from storage at Epping. Took it home and used the air compressor to blow out all the dust and muck inside it. Found the keys for it. Attempted to power it up. Initially saw thick vertical lines on the CRT, but they did not come back after cycling the power. Now all I get is a blank screen. CRT raster is working (you can see it come up and wind down when the power is cycled). Audio makes a “thunk” noise when power turned on. Occasionally you get space invader sound effects when the power is cycled on. +5V on the game board only reading 4.55V. No other tests done.
01-JAN-2017: Did some further tests at its new home in the factory in Dandenong. Adjusted the +5V up, but still no better. There seems to be about 0.5V being lost between the power supply and game board. This needs investigating further. At the power supply I tested the -5V and +12V rails. The -5V is fine, but the +12V is only reading 0.28V. The 7812 regulator has +21V at its input and 0.28V output. Looks like the 7812 is faulty (or perhaps output shorted by faulty tantalum capacitor?). Will change the +12V regulator and try again. Bought a spare boardset on U.S. eBay for $50 today.
03-JAN-2017: Replaced the +12V regulator (7812) and removed the 22uF capacitor that was shorting the +12V line to GND. Powered up the machine and surprisingly it is now working 100%. It appears there are no further electronic faults. Next steps are to disassemble the mirrors and backdrop to thoroughly clean everything. There are also some scratches on the plexiglass front, which I will try to polish out.