Now that the internals were up and running, I wanted to concentrate on the cosmetic issues. First I had to patch all the holes with aluminum brazing (amazing stuff).

The process of brazing was actually a lot more challenging than what I initially thought it would be.

1. •It's very very hot. Don't be fooled into thinking you are dealing with solder temperature even though it looks similar when molten.
   

2. •After the braze cools it can leave very sharp burs which can cut skin, as I found out as I now have a large gouge out of my thumb.
   

3. •The aluminum after heating and cooling is now VERY pliable. I think I need to reheat the area again and do a rapid cool with a dousing of cool water to harden the metal.
   

4. •The braze is a much harder material than the aluminum and this means it's very difficult to file.
   

5. •Although the braze when molten has a lot of surface tension, there is a limit to the maximum size hole you can bridge, as I found out.
   

6. •If you put the braze rod down, don't pick it up by the wrong end. Ouch!
   

I had to put a small bridge of aluminum in the big hole that couldn't be filled, and fill the hole with braze again.

I had some small problems and minor injuries, but overall I think the brazing when ok. The pliability of the original aluminum was a bit worrisome. It did however seemed to stiffen up after a few weeks.

The grinding of the braze was not easy at all, but regardless I made good progress. I used a hand grinder that I borrowed from a friend. The grinder was a lifesaver as I couldn't imagine doing it with a hand file as the braze is very very hard. Next I just had to tidy up any rough grinder marks and then send it off for sand blasting and powder coating. The final results were quite nice, and although there were still a couple of small places where the finish wasn't perfect, and the color wasn't’ a a 100% match for the original. the final results were very pleasing.

The machine arrived and smelt bad like mouse poop. I decided the first thing was to take it apart and wash it thoroughly. The machine was in rough physical condition with 12 holed drilled in the case for various mods and switches. It was also missing the power supply. I got all the boards out of the machine and on my workbench and powered everything by a switching power supply. After a bit of fiddling around I was able to get a screen of random characters, although the image was very fuzzy. I realized that I was using the video feed from the color composite line. Swapping the wires around to the monochrome line produced a clear crisp picture. After a bit of confusion over how to rest the machine and having to reseat the boards and move them around to ensure good connections to the back plane, I was able to get the H/D/M prompt. The characters where however somewhat deformed, having an extra line of pixels on the right side. After some checking I realized that I needed to up the voltage from the switching power supply by about 0.5V to around 5.5V, and only then did the character deformation disappear.The switching power supply seemed to work quite well, but it did seem to get very hot. I soon realized why...

I was still having issues with one of the 24k board. However I realized that my present switcher supplies 5V @ 5A and it wasn't enough to power all the cards once, especially with the second RAM card installed. I ordered another PSU from eBay (cheap $25 one from China) and is is rated at 5V 6A. Once it arrived and was hooked up it fixed the issue with with the RAM card, and I was able to run all four cards with no issues.

I typed in a memory test program I found online. The program displayed an error in the 4400-47FF range. Replacing the 2114 RAM at B7 fixed the problem.

Next I tried to figure out how to get the machine working with a serial connection so I could transfer some programs from my PC. When I got the machine it didn't have a DB25 connector installed. I was told that it should be a very simple operation to add a few jumper wires and make a cable, but I had trouble finding out where to jumper the board. It turned out that the board was already jumpered correctly.

I wired up a serial connection as the board was already setup for serial connection I presumed I would have no troubles, however that was not the case. Firstly I did not have the -9V power supply that was needed for the serial out. after discussing this issue with Dave from osiweb.org it was decided I could do a simple hack to reroute the GND signal as a replacement for the missing -9V in the serial circuit. I did that and still could not get it working. After some troubleshooting I discovered that one of the transistors was dead. I replaced this with a new one. The new one had a Base, Collector and Emitter configuration that was in a different order than the original. A little leg twisting fixed that. After the fix the serial connection worked! I was able to transfer programs from the PC at 2400 baud.

Next I wanted to change the serial speed from 2400 baud to 9600 baud. Looking at the schematics it seemed that all I had to do was cut a single trace and add one jumper. After doing the mod, the machine was able to transfer and receive data at 9600 baud with no issues.

neoncluster.com

Some issues with corrupt characters were fixed by upping the voltage from the switching power supply by about 0.5V.

After the voltage from the power supply was

upped from 5V to around 5.5V.

After replacing a dead transistor and doing the 9600 baud hack, I was able load programs at a decent speed.

Finally I built a bracket to mount the new switcher power supply. and made a backing brace from acrylic for the back plane. I purchased a slew of new parts including a new on/off switch, fuse box, and video and audio connectors. The machine would not be perfectly original, but it would be clean and hopefully reliable. finally I was able to assemble everything, and wire up all the connectors. The last thing I need to do is fabricate a ‘CHALLENGER 2P’ Badge. I’m still not sure how I’m going to go about this but I have already draw up the artwork.

Conclusion

Another project that took considerable time. Although electronically the machine was in ok condition, with only two dead component found, physically it was a nightmare, with missing power supply and multiple (12?) holed drilled in case. This time I learnt more about how to work with aluminum braze than I did about electronics. Still looking for a cheap way to reproduce the Challenger 2P case badge, and I have already redrawn the artwork in anticipation. Ultimately a very enjoyable restoration project.. Much thanks to Dave from osiweb.org for all his help and advice.

Images showing the case after grinding the aluminum braze.