The restoration process continued:

Electronic Repairs Continued:

Monitor Main Board:

The X-Y monitor of a Cinematronics game is a very different critter compared to other vector monitors and operates completely differently than any sort of raster monitor. First off, the monitor accepts parallel digital data words 12-bits wide as its input. There are also several control lines. All of these signals are delivered from the CCPU board through the flat ribbon cable. My monitor board was filthy mess as you saw in the previous pictures, but just look at how it turned out. Regardless, the first time I powered up the monitor, several of the old tantalum capacitors instantly exploded. Also, the breakers popped immediately. I decided that I wouldn't waste time trying to figure out exactly which parts were bad. I thought that even if I replaced them, other parts would soon fail in which case I would have more down time than play time in the future. Instead, I opted to try the shotgun approach. This is where you just replace everything. I spent a couple of days pouring over the schematics and the board to produce a comprehensive replacement part list for the monitor main board. I ordered all of the new parts from Digikey and Newark. When my kit of new parts arrived, I set to work and replaced pretty much every part on the board.

The first time I fired up the completed machine, I could play it blind, but there was still no picture. After the extensive amount of work that I'd put into the cabinet, the wiring, the boards, and the monitor, I frankly just couldn't believe it and had to sit down for a while. At that point, I knew beyond any doubt that the universe was conspiring against me and that I was the butt of a cosmic joke and that it was going to make me squirm and suffer every inch of the way. What the universe didn't know was that I knew that I was being played with and that, my friends, means that the universe is in for a rude awakening.

Here's the finished monitor. Notice that the entire metal frame has been painted with galvanizing primer.

Note the flat ribbon cable below. It's brand new and so is the PCB header.

More than 90% of the components have been replaced with new ones. The DAC80 chips are originals. I have new ones but these ones worked so I kept them. They are the ceramic packages with the gold leads so they are good despite their age.

Here's the freshly cleaned CRT yoke assembly.

Here's the monitor installed in the refurbished cabinet. During the installation, I also included a new plastic CRT shroud purchased from Mike's Arcade.

 

A few hundred words about replacing a few monitor mainboard components with different values...

If you replace all 4 of the 2.2uF/35V tantalum caps with new parts the same value, the monitor will work fine. In the old days, tantalum parts failed quite often, usually rather explosively, and for no apparent reason. Still, they did handle high temperatures and ripple currents better than electrolytic caps and so tended to last longer than electrolytic caps of the time. However, these days, tantalum caps have dramatically improved their reliability. At the same time, so have electrolytic caps. Where once a tantalum cap was preferred over an electrolytic cap due to the tantalum having a lower ESR value, modern electrolytic caps have extremely good ESR values also. Generally speaking, you'll get good service from either type of cap if it is purchased new. However, it remains my personal preference to replace the old tantalum caps with new electrolytic parts. Not only are the electrolytic caps much cheaper, they work great, look nice, and are very reliable. My parts list recommends very good quality parts with a proven track record and excellent manufacturing specs.

I have studied the Cineamatronics monitor schematics closely over the past few years and I have gone through most (if not all) of the FAQs. I haven't found any information that is completely wacky or unfounded. Most of the time there are reasons for the information or the recommended modifications. For example, you are supposed to add a small diode under one of the LF13331 quad switch IC. It was actually recommended by the manufacturer as well as operators in the field (who learned the hard way through hard earned experience). At first glance, it appeared to some people that the diode was useless but in fact it protected the chip from blowing if the negative side of the power supply came up faster than the positive side. This could happen since generally the positive side of a power supply usually had to feed a larger load than the negative side and the increased current draw had a tendency to slow the voltage rise. In the case of the monitor, the two sides are fairly evenly loaded but it only required a short-lived and slight negative bias to damage the IC. Adding the diode cured the problem by shunting the reverse voltage past the chip safely. It was such a good fix that it appeared in the next revision of the monitor design.

With respect to replacing C4, C10, C23 (originally 2.2uF, 35V tant radial caps) with 33 uF, 35V or higher, electrolytic caps... you do not absolutely have to. I did use 33uF, 50V electrolytic caps on my monitors and they work great and I've never had a problem with them. Where they are located in the schematic, I can see that some extra voltage filtering in those places will not hurt a thing but in fact may well be beneficial.

For example, C10 is filtering a completely unregulated noisy voltage input from the +25V supply. If that voltage is lost, the Q2 transistor shuts the beam off. The added capacitance value simply helps hold that Q2 on nice and steady during any transient that might occur on the power line. It will still quickly shut things down appropriately when the power really goes off but there's no need to make it hypersensitive.

C23 will do two things. For one, it'll delay the turn on of Q8 for a short time after powerup and that will prevent the beam from turning on during that time. I don't see that having much of a positive effect except in cases where the machine might be turned off and right back on again (i.e. power blip) in which case it takes a short time for the game logic to reset and to regain control of things. Maybe it just prevents the beam from turning on before deflection begins. The second thing is that it will help provide a nice smooth supply for the base of Q8. Q8 feeds power to the pull-up resistors that control beam brightness. The larger 33uF value of C23 might make the beam brightness appear more steady. Some people may have observed some flicker. If the +5V supply filter caps were bad, then such flicker might have been more obvious. Also, beam brightness may change from one line segment to the next. That means that the current draw on the pull-up resistors R1 & R2 at IC7 (which are fed by Q8) is changing very rapidly. A sudden increase in current draw from Q8 might require a surge in base current that cannot be supplied by the fixed value base resistor R19. In this case, C23 is there to supply that surge of base current. Say a line segment being drawn is dim and that the next one will be bright. You wouldn't what the next line to start out dim and then get brighter as it is drawn. You would expect the entire line from beginning to end to be entirely bright and uniform. A larger value for C23 might provide that. As in the case of C10, the higher value cap can't hurt.

As for C4 being increased to 33uF, again, it can't hurt. Over-filtering the power input to the -15V regulator will not hurt anything. But I'll tell you this much... Remember the diode modification I spoke of earlier? That was added just in case the negative power came up to much faster than the positive supply. What is another way to make sure that this will not happen? That's right, filter the negative supply side more than the positive side. Now check the original values for C1 and C4. You'll see that C1 was 0.33uF and C4 was 2.2uF. See that? Probably before the diode mod came along, someone figured out that increasing the value of C4 helped make sure that the positive side came up faster than the negative side. Here's the thing though. This only really helps matters if the huge caps in the power supply itself are all good and are also all the same value. Here's the catch. Even new caps that size have a huge sloppy tolerance on their value so unless you measure them before installation and deliberately stack the high-value caps all on the positive side, then it could go either way. Increasing the value of C4 definitely would improve the odds so in some cases it might have been just enough. You could make it 1000uF and it would help even more and it still wouldn't adversely affect anything. These are my thoughts. If I am completely off base, (I don't think I am, but if I am) then I install the 33 uF part as recommended. I've done it and it works and my monitors have no problems so I stick with it.

Continue to Star Castle page 4 of 8, the monitor high voltage supply

Back to Star Castle page 2 of 8

Arcade List ---- or ---- Home