After thinking about purchasing a pinball machine of my own for some time,
my wife decided to buy one for me for Father's day 2005. The machine is the
"Space Shuttle" by the Williams Pinball Company.
Some photos on this page are by the person from whom I purchased the
The backglass of the unit.
Note the name of the Space Shuttle is 'Defender'.
The playfield of the game. The paint is in very good shape with only one small
bare patch under the right most pop bumper. Photo by DS.
It all started with my cousin Ron
Canada. He purchased a Space Shuttle pinball machine
and put it in his basement rec room. Every time I visited, I found that the game appealed
to me more and more. I realized that the game would fit in well
in my rec room as it is decorated in a space theme. Also, the
playing field features a picture of the Hubble Space Telescope being
serviced , and considering my past history with the HST Servicing Missions, I thought
seriously about buying one too. In addition, the game really
plays well. It has a speech synthesizer, cool music, the
customary flashing lights, and a nice multi-ball feature.
Close-up of the image of the Hubble Space Telescope on the Playing
The game was built in 1984, and it is remarkable to me that
the designers (Barry Oursler and Mark Sprenger) had the foresight to feature
on the playing field 6 years
before HST was launched!
Doing some research on the web, it
became clear that this game deserves a prominent place in
history. It is credited with saving
the Williams Pinball company from bankruptcy and with saving the
pinball industry in general. I started to search for machines on
(where else?) ebay. I found one (opening bid of $500, but playing
field had numerous bare patches), and bid a price of $750 for it.
I did not win the auction, and continued to search with Google. I
found a reconditioned machine in Minnesota for $1100, and
it could be shipped to Baltimore for about $150. Just before
buying it, I did a search on Google newsgroups and found Diane. She had a
machine that had not been reconditioned (shopped) and was offering it
for $600. Most importantly, it was in New Jersey, and I could see
and play the machine before buying
The machine, on the day we picked it up at Dianne's shop. It was a bit dusty
inside and out, and one drop-target bank did not reset, but since the paint was good,
I knew I could take care of the electronics and mechanical parts.
We drove to Hamilton Square, NJ on
7/16/2005, and picked up the machine. It is very heavy and took
some effort to get it down the stairs, but it made it home
safely. The machine needs some work and cleaning, but it plays
great and I look forward to tinkering with it in the coming
years. The near constant temperature and low UV ray environment
of my rec room should be a good place to have the machine.
The game, set up in our basement. The kids have been extremely excited about
having the game.
Original condiiton of the machine near the ramp entrance.
of rubber parts, balls, and burned out bulbs
I ordered new parts from Pinball Resource to replace the worn/broken items. Total time for the items below was about 6 hours.
Bulbs. I ordered twenty #47 bulbs to replace the burned out
ones. These are the same socket as #44 bulbs, but consume less
current and run cooler. I thought I would need only three or
four, but in the end I replaced ten bulbs! Some of the burned out
ones were hidden behind plastic parts, and were not obvious at
first. I also washed and wiped the glass on all the remaining
bulbs. Some looked like they had twenty years of dust and soot on
them. As a result, the playfield looks much brighter and more
beautiful than before. Bulbs were $1.50/10. The machine
also uses #89 bulbs ($2.20/10), but none of those were burned out.
Rubber parts. I replaced all the rubber parts including the
shooter tip and the flippers (great kit from Pinball Resource for $17).
I had to be careful to have clean hands when handling them as they are
Waxed playfield. I used some Mill Wax to put several layers of wax on the field and the plastic parts. This made them shine and should protect them from wear.
New balls. The machine uses three, and the new smooth ones
(apart from looking better) will reduce wear on the playfield.
After these repairs, I checked every
solenoid and switch for proper operation using the diagnostics built
into the machine. With the brighter playfield and balls, the
smoother rubber parts, and the waxed playfield, the whole game played
so much better.
On one occasion, I happened to
operate the flippers with the
coin door open. I was able to look into the cabinet and
discovered a bright spark coming from the flipper
solenoids every time they were activated. A site I found
recommended the installation of a
capacitor to absorb the coil energy to reduce the dV/dt when the switch
is opened. This is very similar to the enhancement we added to
Space Telescope power relays in 2002 by adding installing the DBC2.
Photo of the flipper solenoid with the new capacitor. Note the
white and black
residue near the End-Of-Stroke switch due to the thousands of arcs over the
years. Not good! The triangular piece of plastic isolates the high voltage
solenoid circuit from the CPU sensor contact that changes the U-S-A lights on the playfield.
The flipper solenoid is a high power
unit, and pulls in with a great
deal of current. Once the flipper hits the end-of-stroke (EOS), a
lower amount of holding current is applied. This creates such a
high inductive kick that an fierce arc occurs at the switch.
After I installed the capacitor, a test showed that the situation had
improved, but sparks were not completely eliminated. About one
out of five strokes will still produced a small spark. But I
think the EOS switch will now last a lot longer.
'T' drop target decal
This target is in the middle of the
playfield and gets a lot of
hits. As a result, it is not surprising that its decal (of a
satellite dish, same as the 3-bank) is very worn. For the
repair, I removed the best one from the machine, and scanned it at 1200
dpi on my flatbed scanner.
Scan of drop target. It was originally created with iridescent
paint, some of which has
worn off where the ball hit the decal. I electronically airbrushed the defects out.
After scanning the the decal, I
it on glossy photo paper and covered it with two layers of clear
tape. I then glued it onto the drop target. This was also a
good opportunity to clean the drop target mechanism and apply a very
coat of oil.
The old condition (left) and the repaired decal on the drop target
It looks much better in real life as the reflection from the camera's flash
caused some blooming in the image. I was able to clean the
red plastic of the target body as you can see in the right image.
Note that the tip of the shuttle has been touched up. The camera's
flash made the metallic paint light up, but is not very
obvious under normal lighting.
When I received the cabinet, there
were several gouges on the paint of
the back unit, and the usual dings and blemishes here and there.
Using paint sample chips, we found a reasonably close match for the
red, white and blue paint colors. A closer match could be
obtained if I carried part of the cabinet (back unit) to a place like
Lowe's to perform a color match.
Before (left, sorry blurry) and after (right) pictures of the repainting of the cabinet.
As you can see, the match is not perfect, but it can only be seen under very bright illumination such as that from a camera's flash.
Adding a powered woofer
Sound is a very important part of the
overall pinball experience.
that the two existing speakers are clearly too small to have much
bass response. I decided to analyze the frequency response to see
how it can be improved.
The speaker amplifier on the CPU board. The driver is a
UPC2002. Although this
schematic shows an 8 Ohm load, the two 4 Ohm speakers are
actually in parallel (2 Ohms total).
This is the expected frequency response of the above circuit into a 2 Ohm
load (2 speakers). As you can see, the circuit rolls off to its -3dB point
at about 113 Hz. The speakers themselves probably roll off
higher than that. Also, the amplifier chip has a low frequency pole at 40 Hz.
Since I had an unused powered speaker
system (for a multimedia PC), I decided to use that as the amplified
sound source. I wired the connection to a 1/8" phono jack that I
one of the holes of the speaker grill. The advantage is that this
mod can easily be reversed without damaging or altering the cabinet.
I disconnected the cabinet speaker and soldered its wires directly to a
small phono jack.
It fit neatly in one of the speaker grill holes. The output is in parallel with the
Picture on the outside of the cabinet. It results in a very clean installation that
can easily be reversed
The amplified speaker plugs into the phono jack and results in a very
The whole system was buttoned up and the
resulting sound is amazing. There is a satisfying rumble when the
Shuttle launches (when the ball is shot), and all the music and sound
effects are fuller and deeper. Since the amplified speaker system
consumes only 2
Watts quiescent (measured with my Kill-A-What meter), I leave it turned on all
Checkout of the spare display driver board.
I purchased a spare display driver on
ebay. These boards contain the UDN7180 chip, which is no longer
made and can be very expensive (as
high as $25 each). The part
number on the board I received is 5760-09461-00, while the one inside
my System 9 machine is 5760-09461-02 (for a
picture of the -00 board go here; top
image). In doing a Google search, I found out that the -00 board
is used in System 7 machines, and I became unsure if I could use it or
not. I downloaded a copy of "Black Knight" to examine the
schematics and noticed that the input power connector has a slight
difference. On the -00 board, -100V comes in on two pins on J6,
while the -02 boards only use one pin. I looked at the wiring in
my back box and was relieved to realize that it will support either
Performing the safe-to-mate on the spare display board.
The demated connctors normally connect to the existing display board.
Note the little loop of orange wire on the connector that plugs into
the board. That loops the -100V to two pins,
allowing -00 and -02 type boards to be used.
I decided to be careful in checking
out the new board as it comes
untested. First, I disconnected the existing display driver
board, and verified that the new board had no shorts on its input power
connector. Then, I plugged in only the power connector, and
verified the voltages on the new board were good. I then verified
that none of the CPU connectors on the board had any voltage on
them. This was to prevent damage to my CPU board if for some
reason the 100V was connected to the CPU control lines. I then
powered down and connected a spare 4digit display and the CPU
connectors to the new driver board. And powered the game back up.
The result was....success. I was lucky and the new board works
100%. All 32 digits on all five displays work fine.
Result of the spare display board with a spare 7-digit display.
spare 4-digit display that I used can also be seen. I repaired this
7-digit display as described below.
Repair of plasma displays
From ebay, I bought six plasma
displays that were untested. They
included two of the hard-to-find 4-digit units, and four 7-digit
units. One of the latter units had a non-functional 10,000 digit,
and I decided to attempt a repair. The connections to this
digit is shown in detail below.
Tight shot of the connections of the digit that was not working.
They are the ones that lead to the
orange wires that underline the digit visible at the top of the picture.
As one can see, there are two pins
that connect to this digit's anode connection (third and fifth from the
right pin in the above image). These pins are shorted together on
the circuit board, and I desoldered the pin on one connection and
measured an open from one pin to the other. It seemed to me that
the open connection was most likely in the interface between the metal
pin and the orange trace. Using some Nickel
Print (conductive paint used to repair printed circuit boards), I
placed some small drops on the edge between the two
glass plates right at the point where the pin enters the glass (red
ovals in the image above).
After it dried, I found that my repair was successful. The
repaired display can be seen in the image above.
Of the six displays in this lot, only three work. However, I will continue to work on the three that don't in my spare time.
The added fuse holders are on the right panel. The wires that
connected to the bridges
are the red and blue wires on the right. See Clay's version.
Location of the added fuses on the schematic (red circles).
They prevent major damage should
the bridges or main filter capacitor short out.
Clay recommends the use of 8 Amp Slo-Blo
fuses, but all I had was 6 Amp Slo-Blo units. The machine powered
up and played just fine.
Repairing a numeric display problem
One day, the kids notified me that the numeric score displays on the backglass were all dark. After some snooping, I found that the culprit was in the -100V supply on the high voltage supply board (part number 5765-09466-01).
Excerpt from the high voltage supply schematic. There was zero
volt drop across ZR3, showing that R4 was open.
Close-up of the high voltage board with test clips installed on R4.
Note the brown/grey color on the left end.
This resistor tested open after removal from circuit.
Click here for Clay's version.
This resistor (R4) is a 1 Watt 39kOhm
resistor. By design it dissipates about 0.35Watts. Over the
years, the heating must have opened it up. Having located the
problem, I decided to also replace R1, which is the matching resistor
on the +100V side. Once R1 was out, I measured its value, and was
surprised to learn that its value had risen to 54kOhms. It would
only be a matter of time before that one too would have opened
up. If I replaced R1 with its original value, it would increase
the high voltage slightly. So in order to reduce stress on R1/R4
further and not increase the high voltage very much, I changed them
from a single 39k resistor to four series 1/2 Watt 13k resistors.
This drops the total power to 0.25W.
In the future, I intend to lower the high voltage by replacing the entire high
voltage section, swapping out the regulating
zener, or using a series
dropping resistor board. This will cause the plasma
displays to last longer. Clay's site has discussion on this
This is apparently a common problem with these units.
Replacement of the Shuttle toy
When I received the machine, the
Shuttle toy had turned a yellow color
in the areas that were not covered by the stickers. In addition,
the mounting screws in the wing flaps were gooped up with some kind of
glue in a crude attempt to repair the damage from ball strikes. Even a
casual observer that did not know very much about pinball machines
noticed this kludgy repair one day, and I resolved to address this.
Original Shuttle toy at the top of the playfield. Notice the
yellow color and the
gooped up damage in the wing flaps.
I received a replacement Shuttle toy
from Greg Deakin along with my purchase of a backglass, and although it
too had damage on the wing flaps (must be a generic problem), the color
had not faded. I decided to go ahead and swap the toys. The
holes of the replacement unit had damage and were enlarged due to too
many ball strikes from balls that come up the middle ramp.
Instead of using washers to hold the toy down, I made covers with the
same size as the
entire wing flap that were cut out of white shiny plastic. I
think this large
surface will prevent future damage to the toy Shuttle. I pulled
off the stickers from the old (yellow)
toy and scanned them to reproduce the stickers for the wing flaps.
The replaced Shuttle toy. Note that the wing flaps have
been covered using white plastic, and then
reproduction stickers were glued on them. Also note the
starboard wing's screw has been painted white.
The port wing's screw is located in the black area, and left unpainted.
The raw scanned decal on the left, and the cleaned up one on the right.
Oh, the wonders of Photoshop.
Here is another approach to this central feature of the machine.
This Shuttle is made of translucent plastic and is lit up from below using white LEDs.
It has engines that flicker red just like the artwork on the backglass.
the CPU board
When I received the machine, I noticed a few quirks with the attract mode music. This is the sound that the machine makes when it is in attract mode (before game play begins). It consists of tones, buzzes and other sounds that are perfectly timed with the light show on the playfield. The kids and I found it quite futuristic sounding. However, the odd thing was that these sounds were not controlled by the volume control on the left side of the cabinet, and I also noticed that at times (during game play) it would change into a loud annoying hum.
Closeup of one of the screws holding the CPU board.
One day while working on the
machine, I noticed that the CPU board only had only two of its possible
eight 8-32 screws that hold the board down. In addition, the two
that were used were not tight. After tightening the screws and
completing the set, the attract mode music
went quiet. I did not understand why at first and after a few
hours of checking realized that the better grounding eliminated the
attract mode sounds. Also, overall, the machine plays quieter now
the constant 'music' in the distant background. It appears that
the music we were hearing all along was due to poor grounding.
the pin show
Josh Kaplan (aka pingeek) paid
me a visit in July 2006 and shot a DVD of my Space Shuttle machine and
the CPU tester. At his subsequent show, the Pinball Fantasies Show,
he sold many copies of the movie "Saving the Space Shuttle". The
DVD shows the results of my playfield restoration,
and a System 9 board being tested in the CPU tester.
Josh selling his DVDs at the Pinball Fantasies Show. Note my
on the monitor next to him.
July 16, 2005 - Purchased machine from Diane and moved it into the rec room. Installed fresh AA batteries on the CPU board.
July 29, 2005 - From Usenet, figured out how to put the game in free play mode. Surprisingly, this is not in the manual. The way to do it is to change 'Function 18' (Maximum Credits) to zero.
With all diagnostic buttons UP.
Press 'Advance' until 'Ball' display reads 18.
Press middle diagnostic button DOWN.
Press 'Start' button until 'Credit' display goes to zero.
Power cycle game to clear book keeping mode.
August 6, 2005 - On ebay, purchased an original full-color flyer promoting the Space Shuttle pinball game. It hangs in a frame in the rec room.
August 14, 2005 - Purchased a very nice backglass from Greg Deakin (GJD999.at.aol.com) for $160 (+$34 shipping). It has a few very small scratches on the back, but nothing visible from front. Greg was nice enough to throw in a Space Shuttle toy so that I can reproduce them with this vacuum forming table. Incidently, he made a mistake with the shipping charge, but did not charge me the difference. What a guy.
August 15, 2005 - On ebay, purchased a working set of plasma
displays for $21 (+$6 shipping). These displays
cost about $40 each when new.
August 16, 2005 - After I agreed on the above backglass purchase,
I found a second backglass that is in similar good shape. I was
not going to purchase it, but the seller bundled an old playfield, a
set of CPU ROMs (worth $10
each), and free shipping (crated) for $350. The main selling
point is that the playfield was mylared, which would allow a complete
restoration. Details here.
Seller was Dave Astill.
August 18, 2005 - On ebay, purchased a tested and working sound
and speech board with the three speech ROMs for Space Shuttle for $35
(+$6 shipping). The speech chips are typically $10
each, excluding deposit.
August 20, 2005 - On ebay, purchased a display driver board
(located behind the backglass for the plasma displays) for $10 (+$7.50
shipping). This board includes the obsolete UDN7180 drivers,
which are worth about $25 each. Tested them here.
August 22, 2005 - On ebay, purchased six plasma displays and four blank display circuit boards for $27 (+$10 shipping). Tested them here.
August 26, 2005 - Received Dave Astill's playfield and backglass. Installed his ROMs to bring my system to rev L-7 (the latest rev). One small difference with the new rev is that the coin door coin reject solenoid goes 'click' two seconds after you power up the machine. No other differences noticed.
August 29, 2005 - I purchased a second machine on ebay.
really. It had no backglass, no displays, no CPU and no power
supply board. However, it had a complete playfield, three circuit
boards (display driver, speech, flipper power), and the harness.
I had the seller send me these last items. I now have a large
set of spare parts to perform my playfield
restoration. Cost was $75 (+$50 shipping), although some
parts have since been sold off to offset the initial cost. See
Machine that held playfield C.
Click on image for full size.
September 20, 2005 - Sold Backglass A to a person in Australia for $75+$102.
September 29, 2005 - Purchased a promo keychain for my machine for $25+$2.5 from pinballpromos.com. This item is rather rare as it was not sold with the machine.
Promo keychain for this machine.
October 10, 2005 - Purchased main power supply board from a person in Belgium ($11+$24). Board is clean and passed functional tests. All I am missing is a CPU board for a complete set of spare electronics.
October 27, 2005 - Purchased a topper/header for my machine for $30+$5 (from Bruce Johnston). This device is often not included with used machines as collectors like to keep them for themselves. My collection of plastics is now complete.
Header for the machine. It was in pristine condition with the
protective plastic still attached.
These are often offered on ebay with $50-$75 starting price.
July 22, 2006 - Josh Kaplan makes a DVD of this restoration project called "Saving Space Shuttle" along with the CPU tester and the coffee table. It is sold on his site.
August 8, 2006 - Purchased an original manual with all the other paper for this machine for only $18+$2.
Original manual and all the other paperwork for the Space Shuttle.
Quite a unique find.
August 12, 2006 - An ebay
auction on a very nice Space Shuttle closes at $1101 (item 230015636374). This unit is in very nice
shape with only a little wear above the U-S-A lanes, 'T' target and
bulls-eye targets. The backglass looks very sharp, and the
Shuttle toy was snowy white. None of the usual wear on the insert
area. Activity was brisk, with 15 bids.
October 2008 - I was able to get the crew of STS-125 (Fourth Servicing Mission to Hubble) to sign one of my playfield plastics (top left one).
November 2009 - See page 2 for more (link below).
(c) Edward Cheung 2005, all rights reserved.