Renovation of a Space Shuttle Playfield (continued)

See here for Part 1 of Playfield Renovation

See here for Part 2 of Playfield Renovation

See here for Part 3 of Playfield Renovation (this page)

See here for Part 4 of the Playfield Renovation

Ramp Repair

The various ramps on this machine are a single piece of black plastic.  The ones from Playfield B and C had been cracked and some small parts missing due to the numerous ball strikes and repair hacks over the years.  I decided to use the worst one of them to develop a technique for repairing them.  I found that using sticky fiberglass mesh tape and two-part epoxy were very easy and effective.  This tape is normally used for drywall seams, and is very strong and allows the flow of material (dry wall mud) through their opening.

Many ramps on newer machines are clear and can be flame polished to their original luster.  Although the Space Shuttle ramps cannot receive this treatment (it bubbles on application of heat), the black plastic is more forgiving to repairs as the back is hidden, and the repairs can be painted black to match the rest of the ramp.

Repair steps that I used:
  1. Sand the area with rough grit sandpaper such as 60 grade.
  2. Wipe down the area with alcohol to remove sanding dust and oils.  Repeat several times.
  3. Apply sections of mesh tape using the sticky feature to hold them in place.  Use two layers for high stress areas offset by one half the grid spacing.
  4. Apply epoxy by pressing it with a flat blade through the mesh holes to get the epoxy onto the black plastic.
  5. When dry, sand and then paint with flat black acrylic.

Step 3 above.  The ramp entrances get very beat up.  The sticky fiberglass mesh
is very easy to apply and stays put.  I applied two layers here.  On the right, you
can see a big chunk missing from the front of the ramp.

Step 4.  Application of epoxy.  This area was split from right to left.  Probably due to a someone not properly
raising the playfield and pushing the ramp against the metal bracket that holds the field down.

Image of the repaired ramp entrance.  I do not have a "before" picture, but
the end result is very good.  With the double layer of fiberglass mesh
and the epoxy, it feels much stiffer and stronger.

This repair technique was modified to reconstruct the exit of one of the ramps that had a large piece missing.  After some thinking about it, I hit upon the idea of using "Model Magic" to form a mold.  This substance can be purchased in an airtight container, and is soft and moist.  After it dries (takes a few days), it firms up, but does not harden completely.  This is an important feature as I could then transfer the mold to the broken ramp, which could be shaped slightly differently.  After I lined the mold with clear tape as a release agent, I pressed it into the broken ramp.  Then I used several layers of mesh tape and epoxy to reconstruct the missing portion.  After curing, the mold was removed, and I then sanded, trimmed and painted as before.

Reconstruction of the exit spout on the ramp using "Model Magic".
A large chunk was missing when I received the ramp.  The yellow mold was formed
on a good ramp and then allowed to dry.  Unlike the case of the Shuttle toy, Aluminum
flashing cannot be used here as the new substrate because of the complex compound curves.

After the epoxy had cured, I used my mini-benchtop grinder with the handtool attachment
to trim the repair to match the reference ramp.  After this: sand and paint.

Can you tell, which ramp was repaired?  It was the top one.  The missing part
is the left part of the ramp exit.  The black shiny ramp is difficult to photograph.
Since this is the exit of the ramp, and is located high above the playfield,
it will not see any ball strikes.

Also see here how I restored my Indiana Jones idol ruins

And Pinbot ramp repairs

Clearcoating the playfield

In preparation for this important step, I sanded the playfield with 1000 grit sandpaper.  After some sanding it was apparent that the acrylic touchups are much softer and protrude above the existing paint.  As a result, I had to retouchup some places, especially around the inserts in the middle of the playfield.  The lesson here is clear: sand before you touchup.  This was prescribed in the various write-ups I read, but I overlooked that point.  I did not sand very much near the ball locks in order to prevent any damage to the overlay and the touchups.  After retouchup, I sanded more selectively.

I then wiped down the entire field multiple times with a clean T-shirt dampened with Naphtha.  This is the final cleaning to remove all oils, waxes and sanding dust.  I found that Magic Eraser does not work well because the sanding dust does not get picked up.  It will dry in streaks.  I found that the bulb openings in the field are 1/4", and I could use #10 washers to cover them up.  In some places where the #10 washers are not usable, I found that old coil sleeves are perfect for covering up the bulb sockets.  See the left edge of the picture below.  I use the latter method sparingly as they make sanding between coats more difficult.

Sample of the playfield just before clearcoating.  Note the fine cracks in the
paint in this picture.  It took a bit of experimentation to make them
visible in pictures.  Most likely the crazing is superficial.  I think that if I
didn't clearcoat, but only waxed the playfield that these would be accentuated. 
This gave me not much choice but to clearcoat.

After one coat of clear.  The darks have darkened and the fine cracks are much
less visible.  I used the Varathane that is recommended by Clay
One of the overlays I made bled slightly, so I need to remember to
go easy on the first coat in the future.  After only half an hour, the
playfield was dry to the touch.

In applying the clear, I learned that the technique that worked best for me was to make sure the playfield is completely level and then to apply heavy wet coats to allow the clearcoat to sort of flow in a sheet.  Apply enough so that it starts to just cloud a little.  This clouding will disappear and the clearcoat will dry transparent and smooth.  During the spraying process, if any texture remains on the surface, hit that area with another burst. After about half an hour, the clearcoat will be dry to the touch.  The exception is the first coat, as it ought to be applied lightly to prevent the bleeding of the overlays printed on decal paper. 

It is fairly easy to tell areas of the playfield that are even slightly depressed from the others as they will be the last to uncloud.  For example, one can see a thin (0.1") ring of cloudiness around each pop bumper.  This ring coincides with the edge of the circular mylar that protected the pop bumper area.

Between each coat, I sanded with 600 grit (after the first coat, I used only one pass of 1000 grit).  Afterwards, I wiped the playfield carefully with a T-shirt dampened with Naphtha to remove the sanding dust.  On a separate test plank, I found out that the IPA (alcohol) dissolved the Varathane.  I like the thought of a readily available solvent that will leave the original paint unharmed, but that will remove acrylic paint and the clearcoat in case I need to restore the playfield in the future.  It remains to be seen if the Varathane will still dissolve in alcohol after a long time.

Each coat was allowed to dry 24 hours before the next one was applied.  Don't forget to spray the can upside down to clear the nozzle after spraying.  These pictures were shot at the end of the drying period.

Full resolution of the above image (after one coat).  Note the remaining
speckling in the black background and the crazing in the blue windows
of Spaceship 8.  This area looked really bad before any clear coat.
Click here for a picture of this area before clearcoat.

After three coats.  The problems are now noticeable only in high magnification. 
The speckles in the black background are barely perceptible, and so is the
crazing in the Spaceship 8 windows.  The degree with which clearcoat is
able to hide the crazing and imperfections in the paint was completely unexpected.

After spraying five coats, the surface looked great except there were still a few inserts where their edges showed a thin depressed ring.  This was left over from when the playfield was bare, and some inserts had wear around them.  I spent the next two days filling in the edges with a small paintbrush dipped in the clearcoat and then sanding them flat with the rest of the playfield the next day.  I obtained small amounts of clearcoat by spraying into a small glass bottle.  Finally, I sprayed the playfield with a sixth and last coat.  Once I was done, the inserts were flat with the rest of the playfield.

Five minutes after the final (sixth) coat.  Note the slight white clouding.  This shade is my gauge to tell how thick to apply the clear.

This is the clear I used.  Varathane Diamond Polyurethane
in the Clear Gloss version.  The project required two cans.

Shots of the playfield after all six coats have been applied.  After this, the field
will be allowed to dry for three weeks, and then the final buffing will take place
with my random orbit buffer.  If you look closely, you will see the
#10 washers (with tape covering the center hole) that protected the light
bulb sockets from spray.

Another shot of the playfield.  I am very pleased with the final result.  The result have exceeded my expectations.

After waiting three weeks for the clearcoat to dry, I performed the finish work here.

Many thanks to the contributors of RGP for their assistance
with clearcoating, especially Clay Harrell.

Reassembly of playfield actuators

While the playfield dried I started to reassemble the playfield actuators (slingshots, flippers. popbumpers, etc).  I ordered various new parts from and such as coil sleeves, a new rubber kit, flipper parts, etc.  Each of the actuators were completely disassembled, sanded with 400 grit paper, tumbled in the vibratory tumbler overnight, and then reassembled with new coil sleeves.  Some coils were replaced, and the targets had new decals applied to them.

Sampling of the major actuators for the playfield after refurbishment and
reassembly.  "T" target, ball saver post, 3-bank target,
and the two flippers.  Due to missing or worn parts on the playfield
almost all of the parts on the flippers are new.
Note the connector I devised for the switches on the 3-bank drop target. 
This allows me to reinstall it more quickly.

By using the C-clamps and some solid teflon blocks, I can flip the playfield over to work on the bottom side
without fear of marking the clearcoat.  The white mass in the photo is the teflon block.

Pinball Playfield Tester

Since I did not know the operational state of Playfield B when I received it, and also due to the many parts that were removed and refurbished, I decided to build a standalone playfield tester.  If I had miswired a solenoid, it could cause damage to the CPU board of the machine.  Also, the machine is several feet away from the playfield rotisserie, and it would be clumsy to move the two closer for a test.  The tester will allow me to verify every switch, lamp, solenoid and associated circuitry and harnessing.  I was surprised to find no similar device when doing a Google search.

Circuit diagram of the playfield tester.  All circuits on the playfield can be tested by this unit.  Selector switches access the rows and column of the two
matrices just like the CPU would.  In addition, solenoids and flash lamps can be fired individually.

The completed tester on my workbench.  Note the four
playfield connectors on the left and that the
cover overhangs the left side of the box. 
The purpose of that is explained below.

Due to the overhang on the top cover, the tester can be mounted on one of the
beams of my playfield rotisserie, or it can be hung on the
sidewall of my pinball machine's cabinet to test
a playfield that is still installed inside a machine.

As one can see in the above image, the tester features a copy of the two matrices from the user manual of the Space Shuttle.  A particular switch or light bulb can be selected by turning the row and column selector rotary switches to point to that item.  Power is provided by a separate current limited adjustable power supply so that I can limit the amount of current to prevent damage should there be a short in the wiring.

The connectors to the playfield are four 0.062" style Molex.  List of key connector parts:
Digikey Part Number
Mfg. Part Number
Pin Extraction Tool
Pin Crimp Tool
Male Pin
Female Pin
WM1001-ND 02-06-1103
6 pin shell
15 pin shell
WM1238-ND 03-06-1152
24 pin shell
WM1218-ND 03-06-1241
36 pin shell

Note that the printed matrix in the user manual has colors associated with each row and column (see picture below).  It turns out that these colors are those of the wires on the playfield.  So although there was no schematic of the playfield itself, I could tell how to connect the selector switches based on the color mentioned in the columns and rows.  When I first fired up the matrix, I found that few of the lights worked; perhaps 25%.  Considering the large amount of work to be done, I was glad to have the tester so that I could work on the playfield from the comfort of my work chair instead of standing next to the pinball machine.

A particular light can be turned on by selecting its row and column in the
matrix of the user interface panel. 
One of the few lights that worked is shown in the photograph. 
I would agree that this tester is very "SPECIAL" 8-).

The bottom portion of the tester.  This allows me to test every switch on the playfield
including the special switches (the three pops, and the two slings).  When a particular
switch is selected and then closed, a beep can be heard.  If the switch is noisy
or intermittent, the beep will not be clear but raspy.  This helps me find
switches that make intermittent contact and that need to be cleaned.

Any of the 18 Solenoids can be tested by clipping the ground terminal of
that solenoid to either continuous ground or
switched ground.  In the latter case, I can use a momentary
toggle to fire the solenoid. BLAM!

Solenoids are tested by grounding their return terminal.  A momentary switch is provided to allow pulsed operation.  I also intend to test the diodes on all the coils by reversing the polarity of the power and verifying that the coil does not fire.  This is important as an open diode will not suppress the large EMF kick of the coil, and could damage the driver on the CPU board.

The insides of the tester.  I intentionally left space in the bottom
right corner in case I one day
decide to mount a microprocessor in there to
scan the matrices automatically.

Update: In February 2006, I sold the above playfield tester for $60.

See here for Part 1 of Playfield Renovation

See here for Part 2 of Playfield Renovation

See here for Part 3 of Playfield Renovation (this page)

See here for Part 4 of the Playfield Renovation

Main Space Shuttle Pinball page

Back Home

(c) Edward Cheung 2005, all rights reserved.