 I wish this blog post meant that I had bought an Attack from Mars (Bally, 1995) and was restoring it for my own pleasure. But alas, no! This was a mini restoration for a customer who had purchased the game, taken it apart in preparation for restoration, but was never able to complete the project. So, I offered to help and get the game back together for the first time in over ten years. This project was a little different to previous restorations, as I was starting with a blank playfield and boxes of parts, with no reference photos of my own. Plenty of fun to be had. It helps that Fiona also loves Attack from Mars, so this was a good chance to give it a thorough playtest! The game had already had some work done to it by a previous owner. The cabinet had new decals, and looked great. The cabinet hardware (legs, coin door) also looked good. So the focus of this restoration was going to be on the playfield, and the electronics, as nothing else really needed to be done Cabinet
Great condition overall.
Above Playfield
Poor condition overall.
Under Playfield
Good condition overall.
Electrical
Good condition overall.
While the game did boot up, it was going to be difficult to actually determine what problems were present until the playfield was back together. All I could do at this stage was test a few playfield mechanisms, lamps, and switches. Some of these were not working, which required some further investigation. The playfield looked to be in average condition, with some inserts having been damaged and repaired with insert decals. Some playfield wear around the Stroke of Luck scoop had also been bogged with wood putty, and repainted. Playfield artwork repairs were not really the focus of this restoration, so I didn't worry too much about these areas. There was no real disassembly stage to this restoration, as it had already been done for me! So, instead, I set about cleaning all of the parts in the boxes I had been provided, and laying them all out for assessment, pictures of which appear below. Then, I could begin to piece together what was there, and figure out what repairs were needed.
 All provided parts laid out for assessment. After parts stocktake, the game went through my standard restoration process to get it playing and looking like new. During the restoration process, I dealt with a number of issues, described below. Tips & Troubleshooting (click on sections below to view details) Coin door constantly always registering as open
After booting up the game and letting it go into attract mode, the game would immediately flash a message on the display saying that the coin door was open even though it was closed. When you open the door, you'll notice there are two buttons on the frame of the coin door at the bottom left. The upper button is the high voltage cut off switch, which kills power to the coil and flashlamp circuits when the door is open. The bottom switch is the memory protect switch, and is the one connected to the game's MPU so it can determine whether the coin door is open or not. Both buttons are described in the first pages of the game manual.
Excerpt from the game manual describing the interlock switch assembly by the coin door.
Some fiddling around with the switch assembly eventually revealed the issue. The green-red wire connected to this switch had broken. This wire connects the switch to the coin door interface board (J9-5). With this wire broken, the switch was hidden from the CPU. A bit of solder and heat shrink tubing fixed the issue. The coin door now registered as closed like it should.
General illumination circuit upgrades
I had some issues with the general illumination lamps both on the playfield and in the backbox. Let's start with the backbox issues.
I hate WPC-95 backboxes with the plastic insert panel and IDC-style sockets. These sockets are terribly unreliable, and are often fragile or broken. I had to retention a number of the sockets as they were holding onto the lamps too loosely. The backbox lamps are on three separate GI circuits (green, violet, brown). None of the lamps on the green circuit were lighting up . The lamps themselves were OK, and the sockets didn't have any physical issues. Looking at the power driver board, I found the connector at J106, which supplies the backbox lamps, to be pretty badly burned. The white-green wire had fallen out of its slot, which was causing the lamps on that circuit not to light. I replaced the connector, as well as the general illumination J105 connector for good measure. The lights on all circuits then worked fine.
At this point, I thought it was worthwhile removing the power driver board to replace the header pins for J105 and J106, as I had just replaced the connectors and there was no point only replacing half of the connection system. I replaced the header pins, and while I was at it, I also removed the general illumination lamp diodes that are also in this area. This is a recommended upgrade as per Pinwiki. Jumper wires on the back bypassed the (removed) diodes, so this part of the board won't get any more charred than it already is.
Then the next general illumination problem turned up. The symptom of this problem was a blown F114 fuse, which supplies the white-brown general illumination circuit on J105. According to the lamp table, this circuit services the lower playfield general illumination lamps. It helped to disconnect J105 and J106 one at a time to narrow down the issue to the playfield instead of the backbox. It took me a while (and a couple of fuses) to finally figure out the problem. The two AC supply wires for this lamp circuit were shorted together at one of the lamp sockets on the lower left of the playfield, near the inlane. This was blowing the fuse. I moved the solder tabs further apart so they were no longer touching, and the problem went away. 
Lamp socket with shorted supply wires.
3-Bank moving target assembly not registering hits
The 3-bank moving target assembly sits in front of the saucer, and hits to the targets lower the bank to allow shots into the saucer scoop. It is the main playfield mechanism on Attack from Mars, and if it's not working, the game is basically unplayable. (Un)fortunately for me, none of the targets were registering hits during the game, while all other switches in the game were operating normally. That pointed to an issue with the assembly itself, so it was time to do some disassembly.
The target assembly comes out of the playfield easily, however it is easier to remove the motor assembly first (four screws) before removing the target assembly (two screws). I found it best to remove the lamp board adjacent to the motor assembly to make things easier. Once the assembly was out, I noticed the problem. None of the switches were actually wired up. A single white-red wire was attached to one of the switches, but that was it. There were no other wires or connectors to actually interface the switches to the game, and the single white-red wire was also broken. 
3-bank target assembly after removal. Single, broken white-red wire visible.
Time to check the switch matrix in the manual. The 3-bank target switches are in the same switch column, which is supplied by a green-yellow wire. This should be connected to all of the switches on the banded side of the diodes. The target switches are connected to three separate rows (white-green, white-blue, white-violet). So the left, centre and right switches should be wired to those, respectively. First, I had a look around the cabinet for a connector that had these wire colours. It didn't take me long to find it, just above and to the left of where the assembly usually sits. This was a four-position connector with all column and row connections needed for the 3-bank. It looked like the wiring harness that connected the targets to this connector had simply vanished. It was not in the boxes of parts I received with the game, and was nowhere to be found in the cabinet. 
Switch matrix with 3-bank targets highlighted.
Luckily, I had a similar wiring harness in my spare parts. Of course, the wire colours didn't match, but I'll put a label on the harness to indicate that the colours aren't original. If I didn't have a spare harness, it would be easy enough to make some with the wires, a 4-position plug (element14) and pins (element14). All I needed to do was cut off one end of the harness, and solder it to the appropriate switches. I also installed a small wire "C" clip to make sure the wires don't get caught up in the carrier mechanism.
At this point, the switches were working again and registering correctly. Yay! 3-Bank motor assembly repairs
So, I had fixed the target assembly (above) but I still had another problem with it. The "moving" target assembly would not move! I was also getting an error message relating to the "up" switch in the motor assembly being broken. This was because the assembly was stuck in the "up" position, and the motor wasn't able to move it out of this position, hence the error report. To be clear, the switch was working fine, but as the assembly needs to move to open the switch, the game assumed it was faulty and threw up the error.
I already had the motor assembly out, so I hooked up the motor to a 12VDC power supply. No dice. I could see the motor engage as it if it was trying to spin, but there was no actual movement. I couldn't move the lift cam manually, either. So, time to do some gearbox servicing! Getting to the gearbox is a breeze on this game. Remove the lift cam (grub screw), unbolt the gearbox from the mounting bracket (two screws), and undo the screws that secure the gearbox casing (three screws). That's it! You're in! The gallery of photos below show the stages of disassembly and the positions of the gears. Once I started taking the gears apart, it became evident that the old grease had gummed up and solidified. None of the gears were able to move. Once of the spacers exemplified this well; the grease had solidified into a thick layer that came off it in one piece. 
Solidified grease on one of the spacers.
Once I got down to the base of the gearbox, I could undo the two screws that held the motor in place. With the motor free from the gearbox, I could test it again with my power supply and verified that it did indeed work. So, the problem was definitely the gears. Then came the job of degreasing and cleaning each of the gears and spacers until they were nice and shiny. Then, a new coat of grease over everything to lube it up again. Once it was back together, I hooked it up to my 12V supply again and the whole thing came to life. We had movement! I reinstalled everything into the assembly, and mated it with the moving target bank. The bank now moved up and down like it should. Motor repair success!
Mini saucer LED mod repairs
A very common mod to do on Attack from Mars is Bill Ung's mini saucer LED mod, a very well designed mod that integrates a series of LEDs, installed on the six mini saucers on the playfield, with the game. They flash and cycle in accordance with the current game mode, and look great. It's a fantastic mod, but a very complex one from a design standpoint. It has an entire controller board that you install in the backbox to control the LEDs. Not many mods have that level of design effort put into them!
This game had the "Temu version" of the saucer LED mod. I spent a couple of hours studying how it worked and wiring it back up correctly so that it would operate properly, as it was missing power cables and had some wires broken when I first got the game. Like the Bill Ung version, each saucer on the playfield has an array of LEDs installed in the mould holes around the flasher dome. However, with this mod, the LEDs are connected to a 555 timer and a CD4022 octal counter integrated circuit mounted under each saucer. The basic design of the circuitry associated with each saucer is described here. It's a simple circuit which pulses each LED in sequence, like a chase light. It's a cool effect, although the implementation (i.e. how it is installed under the saucers) is a bit haphazard.
Each of the saucers are connected to power via an RCA cable. In fact, any simple wiring method with at least two conductors will work for this mod. I found one of the saucer wires had been cut, probably so the ramp the saucer was on could be removed from the game. I installed a connector on it to make removal easier in future. I also had to resolder a couple of wires and LEDs within a couple of the saucers, as the wires are very thin and fragile. Once I got this done, each of the saucers was getting power and the LEDs were lighting up as they should. I did need to buy an additional 2-in-1 RCA cable (Jaycar) as I was one power supply socket short.
The RCA cables are all supplied power via an MCT7805 5-volt voltage regulator, which was soldered to the nearby opto board under the playfield. This supplies the regulator with a 12-volt input voltage and ground. Each saucer is then simply connected to the regulator output and ground, with the saucers being daisy-chained together.
Flipper rebuilds
Each of the flippers got a rebuild, too. After a few play tests, one of the flipper bats failed catastrophically, so I replaced both of them for good measure. The old flipper shafts were also glued into the crank assemblies (what the hell?) and needed some real tapping force to free them. After the rebuild, the flippers were nice and strong.
I posit that the FL-11629 coils used here are actually a little too strong. They send the balls into the "T" and "I" martian targets at high speeds which cause the balls to deflect off the targets and go airborne. A cool effect, to be sure, but I can see why these targets, and the Stroke of Luck scoop, are commonly broken on these games. Martian figurines spinning around
The martian figurines are the signature feature of Attack from Mars (even more than the saucer, I would argue). They are hilarious when they go off during martian attack mode, jumping all over the place. This machine was actually missing a couple of martian figurines, as well as the brackets they attach to. Luckily, both the martians (PSPA, RTBB, Mr Pinball) and the brackets (PSPA, Mr Pinball) are available from multiple sources.
While the jumping around is cool, there was a problem. The martian on the left side of the playfield was actually spinning around on the shaft when jumping around, such that he would turn into the wireform ramp and block the balls. The martians are not supposed to spin around like this; they are meant to be locked in place so they only move vertically up and down, but stay facing the one direction. The assembly is constructed such that the martians are mounted to a long shaft that is kept in place by the main mounting bracket/weldment. This weldment has a notched hole while the shaft sits in. The shaft is also notched, so that it cannot "spin" in the hole. However, over the course of many years, the shaft wears out the hole in the weldment, making it wider. This provides enough space for the shaft to rotate, as the notch is basically worn away. So, the shaft will spin, and the martian will spin with it. One solution is to buy a new bracket/weldment (PSPA). However I decided on a simpler solution. I cut some sheet metal to the same size as the topside of the weldment. This new piece would sit between the weldment and the underside of the playfield. Then, I cut a new hole in the piece which matched the size and notch arrangement of the shaft. In effect, I was replacing the worn out hole in the weldment. After some test fitting and drilling new holes for the mounting screws, it was done. I had to make up three plates in total, as three of the martian assemblies were spinning around. Once the new plates were in, they stopped spinning around. Similar pre-fabricated plates are also available on eBay.
Broken martian switches
The MARTIAN switches, particularly the T and I switches, get absolutely hammered. They are hit straight off the flippers, and the flipper coils in this game are a little too strong in my opinion. This results in these switches getting smashed to bits, as well as the mounting holes in the playfield for the switches getting destroyed. The T switch on this game was broken, and the front face of the switch was falling off. Replacement is the best option here (PSPA, Mr Pinball). I transferred the decal and applied it to the new target with double-sided tape.
Ramp spacer replacement
One of the spacers that holds up the left wireform ramp was damaged. Both ends of the spacer are threaded, which allows is to screw into the playfield, and allows a nut to secure the wireform ramp to it. The top thread on this spacer had been destroyed. Initially I cut off the broken thread, ground down the top of the spacer, and tapped a new hole into it (8/32" thread). However, I found the same spacer in my spare parts pile, so it was easier just to replace the whole thing. These spacers are no longer available, so your best bet is to combine some smaller spacers to the required height. There is some interesting discussion on this particular spacer on Pinside.
What causes this spacer to break? I think the coil in the left popper is way too strong, because it kicks out balls with enough force onto the wireform ramp that the whole playfield shakes. This much force has to go somewhere, and the thread on this spacer happens to be the closest anchor point that absorbs the brunt of the force. Loop gates missing parts
The top of the playfield features two gates. These are coil-activated gates, which are normally one-way gates, but can open up to be two-way gates when the coil activates. Both gates were present on this game but were missing parts. Namely, the solid, sping steel gate flap. This is the flap that is actuated by the coil to allow balls to pass in both directions. These flaps are important, because balls are only sent to the pop bumper area when one of these gate flaps blocks them from going all the way around the orbit. Without the flap, no more pop bumper action. Sad!
Complete gate assembly on the left, and gate assembly missing the flap on the right.
The only difference between the left and right gate assemblies is the position of the flap, and the shape of the gate wire. The flap and wire can be swapped around to turn a left-hand gate into a right-hand gate. Parts to do this are available, including the flap (Mr Pinball), left gate (PSPA, Mr Pinball), right gate (PSPA, Mr Pinball), and swaged hinge wire (PSPA, Mr Pinball). I bought all of these parts, installed the missing parts, and the gates were operational again. 
Complete gate assemblies reinstalled into the game.
Centre trough switch issues
The switch in the centre trough (below the saucer) is a critical switch in the game. If it's not registering properly, you can't destroy saucers, which is the whole point of the game. The switch in this trough was not registering hits, and balls were occasionally getting stuck in the trough. Not good! I realised that balls were getting stuck as soon as they fell into the trough, and were sitting just behind the drop target instead of rolling towards the popper.
First, taking out the trough to have a good look at it revealed the first potential problem. The switch slot had been dented by thousands of balls hitting it over the years, creating a bit of a divot. The ball would then sit in this divot, and couldn't get out. A hammer and punch tool to bash the divot back to a flat plane solved this problem. 
Divot beaten into the switch slot of the centre trough.
But the issue remained, and the ball kept getting stuck in the switch slot of the trough. Further investigation revealed that the rollover switch actuator was not being depressed by the ball, and it was "holding" the ball in place if it dropped into the trough at a certain angle. The bend in the actuator creates a high point, which the ball can get stuck behind. Because the switch is mounted leftwards, rather than rightwards, simple gravity and momentum was not enough for the ball to roll over the switch. It needed an extra push to overcome the natural resistance of the switch mechanism (which is minimal at the best of times). 
Switch mounting position and actuator shape. The apex of the actuator is too far to the right.
Ultimately, this switch mounting bracket is poorly placed, and the switch actuator is the wrong shape. I had to flatten out the switch actuator so that it covered more of the switch slot, which meant that balls would fall on top of it and close the switch more easily by falling into the trough. It's much easier for gravity to do all of the switch closing work here, compared to the minor force of a rolling ball. 
Switch actuator after bending to a more reliable shape.
Installation of Stroke of Luck hole protectors
Cliffy Protectors are available for Attack from Mars. However, this game came with a couple of homemade protectors made of brass for the Stroke of Luck hole. So, I installed these on the front and back sides of the hole. First, I needed to add some timber repair filler to repair the damaged wood and provide a flat surface for the protectors to mount to.
After some initial fiddling around with proper positioning and some bends and adjustments to the protectors, they were ready to fit. There was a hole cut into the front protector already, but I needed to add some holes to the rear protector which covered up a couple of mounting holes for a ball guide and a post.
Once they were installed, they covered up the damaged areas reasonably well. A quick and cost effective alternative to brand-name protectors! Ball guide repairs and adjustments
I needed to adjust the metal ball guide on the left side of the playfield a little. Balls coming down this orbit would strike the rubber post, which sent them straight down the middle. The ball guide needed to be bent towards the centre of the playfield to keep the ball away from the post. Some pliers and protective matting let me bend the guide without damaging it. 
Ball guide which needed adjustment so balls wouldn't hit the rubber post.
One additional ball guide repair was needed. The mounting tab for the ball guide closest to the shooter lane had broken off. The autoplunger coil bracket was misaligned, which meant balls were being fired to the left side of the shooter lane instead of straight up it. The ball guide was taking most of this force, which eventually snapped the mounting tab off. A few small welds got it back together again. 
Ball guide mounting tab after repair.
Left slingshot not working
This was an easy one. The left slingshot was working intermittently. On closer inspection of the slingshot assembly, I found that the coil winding had become detached from the lug, creating an open circuit. The coil windings are coated with enamel; sand this off to create a substrate for the solder to stick to. Then, simply reattach to the coil lug. However, be careful if resoldering wires to this lug in future as the coil winding may detach again.
Balls getting stuck in trough
This problem manifested primarily during multiball, and resulted in no additional balls being kicked out onto the playfield. Which, by the way, is the whole point of "multi" ball. Closer inspection showed that balls were being held up in the trough at position 2, and not falling to the bottom of the trough so they could be sent to the playfield by the trough kicker coil.
This was caused by divots in the metal trough trapping the balls and preventing them from rolling down the assembly as they should naturally be inclined to do. A simple, effective solution to this, short of purchasing a brand new trough assembly (RTBB, Mr Pinball, Pinball Haus), is to purchase a shim (RTBB, PSPA, Pinball Haus) which slides into the trough and sits below the balls, creating a smooth, even surface that helps the balls roll down towards the trough coil. This solved the ball hang-ups in my case. 
Ball trough shim installed.
Reassembly As this game had been disassembled before I go it, reassembly was interesting because I had to reply on my own knowledge about how Williams and Bally playfields were constructed to put it back together. I found some detailed teardown pictures which also helped the process. I noticed one main issue with the way the game was put together that needed rectification. The backbox lamp insert panel was a little odd. It connected to the lamp connector on the right side, whereas most games connect to it on the left side. Also, the positions of the lamps seemed a little "strange". The flashers, in particular, were not where they were supposed to be. Several flashers are meant to light up the explosion on the bottom right of the backglass, but these were not present at all. Turns out, this lamp insert panel was actually installed upside-down. Someone had replaced the plastic trim on the edges, and had put the panel back on incorrectly. Flipping it upside-down put the flashers back into their correct positions relative to the imagery on the backglass, and also put the connector on the correct (left) side of the game. This post on Pinside helped me figure out that mine was the wrong way around.  Backblox lamp insert installed upside-down. Conclusion Attack from Mars is always at the top of most peoples' most wanted pinball lists, and I can see why. It's a damn fun game. Destroying saucers is one of the most fun things to do in a pinball machine, and the saucer is one of the best integrated toys in pinball machines of the 1990s. Plus, the jumping martians are a lot of fun (particularly when they're not spinning around and blocking the balls!). This is a very quick and deadly game. Shots to the centre bank or the martian targets send the ball straight back to the flippers. Ramp and orbit shots are fast and mean. A good game for fast thinking players, and those who like to "run and gun". However, it's not a particularly deep game, and after a while the game can feel a little like chopping wood due to the repetition of certain shots. The 3-bank targets and the saucer are the only real shots you need to make to get to the end of the game, so it's probably not a good game to have in a small collection. But it's the perfect game to just pick-up and play, particularly when you don't want to spend a lot of time progressing through modes and navigating a deep ruleset. The customer was very happy with the finished product, not least because he could finally play one of his favourite game after ten years of it sitting idle in his house in pieces!
0 Comments
Leave a Reply. |
About
Here you will find logs of our pinball and arcade machine restorations, repairs, discussion about general pinball and arcade topics, as well as recounts of our random pinball adventures.
Check back regularly for updates! Blog updates
Archives
November 2024
Categories
All
Donate
Running this website is a hobby for me (just like pinball!). I like being able to show off my restoration work so everyone can learn from it and potentially fix their own machines. If you enjoy reading the site's content or it has been helpful to you, please consider donating to offset some of the website's operating costs. |
RSS Feed