Despite my DM project from last year's pinfest not really being complete yet, I picked up another inadvisable project at this year's pinfest: a Gamatron

pioneer bg on (resized).jpg

Unloading it I already ran into problems, as the cabinet (originally a Lost World before Gamatron conversion) was in bad shape, with half the bottom falling out:
IMG_20180505_112842 (resized).jpg

I took out the playfield and cleaned out the inside of the cabinet, then glued the bottom back into place.

IMG_20180505_113602 (resized).jpg

Taking the playfield out wasn't as easy as expected, as some of the wires were spliced into the cabinet wiring harness. I just pulled them apart for now (they were just twisted together without even wire nuts), but they (and many more spliced and electrical taped wires) will need to be investigated

IMG_20180506_194152 (resized).jpgIMG_20180506_194104 (resized).jpgIMG_20180506_194059 (resized).jpgIMG_20180506_194107 (resized).jpgIMG_20180506_194147 (resized).jpgIMG_20180506_194054 (resized).jpg

There's enough hackery going on here that I don't feel confident even plugging the rectifier board in to test it as is. I think I'll start by comparing all the wiring with the stock wiring in the Lost World schematics, and hopefully I can identify any specific modifications that were part of the gamatron instructions (thankfully available on ipdb), and attempt to clean them up while reversing anything else I can find.

When I originally read about Gamatron the conversion sounded like a really cool idea. A drop in daughterboard (the PSM) basically containing a stern sound board and all the MPU-200 specific circuitry so you could just plug it into the CPU socket of any bally/stern game to upgrade it as necessary? But reading the installation instructions it becomes clear that the conversion isn't that clean. Besides from the CPU socket, you also need to remove the roms and CMOS ram (which I expected), as well as clip the output pin of the clock generator chip (as MPU-200 uses a faster clock than ballys do), and run a single wire connector from the PSM to the J5 diagnostic connector on the MPU to send that new faster clock back into it.

Splicing the 5+12V from the MPU connector to send to the PSM also makes sense in retrospect, but then you get to things like shifting all the wires in the display connector over one place to enable 'seven digit' scoring, with a fake 0 cutout and an added #44 lamp to light it. Curious to see how that will look in action.

It'd be nice to find a new cabinet, as this one is pretty beat up. The original conversion called for painting the whole cabinet black, then applying some stickers:
IMG_20180505_113531 (resized).jpgIMG_20180506_194032 (resized).jpg

Sadly they seem to have fallen off one side:
IMG_20180505_112943 (resized).jpg

I'll scan them in and hopefully I can get some more made. At that point taking any donor cab and converting it should be pretty simple.

The game came complete with all boards, but the MPU and driver were just sitting loose for some reason. MPU had some mild corrosion, and battery had already been removed (the PSM has a build in memory capacitor stock, which is cool). On the bench, it booted just fine once I replaced a bad cmos ram, but that won't be needed anyway since those chips are located on the PSM board anyway. So why was one in there? Some rom chips were also installed, which made me suspect these were just some random boards thrown in by the seller or someone down the line, but when I checked the roms, it turned out they were from Lost World! Probably the original MPU board pre-conversion, or some weird coincidence? So where'd the converted MPU go? I'll have to re-convert this one, but luckily it does boot, so that's less trouble to deal with. The PSM also looks to be in okay condition (thanks to that memory cap), but the schematics on it are sparse (and hand drawn), so troubleshooting any problems with it will be interesting.

IMG_20180506_194200 (resized).jpg

One problem has presented itself already: the 40-pin connector thing that plugs into the CPU socket has a broken pin

IMG_20180506_194137 (resized).jpg

Looks to be pin 21, the R/W pin on the CPU, so sadly an important one. Not sure if I'll be able to find a replacement for this weird connector. Not sure what to even call it to try to find one, but maybe GPE can save me. If not I'll just have to run a separate wire I guess...

Went through all the connectors today, comparing with Lost World schematic. Besides from the one splicing of the playfield and cab wiring together that I still need to figure out, all the wiring matched the manual + Gamatron modifications. On the other hand most of the cabinet connectors seemed to have been cut off and then another connector from another game spliced on, every single wire twisted together and wrapped in electrical tape. Not sure why they did that, but it'll be easy enough to put in a new connector before the twists and clean it up. All the loose wires in the bottom left look to have been cut out of the connector for the controlled lamps on the insert panel, probably from adjusting the wiring to use the right pins for a different game. Not sure how well this was done, but for now I will just leave this connector disconnected.

Although most connectors will need re-doing, the J2 connector on the rectifier board from the power switch was fine, so I plugged it in an tested the rectifier board, which worked fine, all voltages as expected. The Gamatron manual notes a change to the jumpering of the transformer to make the coils stronger that looks like it's just high tapping, so I will need to check the wiring of this transformer and see if it has been high tapped, and then decide whether I want it to be or not.

I also pulled the PSM to give it a look over, and everything looked clean. I pulled the CPU that was already in it and replaced it with a known good one, leaving the 6116 RAM (since I don't have any replacements or board to test it in) and 2764 rom (since the game will test that itself). I then removed the 5101 U8 rom, ram chips, and cpu from the -35 MPU, and desoldered pin 10 of U16 to prevent the 500MHz clock from reaching the rest of the chips. I plugged in the 40-pin dip connector (minus pin 21), and ran an alligator clip between the R/W signals on each board to make up for the broken pin, but when I powered up the board it stopped with the led on after the second flash, indicating a ram failure. I replaced U7 with a known good chip (despite that the first chip had passed Lost World's self test fine), but that didn't make any difference. Since the rom test passed but ram failed, I figured something was still wrong with my R/W jumper causing the CPU to fail to write to ram (even though it could read fine), and went through triple checking things before realizing that I'd looked at the wrong pin on the schematics. Pin 21 of the CPU isn't R/W, it's just ground, which is already connected via some of the connector splicing, so my broken dip cable is fine. Removing my jumper, the board booted successfully, so it seems they're good to go.

Any idea how the rules compare? Does it still run on bally boards?

Even all the mechs are different, dang. I'd be so cool to get two of these set up next to each other!

Taking a look at the bottom of the playfield to try and map out the solenoids, I noticed these cool flipper mechs with plastic bases. I always like seeing what weird stuff Stern managed to make out of plastic, seems this was the next step after pop bumper rings.

Weirder though is the double EOS switch. Originally I thought it might be lane change or something, but both of them are NC and wired in parallel. Seems like a horrible idea reliability wise, but hard to imagine an op going to all the work of wiring them together with longer screws. Will need to investigate how this was originally.

IMG_20180508_193139 (resized).jpg

Do you think replacing them with a single new EOS would be fine, or should I stick with the 2? These look hammered... They do seem really concerned about the game playing stupid fast though. 3" levelers, high tapped, extra EOS, new cab switches, giant diodes....

That reminds me, I'm still trying to figure out the flipper wiring on the SDB. The manual (step 13) says to do something with some wires coming from the playfield, I think from the flippers, and a few other things (originally self test return and coin lockout?) and swapping wires with J2 that I can't quite follow. My game has these as just bare wires twisted into the cabinet harness, it almost seems like they're bypassing the flipper relay. Do you know anything about how it's wired on yours?

I guess my confusion is, it sounds like they want me to take those four wires that were in the original game's J2 connector and replace them with ones from the playfield harness, but the playfield already has 2 15-pin connectors, one of which matches the keying and wiring I'd expect for J2, including wires matching all the colors and positions specified in the manual, but then there's also another set of those four wires just loose? There's no other J2 from the original game to insert them into either. I think usually J2 would be the cabinet solenoids and J5 the playfield, but there's no cabinet solenoids in this game left, and instead (judging by the F2k schematics) it just uses them all for playfield coils

Went to reinstall the tested MPU + PSM in the head, and realized that the reason the boards were loose in the bottom when I got the game is because there are no mounting plates for it? IMG_20180509_190033 (resized).jpg

Also checked out the wiring for the speaker. The schematics for the sound section of the PSM on IPDB are too blurry to read, but they seem to match the schematics for the sound board in F2K, so based off that I confirmed that two of the wires were going to a volume potentiometer in the cabinet, but strangely the other two wires which should have gone to the speaker.... also went to another, larger potentiometer?

IMG_20180509_190008 (resized).jpg

The game also is missing its stock speaker, and had a random 6x3 car speaker rolling around in the cab instead, but not wired into anything.

Began tearing down the playfield for cleaning:
IMG_20180509_192018 (resized).jpg
IMG_20180509_192024 (resized).jpg

The ball lock area is a nightmare, absolutely horrible design. Especially this plastic, which might be the worst designed plastic I have ever seen:
IMG_20180509_191213 (resized).jpg

Can't believe the thing isn't broken.

Will need to order new lane guides for the top lanes, they're all smashed and cracked for some reason.

Tore down, cleaned, shopped, and rebuild the playfield. Was dirtier than I expected.

32349078_2166403240305566_7713845784277417984_n (resized).jpg

IMG_20180511_192028 (resized).jpg

While putting new rubbers, I noticed there was a lot of play in the flippers, which I hadn't examined yet. I disassembled one, and found that the links were very worn. Not a big deal, I'll just order new plungers+links. Then I noticed the first real problem, which is that this fancy plastic base is also the bushing. There's no removable part, and these are very worn, probably beyond the point of being enjoyably playable. There's actually two small brass washers under the flipper bat on top of the 'bushing' top to give it space to not drag the playfield, although I don't know if those are stock or were added by the operator as the top of the bushing got worn down.

Further investigation revealed that the coil stop is not like any other coil stops available, and thus is also not replaceable (not a big deal though, it looks to still be in good shape), however the plungers are also 1/4" shorter than the replacements I have on hand. I don't know of a different length ever being used, so these may also be specific to this new flipper design? I double checked the flipper throw however, and it's the same on 2nd gen Stern flippers as on these '3rd' generation ones. It looks like I may need to order two complete new mechs though, as these don't look like they'll be salvageable, which is too bad.

Another weird post-stern oddity:
IMG_20180513_150341 (resized).jpg

How the hell do I get this pop bumper cap off? There's no screws! I tried squeezing it various ways like a gottlieb, but nothing seemed to work. Maybe I'll get lucky and the light will still work... The plastic pop bumper ring is luckily still intact, since those aren't available anywhere either, although they've still got the annoying design where you can't replace the yoke

Doesn't seem like it. It has some wiggle to it. Unless the entire body is loose... There might be some kind of tab, but I can't see it well enough to figure out how to release it. And can't remove the assembly to get a better look because the cap is on it.

'Build'? Do you have just the PF or?

Curious that the sound is different, I wonder what the difference is... The schematics on ipdb are very hard to read, it'd be great to have a better scan. If I did it might be worth trying to recreate the full schematics (the non-sound part seems very simple) and then you could get a new board made or something...

Ah, they do seem to look the same to the ones on Af-Tor on IPDB. Sadly I can't seem to find any flipper parts for it either...

Another interesting oddity: noticed one of the three slingshot coils was different than the others (two are original pinstar). The third was a Wico replacement. Didn't think anything of it at first until I recognized the winding count: it's a williams slingshot coil! Eg, 25V, not 50V. That explains why the kicker arm on that side was broken off... Looked up the right coil (thankfully the windings are specified in the manual), expecting it to be a standard bally sling coil I could just steal from a parts PF I've got, but it's actually weaker than that. Williams was about 5ohm, bally 10, gamatron 15. I would have expected, if anything, that it'd be a bit stronger, like the flipper coils are.

Fixed up all the connectors. Resolved that the wires that were originally tied together between cabinet and playfield harness probably were supposed to be like that and added a connector for them, as well as the spot where the game over, highscore, etc lights (cabinet) go into one of the playfield lamp connectors. Older games had one connector for cabinet and one for playfield, but over time more and more things were needed on the PF, so those connectors now go there instead. Still can't understand the instructions in the manual, but this way the wiring makes sense.

Put the included driver board in and powered it up to test the voltages. Measured 4V at the 5V and 10V at the 12V (HV was fine), so it looks like something is pulling it down. Didn't have any LM323s on hand to try rebuilding (though I guess ti could have just been a bad capacitor), so I swapped in a new Alltek board. All voltages now checked out, though I'm a bit annoyed that the fancy Alltek board has its own dedicated CPU to monitor things but couldn't be bothered with a pot to adjust the HV level. It seems to be putting out 182V, while I prefer to keep mine around 170...

Hooked up the MPU and it booted up fine, so I plugged in the player 1 display as well, turned it on again. Led was on solid. Rebooted a few more times, stayed solid. Weird. Put MPU back on the bench, booted up just fine. Put it back in the game, worked, got digits on P1 display. Rebooted, led was solid. Checked voltages, all fine. Very confused now. Replaced sockets that connect PSM and MPU together, replaced J4 header, same symptoms. First boot works fine, after that it won't boot until you let it sit for a while. Swapped in a regular MPU from another game, it boots fine every time. Put Gamatron MPU+PSM in the other game, boots fine every time. Swapped Alltek driver from Gamatron to other game and other game's (rebuilt) driver into Gamatron, now Gamatron boots fine? Grrrr.

Got creative not wanting to leave it sitting for a while while I wait for MPU mounting rails:
IMG_20180518_193739 (resized).jpgIMG_20180518_193749 (resized).jpg

After that, I plugged in all the connectors and booted up again, getting to attract mode. Lots of lamps out of course. The self test switch worked, which means my added connector is probably right. All coils fire in coil test. My credit display isn't showing any digits for some reason despite the display being good. Amusingly, the coil and switch numbers are missing the first digit (since it's just a light up zero). Great thinking there...

IMG_20180518_194957 (resized).jpg
IMG_20180518_194932 (resized).jpg

Next will need to wire up the speaker and track some balls down

Connected up a spare speaker and sounds work fine. Installed a replacement kicker for the right slingshot and some MPU rails a fellow pinsider sent me. With some tweaking of switches and a lot of new lamps it was up and running! Surprisingly all the lamp sockets worked, which was nice. Credit display turned out to be a bad pin in the connector for the 5V.

Need to track down some more displays and order some parts to rebuild whatever parts of the flippers I can (probably just sleeves and links), plus EOS and cabinet switches, and then look into getting new stickers made for the side.

15271209324221899546020 (resized).jpg

Gamatron has been doing well. I put it at a local barcade and some months it out earns some modern Sterns. Last month it lost all its memory for some reason, which seems like it shouldn't be possible given the memory cap on the psm

No, didn't think of that. This is on every day though so I don't think it should be an issue?