Im having a troublesome issue with my Stargate.

The good:

Tournament mode works well, and the game is set on free play.

Mostly all LED lights.
New shooter lane switch, sarcophagus switch, outhole switch, trough switch.
A "light" spring was added to the trough gate to keep balls from bouncing back and triggering false ball counts.

New outhole coil and ball release coil.

New cabinet flipper switches.

Flippers completely rebuilt with complete kits from PBR.

Optos checked and cleaned

Drop target assemblies cleaned and inspected.

Kicker targets cleaned and rebuilt.

All connectors in the game have been reseated, all chips reseated, V4 game ROM and V3 DISPROM installed.

ColorDMD installed with German repro "clear" speaker panel plastic.

Sound is crisp and minimal hum from the cab speaker.

The game plays fast and powerfully and is a "joy" to play!

All switch test and lamp tests are perfect.

The bad:

Problem #1:

During extended gameplay, during multiball, the game loses track of the ball count.

The multiball modes end early, with as many as 3 balls still in play.

The flippers die and the balls drain out.

The it sets up ready for the next ball.

This doesnt happen in all games, and it happens on random mutliballs, sometimes twice in the same game.

It is mostly apparent in a Ra's bracelet mode when all 4 balls are in play, drain one ball and the mode ends draining the other 3.

Again its random and can happen in the glidercraft multiball but only have 2 balls during the mode but when 1 drains the flippers die and the mode ends.

This never happens in a short game, but always happens in a long "great" game.

This happens with the original trough switch as well as the new one recently installed.

It does not seem to be a TILT or a SLAM as there is no indication on the screen.

Problem #2 (could be related to problem #1)

If I take out all 4 balls and replace them "one by one", the outhole solenoid hesitates for up to 3 seconds to fire on the first ball but seems fairly instant on the last 3.

The outhole coil is plenty strong (and its the correct coil and resistance is perfect on it). but it isnt strong enough to clear 2 balls in the lower side of the trough.

It seems that the outhole kicker might also die when the flippers die?

This is somewhat of a problem as, when the mode ends early and flippers die, sometimes 2 balls can get trapped in the lower part of the trough and then the game is unplayable until I remove the glass and manually help one of the balls over the top to the higher side.

Wall voltage is a fairly continuous 117v to 121v.

I havent done any "ground" mods.

I have gad the game for a few months and this behavior is recent.

Any insight to this would be helpful

Not helpful.

Thanks for the bump!

New developments:

I left the game off overnight in my unheated but insulated shop.

Temperature was 60 degrees.

Turned on the game, waited 5 mins to warm it up.

I played 2 pefect games with no issues.

I left it sit powered up for 2 hours and allowed the room to warm up a bit.

Played one game and had 1 issue of a multiball ending early.

Waited 2 more hours,, played 2 games and each had at leat one mutiball failure.

Waited 2 more hours, all multiballs had failures and the game was unplayable.

Sound like a temperature sensitive issue,

Ill do the ground mods on monday and report back.

Meanwhile I ordered a new sytsem 3 MPU board from Barakandl/Weebly off of Ebay.

There are 2 available, one from Boston Pinball and the one from Barakandl/Weebly.

Im sure that they are both good products but the Barakandl one was $50 cheaper.

It will be here next week to try out.

Ill be doing the extreme ground mod tomorrow (cutting off connectors and installing soldered ring connectors)

My optos test properly, but one never knows what happens under load.

Id be inclined to just replace all of them at this 30 year old point in time.

One thing I did (testerday) was inspect the auto launcher and get the ball to rest on "it" instead of the tip of the shooter rod.

It was easy to replace the barrel spring and add a shooter washer between the spring and shooter housing.

This pulled the tip of the shooter rod back enough so that it sits about half a millimeter behind the launcher arm.

I also squeezed and adjusted the launcher arms so that they were centered on the ball in the playfield gtoove.

The launcher now, always makes it up the ramp with very few roll-backs.

I was reading about how an old pyramid motor/gear box, draws more current than a new one.

Extra current draw is a bad thing in a matrixed system.

Old motors seem to draw more current if they are just "gummed" up and need cleaning.

No replacements are readily available and I doubt that the windings are shot, but who knows!

When I have it apart I will certainly silicone grease it and clean out the electrical motor with spray.

I was reading that the nylon outer caps on motors can be removed and totally clean out old "gunk" and basically re-new the motor.

We'll see if I actually do this instead of just buying a new one, down the road.

I did look at this!

Sadly, nothing was acting up.

Its possible that the grounds are interfering with the data returns to the MPU board?

Or, its also possible that the MPU board might be having trouble "seeing" fast switch actuations, due to faulty connections or impending component failure.

I would like to hear about this more.

I will have more info, good or bad, tomorrow.

I removed the little ground board.

I absolutely agree that having the ENTIRE game running thru two little foil traces seens crazy.

Thay said, my board is in PERFECT shape, no cracks/hacks or wear at any level.

Most I can say is that the grounding areas are very slightly oxidised.

I seriously DOUBT that there are any severe grounding problems on my game.

20221114_130501 (resized).jpg

20221114_130511 (resized).jpg

Just to future-proof my game, I decided to go ahead and remove the connectors to the ground board and place them on the ground plane of the transformer.

I stripped them, twisted them together, soldered them together, crimped them onto the ring terminals and then soldered them to the ring terminals.

These ring terminals accept solder fairly readily with a drop of flux.

They are absolutely ROCK SOLID.
20221114_153555 (resized).jpg20221114_153726 (resized).jpg20221114_153826 (resized).jpg20221114_163011 (resized).jpg

20221114_164013 (resized).jpg20221114_162956 (resized).jpg

After the ground mod I played a few games.

The game plays "more sprightly"!

Game one:

Short game with no malfunctions/

Game 2:

Longer game, no extended multiballs. no malfunctions.

Game 3:

Longer game a couple 2 ball multiballs, during glidercraft the game ended early after draining one of the 3 balls in play.

Flippers died, balls drained the game ended with 2 balls still in play.

So, as I expected, the ground mod did nothing really (except maybe playing a bit more brightly) and Im back to suspecting the MPU board.

Yes, nothing out of the ordinary,

My thought is that it may be some sort of logic issue?

Im hoping that replacing the MPU board will have a positive effect.

I dont game any spare Gottlieb parts, so Im not adverse to picking up a spare CPU and driver board,

The new boards have a "host" of features including NVRAM and no battery.

I really like the new power supply boards too.

Bump

Any Ideas are good Ideas at this point !

My new MPU board arrives today.

Ill install it friday and report back.

If this solves everything, I would be surprised!

It would just be too simple...

I have discounted current draw from the motor as being an issue due to the game losing the ball count when the electric motors are not running.

Solenoids on the pivot targets are something to be explored.

My game does not have extensive plays on it. A lot for sure, but its not beat down to rubble like some of my other project pins.

So, Im not thinking that the smart switches (or optos for that matter) are worn out yet.

They seem fast and accurate in tests.

While waiting for the new MPU board, I was looking at the strobe lines for the lamps and switches.

Its quite a system!

However its not too bad to check the switch and lamp rows and columns at the same time.

Strobe 2 seems to be the common strobe.

Certainly, I guess a bad diode on a lamp could cause false closures in a row or column.

But, If that were the case, I would think that I would see additional lamps lighting up in single lamp test.

The game has all comet sunlight 2 SMD frosted lamps and 5 SMD tower flashers as well as some 8 SMD tower flashers for the #67 lamps.

I dont see any LED ghosting in the system at all.

I was thinking that ghosting might affect the switch performance, but that hasnt been reported anywhere yet.

And, Im not seeing any at this time.

Im thinking a lot about power today.

Specifically the 5v and 12v systems.

Im also thinking about resistance and age in older architectures.

I imagine that pumping up the 5v via the trim pot might have beneficial results.

At the same time, Im working the problem from the bottom and the top at the same time.

My feeling is that if all switches are good, then there is a power/signal/return problem when the game is under load.

I will replace the 12v bridge rectifier and the 10k cap on friday as well as install the new MPU board, since I have the parts in stock.

Ive already replaced the other 2 rectifiers just because I had them (the old ones tested good).

My thoughts are that an anemic 5v, with low current flow would probably manifest itself as phantom switch closures and erratic performance.

I ordered a Boston Pinball 5v supply to add into the mix just for future proofing.

I will likely order a driver board next.

I really think that if replacements for 30 year old boards are available (and they actually work) one should get them.

My experience is that older electronics are not as stable as newer more robust parts.

Im not really a fan of "cute" authenticity, if the system works the same.

Here is a small note from pinrepair.com:

"One chip that is critical to the timing of the CPU board is at U11. This chip must be a 74HC123AN or the game will not boot (or will constantly reboot). U11 must be this exact HC variety (a 74LS123 will not work)."

"74HC123AN: used on the CPU board at U11. Note this chip *must* be this exact HC variety (the AN suffix is important too so it appears). It's a strange situation where if this chip is suspect, it can cause boot up problems (or issues when the game is warm.)"

I'm definitely having issues when the game warms up.

If swapping the MPU board solves the problem then this might be the suspected failure.

The game plays flawlessly when its cold.

I received my new MPU board.

I really like it, it boots up twice as fast, and now I can actually see the eprom version again on my colorDMD.

Oddly the first of the 2 startup beeps/squeals goes by so fast its just a slight pop.

The game processing is faster and cleaner, no more lag!

The new board has NVRAM, so no more battery!

I was going to mod my original board with a coin battery but this is way better, and again, future proofing...

Sadly this did not totally solve the lost ball issue Im having.

However, oddly, its very reduced and the game does not become unplayable after several hours of run time.

So, its a lot better! But not perfect yet.

Time to move on the the next item...

One of the recommended mods on Gottlieb system 3 is to do something about the terrible power supply board.

While trying to track down my switch issues, I started looking at the power.

Testing the output of the 5v power yielded not so surprising results.

Im getting 4.94 at the supply board and 4.25 at the MPU. Oddly the new MPU still funcions at this low voltage.

I expected a .5v drop from the cabling and connectors.

The little adjustment pot is VERY finicky and near IMPOSSIBLE to get right.

If I just tap it it goes way up or down but not where I want it.

Replacing the POT with a 10 turn or better, is probably an ok solution but I bought a new power supply board instead.

I bought the Boston Pinball power board as they removed the POT and put in DIP switches.

Its a great set up.

The new Power supply board is outputting 4.97v and reads 4.6v at the MPU.

I clicked the DIP switches one at a time (as recommended from Boston Pinball) and now have a ROCK SOLID 5.13 at the MPU board.

WOW, this really woke up the game YET AGAIN!

My game is betting faster and stonger every day.

Im going to have to make some mechanical adjustments.

Im now getting air-balls off the flippers.

Balls are hitting the glass, and flying onto the wire ramps, or all the way into the shooter lane!

Game play is fast and furious! Ramps are a breeze to make. Hitting the pyramid from the upper right flipper is fast and accurate!

My game is getting to be AMAZING!

But, still loses track of the ball count once in awhile. This is too frequently for me.

I read about checking fuse holders in the repair guides.

As I was checking the holders, I found an 8 amp fuse installed in the trough coil fuse holder.

THIS IS EXTREMELY BAD !

Finally it has dawned on me that this ball count and trough issue has been going on since before I bought the game.

Now it makes sense that the ball release and trough kicker coils were dragging and had to be replaced.

Something really BAD had happened in the past.

I replaced the 8 amp fuse with the 0.5 amp recommended one.

Powered up the game several times and the fuse is holding up perfectly.

Still however with all of the changes Ive made, the outhole kicker is sluggish to respond after the switch is actuated.

It can take up to 5 seconds for the kicker to respond after a ball lands on the switch.

I think something is amiss here....

Time to look at the driver board.

After seeing the 8 amp fuse incorrectly placed.

Im very concerned about serious damage the the driver board.

Sure enough, both the ball release and the trough kicker MOSFETS (q27 and q29) have been badly hacked and worked on.

I think Im finally getting very close to the ball count problem.

The board is a mess.

I found a damaged trace on the outhole kicker MOSFET.

I added the jumper wire (with a little "wire wrap" wire) as the trace under it was cut 3/4 the way across and scored deeply. Possibly from the original repair guy trying to salvage his terrible work, or the next guy checking for problems, as I am doing.

There is no "easy" fix for the blown VIA's and overheating of the fiberglass.

Adding the jumper really stabilized the game issues when the game is totally warned up.

But it still didnt fix the ball count issues completely.

After seeing the over-fusing and damaged coils and the bad MOSFET board work, I suspect that the chips at U3 and U4 are heat damaged or compromised in some way.

You can only barely see from my photo, but the top metal edge of MOFET 27 is scarred from countless ground firing tests.

Also notable is that Q27 looks like it has been replaced a couple times (possibly searching for the problem my game has?), due to the way its been soldered on top of the existing pins of the prior "hack".

I have no complaint about this method, it has saved a ton of boards from being ruined by inexperienced board repair technicians.

The replacement? MOSFETS are identical the existing ones. I'm confident that they are "pulls" from another board.

So, a bad hack job using sketchy unproven parts? Welcome to pin-repair, lol.

I have ordered and new driver board from Barakandl, it'll be here in a week.

20221119_135047 (resized).jpg20221119_135317 (resized).jpg

The delayed response from the outhole switch to the coil actually firing is troubling.

5 and sometimes 8 seconds, really? Lol. If a slingshot did this it would really be noticable, and terrible.

No other coil in the game has this problem.

Im crossing my fingers that the replacement driver board with solve this.

Im still waiting on the new driver board.

Meanwhile, Im continuing to molex mechs as I remove and inspect them.

Nothing is particularly worn out, or even that dirty for that matter.

However, it was fun to clean up the 3 drop target assemblies and add tiny bits of lubrication via the super lube pen.

Cleaning/lubing the kicking stand-up targets was particularly rewarding as now they are TWICE as powerful and fast!

Tiny amounts of lubrication on metal to metal friction areas with a dry lube or a non-drying lube like super-lube silicone oil is recommended by the manufacturer.

The key is tiny amounts, and wipe off all the excess. With these modern lubricants, the tiniest amount is all you need.

I will get back on this project after the holiday.

Thanks for listening to me rant, lol.

Yes, I have rivets and eyelets for this type of repair!

Thanks for the recommendation!

But, for the money and time, PLUS getting "improved" electronics that are NOT 30 years old, I bought a new board.

I lost confidence in my driver board, mostly due to over temp, over-fused ( 8 amp fuse for the trough coil) circuits from the prior experts who worked on this game (and having to replace both swollen coils in the trough).

I believe that these same experts drilled the hole in my coin door, to put in a credit switch.

It was very neatly done, but the loose wire that fell off the center kicking target, was drooped down and easy to spot.

This was the only problem that kept it out of tournament mode for free play.

Just like installing the replacement MPU board, I would be shocked if swapping the driver board totally solved the problems with this game.

The outhole "delayed response" problem is troubling...

Yes sol 28 and sol 27 were very damaged (The ball release coil, switch and plunger and spring have been replaced and are NEW). The ball release seems to work properly and quickly enough.

This is indeed mysterious!

The switch on the outhole is a regular DB5 or DB3 microswitch (I have bags of those) with a flatblade actuator, mounted on a plastic bracket (I have spares of those too).

The switch registers promptly and immediately in test, and no other switches activate.

(During the "playfield smack-down thunk" test, I did get a closure on the sarcophagus drop target. I adjusted that properly when I recently rebuilt the mech).

I replaced the outhole switch "twice" with new ones thinking that maybe I put in a bad switch (it happens lol). The ones I pulled out work perfectly.

The outhole coil was heat damaged, swollen, and dragging on the coil (evidence of a problem in the past).

I replaced the coil, spring, plunger and link. I removed (pulled one leg off) and tested the diode on the NEW coil, it tested good. Then I removed that diode and installed a new "tested" IN4007 diode, just to be sure.

I havent traced the wires from the coil and the switch yet for shorts or breakage, but moving the harness around hasnt made any changes.

The NEW driver board may arrive early today. Ill install it and report back.

The next step I can do is trace the wires from the switch and the coil to see if there is any mischief going on there.

All of the boards I bought were really just to future-proof the game, with needed upgrades (the MPU and power supply specifically).

Again, the OLD driver board can be repaired, but a new one isnt expensive and I feel better about newer parts.

So, Im happy with them.

I received the new Driver board.

It installed with no problems. Its a really great and clean board! Very robust and well made.

Im happy with the new non-hacked board.

Sadly this did nothing for my odd game behaviors, its still the same and unplayable.

Outhole coil is still sluggish to respond.

All multiballs end early, flippers die and sets up the next ball in the shooter lane.

Next to check are the optos.

Ill clean and reflow the solder on the little boards as well as check the connectors.

I do have a spare 4 opto board and a spare filter capacitor that Im going to install.

Ill report back with the result.

Going over the switch matrix.

Today Im swapping the right outside smart switch with the left one.

Given time, I will pull the wireforms and reflow the opto boards.

Right now Im rebuilding the left kicking target, and the bullseye targets since they are on or near the strobe lines for the trough switch.

The bullseye switches have capacitors on them that might be bad or missing, I will just put new ones on.

I will attempt to get through to PBR to get 5 new pairs of opto boards on my next order.

I still haven been able to get through by phone.

The optos dont seem to be a part of the trough circuit. But its been reported that extra balls can be added from a flaky opto on the ramps.

The game has 5 pairs of optos, but only 4 of them appear on the matrix chart. Also the subway opto doesnt appear on the chart.

I imagine that the subway one is just piggy-backed onto the left VUK opto since the opto board only supports 4 opto pairs.

20221209_181153 (resized).jpg

Going down the switch list from top to bottom, and side to side looking for common problen causes.

Examining the left kicking target.

I found a loose wire hanging by a thread!

Its attached, but unlikely it carries much current or signal.

Its on a return line, not a strobe.

But its important nonetheless!
20221210_164049 (resized).jpg

I fixed it by adding a molex and crimping the 2 wires together onto the same pin instead of the brittle solder joint.
20221212_123719 (resized).jpg

This wire often falls off.

Its connected to the tourament switch.

This repair should help solve that problem in the future maybe.

Since I molexed the bullseye switch and the left kicking switch (and its solenoid), I can now easily remove them for cleaning and maintenance.

Both are pretty crusty.

I always molex everything I pull off of a pinball machine these days.

Its just a better approach to the work in general.

I got my marco order.

I replaced all 4 sets of optos and the 4 opto controller board.

The subway opto (pyramid top) is a smaller opto than the rest. I didnt order the pair but they work perfectly.

During this opto replacement job I found MANY marginal solder connections.

I also took the time to rebuild all 3 VUK's.

All 3 were missing springs and needed coil sleeves and new coil stop brackets.

They are all NEW now and work perfectly!

The new optos are much faster responding than the old ones. This was a great upgrade!

Still however the game is unplayable.

Multiballs end early, flippers die and a new ball is served until the game ends.

Today fuse F21 blew and the right kicking target coil is locked on.

I pulled out the coil, checked the diode and tested the resistance of the coil (A-5195).

The diode tested perfectly but I replaced it with a new IN4007.

The coil resistance is 15 ohms. It should be 11.5 without the diode.

Its not too bad and pretty close to spec, but Ill replace it anyway.

I orderd a couple new coils from Marco.

My new driver board evidently blew a mosfet! Theres nothing wrong in the circuit.

I put my old driver board in and the coil is not locked on.

So, I orderd some IRL540 mosfets from Marco.

Ill repair the "new" board in a few days.

I read that also, so I disconnected both of the motors and played a few games but the problem persists.

Im having positive results by making extreme adjustments to the trough switch.

One thing I noticed is that the "new" trough switch's spring tension is much "lighter" than the original switch.

Its extremely sensitive. It can be triggered by "thunking" the playfield.

Since I have gone thru the entire game now, and fixed dozens of problems that I wasnt aware of, the game plays FAST and powerful !

De-bouncing the trough switch seems to be the last item to solve.

I put the old trough switch back in yesterday and had positive results.

The game played perfectly until the Ra's temple mode.

This is a 4 ball multiball.

With the ball trough fully emptied.

The mode ended early with 1 ball still in play, flippers died, ball drained, next ball set up in the shooter lane.

Im going to rig up a detent for the trough switch to de-bounce it a little and see if it has a positive result.

Oddly, for the moment, only the Ra's temple mode is ending early.

The final 7 segment mode did not end early.

Im running version 4 of the game ROM right now.

Yesterday I replaced the Right Outlane "Smart Switch" as it was last on the switch column that includes the trough switch.

It did not solve the problem.

So, as of now, all related switches (plus a lot more) have been replaced. As well as several "out of spec" coils.

Yestereday the backbox "Rope Light" stopped working.

It was a simple fix.

The top side of the far right resister on the light insert panel was loose and had become unsoldered from the heat.

It was very oxidized. Some cleaning with a fiberglass brush, flux and resoldering, fixed it quickly.

Thats a great Idea!

Since the lamps and switches use the same strobe lines, its certainly a possiblility!

Ill look at it today.

I had the idea to de-bounce the trough switch to see if it had a positive effect.

20221229_103121 (resized).jpg

De-bouncing the switch limits the time that the switch can be actuated as well as insuring no false mechanical hits.

The extreme bend in the front edge was to keep the actuator in-line with the balls to avoid a secondary switch closure from balls bouncing back.

This had a VERY positive effect.

Most multiballs work as they are supposed to.

Ra's temple 4 ball multiball ended early with 1 ball still in play.

Any 4 ball multiball will end early.

The Glider 3 ball mutiball will end early 30% of the time.

Next step is to debounce the switch electronically.

To completely eliminate the possibility of having 2 bad trough switches (the original plus the new one I bought), I built a new one out of a spare outhole switch.
Being a completely different switch and having a much stronger spring in it is a positive.
20221229_120044 (resized).jpg

Sadly this did not solve my issue at all.

It actually worked extremely well but with out the foam de-bouncer, games are unplayable.

I decided to add an electronic debouncer to my test switch to "emulate" the foam that I put on the original switch.

A simple R-C (resistor+capacitor) circuit works well for this. The capacitor holds down the signal for a little while, the resister drains the capacitor to get it ready for the next switch actuation.

Similar setups are the staple of pinball, you see it all the time over the generations

It worked perfectly, and as well as the foam applied to the original switch.

Sadly however, its not 100% and STILL there are multiballs ending early. Although not as much as without the de-bouncer.
20221229_141741 (resized).jpg

Why am I debouncing a switch?

Why does de-bouncing the switch have a positive effect?

Why do all the other system 3 games "NOT" have to have a single switch de-bouncer?

Getting multiple switch closures and extreme switch sensitivity lead me to see that the whole game possibly needs de-bouncing.

This would mean that the de-bouncing parameters in the existing code are being pushed aside do to an electronic problem.

I think the only thing that would do this is, "DC ripple".

DC ripple in the CPU 5v as well as the switch matrix.

I replaced the 12v bridge rectifier early on. This didnt solve the issue.

I did "NOT" replace the 10,000 mfd filter capacitor for the 12v bridge.

The 10,000 capacitor is there to specifically address DC ripple and a host of problems associated with partially rectified AC power.

Extra pulses from DC ripple would absolutely fool the CPU that there was a switch closure, a phantom switch closure, or a double closure from a single actuation.

This may account for the long time it takes for the outhole coil to respond when balls drain.

Extra pulses make it hard for the CPU to recognize that the switch was pressed.

System 3 games have a software detent so that if a coil is machine gunning, it will disable the coil until you power down the game and re-boot it.

Im curious if this effects a switch that pulses wildly?

I dont have a 10,000 mfd 25v capacitor in stock.

So, I can put in a 15,000 mfd 35v instead.

The rule of thumb is that you can go up as much as 100% in mfd value, and anything higher in voltage.

You just cant go lower by much. Plus or minus 10% is the specification.

Worst case scenario is that it will smooth out the voltage a little more and last longer.

And possibly not solve my problem.

I have a box of 15,000 mfd snap ins, this will be just fine.
20221230_105355 (resized).jpg20221230_105447 (resized).jpg

Since my game worked perfectly well for months, and got progressively worse until the game is unplayable, It sounds and feels, like a failing electronic component.

I have replaced:

CPU board
Power supplu board
Driver board
All 3 bridge rectifiers (replaced with 1000v versions for durability)
Diode board
Opto board plus 4 sets of optos
Playfield filter board
DMD display (replaced wth colorDMD LCD version)
Leaf switches
Smart Switches
Rope light
Coils
Coil sleeves
New springs for all plungers and drop targets.
Drop targets totally cleaned and rebuilt.
Flipper complete kits with coils and bats plus flipper shafts.
Outhole switches, coils and plungers for the outhole and autolauncher.
New coil stop Z brackets, plungers sleeves and diodes for all the mechs I rebuilt.

The game came with Comet frosted 2smd LED's pretty much throughout. I added some 5smd tower flashers for the burnt out #67 bulbs and repaired a couple sketchy 3 and 4 lamp boards that had cracked sockets.

Extreme ground mods are done.

I also Molexed every mech and assembly that I took out for easier servicing.

I have also found VERY numerous sketchy wire connections, over fusing, old 1st gen ROM, and over-all light corrosion that needed some TLC and lubrication.

The game play is fast and POWERFUL!

Its super easy to make the ramps, the ball action is lightning fast!

This is a best playing Stargate I have ever seen.

It plays pretty much like a brand new game.

All of this work is part of the regular restoration process that I do on all of my games, so its not anything more than another resto or out of the ordinary for me.

Next steps will likely be a playfield swap and possibly cabinet refresh and plate all the brass pieces.

Hopefully I can get the mutiball problem sorted out and play the game a bit before the official tear down.

15000 mfd filter capacitor is installed and works perfectly as expected.

These capacitors are held on with a nylon tie and double sided foam tape.
The tape was dried out. I used a little strip of 3/8" gray VHB tape on the reinstall. It will last longer than the foam.

This is actually an upgrade and should be better than the stock one.
new 15000 cap (resized).jpg

POSITIVE RESULT ! ! !

I played 4 games with the new filter capacitor, and all of the muliballs finished properly.

I cant say yet if its totally fixed, as I want to put the proper switch back in that has no R-C mod on it.

This will be the definitive test as the game was unplayable with an unmodified switch.

I can do this in a couple days, after the holiday.

Crossing my fingers !

Some observations from the 4 test games I played.

1) Left upkicker is sometimes taking a couple of tries to get the ball up and into the wireform.
It didnt do this prior to rebuilding the mech.

2) 2 balls ended up on the outhole.

On issue 1:

When rebuilding the mech and replacing the optos, the plunger spring was smashed, rusted and completely destroyed.

I replaced the missing/ruined spring with the correct one.
This actually gives the coil slightly less power than it had without a spring.

Its actually somewhat normal for the left upkicker to struggle a bit.

The official solution from pinrepair and the pinwiki is to put in a stronger coil to fix it.

I think that I will instead pull off enough coils of wire off the coil to lower the resistance 10-15%.

Less wire = lower induction = more power! measuring the resistance and windings of the coil is a good guide as to its over-all power.

In my experience, removing 50 to 100 turns of wire, or so, will increase power 10% or more depending on the coil and thickness of the wire. Gottlieb coils use very fine wire so It might take removing a few more turns to achieve my desired outcome.

Ill report back on the new resistance measurement that I arrive at.

Issue #2:

Its well known that the system 3 outhole system is anemic at best.

It barely has enough power to kick a single ball over the trough hump to the resting area.

If you get 2 balls on the outhole side, the mech will clear them, but it will take many tries to do it.

This was further evidenced by abnormal wear of the outhole kicker and overheated "bulging" coil on my game.

If you get 3 balls on the outhole side, the mech will not clear them and the game will be out of service until they are manually cleared.

I think that this is a common problem, as I see a lot of system 3 games with an incorrect williams 23-800 coil installed.

Operators know that this coil is commonly used on most Bally/Williams outhole kickers and is the workhorse for this application plus they usually have them on hand as they are used a lot as a generic coil.

The problem with the 23-800 coil is that its way too powerful and balls get forced out of the trough and into the shooter lane.

So, though this "fix" works, its not quite right.

What happens when you have 2 or 3 balls drain at exactly the same time?

Sometimes the ball will "not" make it over the hump and roll back down to the outhole switch. This activity does not affect the ball count and the ball will be simply kicked up again .

However while a ball is rolled up the trough, another ball can immediately roll onto the switch at the bottom.

This can lead to 3 balls in the outhole side of the trough.

I think that when the game is under a heavy load (long multiball), there is less power available to fire the outhole coil enough to work 100%.

A coil upgrade capacitor system is used on API and Spooky games to store some energy for this exact problem.

On my API Houdini it has worked wonders to give just a little more stored energy when needed.

I see that there is a small, under playfield filter capacitor board (that I replaced) on system 3 games.

There ia not a lot of documentation of why its there. Is it there to prop up performance?

Im curious that if I replace it with the 4 capacitor Spooky board, will this anemic behavior improve or be solved?

It would be a good test and possibly solve the fade problems that system 3 games have on extended multiballs.

It may also solve the weak left upkicker issues.

I can also just remove enough wire off of the existing coil to get it a bit more robust, but Im also interested in a performance upgrade over the whole game.

Ill have to look at the schematic to see what the filter capacitor board is doing.

Replacing the 12v filter cap had made the game go from UNPLAYABLE to VERY playable.

Without using an osciloscope to go into it in depth and visually capture the rough voltage, I made the right call.

The game is vastly improved, but still can be annoying.

That said, Im now looking at the 33000 mfd cap to stabilize the lighting circuits.

Since the light matrix and switch matrix share the same strobe lines, it seems like a good Idea to go ahead and replace the 33000 mfd cap at this point.

Bad/dirty/rough power in the lighting strobes will also negatively affect the switches.

They are only $6 on amazon+ shipping (cheaper than $17 plus $9 at marco).

The game loses count on a multiball only on long games, or only on a super long multiball.

Interestingly, multiballs that end early are now random and seem to be related to their duration other than the actual mode.

Careful adjustments to the trough switch also has a positive affect, but isnt perfect.

33000 cap comes in a couple days.

Im hoping it has a further positive affect.

The other capacitor arrived today.

Im hoping to install it tomorrow if it isnt too busy at my shop.

Since I replaced everything on the top (all the pc boards) and the bottom (all the bridge rectifiers and the EXTREME ground mods), and replaced all the affected switches and connectors,

The issue should be dirty power that can be caused by a variety of things with the switches and lamps sharing the same strobe lines.

Basically the probelm is related to strobe 2, or the whole game if its DC ripple. Strobe 2 is common to all of the related switches, and a few of the lamps.

I have inspected all of the lamps and lamp boards on strobe 2. They all seem in good condition, with no glaring problems.

All of the affected switches have been replaced as well as a bunch of other ones that just seemed old and worn.

The pyramid lamps might be suspect.

But in single lamp test, in a darkened room, there is no ghosting or odd behavior on any of the lamps.

Ill take a look at the flashers, it sounds like a good idea.

With a new driver board and a new CPU board and new connectors, as well as a ton of new switches, The problem is getting narrowed down.

Pulling out the flasher bulbs might have a positive effect!

Thanks for this!

Gottlieb uses #67 flasher bulbs. They are more durable than the #906 ones.

Most of my flashers have been changed to 5smd tower flashers.

All of my regular bulbs have been changed to 2smd fosted sunlight.

I did have a question of whether changing to LED's disrupts the strobe lines or not,

The concensus is that it shouldnt, as other games do not exhibit this behavior.

After seeing the very positive result of changing out the 12v filter capacitor, Its looking like dirty power is the cause of my trouble.

Getting to the bottom of it and finding the cause, is quite troublesome.

Well, so far it seems to be extra pulses in the strobe line to affect the trough switch.

I believe I'm getting multiple closures that is throwing the ball count off.

Changing out the switch had no affect, but debouncing it had a dramatic affect.

Let see if changing out the 33000 mfd cap does anything at all...

First off, Thank you for taking an interest in my little problem!
Getting feedback and IDEAS from experienced repair people is what Pinside is all about.

That said, I had that exact thought a while back.

Since all switch diodes are on the dedicated diode board on Gottlieb system 3 games, I replaced the diode board.

This had ZERO effect.

However, this lead to the thought that something else is causing trouble.

10000 mfd capacitor?

33000 mfd capacitor?

Bad diode on a cotrolled lamp board? Right now Im making a list of all lamps on strobe 2.

Bad/shorted connction in the flashers?

Bad resistor board?

Since all the main boards are new (including power sullpy), and EXTREME ground mods are done.

I can focus on what's in between!

Replacing the 10000 capacitor has had the most dramatic effect.

I will likely replace the 33000 mfd cap and the resistor board (replacing the board just because its old for future-proofing, it wont address my problem) tomorrow and report back.

I really liked your suggestion about the flashers! Its simple enough to pull them all out and see if the game stabilizes.

However, the 33000 cap does control the power to the flashers.

These 2 ideas are on the same path.

We'll see if the cap replacement has a positive effect.

This is something to check.

The credit button is on the list of suspect switches in the switch matrix. Im not clear on whether its diode is on the diode board, but it probably is so it may not be the issue.

Either way, I havent looked at it yet.

Thanks for this!

My new power supply board is amazing!

I have 2 of the dip switches tripped out of the 5 or 6...NO MORE POTS, lol.

Currently it reads at just over 5.2v (at the cpu), and its STABLE!

My new 33000 mfd cap arrived and I finally had time to install it.
33000 mfd cap (resized).jpg

Now both caps have been replaced and installed.

I cant be sure, but it looks like the 33000 mfd cap has been replaced with a used one, in the past.
Its a bit rough and used looking. It also has some kind of white overspray on it.

Also it wasnt stuck to the tape like the 10k one was. It was loose and mounted with a red nylon tie vs a regular white one.

You just never know...
old caps (resized).jpg

After installing the new 33000 mfd cap, the results are quite POSITIVE!

I put 10 games on it and had no multiball problems at all.

Game 11:

The game lost a ball in the left VUK.

During ball search the game inserted an extra ball into play.

The missing ball was found during play and when a multiball ended, the flippers died prematurely and ended the ball.

On the next ball (in the same game) flippers died again during multiball.

The ball count was still off during the rest of the game.

I did NOT play another game to see if the fault persisted to the next game.

I rebooted the game instead, and the problem went away and played perfectly again.

So, it seems that in some cases a missing ball "correction" in the code, will persist between balls until the game is rebooted or a new game is started?

Gottlieb doesnt share code fixes.....so who knows what they did by version 5 (im on version 4 for now)

Or

Im getting very close to perfection and still there is something mildly affecting the strobe line?

Im not clear as to how a ball can be lost in the left VUK.

This has only happened once since Ive owned the game, But I did replace all the missing VUK plunger springs.

Obviously its hung on something, or the opto didnt recognize a ball was there.

I dont think the ball is settling and stopping in the subway. It looks brand new and there are no dents in it.

Certainly I can improve its angle with a washer under the left side to increase the ball speed, if its not too thick and hangs on the VUK.

More likely the ball is resting on the edge gap of subway to the VUK wireform.

I havent yet pulled off wire on the VUK and trough coils to improve their power.

When I pull the left VUK apart for inspection, I will pull off some coils of wire.

So far however, the game is more or less fixed.

Now Im just going thru and checking more age related items.

Im likely replacing the aux driver board to freshen up the flashers.

Molexing everything in the game, sure makes regular inspections EASY!

Currently the game is playing so fast, balls are flipping over the bats on a hard rebound.

This happens about once in 5 games.

The flippers are not set too low as I checked them with a ball.

The center of the bat is above the centerline of the ball.

It might just be the narrow flipper bands allowing some upwards deflection?

There might be a little more room on the shaft notch to raise the bats a tiny bit more.
20230113_135547 (resized).jpg

I still have a way to go.

Im looking at the flashers right now.

Im ordering a new auxillary driver board for them.

I bet the transistors are old and choppy.

Im tracking down the slow/sluggish response of the outhole kicker after the switch is closed.

I found and installed a new resistor board, that I forgot I bought awhile ago (it has diodes on it).

This did not solve the issue.

Finally got around to replacing Q7 on my new driver board.

Q7 controls the right kicking target.

these new boards are super easy.

Just cut the leads to the existing bad part, either leave it on board on remove it.

I just left it in place and soldered a new MOSFET on.

It works perfectly again.
20230114_113830 (resized).jpg

My complaint about the outhole kicker is interesting.

In attract mode the outhole kicker is slow and methodical?

"In game" it works as it should.

However when the game is depleted during very high activity, the outhole kicker can fail to launch the ball hard enough to get it over the "hump" in the trough.

During this high stress and high volume session, after many successive multiballs, the outhole kicker and the TROUGH switch can fail to activate properly.

This is simply tested out by going thru the modes with the glass off and catching the balls in ones hand.

I can hit every target, activate every switch and opto.

then drop all the balls except one into the outhole.

This works flawlessly with no errors.

I can do this with every mode including Ra's temple in the same session.

No errors whatsoever.

My game is jumpered to 110v (red/white jumper). It came that way from the factory, jumpers still stapled to the side of the cabinet.

Next I going to change the main power jumper to 120v (orange/white jumper from bag) and see what it does.

John Robertson (John's Jukes) recommends the 120v jumper for a variety of reasons.

It is possible that the game gets buggy when set to 110v and actually delivering 115-121v.

I can see how this might cause some issues.

Ill report any changes.

I installed the 120v jumper.

ALL HELL BROKE LOOSE ! ! ! !

WTF.

2 balls get kicked into the shooter lane for each ball served, WTF ! ! !

Multiballs last 1 ball....WTF !

No ball gets past the outhole kicker without 3 or 4 tries, or it fails completely and sits with 2 or 3 balls kicking endlessly....

Sounds bad, VERY bad....

But its NOT!

Finally I have something to work on in real time.

The game should work perfectly with either the 110v or 120v jumper. This was tested at the factory and MANY games have it installed with no ill effect.

Lets look at it and see whats going on.

This could lead to actually fixing my game.

YAY! (I think...)

Lets look at the outhole kicker problem.

Originally when I got my game, the outhole kicker was weak.

It was dirty, grimy, and desperately needed a rebuild.

On inspection the coil sleeve was permanently stuck in and the coil had some serious heat damage.

On test it read 137 ohms. It was shorted a bit.

I unwound enough wire off of it to get it down to 65 ohms (as is the spec for the coil) and reinstalled it while waiting for a new coil from PBR.

I installed the new coil when it came and thought nothing of it, as it was new.

Time to look closer.

I pulled out the "new" coil and checked its resistance (it should be 65.8 ohms or close to it).
new a-26451 ohms (resized).jpg

I clipped one side of the diode to get a true reading (diodes add about 1 ohm).

WTF ! Its supposed to be 65.8.

71 ohms makes it out of spec by near 10%.

If the coil was within spec it would be more powerful and likely work correctly.

Lesson learned.

Check new coil resistance before installing all new coils....

The bible for pinball coil restance measurement:

https://flippers.com/coil-resistance.html

I decided to add a little oomph to my outhole coil since its out of the game and out of spec anyway.

I unwound 400 turns of copper:

new coil lower resistance (resized).jpg
Dropping the resistance 10% from spec should give the coil the extra boost it needs, without going overboard with a Williams AE 23-800 coil.

I reinstalled it into my game.

IT WORKS perfectly!

Now, I can recommend that unwinding a few turns of copper from your outhole coil will solve the anemic/weak performance that it traditionally has.

Outhole problem is now permanently solved.

Im looking at the pop bumper section of the playfield today.

It looks ok.

The left kicking target needs attention. I pulled it out, cleaned off the corrosion, cleaned the switch contacts and lubed its shaft.

It works better than ever now, SUPER strong and responsive.

Pop bumpers need cleaning and inspection.

Game is still kicking out 2 balls for one on the ball serve.

As CaptainNeo has suggested (thank you so much for this), its time to look at the flashers!

Flashers are controlled by the auxiliary driver board (there are no diodes on them to go bad).

I ordered a new one just for the heck of it, they are inexpensive.

Meanwhile I can look at sockets and wires to see if anything is out of sorts.

Everything looks and tests properly.

But look here!
rebellion flasher wire (resized).jpg

A wire has fallen off the Rebellion flasher.

This flasher is not like the rest of them. Its the main 20v supply.

flasher schematic (resized).jpg

I soldered the wire back on.
rebellion flasher repaired (resized).jpg

The game has, stopped kicking out 2 balls......WTF?????

Absolutely. Again, thank you so much for the suggestion!

The game stopped kicking out extra balls already!

Next we will see if the ball count issue is fixed.

Thanks!

Its a puzzle for sure.

I ordered some pop bumper parts and smart switches.

I will freshen them up a bit while Im inspecting that section.

After that, the pivot targets likely need attention.

Smart switches are easy to test out as you can just unplug them to take them out of circuit.

Unplugging all of the smart switches does NOT solve the ball count issue.

Im leaving the 120v jumper in place for now.

All of the game problems seem exaggerated.

This might help to narrow down the culprit.

I played a game to check out its current staus.

The game lost the ball count on ball one.

Kicked out an extra ball on ball 2.

Played 4 balls on glider craft.

Mode ended early with 4 balls in play.

All 4 balls exited the outhole.

The game couldnt clear them and they all stayed trapped in the outhole.
My modified coil still isnt strong enough to clear 2 balls. Ill work on this a bit more and get it to clear 2 balls at once in test.

Putting a 23-800 coil seems extreme.

Possibly the "blue" wrapper coil thats listed at 42 ohms will work well enough?
4 balls in outhole (resized).jpg

Today Im going to manually test the game modes to see if a switch or mode can shine light on the problems.

The plan is to start multiballs and remove the balls in play as they come out.

Manually trip the switches with a ball and see if I can get the mode to end early while dropping a ball into the outhole once in awhile to simulate game play.

Also see if a mode times out.

In theory a mode shouldnt time out until there is only one ball left in play.

If something wrong happens, I should be able to reproduce it.

This should be interesting.

Before getting into something exotic, I was thinking about the dramatic change that the game went through by changing the power jumper.

Maybe its already in a low power state?

Are the main fuses loose or corroded?

Time to check the power box MA-1928

stargate power box (resized).jpg

5 amp fuse holder (F2) has a cracked nut and the top is wiggly.

The fuse is loose in the holder and doesnt firmly touch the bottom contact.

broken nut (resized).jpg

There doesnt seem to be a Gottlieb fuse holder per se.

The holder has Buss HTB molded on to it. Searching for that yielded good results.

This one is perfect:

https://www.amazon.com/Blue-Sea-Systems-Panel-Holder/dp/B000XBB8YI/ref=asc_df_B000XBB8YI/

I bought a couple at West Marine in reno.

fuse holder (resized).jpg

Evidently boats use them all the time.

Its a perfect fit. Although it inserts from the bottom instead of the top.

I doubt that this will change the game much.

All Done:

installed bottom (resized).jpg

installed top (resized).jpg

Fuses fit snug and the tops are NOT wiggly!

Much better than the 28 year old crispy ones.

The power box looks great and un-tampered with.

Fuse holders are solid!

Still waiting on Pop-bumper parts to get those sorted out.

Meanwhile my Aux driver board arrived!

Im certain it will have little to no effect on my ball count issues, but freshening 28 year old boards can only be a good thing.

old board:
old aux driver (resized).jpg

new board:
new aux driver board (resized).jpg

2 min install and it works perfectly!

While waiting for parts, I noticed that the pivot targets "squeek" and one has a cracked switch.

I "do" have those switches in stock!

Its super easy to lube the pivot points, and replace the switch. I might as well replace the coil sleeve while Im at it.

left pivot target (resized).jpg

Oiled all the pivots on both pivot targets with a tiny drop of oil, but not just any oil:

Super lube precision oiler part# 51014. Dries out leaves no gummy residue, its primarily PTFE (teflon):

https://www.homedepot.com/p/Super-Lube-7ml-Precision-Oiler-Oil-with-Syncolon-PTFE-Lubricant-Per-Each-51010/202932717?source=shoppingads&locale=en-US&srsltid=AeTunco7CTEEzMQcBZXPE5Yw6vsCWLf4RVZH0q7uH2Nw_0JeFNrIKlr_lZw

Or use any "dry" lube. this works well.

oil pivot points (resized).jpg

Squeek and roughness all gone. Works buttery smooth now.

New switches both pivots.

F#ckn Marco sent one that was missing the bolt plate.... Its ok just move the old one to the new switch.

new switch missing part (resized).jpg

smart mounting bracket (resized).jpg

10 mins to replace both switches! They work perfectly in test with a ball.

Still waiting for pop bumper parts, maybe monday.

Meanwhile, my left VUK is sticking up.

Time to examine:
VUK cap (resized).jpg

Looks like the cap has been replaced and a bent nail is used to secure it.

The Gottlieb 5-40 retaining shaft is not to be found anywhere.

McMaster has a suitable 5-40 shoulder bolt so I ordered a couple. (not cheap)
shoulder bolt (resized).jpg

The plunger cap is badly mushroomed and worn down at least 3/32"
This will make the cap rest too high and hang balls on it.
It will also make the plunger poke thru too much and cause it to settle too high up.

I temporarily added a fender washer will take up the slack and fix it.
Also the return spring is incorrect, but if it works it will be ok. I ordered the correct one.

washe installed (resized).jpg

All new VUK parts ordered from Marco!

they will be here next week.

MASSIVE UPDATE ! ! !

Part one.

Before I rebuilt the pivot targets and replaced the pivot smart switches:

On a hunch I re-installed the 110v jumper and played a test game.

Multiballs were grabbed by hand and held.

All of the modes ended correctly except one, Rebellion.

In rebellion the only active targets are the pivot, pop bumpers and the left kicking target.

Manually flinging balls at the pivot target and letting go thru the pops was ok until, one of the balls drained and wouldnt make it up the outhole trough ramp easily, it took 3 tries.

Suddenly the game launched another ball?

I let the balls drain one by one, the mode ended early, flippers died and the ball ended with 2 balls still in my hand.

I put the 2 balls into the outhole and they wouldnt make it up the trough ramp.

Something is going on in the pop bumper/pivot target/left kicking target area of the game.

I already found the loose wire on the Rebellion flasher, and fixed it.

This is the first time that I could manually see something very wrong.

Pretty exciting!

MASSIVE UPDATE ! ! !

Part two:

I suspected that the pivot target smart switch was the culprit, even though the pop bumpers are crusty and tend to stick when actuated manually.

I havent rebuilt the pops yet or replaced their smart switches yet.

The only oddball part is the left pivot target smart switch (and it has a crack in it), its part of the group that failed the Rebellion mode.

Pivot smart switches are different than the other ones, they are not interchangeable.

After cleaning, lubing the pivot points, and then replacing the "smart" switches on the pivot targets I played a game.

Im still on the 110v jumper and will leave it that way. It came from the factory with it on.

On initial game start:

The game immediately plays differently. Peppier, stronger, faster responses.

Mulitball after multiball modes end properly.

Flippers dont die, the game plays flawlessly.

I made it past Ra's bracelet with a 4 ball mutiball, no errors (this is always an iffy mode on my game and rarely finishes properly).

I made it all the way to Eye of Ra, massive multiball that went on forever!

No errors, drained down to 1 ball and started a few more modes.

NO BALLS collected in the trough, ALL balls made it up the trough ramp on the first try.

Thinking this was a fluke, I played 5 more games, ZERO ERRORS.

Still not trusting the game, I left it in attract mode for 4 hours.

Played 2 fast games, ZERO ERRORS ! ! !

So, I truly believe that the game is fixed now!

This is the best that the game has performed since I bought it.

So it appears that the source of the ball count errors were due to a faulty smart switch on the left pivot target.

On the pivot targets there are a lot of pivot points.

Any metal on metal pivot points in a pinball machine should be lubed with tiny amounts of silicone oil (or sewing machine oil in the old days).

***NOTE*** I said TINY amounts of oil, it shouldnt be DRIPPING off and any excess should be wiped off. This will do a good job of "dry" lubing.

In my picture earlier, I tried to show one of the axle pins of the pivot target being externally lubed with a tiny drop of silicone oil.

This includes slingshot bearings, spinner targets or any axle pin in a game.

On some of the axle pins and press in bushings I use silicone grease.

The parts will last forever and the game will play faster.

On the left Horus guard there is a Smart switch behind the round white target face.

Its not a regular "leaf" switch that is common in pinball games.

These are NOT adjustable, do NOT need lubrication and for the most part, dont fail (unless you are just unlucky )

left horus guard (resized).jpg

Why it has failed and affected the ball count is a mystery.

Its not on the same strobe line, nor is it a part of the column or row that contains any of the trough switches or the shooter lane switch.

Its also not in any way, a part of the outlane or trough coils circuits that I can see so far.

However is does report to the CPU and I guess can WREAK HAVOC if it fails in just the right way...

YES!

You did!

But, they never go bad, even after 28 years????????? LOL !!!

I am replacing them as I go. Next are the pop bumper ones.

Gottlieb games are very robust and relatively trouble free compared to the other game manufacturers.

The main reason I created this thread was to document problems and solutions.

Comparatively little is documented about Stargate.

I will continue adding my repairs to this thread as I go through the game.

I have a fix for the sticky/weak left VUK next.

It was a tough one thats for sure.

Only 72 days to figure it out.

It might have KILLED a lesser man etc....

Now onto the rest of my examination of the game.

Im nearly through all of it.

Pops and pivots were nearly the last items to check.

Oddly, things are always found in the last place one looks...

In the procees, VERY many marginal/hacked/broken/incorrect items were found and addressed or replaced.

Plus all new electronics! Gottlieb parts are at bargain prices compared to Williams.

For a few hundred dollars I have rebuilt most of the game with all new parts.

The game plays like a dream!