(Topic ID: 50864)

Space Shuttle won't boot - error code 7 (Williams system 9)

By Collin

10 years ago


Topic Heartbeat

Topic Stats

  • 41 posts
  • 6 Pinsiders participating
  • Latest reply 10 years ago by vid1900
  • Topic is favorited by 6 Pinsiders

You

Linked Games

Topic Gallery

View topic image gallery

ss_crystal.jpg
1.jpg

#1 10 years ago

Hello all,

I picked up a Space Shuttle last night that won't boot. The mpu display shows a 7, which PinWiki identifies as:
"System failure. Occasionally, the following components can contribute to this problem: U21 (4MhZ crystal); components in the IRQ circuit; broken leads on C9 (22uF) in the reset section; loosely seated ICs on the CPU board"
Any thoughts on what I should check first? I've hear bad or poorly seated ROMs can cause this problem.

Maybe worth noting, this game spent a year in a garage that's not climate controlled but played fine before that.

Thanks,

Collin

#2 10 years ago

Pull hard on all the ICs with your fingers and see if you can pull them partially out, if you can, replace socket.

If any socket says SCANBE, replace it immediately!

1.jpg1.jpg

#3 10 years ago

Thanks! I'll check the sockets this evening. Looks like I may be in for my first real board repair.

#4 10 years ago

Seems like the chips are seating properly, and the sockets are not Scanbe. I've got a local friend burning the ROMs for me this weekend... Will see if that helps.

What is the IRQ circuit mentioned on Pin Wiki? Also, is there any good way of testing the crystal?

#5 10 years ago

Need an oscilliscope to test the crystal properly, otherwise you can just swap it. It's a sub $1 part.

IRQ is a bit more complex, and again you really need a scope to test it properly.

-Hans

#6 10 years ago

Okay, my friend has re-burned the ROMs for me - they tested bad, but once erased and re-burned tested fine. Powered the game on, and may have seen a 0 momentarily (not sure on this.) but ultimately am still getting error code 7.

I'm thinking I'll go ahead and replace the crystal and capacitor mentioned on Pin Wiki and see what that does...

I can borrow a scope to test the IRQ circuit if needed. Have never messed with that before; is it a poor idea to learn on this board?

#7 10 years ago

I replaced all the caps on the power supply and that fixed my problem. My Space Shuttle was showing a 7 or 9..

#8 10 years ago

Collin,

I assume the crystal is for the CPU clock. Instead of trying to test the crystal you should be able to see if the clock is running. Using the schematics find the clock pin on the CPU (i think it is Pin 37). I would guess on a DMM you would see around 2.5vdc. This is not a full proof test, but if it is all the way low(0 vdc) or all the way high (5 vdc), that is a problem.

Failing filter cap like Choggard mentions can crash out the MPU. If the cap is failing it will start leaking AC voltage(ripple) into the 5v line. You can test the filter cap. Power on, DMM set to AC voltage. Put the leads across the 5v filter cap. You want to see almost no AC voltage. 0.25 VAC or more the cap needs replaced. Good idea to replace this cap anyways.

Andrew

#9 10 years ago

to troubleshoot any further you need to get the leon test rom - anything else is a waste of time.

get leon test rom booting - if it wont boot issue is RESET/MPU/XTAL possible (caps on the reset area of the board, the 555 chip, your IRQ chips all can be blown or bad)

once leon is booting use a o-scope or logic probe to test all PIA's

basically your game isnt able to load the "game rom code" you need to find out why

http://www.pinwiki.com/wiki/index.php?title=Leon_Borre_WMS_System_9_Repair

this system is fine to learn all this on.. just be confident in your soldering PCB skills.. you wont be getting a new system9 PCB easy/cheap. (however worse case I think you can hack sys11 to run SS code)

also it helps to know WHY.. this might have happened.. one large warning that sys9 has two large molex plugs that plug playfield into the MPU/Driver if you reverse these you will blow PIA's and more

#10 10 years ago

Okay, so I was looking on Page 28 at the section regarding Error Code 7 - it says
"Check for pulsing at any data pin of microprocessor-chip U17 on the CPU board."
It goes on to say the crystal or capacitor C41 may be defective. Capacitor C41 is reading 4.82v DC and .002v AC.

With the power on, C9, which is reference in the Pin Wiki section under error 7, initially showed like .6-.8v AC, and some small amount of DC voltage. Then it stopped showing any consequential voltage whatsoever...

#11 10 years ago

I was directed to this guide by a KLOV member:
http://gamearchive.askey.org/Pinball/Manufacturers/Williams/pdfs/pinball_troubleshooting_sys9.pdf
which I've used to try to troubleshoot as well.

Quoted from barakandl:

Collin,
I assume the crystal is for the CPU clock
Andrew

I believe that's correct, based on what I've read.

Quoted from barakandl:

Instead of trying to test the crystal you should be able to see if the clock is running. Using the schematics find the clock pin on the CPU (i think it is Pin 37). I would guess on a DMM you would see around 2.5vdc. This is not a full proof test, but if it is all the way low(0 vdc) or all the way high (5 vdc), that is a problem.

Quoted from barakandl:

Failing filter cap like Choggard mentions can crash out the MPU. If the cap is failing it will start leaking AC voltage(ripple) into the 5v line. You can test the filter cap. Power on, DMM set to AC voltage. Put the leads across the 5v filter cap. You want to see almost no AC voltage. 0.25 VAC or more the cap needs replaced. Good idea to replace this cap anyways.
Andrew

Would that necessarily be C41 probably? That's the only 5v cap as far as I'm aware.

I've seen C9 (10v) and C41 (5v) reference in the couple guides I've looked at. C9 is giving some odd readings, it initially showed like .6-.8v AC, and some small amount of DC voltage, then stopped showing any noteworthy voltage period.

#12 10 years ago
Quoted from Spudgunman:

to troubleshoot any further you need to get the leon test rom - anything else is a waste of time.
get leon test rom booting - if it wont boot issue is RESET/MPU/XTAL possible (caps on the reset area of the board, the 555 chip, your IRQ chips all can be blown or bad)
once leon is booting use a o-scope or logic probe to test all PIA's
basically your game isnt able to load the "game rom code" you need to find out why
http://www.pinwiki.com/wiki/index.php?title=Leon_Borre_WMS_System_9_Repair
this system is fine to learn all this on.. just be confident in your soldering PCB skills.. you wont be getting a new system9 PCB easy/cheap. (however worse case I think you can hack sys11 to run SS code)
also it helps to know WHY.. this might have happened.. one large warning that sys9 has two large molex plugs that plug playfield into the MPU/Driver if you reverse these you will blow PIA's and more

Cool, thanks for the info! I'll buy some EPROMs, and hopefully I can buy Andrew a case of beer to burn those for me.

As far as I know, the two large molex plugs weren't reversed, and it sounds like the previous owner never really had it disassembled, but the game to me dead like this, so I suppose I can't be sure. His explanation was that it was in a garage that wasn't insulated or heated or anything over a winter, and when he powered it on in the Spring, it was dead. This makes me suspect the crystal, as it seems they may be somewhat temperature-sensitive, though I've yet to understand how to test that.

#13 10 years ago

HOLY SHIT, I think I diagnosed something!
Regarding Andrew's suggestion of checking the crystal, I found this on here:http://illinoispinball.webs.com/williamssys911part2.htm
"If you have good RAM and EPROMs, and +5 volts at the CPU board, now is time to check the "clock signal". The clock signal is generated by the CPU board's crystal and some other components. The clock signal ends up at the U15 6808 CPU at pin 39. Using an oscilloscope or logic probe, check for a signal here with the game on. It should be a square wave. If all you get is a constant low or high, you have a problem in the clock section of the CPU board."

Pin 39 on U15 shows 4.82v constantly.

That means the crystal/clock isn't functioning, right? Am I correct in thinking the logical next step from here is to replace the crystal?

#14 10 years ago

I just reviewed the schematic. If pin 39 on the cpu is 5v(constant high) there is a problem. With a DMM i imagine you should see 2-3vdc on CPU pin 4(irq), 39(clock from crystal), pin5 (Valid memory address), Pin 37 (marked as E, i assume this is also a clock signal).

It could be the CPU, but likely that crystal is bad. There is also some logic gates involved. The schematics i am looking at suck and can't identify the pin / IC numbers.

Andrew

#15 10 years ago

Okay, cool. That fits with what you were saying earlier.

I'm working under the assumption that the bottom left pin is #1, and it looks like the bottom goes 1-20 left-right, and the top goes 21-40 right-left. If that's the case, then pin 39 is a constant 5v.

I'll go ahead and replace that crystal, and see where that gets me.

It's a 4 mhz crystal, as you'd said; do I need to worry about specifics beyond that? The markings on it are as follows:
50004-116
4.000000 mhz

ss_crystal.jpgss_crystal.jpg

#16 10 years ago

It is a standard part. They are super cheap that is part of the reason why i would say just replace it. We can also look at the clock signal as it goes through the logic gates.

ebay.com link: itm

#17 10 years ago

Thanks, Andrew! I grabbed a few of those. I was just uncertain where to find it or how specific a part it was.

#18 10 years ago
Quoted from barakandl:

The schematics i am looking at suck and can't identify the pin / IC numbers.

http://pinside.com/pinball/forum/topic/new-space-shuttle-1984-williams-manual-done

Here is good manual

#19 10 years ago

Thanks Vid. Much better.

I was wrong, there is no logic gates between the crystal and the CPU. That simplifies things and points at the 6802/6808 cpu or the crystal.

Andrew

#20 10 years ago

Also the sorceor manual is good

Telling you man - don't touch that board with a solder iron till you boot a test EEPROM

Sometimes a thump on the xtal will bring it back.
And a oscope will check your xtal

A error 7 can mean fail to load EEPROM also easily caused by old crappy connectors in a barn.
It just means your CPU isn't running code. Not that the CPU isn't running or dead clock. If blanking or Irq is faulty - so is CPU boot. If any one PIA is bad so is boot with original roms. You can not troubleshoot with game rom, it is risky!

A lot of stress avoided by simply putting a rom chip in there.

Seriously - just replacing parts blind is a risky way to operate. Get proper tools.

#21 10 years ago

Running the test rom, bad roms, or even no roms, the clock should not be a solid high. The test ROM will not even run in this situation. There is a problem with the clock signal and this is not a blind assumption. The clock circuit is simple enough and replacing a 10 cent component is not risky with a good desoldering tool.

#22 10 years ago

and you validated this "solid high" with a DMM? because that is the proper tool to test a clock with. right. also you didnt know what pin or how the clock worked on this MPU untill about 2 posts ago.

ok I will back out.. you can fix this and rub it in my face when you get lucky by shotgun fixing something.

#23 10 years ago

Put a DMM on a clock pin of any 680X cpu and it will not be 5v on a dmm if it is functioning correctly. DMM is not the ideal tool but we working within the means of the original poster. This is not a shotgun fix. I am a little green on sys9, but i do mail order repair on sys 3-7 and bally MPUs on a daily basis.

Ideally should check the clock phase 2, which they call "E "line at pin37. The DMM lead (or scope lead) will effect the crystal.

Collin, if you want you can bring the MPU over and we can scope things out on my bench.

#24 10 years ago

its all good man, just remember to come back and say I told you so.

unlike the sys7 crap crystal the sys9/11 had a more stable osclator/crystal that isn't as old and crappy or prone to failure and mandatory replacement like the older generation. (thus fixing it with out proper testing is a guess, using a tool that isnt good for testing. it also happens that the pin for xtal sits right next to a juicy VCC, so testing from OP is very easily resulting in bad test, while not checking /asking about blanking and hold lines.. hence.. a shot gun fix.)

also sys9 is sys7 just no interconnect bullshit. so you should feel at home.. the clock being a major difference that the sys9 oscillator is built into the U21 and on <sys9 its not.

#25 10 years ago

Andrew, if you don't mind, it'd be nice to test it. I may see if I can pick up a decent used oscilloscope myself, otherwise I'd like a chance to bring it over.

#26 10 years ago

Do you have access to clays guides?
They are easy to locate find the repairing sys11/9 and start troubleshooting it as he suggests

Step 1) remove from game
Step 2) follow guide and test your reset and blanking lines
Step 3) burn Leon rom and make 100% sure you burned it correctly then follow its testing.

Only fancy tool you need is to know someone with a burner - if you PM me I will even mail you a verified 100% working chip. If your gonna repair electrical stuff a oscope is a good investment - however a logic probe is a better one .. If you don't think a $300 oscope is a good use of cash.

#27 10 years ago

Here is where we are....

Collin brought over the board today. Test rom did not run. The clock was fine but the VMA was stuck high. Replaced a few logic gates and got the VMA pulsing as it should. Then the test was ROM was booting. Checked all the PIAs and replaced one that was dead.

Putting in the space shuttle ROMs the board gives an error code 1 (6116 ram). Socketed and swapped the 6116 ram and still error 1. Checked Q52. Verified ROMs. Now we stuck for the moment.

#28 10 years ago

I am thinking we have a memory protect issue. When i get home from work i am going to trace the memory protect out.

Reviewing leon's instructions pressing the diag switch should initiate a RAM test. Pushing the button last night on the bench with the test rom did nothing, the diag digit just continued cycling 7 and 0.

Also with game ROMs. Right at power on the diag digit goes to a 1. Should i get the error code right at power on, or should i have to press the diag button first before getting the error code?

#29 10 years ago

Thanks again for all your help Andrew! I'm surprised this board is such a cluster.

My understanding from what I've read is that error codes show up immediately on boot. Perhaps someone with personal experience can verify.

#30 10 years ago

you getting good o-scope data on all CPU address lines and memory chip address lines (AKA the same thing on each buss line leg to leg -verify that a chip socket or trace isnt making a mess)
are lines from the game rom to the MPU good on scope? (same deal)

.. its been a while since I tested with this spicific rom.. but I believe there is a clear ..pause.. then it will continue the flip.flop.. I can pop a chip in a board I have and video if you need to have good test reference..

#31 10 years ago

this sounds so much like when the playfield was plugged in at some point the wires were reversed. this "mess" is classic. specially with the blown PIA. just make damn sure when its booted again to double and triple check the MPU-> Playfield molex (black and white plugs) so you dont go backwards again.

#32 10 years ago

CPU is running fine. No crashed address or data lines. I didnt check on the other side of the buffers yet though, but i can assume they are fine since the test rom runs (maybe not?). I doubt the test rom would even run with a crashed address or data line.

The gate crashing the VMA signal was well isolated from any connectors. Unlikely a connector mishap could have killed it, but i suppose it is possible. The sound section PIA was the one bad. I am not sure if Leon's test rom tests this PIA, but it was definatley bad. Bunch of floating outputs before being replaced that where active afterwards.

Williams manual says error 1 is Check Memory protect and u18 cmos ram(replaced this with no change). I am going to dig into it here soon after work and report back.

#33 10 years ago

Spudgunman - as soon as I read your most recent post, I had this "Oh crap, he must be right" moment.

The game came to me like this, but the previous owner did say he had the head off once. For this much to be messed up, it almost seems like that has to be the case. Do you happen to know of anywhere there's a list of components that get damaged by swapping the black and white connectors? Assuming that's what happened, it would probably save us a fair bit of troubleshooting.

Thanks,

Collin

#34 10 years ago
Quoted from Collin:

Do you happen to know of anywhere there's a list of components that get damaged by swapping the black and white connectors? Assuming that's what happened, it would probably save us a fair bit of troubleshooting.

no.. we are dooin itbasically it throws HV down the logic outcome is spastic and insane.

if the sound PIA was the only one shot hard to say it was the connections, but its so darn easy to do. and causes problems just like this.

check the lines, cuz thats worth the time, and easy. also those 6116's can get ESD like a b*** you darn tootin' its quality chip. insane - but not improbable.. always the easy thing you don't expect.. especially when its new parts that you think are good. (these ram-1 errors are normally not problematic)

#35 10 years ago

The connectors did not get mixed. Just a handful of ICs where bad for whatever reason.

U43 had one bad AND gate on the CE_CMOS line that was making the board give error code 1. The orginal 6116 was good so it can be reused. Now it is booting up to a zero which should be good to go. Test ROM still runs fine and passes memory test now. Won't know for sure until Collin play tests it, but it appears to fully booting now.

#36 10 years ago

Awesome! If you don't mind, I might stop by tonight to pick it up

-1
#37 10 years ago
Quoted from barakandl:

The connectors did not get mixed. Just a handful of ICs where bad for whatever reason.

which is possibly caused by reversed connectors. or that "whatever reason" which causes barn sitting pins owned by people that could make a simple mistake of reversing two plugs that look the exact same causing failure that is similar to what you experienced. but your right, its impossible. seeing as you have repaired 1 system 9 funny you should say it can not happen. seeing as this was also "for sure a simple clock issue." I forgot.. mail order you fix them all the time.. sorry for wasting the SQL post space.

glad its working

#38 10 years ago

Check the schematics and look where U43 and U36 are. They are no where near where those connectors would send high voltage to. Those ICs do share some common inputs and outputs which is suspect, but if HV killed them it would have taken out all the other TTL chips in the path including the static ram which would be first to go.

This is a tech support forum. If you got beef with me (which i can't figure out) take it to PM.

#39 10 years ago

no beef, you clearly know what happened.

#40 10 years ago

All I know is that thanks to Andrew, the game is working!

There were several bad chips, and despite not being familiar with the hardware, it took Andrew about 2 hours at most to diagnose and repair it.

#41 10 years ago

Usually on system9 boards, if the black and white connectors get mixed, the game roms, sound roms and U9 will all be killed.

Promoted items from the Pinside Marketplace
$ 44.99
Cabinet - Shooter Rods
Pinball Shark
Shooter rods
$ 69.00
Gameroom - Decorations
Pinball Pimp
Decorations
$ 129.00
Cabinet Parts
Bob's Pinball Stuff
Cabinet parts
From: $ 1.25
Playfield - Other
Rocket City Pinball
Other
From: $ 35.00
Cabinet - Other
Rocket City Pinball
Other
$ 50.00
Cabinet - Toppers
Slipstream Mod Shop
Toppers
$ 18.95
Eproms
Pinballrom
Eproms
$ 44.99
Cabinet - Shooter Rods
Pinball Shark
Shooter rods
From: $ 33.00
Gameroom - Decorations
Rocket City Pinball
Decorations
$ 12.50
Lighting - Led
RoyGBev Pinball
Led
$ 33.95
Eproms
Pinballrom
Eproms
2,400
Machine - For Sale
Allentown, PA
2,500 (OBO)
Machine - For Sale
St Charles, IL
$ 69.00
Gameroom - Decorations
Pinball Pimp
Decorations
From: $ 1.00
Playfield - Other
Rocket City Pinball
Other

Reply

Wanna join the discussion? Please sign in to reply to this topic.

Hey there! Welcome to Pinside!

Donate to Pinside

Great to see you're enjoying Pinside! Did you know Pinside is able to run without any 3rd-party banners or ads, thanks to the support from our visitors? Please consider a donation to Pinside and get anext to your username to show for it! Or better yet, subscribe to Pinside+!


This page was printed from https://pinside.com/pinball/forum/topic/space-shuttle-wont-boot-error-code-7-williams-system-9 and we tried optimising it for printing. Some page elements may have been deliberately hidden.

Scan the QR code on the left to jump to the URL this document was printed from.