In Progress Firepower – From Powerless to Mission accomplished

[Moonbus]

My first ‘fixing it’ documentation for you guys, I now have 3 games, 2 of which are machines in various stages of ‘getting there’.

Magic Circle – 1960’s Bally EM – Working but in need of a total overhaul.
Mr & Mrs Pacman – 1980’s Bally SS – Bought as a box of bits minus cab.
Firepower – 1980’s Williams SS – Bought from Andy on here.

Only been in the game a few years and got into it as I have always liked tinkering and finding solutions to problems, electrical and mechanical.

My history of pinball abuse can be found at the following locations if you are so inclined.

Magic circle story can be found here….
https://pinside.com/pinball/forum/topic/bally-magic-circle-semi-sympathetic-playfield-restoration

Pacman story can be found here…
https://pinside.com/pinball/forum/topic/mmpc-first-ss-second-pin-coffee-table

Chapter 1 - Introductions

I’m going to start this off with a warning. This is not going to be a quick project. My time is limited and I have to priorities it in the following order…

Money making Job
2 Boys under 5
2 other machines
This machine
House maintenance
Wife

I am also relegated to doing anything productive outside as space is limited meaning that at this time of year, with the light limited, I am not going to be doing much more then scratching the surface of this project.


So with all that out the way…ONWARDS

This machine has been around a bit. It was originally sold by @johnparker007 to @andy and Andy ended up selling it on to me. Both I gather had intentions of doing it up with the included NOS playfield but in both occasions their missions failed. Is this a jinxed machine, well it did take 2 months for Martin’s BIL to deliver it so it may well be. (Martin was 100% sympathetic to my plight)

Know History: Machine boots into test mode if playfield is unplugged. Plugging in playfield causes playfield fuse to blow.

Initial Findings

Bottom cabinet External​

• Structurally sound, usual knocks and dents found on any wooden box of this size and age
• Joints and surrounding paint not showing any signs of stress or strain that I would expect if the box had been dropped or twisted.
• Section missing from bottom lip on rear chipboard panel. Andy included the missing bits.
• Paint not showing any signs of fading.

Bottom cabinet Internal​

• Dirty. Nothing a wipe down wont initially fix.
• Standard rust on lockdown bar receiver.
• Speaker a little corroded.
• Power switch relocated then removed.
• Improvised relay circuit to trigger 2 counters.

Backbox External​

• Solid but slight movement in joints. Top right hand corner is slightly separated and shows symptoms of being knocked or dropped.
• Additionally it has the usual signs of abuse for a bit of furniture of this age.
• Paint not faded.

Backbox Internal​

• Display panel crooked, as if it has been hung on when open, or maybe from a drop.
• Trim and locking mechanisms look ok.
• Everything is labeled. To me this indicates some troubleshooting has been done.
• Galvanized fencing wire added to GI line on back of display board.
• GI line to display board patched to bypass original connector.

Backglass​

• No cracking and only a minor bit of black missing from around one of the displays.
• Not pure white on the backside but I don’t know what it should look like.
• It may be a different story once it is lit up in the box but it is a good starting point.

Boards​

• 40pin connector replaced with a birds nest.
• Batteries relocated into remote pack.
• Some signs of board work but nothing major.
• Standard things such as resistors and caps replaced.
• Driver board burnt under resistor bank.

Original Playfield Topside

Guess it is the expected wear for this game but the top end of the playfield has been hammered by the top kickout hole.​

Original Playfield Underside​

• Galvanised fencing wire added to GI line.
• Mechanisms all rust free but all lubricated with oil.
• All kicker mechs seized up as oil has hardened.
• Spinner link installed wrong way round.
• Lamp sockets all have additional wire jumpers from solder tags to main lamp body.

Plastics

Most are cooked by the GI lamps.​

NOS Playfield​

• Showing signs of age and a lifetime outside a game.
• Slight planking, not lifting but the wood drying / ageing has split the paint.
• Screen printing is misaligned.
• White is yellowed. This is probably due to the protective layer on the playfield reacting with light.
• Inserts all whole, transparent and imperfection free.

Initial Summary

Generally happy with the machine, bought it as a project and that is what I got. It looks like the machine has had some investigative work done on it, the backbox has been dropped and the GI lamps have cooked the plastics. Someone has tried to rectify this by improvising giant resistors out of galvanized fencing wire.


Sounds like something we can get our teeth into.

Firepower Photo Album
https://1drv.ms/a/s!Aqp7OhVuaaqngeVUA0SC3fuZotO4Bw

 

[Moonbus]

OK, where can i find info on embedding images as what I did above did not seem to work

Edit: I'm trying to add individual images from the OneDrive album linked at bottom of first post.

 

[kevlar]

I don't know how to do what you're trying to do, usually if people want to show onedrive album photos they just post the link. Personally I upload the photos directly to the thread using the ' upload a file ' button, this forum is setup to host photos so you don't have to store them elsewhere.

 

[astyy]

I've noticed when you paste images in you can then click on them to enter an image URL.

It's great that you're writing your shop log on here I'm predicting that no airbrushing will occur indoors this time

 

[astyy]

What's going on here - GI OCD (image has link btw which still doesn't work! - @nedrued will know his shoplog pictures link to flikr)

 

[Jay Walker]

Hi,

In the close-up showing the underside of the top right corner of the playfield, that looks like a short circuit right there if it's the one piece of fencing wire. The G.I. of this game is all one common circuit for playfield, backbox insert and coin door, with a 20A fuse - one of the three on that square of fibre behind the large blue capacitor.

 

[Moonbus]

It's great that you're writing your shop log on here I'm predicting that no airbrushing will occur indoors this time

Thought I better get used to this forum and how it is coded. Felt guilty always linking to pinside.

The airbrush is now a garden job

 

[Moonbus]

Hi,

In the close-up showing the underside of the top right corner of the playfield, that looks like a short circuit right there if it's the one piece of fencing wire. The G.I. of this game is all one common circuit for playfield, backbox insert and coin door, with a 20A fuse - one of the three on that square of fibre behind the large blue capacitor.

The spiral of wire is indeed the fencing wire. Whole playfield needs checking as it seemed to be the solenoid fuse that was blowing.

 

[Moonbus]

Chapter 2 – The Plan

• Fix this machine up so it is playable.
• Paint and clear the NOS playfield.
• Patch up and custom spray the cabinet.
• Hopefully end up with a machine that can live in the house with the rest of my family.

List of initial things to do…

• Replace all fuses - Done

Playfield

• Remove fencing wire - Done
• Remove all plastic over GI – Done
• Remove all bulbs - Done
• Check for shorts on GI circuit – Done
• Check for shorts on Switched lighting – Done
• Check diodes are still diodes on all switch lamps – Done
• Check diode polarity on lamps and solenoids – Done
• Check diodes are still diodes on solenoids – Done
• Unsolder all grounds from solenoids – Done
• Check all solenoid resistances against resistance chart – Done
• Check solenoid power line for shorts – Done
• Test each pin in connectors against all other pins for shorts – Done

Backbox

• Test transformer voltages - Done
• Test voltages on PSU board – Done
• Test bridge rectifier voltages – Done
• Check for ground shorts – Done
• Install pre rectifier fuses or new bridge rectifier board –
• After game is working fix 40 pin connector –
• Remove fencing wire from GI line – Done
• Test for shorts on GI circuit – Done
• Fix displays board brackets – Done

Bottom Cab

• Hoover out – Done
• Test grounding – Done
• Box in surge protector and thyristor assembly –
• Add power socket –

 

[Moonbus]

What's going on here - GI OCD (image has link btw which still doesn't work! - @nedrued will know his shoplog pictures link to flikr)

I wanted an easy way to count the bulbs to make sure I had them all. Turned out I missed two on the GI line so when I did the continuity test I had a short.
This resulted in me removing each socket from the GI circuit in turn to test them for shorts only to find bulbs in the last two to check, they were under the metal ball guides by the flippers.

 

[Moonbus]

To expand on some of the processes I have been though above.

Testing the lamp sockets.
As all the lamp sockets had additional bits of wire soldered onto them I wanted to be sure that there were not any shorts. You cant test the sockets with the blubs in so they all had to come out.
With the bulbs out it was just a case of testing continuity across the two solder tags on the lamp older. The diodes could also be checked using the diode function on the multi-meter.

The GI circuit is like a train track. live is one rail, ground braid is the other rail and the bulbs are the sleepers. With all the bulbs removed a continuity test across the live and return wires should reveal an open circuit. I got a buzz.
To isolate where the short is you need to find places where it is convenient to break the railway tracks, usually this is defined by the ground braid running around the playfield. Working from where the GI wires enter the playfield,isolating sections of the GI circuit and testing them for continuity should tell you where the short is. Once a section is isolated each bulb holder in that section can be checked. My buzz was because I had overlooked two GI bulbs under the metal ball guides next to the flippers.

Testing the solenoids
You cant really test solenoids whilst both solder lugs are connected to anything. If the resistance of whatever the coil is connected to is less then the resistance of the coil you will get false readings. Same way as you cant test diodes on coils unless one leg of the diode is detached from the coil.
Once I had the coils prepped I tested their resistance against the cart here...

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

As nothing dodgy had presented itself so far I decided the next thing to do to test for shorts would be to test for shorts on the connectors. So I used the multimeter to test every pin, to every pin on all connectors. I did this in both directions to take into account diodes. This found nothing wrong as well.

So I am now thinking that the fault is probably going to be a transistor on the driver board that is a knock on from an earlier issue on the playfield.
A few coils have been changed. Later I will check the schematics to see if they all have anything in common.

 

[Moonbus]

Ticked these off last night...

• Test transformer voltages
• Test voltages on PSU board
• Test bridge rectifier voltages
• Check for ground shorts

Tested the voltages on the plugs from the transformer and on the GI fuse line.
6.3vac


18Vdc


28Vdc


90Vac


18Vac


I then removed all connectors from the power supply board to other parts of the machine and plugged in the plugs from the transformer. I could then test the outputs from the power supply board. These were all within range.

A note on fuses. Always check them.
This machine had the following in it...

0.5A instead of 0.25A on the High voltage line.
20A instead of 8A on the lamps line.
2.5Asb was missing - the fuse that keeps blowing
20A instead of 10A on the flipper line.

 

[Moonbus]

Some pics of the NOS playfield just to document things that will need addressing in the future. Something curious is the first picture it a true representation on the colour of the white, in the other pics the colours are vibrant and the whites are white. I think this is magic caused by the phone camera. Hopefully this is an indication that the discolouration is in the clear and not the paint.

True white colour


The playfield date, I think the date is Mar 11 1992


Some of the other issues...


Wonky routing ?

 

[Jay Walker]

Those last two pictures might be a faint echo of the prototypes (and early production), which had cut-outs there for rebound switches behind the drop targets which were originally intended.

 

[Moonbus]

Those last two pictures might be a faint echo of the prototypes (and early production), which had cut-outs there for rebound switches behind the drop targets which were originally intended.

I had a similar thought, as if someone was using a drop target playfield as a guide and tried to freehand over the gap. There are a few other stray cuts around the edges but nothing worth worrying about. Did they have automated CNC for these playfield runs or was some bloke in a shed cutting these all out by hand from a template ?

 

[Moonbus]

Ticked these off last night and today...

Playfield

• 
  
  
  • Check diodes are still diodes on solenoids

Backbox

• Remove fencing wire from GI line
• Test for shorts on GI circuit
• Fix displays board brackets

Bottom Cab

• Hoover out
• Test grounding

...and I moved onto turning the power on.

Chapter 3 - Getting a heartbeat

Now most of the preflight check had been ticked off I decided to power the game up and address the know problem of the Solenoids fuse blowing on power up. As I knew the playfield was as it should be I was 100% the issue was in the backbox, either with a connector or something on the boards.

The machine looks like this at the moment. Unconventional I think but its easy to work on and I can reach and see everything.


I plugged the power connector from the base box into the transformer and the solenoid cables from the playfield into the black plug and 2pin in the back-box.
I did not have the other playfield, displays or base box connectors attached as I was only interested in the solenoids.

Made sure nothing else was connected or making contact and flicked the switch and jumped out of my skin as a load of solenoids fired on the playfield at once. Instantly turned power off and the solenoid fuse had blown. 1 fuse down, 9 to go.

Fuse 2
In the quest of achievement sacrifices must be made, Fuse 2 shall be sacrificed in order to greater understand what is going on. We shall do the same as before but this time video it so we can see what solenoids fire. This revealed that the following all fired at the same time, thus blowing the fuse.
Ball Release
Left Ball Save
Ball Ramp Thrower
Right Eject Hole
Left Eject Hole
Upper Right Eject Hole

A quick review of the documentation...
http://www.firepowerpinball.com/downloads/
...revealed that all these solenoids were on the same socket on the driver board and that in the world of firepower solenoids they were all unique in not being special and therefore not switched.

Googling prior to this machine turning up had me prepped to replace a few transistors but I thought having 6 faulty in one session was a little odd so I dug deeper into the circuit.

Solenoid - Connected to - Controlled by
Ball Release - Q15 - IC1
Left Ball Save - Q27 - IC2
Ball Ramp Thrower - Q29 - IC2
Right Eject Hole - Q23 - IC2
Left Eject Hole - Q21 - IC1
Upper Right Eject Hole - Q25 - IC2

Fuse 3
So all faulty solenoids were controlled by two IC's. Lets see if the fuse blows if only two solenoids are powered, one on each IC. Upper right eject hole and Left eject hole were selected as they were easiest to deal with, all the others had their return wires removed.
Power on, the two Mechs fire and lock on. Fuse 3 lives.

I decided to pull the board and change the four transistors associated with the Upper right eject hole and Left eject hole. Power on, no change. Could IC1 and IC2 be bad ?

IC1 had been changed before and there was a rogue bit of red wire next to it, I planned on changing this later once the game was confirmed working but I thought I may as well do it now.

IC1 Topside


IC1 Bottom Side


Cleanup


After


Power on, same happens, Fuse 3 is still alive.

Time for the logic probe. Using the test-points TP9 (+5V) and TP10 (Gnd) on the CPU board to power the probe I tested for the Low to High pulse on test-point 4. If you get the low to high then at least the CPU board is behaving reasonably. This all checked out ok and the same test was done on the driver board for pin 37 on the 40 pin connector to make sure the signal was getting over there.

Whilst doing this test I noticed that the solenoids were popping at the same time the signal changed from low to high on the probe. To me this means the circuit is working so I could assume for now that IC1 and 2 and the associated transistors were not at fault.

Next up the line was the 40 Pin connector. I really didnt want to have to address this now but I thought I better have a look at the back of the CPU board so out it came.
This is what I found...


... a bit hard to see but centre picture there is a floating wire that should be connected to pin 7 (I think). This was soldered back. All headers on the power supply were reflowed as the boards were out and the head was reassembled.
Power on and no solenoids go off, success, or is it, only one way to find out, plug the screens back in.



Hmmmm, Never worked on a williams before GOOGLE, Oh its the pinball version of the Blue Screen of Death but is most likely due to having faulty memory, or you forgot the batteries, Doh.



Hooray.

Once the displays were working I shut down and plugged the bottom box harness in so I could use the test functions in the front door.
Found a video on youtube...

...and set about testing the solenoids and displays.

So far so good.

 

[Jay Walker]

Well done sorting that problem. The "1 497 6" display is the entry point for bookkeeping audits, with 1 being the pre-L7 cpu, FirePower's model number*, and rev.6 software. The "04" signifies the 4th sub-section of the tests, after displays (not numbered), feature lamps, solenoids (inc. audio) and switches.

* this is helpful if the cpu has unknown program chips - I once booted up such a board to find that the program was Flash, #486

 

[Moonbus]

Well done sorting that problem. The "1 497 6" display is the entry point for bookkeeping audits, with 1 being the pre-L7 cpu, FirePower's model number*, and rev.6 software. The "04" signifies the 4th sub-section of the tests, after displays (not numbered), feature lamps, solenoids (inc. audio) and switches.

* this is helpful if the cpu has unknown program chips - I once booted up such a board to find that the program was Flash, #486

When I googled the error code it was a relief as I was still not sure the CPU board was working. All the ROMs on it have heatsinks stuck on so there was little indication of its past apart from 'Firepower 2'' written on the board it in pen. So to know it is firepower is a nice feature.

 

[Moonbus]

Made some temporary plastics today so I can at least have the game running whilst I make some new ones.



Also I got a delivery of LEDs... thank you @alchy for the quick service.

 

[Moonbus]

Decided to finish removing the last of the additional items that had been adde to this machine, two counters that had been mounted above and below the lift tilt rail along with t a nice little circuit board.


The board intercepted the return from two of the coin mechanisms and is powered from the tilt switch next to it. Top counter recorded number of 50p droppes and the lower counter 10p.


Going by the game card I make that £2350 and 40p meaning about 23,504 games, at least 70,512 balls since the counters were added. Probably explains some of the playfield damage. I doubt the counters were added from new either.


Extra cables had also been run through to the coin door, this was pulled out and the original cables reattached.


..and the original cables were joined where the circuit board was...


Finally for this stage of things I used some cardboard to trick the ball trough switches...


...and I powered her up with all connectors plugged in for the first time. Coin counter switches worked when activated internally and the credit button when pressed released the marvellous sound...
F I R E P O W E R

We got a working game, one that is on its side, in 3 sections propped up against a wall, but it shows life, responds and talks. Thats more then the wife so it must be alive.

Next up I will run some switch and lamp tests and then its on to the next phase.

 

[Moonbus]

Quick update.
I laid the playfield flat and propped the top edge to give it some incline. Removed the cardboard used to trick the switches and loaded up 3 balls.
Powered up the came, hit the credit button, a ball was kicked out and firepower was announced. There was not any background sounds.

I ran the ball around the playfield and tested all the interactions, switches and switched lamps and they all seemed fine with pongs and pings but there were still not any of that tension building background audio.

The test button on the sound card gives a lovely demo of speech, sounds and background noises but in play only speech and pongs can be heard.
I'm going to assume that as all the sounds work in test mode that this is going to be a signal issue from the CPU/Driver boards to the sound board and it will get fixed when I do the board rework phase.

During the playfield test I managed to test all game sequences including multiball thus marking the end of chapter 2.

 

[Moonbus]

Chapter 3 – Safety and Reliability
Now the game is certified working, before I put it all back together there is a new list of things I would like to tick off with the aim of making the games safer, reliable and child proof.


Display Board

• Clean out all lamp sockets - Done
  
• Sympathetically LED - Done

Harness

• Replace all IDC sockets - Done
  
• Replace / Re-pin display board GI and SL connector - Done
  
• Add special solenoid fuses -
  
• Tidy up mounting position - Done
  
• Add 2.5mm grounding wire to additionally ground all boards -

Power Board

• Replace all headers - Done
  
• Capacitors, diodes and high voltage sections already fixed by previous owner - X
  
• Install fuses on rectifiers - Done

Sound Board

• Replace all headers - Done
  
• Sounds working so not replacing capacitors at the moment - X

CPU Board

• Replace all headers - Done
  
• Install NVRAM ( thank you - @lukewells ) -Done
  
• Remove remote battery holder - Done
• Check IC sockets - Done
• Double check 6810's and re-socket - Done

Driver Board

• Replace all headers - Done
  
• Rebuild Lamp Blanking section with zero Ohm resistors and IRF9Z34N mosfets ( https://pinside.com/pinball/forum/topic/vids-guide-to-bulletproofing-williams-system-6#post-594559 ) - Done

40 Pin connector

• Replace through hole headers and sockets with a 90 degree system - Done
  
• Mounting system to better secure CPU and driver board as originals have been tweaked - Done

BackBox

• Find a way of adding PETG sheet to protect backglass from kids -
  
• Think about relocating transformer to lower cabinet -

Lower Cabinet

• Box in surge protector and thyristor assembly -
  
• Add power socket –

 

[astyy]

but there were still not any of that tension building background audio.

Perhaps the game adjustment (35) for background sounds has defaulted to OFF.

 

[Moonbus]

Perhaps the game adjustment (35) for background sounds has defaulted to OFF.

See, this is why its great to be a part of a comunity like this. It would probably have been another week or so before I stumbled across that setting.
I will give this a go tonight before I start attacking the boards.

 

[Moonbus]

Perhaps the game adjustment (35) for background sounds has defaulted to OFF.

Turned out to be the DIP switches on the soundboard, documentation I have said to have both set to on. Just found an old thread on pinside where someone was moaning about the same symptoms I have and it turns out turning DIP switch 1 off on the soundboard gives you speech and sound. Should have seen my face, very happy indeed to finally hear the game in motion in my own house. Actually feels like I own a firepower now.

 

[Moonbus]

Anyone got suggestions on cleaning and de-gunking bayonet bulb sockets?
I'm getting fed up with bulbs getting stuck in and pulling their heads off trying to get them out.

 

[Jay Walker]

Turned out to be the DIP switches on the soundboard

AIR one switch is for speech module attached (or not), and the other for simulated chimes or 'proper' sounds

 

[Moonbus]

AIR one switch is for speech module attached (or not), and the other for simulated chimes or 'proper' sounds

Indeed, also seems to depend on the jumper config on the sound board, just checked the docs I got from IPDB and firepowerpinball.com and both say to have both jumpers set to on. It works with DIP 1 off on mine so I am not going to hunt anymore.

 

[Moonbus]

One more problem to address in the board work. Added to the list that the memory test fails and the bottom LED stays lit and displays are blank. Hopefully this will sort itself out when I check the MPU over.

 

[Moonbus]

Future addition to project.
Inspired by this thread on here ( http://www.pinballinfo.com/community/threads/shaker-motor-do-i-need-one.38140/ ) I will add vibrating buttons on multiball.

Using additional switch leafs added to the 3 ball trough switches and the 3 kickout hole switches I shall activate a shaker motor attached to the flipper buttons.

The shaker motors will be similar to those used in console controllers and they will be mounted on a plate that slides under the retaining ridge on the flipper button shaft.