What's new
Pinball info

Register a free account today to become a member! Once signed in, you'll be able to participate on this site by adding your own topics and posts, as well as connect with other members through your own private inbox!

A/C select relay troubleshooting on a System 11 (El PinBotto)

VeeMonroe

VeeMonroe

Registered
Joined
Aug 4, 2021
Messages
2,488
Location
London
We’re troubleshooting the ’El PinBotto‘ (i.e. Cirsa PinBot). None of the side ‘A’ solenoids work.

We’ve found the A/C relay on the underside of the playfield and it appears to produce green flashes from inside the relay when it switches when we use coil test 14 to test the relay. The relay appears to click and something inside of it moves when we switch between the ’A’ and ‘C’ sides during the coil test. The ‘C’ side flashers all work.

We’re using a PinBot (Williams) manual off the internet and some general Wiki sites on System 11s to trouble shoot. We’ve checked all fuses and replaced F4, which had blown.
 
Crewey said:
@VeeMonroe oh no !
In terms of ‘A side’ what exactly isn’t firing ? Incase it’s something I’ve repaired previously or @AlanJ
Click to expand...
None of the ‘A’ side. The immediate problem is the outhole kicker and the ball shooter lane feeder because these being out means it’s impossible to start a game. But the single eject hole, three-bank drop targets, ramp raise/lower, etc. aren’t working either.

The two sides are labelled ‘A’ and ‘C’ and have a relay to switch between them. It’s apparently a feature of System 11s to allow more complex designs without adding additional solenoid wires to the CPU board (we’ve discovered from internet research, having not owned a System 11 before!)
 
If it turns out to be a case of the relay being faulty, please take comfort in the fact I was able to desolder and solder a new relay on my Robocop a few months ago with success. It just takes a bit of patience, and resist the urge to be heavy handed when it doesn’t want to come off the board
 
Crewey said:
If it turns out to be a case of the relay being faulty, please take comfort in the fact I was able to desolder and solder a new relay on my Robocop a few months ago with success. It just takes a bit of patience, and resist the urge to be heavy handed when it doesn’t want to come off the board
Click to expand...
Thanks :D

I’ve posted on Pinside too and they think the relay is likely to be fine, and the problem is with loose connections. So, that’s going to be a big desoldering/soldering job…

If anyone has any advance/alternative to that idea, we’d love to hear it - just incase it turns out not to be loose connections :)
 
Ah fair enough, I’d possibly start checking those thermal resistor boards before following hundreds of wires, the heat they permeate something may have disconnected maybe
 
Last edited:
I'd check you have all the connectors in properly to all the boards.
Edit - hang on need to look at the schematics more........

Sorry - original post was incorrect - I was looking at the flipper enable relay!

Q7 controls the solenoid select A or C relay.

However you say the relay is clicking in the solenoid test - The A side should be switched on when the relay is not energised the C side is switched on when the relay IS energised. - Check to make sure no wire has dropped off.

Is the C side on all the time?
 
Last edited:
Measure continuity across the switched contacts with the pinball off. Reading what you wrote (and what @AlanJ suggests), it may be stuck on the C side. Can happen on a short, the contacts weld together. It will still click but the armature won't be moving. There should also not be any arcing in there as you describe.
 
If the relay has managed to fuse its contacts to the 'C' side, then whatever flashbulbs are the c-side of the outhole, ball feeder, etc, would operate instead. Such as in the solenoid test, is there nothing happening during an 'A' pulse, or is the associated flashbulb circuit pulsing instead? If there's nothing on the A-side pulses, then the relay is operating. It may not be making contact for the A side, but it isn't stuck on the C side. I think that the usual check for power at a solenoid, (grounding the return side of the winding, the tab of the transistor, or any point between) still applies with Williams extended 'A' solenoids. Providing that the relay is Off, and making contact on A, as below.

As Alan points out, the A/C relay isn't switched to pulse the A side, so with the relay dormant the A-side solenoids should operate, driven by their own particular transistors. The relay operates, using Q7, to use these transistors for the C side loads (including the knocker for Pinbot, as 01C) when required. Using the schematic, see which wires of the relay pcb should be connected with the relay Off. Though the manual for Pinbot on ipdb doesn't have the handy explanatory drawing of the A/C relay circuitry that I remember from other games manuals. Maybe look at those for F-14 or other similar age games. Going by the offering for the later Big Guns, the wire colours are Red for Com, Blue for A-side and Orange for C-side.

The ipdb entry for F-14 has a supplement sheet showing the extender pcb, diode pcb, flashbulb resistor board and the explanatory drawing.

There's that diode pcb nearby, isn't there? The solenoid power is supplied to the relays' main COMmon contacts, with the relay Off it should appear at the relay N/C contacts, each A-side solenoid, and also at each A-side diode that has a solenoid connected. With the relay On, Com should switch to the N/O contacts ('C' in these drawings), the flashbulb circuits and C-side diodes.
 
Last edited:
VeeMonroe said:
The two sides are labelled ‘A’ and ‘C’ and have a relay to switch between them. It’s apparently a feature of System 11s to allow more complex designs without adding additional solenoid wires to the CPU board (we’ve discovered from internet research, having not owned a System 11 before!)
Click to expand...

A solenoid extender set-up allows a number of solenoid drive transistors to multi-task, as it were. Rather than produce an extra pcb with more drive transistors, and magic up a method of addressing them to drive more loads, as with feature lamps, using one solenoid drive circuit (maybe not even that, in Bally's method) a relay can switch power between groups of paired solenoids (or more often a solenoid paired with a flashbulb circuit, with Williams games).

It isn't unique to System 11 games, though. Bally used it quite often on their original solid-state hardware, which had 19 solenoid drives. Two of these were committed to flipper relay and coin lockout, so the limit was really 17. Games with more than this, such as Eight Ball Deluxe or M&M Pac-man, had a 'Solenoid Expander' relay fitted under the playfield to switch between groups, which varied in size (as in another recent topic, EBD only has three 'pairs', all solenoids, while Pac-man had a lot more, even including the thumper bumpers). Where Bally differed was in driving the relay; facing a shortage of solenoid drivers a feature lamp Scr drive was programmed for it instead.

Williams occasionally used the technique earlier than System 11 too, though less than Bally. Williams hardware always had 22 solenoid drives, not including flipper relay, but one instance where extension was used is Defender - after all, two intelligent 5-bank drop target banks are 12 solenoids by themselves. Even then, the normal solenoid test doesn't include the extended side.
 
VEE‘S HUSBAND HERE (who has been troubleshooting the pin all day):

Because this is a European clone, it turns out the A/C selector relay is on A when power is off and the relay coil needs energising to switch it to C. The relay is working, as is the board it is on. Now I have found a decent ground, I can track the +48V dc solenoid/flasher power (relative to backbox ground) from the common pins to the A and C pins depending on the position of the relay, and to the solder pads at the edge of the small relay board where the wires are connected.

So it now looks like there are three, count ‘em, electrical faults with this pin.
1) +48V dc is not getting from the A/C selector board to all the A side coils I was able to check, even when the relay is switched appropriately. There is about 0.3V between (both terminals on) the coils and the backbox ground.
2) Sometimes, I get a different failure mode where the relay on the auxiliary power board (which appears to be the relay turning on power to the flippers, although I haven’t done as much troubleshooting on this one) clicks constantly, the A/C relay clicks constantly when it is trying to send power to the C side, and neither the A nor the C side works when I run a solenoid/flasher test. Once the machine switches on in this ”mode”, the only cure seems to be switching off and on again.
3) The ground problem. There is 110V ac between the backbox ground and the true ground (i.e. the neutral and earth pins of the mains input, and the green/yellow earthed bits of metal in the cab (of which there are not many). When I connect them, there is mild sparking and a fault current of about 0.3A. I have no idea why this is, or if it is usual, but I don’t like it and it makes me want to slap a DANGER ELECTRICAL FAULT DO NOT USE sticker on the pin.
 
Last edited:
On ground issue: you have the ground floating in the backbox - as the transformer making the AC voltages has no ground through it you‘ll get something arbitrary as the potential to true ground will vary given the load on the various outputs. Link up the ground braid which will pull the ground potential to where it should be and you‘ll be fine.
Not getting power to A side is a broken connection somewhere. Should be easy to debug as it is on A by default, just follow the wire.
The constant switching is interesting, is that the A/C relay or flipper enable? fix the two above and then we’ll look at that.
 
Husband has just read it. He thinks it’s a heroic effort that makes our problems look positively minor!
 
1. you have a broken wire or dodgy connector somewhere. follow the A side wire until you find the break

2. if you gently press the mpu board in the centre does this happen ? sounds like a something losing contact

3. make sure the earth in the backbox is connected to earth in the cabinet.
 
We have fixed the A side wire issue - it turns out that El Pinboto has an extra fuse on the underside of the playfield that isn’t mentioned in the manual because it doesn’t exist on the Williams Pinbot, and that fuse had blown.

Unfortunately, there appears to be a new fault. The machine is not recognising that the pinballs are in the trough - I get a “Waiting Pinballs” message when I press the credit button and the game won’t start. The optos in the trough (El Pinboto uses optos where the Williams machine uses mechanical switches, but they appear to have the same position in the matrix) appear to be able to be triggered, but with two balls in the trough the “Ball trough 2” switch doesn’t trigger reliably. My current best guess is that something is physically out of place meaning that balls in the trough don’t line up with the optos.
 
VeeMonroe said:
We have fixed the A side wire issue - it turns out that El Pinboto has an extra fuse on the underside of the playfield that isn’t mentioned in the manual because it doesn’t exist on the Williams Pinbot, and that fuse had blown.

Unfortunately, there appears to be a new fault. The machine is not recognising that the pinballs are in the trough - I get a “Waiting Pinballs” message when I press the credit button and the game won’t start. The optos in the trough (El Pinboto uses optos where the Williams machine uses mechanical switches, but they appear to have the same position in the matrix) appear to be able to be triggered, but with two balls in the trough the “Ball trough 2” switch doesn’t trigger reliably. My current best guess is that something is physically out of place meaning that balls in the trough don’t line up with the optos.
Click to expand...
Good progress on the A/C issue

the trough optos are controlled by a pcb under the apron - try reseat connectors to that little board.
 
Good thing you‘re getting there. You are right, you have some sort of an alignment issue if you can interrupt the optos manually.
 
Jon has now photographed the problem with the trough. The PCB under the apron, which doesn’t exist on a US/Williams PinBot, and which is used to drive the trough optos, the LEDs on the PCB signal the ‘position’ of the optos and show all three optos are working just fine. The problem is getting the signal from the Cirsa driver board to the switch matrix.

With considerable effort, he was able to unscrew the PCB board in order to look behind it. And promptly the machine started working again!!!!

96E5A76F-C7FC-4F6D-B42D-2832B25AF873.jpeg
 

Attachments

  • B4194683-86EC-4E59-BD1E-680E676699EA.jpeg
    B4194683-86EC-4E59-BD1E-680E676699EA.jpeg
    180.9 KB · Views: 3
Next update by Jon (not greatly positive).

After the switch test appeared to be working (from unscrewing the opto driver board, reseating the plugs, and putting it back) I restored the pin to service.

First stage was doing something about the ground issue - rather than trying to run ground braid everywhere, I took the shortcut of running an extra wire on the main transformer between the neutral end of the input coil (already verified connected to the cabinet ground) and the 0V end of the 88V output coil (already verified connected to the backbox ground). This looks like it works - I am hoping that the breaker that went off at the house panel a few minutes later is a coincidence given that a fuse should have blown somewhere in the pinball machine before it tripped a breaker.

Final switch test showed that the vortex exit switch had stopped working - this turned out to be a plug that I had pulled as part of the original troubleshooting of the A/C relay.

Screw everything that was unscrewed back in, switch on, GAME STARTS! Left flipper not working - turns out the switch got misaligned with the button while I was messing about under the apron. Put it back. I CAN PLAY!

But drains are not registering correctly - when a ball drains it just pushes a ball into the shooter lane without ending the ball in play. I want to cry because I know how much trouble this fault caused AlanJ and I don’t have the skills to fix it again…

1st port of call is another switch test. It turns out that the outhole switch is completely dead - I can’t trigger it at all. So apron off to expose the driver board again. Driver board LEDs still working, so the actual switch is fine. I check the other switches on that line of the matrix (column 2 - green/red) and it looks like the outlane and inlane rollovers are also not working so the whole column may be out. I also try to troubleshoot the driver board - there is a red-green coming out both ends of it so I wonder if the column wire runs through the board.

Conclusions from taking the board out and trying to understand what it is doing without a schematic (stupid, I know, but needs must):
- The switch matrix wires on the left side of the board (which also has power leads) are switched into each other by the bank of transistors, making the connections the switch matrix is looking for. The bases of those transistors are connected to the output legs of the quad op-amp chip in the middle of the board. The inputs of that chip are connected to the green/x wires on the right side of the board which appear to connect to the opto sensors.
- The white/x wires on the right side of the board carry power to the opto LEDs - they are connected to the +ve rail via protecting resistors
- All header pins on both sides have continuity to where they are supposed to, so given the on-board LEDs are also working I am confident that the problem is not with the board, the headers, or anything on the right hand side (opto LEDs/sensors/cabling connecting them)

There are two sets of green/x wires going through the hole on the right of the apron - it looks like one of them is the feed from the RHS of the board to the opto sensors, and the other is not connected to anything (there are a bunch of loose connectors tied up under the apron). But I am reasonably confident that the LHS wires go to the switch matrix.

One other test - I can trigger the switch matrix by manually sticking wires into the holes in the plug where the header pins go, but only the trough 1 and trough 2, not the outhole.

So now this is a wire troubleshooting problem again. Either the plug needs rewiring, or the green/red switch matrix wire is broken somewhere. The latter is more likely than you would think, because I am pretty certain that a plug pin would only affect one switch, not the whole column.
 
And the wire troubleshooting comes up clean - I have continuity between everything I can find on that wire, including (it appears) from the plug in the backbox to the solder pad on the opto driver board. I am now officially confused.

DC Levels at the left hand of the opto driver board for anyone familiar with switch matrices of this era:
1: Ground (I am still getting readings of a few 10ths of a volt between things which should both be grounded, so I am measuring relative to this as a local ground)
2: 11.3V (power)
3: 0V (not used)
4: 2.8 V (red-greed to switch matrix, faulty)
5: 0V (not used)
6: 4.55V dropping to 3.75V when outhole opto triggered (white-grey to switch matrix, faulty)
7: 4.55V dropping to 4.05V when trough opto 2 triggered (white-red to switch matrix, working)
8: 4.55V dropping to 4.05V when trough opto 1 triggered (white-brown to switch matrix, working)
9: 4.44V (green-yellow to switch matrix, working)
10: Shorted to 9 in plug wiring - this is correct given my understanding of the board setup.
 
VeeMonroe said:
And the wire troubleshooting comes up clean - I have continuity between everything I can find on that wire, including (it appears) from the plug in the backbox to the solder pad on the opto driver board. I am now officially confused.

DC Levels at the left hand of the opto driver board for anyone familiar with switch matrices of this era:
1: Ground (I am still getting readings of a few 10ths of a volt between things which should both be grounded, so I am measuring relative to this as a local ground)
2: 11.3V (power)
3: 0V (not used)
4: 2.8 V (red-greed to switch matrix, faulty)
5: 0V (not used)
6: 4.55V dropping to 3.75V when outhole opto triggered (white-grey to switch matrix, faulty)
7: 4.55V dropping to 4.05V when trough opto 2 triggered (white-red to switch matrix, working)
8: 4.55V dropping to 4.05V when trough opto 1 triggered (white-brown to switch matrix, working)
9: 4.44V (green-yellow to switch matrix, working)
10: Shorted to 9 in plug wiring - this is correct given my understanding of the board setup.
Click to expand...
And I am going to call it as a fault on the CPU board. I unplugged the columns of the switch matrix (1J8) and measured the DC voltage on the header pins with the switch test running. All are at 4.56V (relative to the ground plate of the backbox) except pin 2(green/red, faulty) which is 2.92V and pin 1(green/brown, working) which is 4.62V.

Based on my limited understanding of switch matrices, these measured DC voltages are duty cycle averages between +12V or thereabouts and 0V as the machine cycles through the columns faster than the multimeter can keep up, and the different voltage on pin2 means that the line isn't going to 12V when it should.
 
if a column is faulty, all the switches in the column will be faulty. check some of the others. if so replace the column driver transistor. might be q49?
 
AlanJ said:
if a column is faulty, all the switches in the column will be faulty. check some of the others. if so replace the column driver transistor. might be q49?
Click to expand...
Jon says that makes perfect sense. Thanks for working the Xmas shift on this one 🥰 🤶 🎄
 
Please do the ground properly, there should be a braid running through the whole machine connected to main earth. Reason: if anything in the machine fails short to ground the fuses will blow. If you just jury rig connections you may end up defeating this and create shock hazards. Rant over. As @AlanJ says, make sure the whole column is out. A very easy way to do this is shorting row and column pins in switch test directly at the CPU with a cable with connectors removed. That will tell you if the CPU is OK. Should that be the case and the cabling not broken you have defective components on the opto board. You can short row/column on the opto board to see if that is the case. Careful, there is power there as well which you don’t want in the matrix.
 
You must log in or register to reply here.
Back
Top Bottom