What's new
By:
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!

Strange Williams System 7 End of Ball Issue

Fantazia2

Fantazia2

Registered
5Years
Joined
Nov 3, 2015
Messages
1,994
Location
Darlington, UK
Hi,

anyone come across anything like this before on a Williams System 7 (Jungle Lord)

All solenoids test ok in solenoid test and all switches test ok in switch test, apart from the upper sling scoring switch but that just needs adjusting, its not stuck and manually actuating it works ok.

So the issue is I can start a game, all the drops reset as normal and it kicks the ball out, the game runs ok and everything works including the lower drops and upper pf kick outs as expected, but when the ball drains it doesnt trigger the end of ball sequence, the ball gets kicked out of the out hole back onto the ramp so both balls are there but it doesnt add up the bonus and kick out the next ball.

Testing things on the playfield when this happens all the switches work ok for targets/roll overs etc and score still counts up, the 5 bank drop targets on the upper playfield still work as expected running through the sequence of 1 up, 2 up etc and go al the way through double trouble, etc but the lower 2 sets of drop targets no longer trigger anything happening along with the upper pf kick outs.

Sometimes I can eventually trigger the end of ball by hitting various switches and the upper 5 bank drops over and over again.

Other things I have checked include tacking the apron and then then starting a game, grabbing the ball mid game and sticking it straight onto the ball ramp (bypassing the out hole) which then triggers it kicking a ball back into the shooter lane again, if I take the ball off the playfield mid game and trigger the out hole switch it fires the solenoid to throw the ball back onto the ball ramp but the above issues with the lower drops and kick outs dont happen and the game continues as usual, if I manually trigger the out hole switch and then put the ball on the ball ramp then the issues start again.

Anyone come across this or could suggest what to look at to fix.

I dont think its a switch matrix/PIA issue as all the switches work as they should and the issue doesnt appear to start until the out hole and then the left ball ramp switch is triggered which should then start the end of ball sequence.

Could it be a ROM or a processor issue as it appears to be still running and not locking up as other switches trigger sequences as they should and the score still counts up, almost like its missing the sequence when it should be triggering the end of ball.

5v on the MPU test point has been measured at 4.97v, havent test all the voltages at the ICs themselves yet as need to make a map of where they are on all the different chips.

Diagnostics button on the MPU quickly shows 0 on the 7 segment display and then goes out and the game starts attract mode ok.

Have put an NVRAM in as the batteries have been removed previously, but also tested with the original 5101 and the same happens.

Any ideas anyone?
 
does it only happen when some of the drop targets are down? ie try starting a game , score some points but don’t drop any of the drops. then drain a ball. what happens?
 
Doesnt appear to make any difference, drops work like on Black Knight on a timed mode so usually they are all up when you drain.

Have tried all permutations of this, draining with no drops ever hit, draining after drops hit and reset, and draining when drops hit and still in timer mode. When still in timer mode the flashing light for the timer stops flashing and stays on solid.

Issue seems to start after triggering the out hole and then the left ramp switch.

Have tired tilting the game as well and get the same sort of behaviour in terms of out hole and ramp switch, after tilting if I put the ball in the ramp manually with touching the out hole switch it will kick a ball straight back into the shooter lane, if I trigger the out hole and then the ramp switch it then sits there and does nothing.

Slam tilt works though and resets the game back to attract.

So the out hole switch is definitely triggering as it activates the solenoid, and the left ball ramp switch is definitely triggering as thats when the issue start from it being triggered after the out hole is triggered.

Another thing I have noticed is once it gets stuck the score stops "strobing" (not sure if thats the right terminology) which seems to be normal behaviour when the ball is in play based on youtube videos, guessing to show which player is currently the active player.

Also tested this by manually triggering the out hole switch and the score keeps strobing as per when in game, it doesnt stop until the left ramp switch has been triggered after the out hole, going by game play videos the when the ball drains the player score stops strobing and the bonus score is added before kicking out the next ball into the shooter lane and the score starts flashing, followed by strobing once the ball is in play.

Its likes its getting stuck at the point of adding the bonus points to the players score before it triggers the reset of the playfield and kicks the next ball out.
 
Ive also just tried triggering the auto ball save feature by hitting the turnaround from the left hand side, got that lit for both lanes then rolled the ball over the outlane, which triggered the ball save mode with the lights going out, then put the ball in the outhole and it kicked it over to the ball ramp and then the ball was kicked out into the shooter lane and the game went back to normal mode. Hit a playfield switch to switch the ball save to the other out lane and tested and it did the same again, so it does look to me to be related to when it gets to the point of running the end of ball sequence in the program code and adding the bonus on etc.

Would that be something ROM related or RAM related?

There is an new tricks version of the JL roms that I think I may get a copy of and try that to see if the behaviour changes.

Ive also found this linkwith some primer information for writing your own game code, will have to give it a read later on. http://gamearchive.askey.org/Pinbal...inBuilder/text/williams_lvl7_programming.html
 
Possibly have found the issue, decided to take the ROMs out to reseat and found this in IC26.
Bit hard to see from the first picture but one of the legs was bent under the chip itself. Annoyingly I dropped it before I could picture the underside and the leg dropped off.

Replacement standard and New Tricks Roms ordered and will have to wait for them to arrive to test further.

IMG_20230727_183536.jpg

IMG_20230727_183624.jpg
 

Attachments

  • IMG_20230727_183624.jpg
    IMG_20230727_183624.jpg
    19.8 KB · Views: 0
Your mention of similarity to Black Knight has at last prompted me to recall a possible reason for switch matrix problems; is the driver board an earlier one?

The earlier games had resistors in the column drive circuits, R 204 to 211. In the later driver board for L7 games, these were eliminated. The 'Summary of Features' sheet for Jungle Lord mentions "As in Black Knight, complex playfield features result in numerous closed switches. This requires R 204 thru 211 (if present, as when an older board is fitted) be replaced with zero-Ohm jumpers".
 
Last edited:
Thanks but it wasnt a switch matrix issue, you are right though later boards replace resistors with jumpers to improve responsiveness of the switch matrix, at least that's what I've read and my driver board already has those in place.

Issue appears to be due to the missing leg on one of the game ROMs causing an issue with the code.

Surprising that it passes the ROM tests OK but then it occurs to me you can get an alternative rules set ROM so it's probably not checking the actual full content of the ROM.
 
It is checking the full ROM via a checksum. You lost pin 18 which is address line 11 on the 32k chips. You are probably not using this line and have 16k chips. If one address or data line was missing nothing would be working. Don’t think this is your issue sadly therefore. Have you checked for correct switch activation with balls, not fingers?
 
drhex said:
It is checking the full ROM via a checksum. You lost pin 18 which is address line 11 on the 32k chips. You are probably not using this line and have 16k chips. If one address or data line was missing nothing would be working. Don’t think this is your issue sadly therefore. Have you checked for correct switch activation with balls, not fingers?
Click to expand...
Oh, well thats disappointing to hear, yes have checked with balls as if you get the ball save mode working and the ball drains if saves its and fires it back into the shooter lane.
 
Sounds like you have a problem with switches somewhere. Weird behaviour like this could be a socket problem, or a bad rom, bad PIA, or even corrosion. Did you cheked not havig corrosion in connectors? One time a had a machine with weird switches triggering and the problem was a huge corrosion spread in the BIG White connector between the head and the and playfield. But also, I would try with new rom (flipper rom too), and another pia, just to try to isolate the problem. I would check too the condition of the IC sockets as sockets are always problematic in this williams games as also the interconnector. Then, check the cables under the playfield to see if theres any cable doing a shortcut maybe.
 
Can't see it being a switch issue as they all work correctly and the issue only starts during game once the left ball ramp switch is triggered after the out hole switch and the game not being in ball save mode, if you setup and trigger ball save mode the game does the correct sequence of trigger out hole to put ball on the ball ramp and then it kicks the ball out to the shooter lane to save it.

Hence why to me it's says its a program issue and not the switch matrix/PIA. So either ROM issue or could be the game RAM. It literally seems to trigger everything when the ball drains up to the point where it starts to add the bonus, so the MPU sees that the ball has drained and is on the ramp as the player score stops flashing/strobing like it does during play.
 
You said: "I can eventually trigger the end of ball by hitting various switches and the upper 5 bank drops over and over again". Thats sounds like a strange switch behavior. I would add to check the 4049s.
Dont you have the Leon test rom? With that you can check the ram and test the outputs of the PIA.
 
I have the Leon test ROM on the way with some game Roms and also considering looking into a pincoder.

The reason I don't think it's switch related is I can't always seem to repeat the behaviour to trigger the end of ball very often so was thinking it's getting stuck in the code and eventually doing a load of things gets the code unstuck.

The CPU itself is seeing the out hole switch as it fires the Solenoid and then it is seeing the left ramp switch being closed as it starts to do some of the end of ball sequence like stops the player score scrolling, but you can still activate things like the 5 bank drops and they reset in the correct sequence and still add to the player score along with all the switches other than the mentioned drops and kicks that don't respond but others on the same rows and columns of the matrix trigger ok.

I could do with knowing what the end of ball script/process is to pinpoint where its getting stuck, the link above about the flipper software mentions that several threads run and to me appears one of them is stuck which is doing the end of ball and the others are still running OK as you can trigger events in the game that work like they should like the 5 bank drop targets.

He also mentions there is disassembled code out there which could be very handy to help with this as it would show what the end of ball process is and what it's trying to do.
 
drhex said:
It is checking the full ROM via a checksum. You lost pin 18 which is address line 11 on the 32k chips. You are probably not using this line and have 16k chips. If one address or data line was missing nothing would be working. Don’t think this is your issue sadly therefore. Have you checked for correct switch activation with balls, not fingers?
Click to expand...

Had another check and these are 2716 ROMs and pin 18 is is the EP pin and should be pulled low for read operations. Not sure if having it not attached would cause any problems so it may not be the issue, but all the other legs on it looks a bit shady as well.

I think you might have been referring to a 2532 which pin 18 is one of the address lines.
 
Fantazia2 said:
Had another check and these are 2716 ROMs and pin 18 is is the EP pin and should be pulled low for read operations. Not sure if having it not attached would cause any problems so it may not be the issue, but all the other legs on it looks a bit shady as well.

I think you might have been referring to a 2532 which pin 18 is one of the address lines.
Click to expand...
OK, maybe wasn't clear enough. EP inverted is the standby line on the 2716 EPROMs - this is unused in this application and needed if another chip occupies the same address space as the ROM. On the 2732 chips this is A11 and the board can be jumpered to use these. You can then concatenate ROMs but this hasn't happened on your board as the ROM test would fail for sure. So losing this leg doesn't make a difference in your case. I would therefore be surprised if a new ROM fixes your issue. The ROM could obviously have further issues, but that should normally be flushed out by the testing.
It obviously is possible for the RAM to have a problem but that manifesting in exactly the same way always would be surprising - I don't know the architecture and software well enough to know if the memory image always is the same but I would expect this to affect multiple routines. Also, RAM is tested at startup as well so should throw an error. Still think it is more probable you're having an (intermittent) switch issue therefore. Let's see what the new ROM does and take it from there.
 
An update

Replacement ROMs and test ROM arrived, tried the replacement ROMs first and the same issue still appeared in game.

Stuck the test ROM in and tried the memory test first and it gets stuck at test 2 which points to IC19 CMOS RAM as mentioned by chilpin, it does this on both the new NVRAM and the original RAM chip so having a guess its something to do with that circuit on the PCB or anything thats between the CPU and IC19, it was already socketed when I got it so looks like its back out with the schematic and since the board needs removing anyway to reflow headers and trace an issue with one of the display strobes will come back to it then.
 
Just buzz out all pins on the memory - good chance it is the socket or its soldering which is causing you issues. Still surprised this works at all though.
 
I knew was the ram, those weird behaviours are always the ram. Check any issue in the traces and check the socket, I seen many boards badly socketed. I will try a third ram too. Good Luck
 
Well after probing and realising what an idiot I am, it turns out its probably not IC19, it wasnt passing the RAM tests as I had the coin door closed which triggers the memory protect circuit, opening the coin door and then running the test all three RAMs passed.

Back to the drawing board.
 
Problems on these machines are mostly connector issues.
Just go through the following - It should hopefully save you a lot of time in the end

1) Remove the CPU and driver boards and reflow the solder on ALL header pins (to get rid of any "cracked header" pin problems).

2) Put the boards back and connect it all up again. While the machine is still switched off, use a small flat bladed screwdriver and go round every connector to "wiggle push" each wire more firmly into their 8 pin connector.
You don't have to force them in, just use just enough force to reseat the wires.

3) Put everything back again, and remove the 2 balls.

4) Power up and go into switch test - gently clean, regap and test the switches wherever a ball may rest e.g. 42,10,9,43, 38, 39
Pay attention to the first 4 in that list - make sure they definately work
See the ops manual below (page 12) for locations

Jungle Lord Operators Manual Instruction Booklet

User Manual: Jungle Lord Instruction booklet
usermanual.wiki

Worth a try
 
Last edited:
You must log in or register to reply here.
Back
Top Bottom