Odd problem and odder problem

[poppo]

First let me remind those that don't know that I have my S+ boards modded to where the eeprom is on the MPU. So I can swap boards all day and not have to reset anything. An un-modded board will use the EEPROM on the motherboard.

So I have these two other (10mhz) (un-modded) boards I picked up mainly for spare parts. I did not expect them to work, but figured I would try them. But before I go into those issues, here is what my first oddity was. In order to troubleshoot one of the other boards, I decided to swap CMOS chips between those two spare parts boards. But in my haste, I accidentally removed the CMOS from my good board. So after I put it back in, I got the standard 61 error and lost my credits. No big deal. So I start feeding coins in and let the wins go to credit (as it was set to do). I have a hopper/credit limit of 300. So I had about 25 credits on the machine and I get a win that will pay 6. But instead of going to credit, it paid those 6 out of the hopper. I went into the self test to double check the settings and everything was fine. Actually this game will not let you set the limits below 300. I have not had it happen again. Fluke, or did going into the self test sort of refresh the settings?

Second issue:
Problem board #1. Turn it on, clear the errors and and when it goes to do it's maiden spin, all three reels turn a bit, stop, turn a bit stop, stop and eventually I get a reel tilt. Keep in mind this is a working machine and it is the MPU causing this. The board has known good reel and game chips. Did a clear just for kicks and it did not help. Also tried a good tray, so it's not the trasformer on the MPU. Nothing looks crispy or like it had smoke checked.

Problem board #2. Had an error 12. Swapped the battery from of the other board. Cleared the errors and it works fine.

So any ideas on board#1 and the goofy reel spin?

[poppo]

Ok, problem solved. I should have looked right away, but there were two bent pins on one of the MPU connectors. Straightened them out and this board now works too.  

BTW, I did a search for this other item and came up blank. But in case anyone is interested, a CR123A battery works just fine for the CMOS backup.

I'm still curious about the weird 6 coin hopper payout on the original board though.

[StatFreak]

Good job solving those issues and letting us know what happened.


I've never heard of anyone having a machine payout a small win to the hopper when the hopper limit was well above the current number of credits. 


I wonder how long the CR123A will have enough voltage to support the machine? The mAh rating is fine, but since it's starting at a disadvantage, being 3v instead of 3.6v, it will reach an unusable voltage sooner than the proper battery. On the other hand, the 3.6v batteries last so long in the S+ that it could still last for several years... 

[poppo]

I wonder how long the CR123A will have enough voltage to support the machine? The mAh rating is fine, but since it's starting at a disadvantage, being 3v instead of 3.6v, it will reach an unusable voltage sooner than the proper battery. On the other hand, the 3.6v batteries last so long in the S+ that it could still last for several years... 

The typical 6264 static ram has a 2V min voltage for battery data retention. The S+ MPU determines low battery by using a comparitor circuit. However I don't know what the threshold is that triggers a low battery. So it could be close to 3V or it could have a long way to go. It would be simple to 'fool' the circuit by just connecting R104 to R101 as shown below. You would never get a low battery error since it would always look like the battery was good by reading vcc instead of the battery.  Of course there should be a good battery in there or you will get other errors every time you power up, but CR123A batteries are cheap and easy to come by.

http://www.datasheetcatalog.org/datasheet/HitachiSemiconductor/mXwvwws.pdf

[StatFreak]

As I mentioned, given the efficiency of the S+, the battery may last quite a long time. Let us know how it works out, particularly if it dies prematurely.

[poppo]

As I mentioned, given the efficiency of the S+, the battery may last quite a long time. Let us know how it works out, particularly if it dies prematurely.

I suppose the problem could be if the threshold is right at 3V. If the CR123A dips below that even a little, it could trigger the low battery. I am probably going to mess with bypassing the voltage comparator. I'm sure that CR123A will last longer than me.  

Years ago we used to make solid state hard drives with banks of 6264 static rams. We would use 2 AA batteries for backup and they lasted forever. Those chips only draw between 1-100 micro-amps when in stanby (1 typical, 100 max). That is .000001 amp (1 micro-amp). A CR123A rated at 1300ma would last between 1.4 years worse case and 148 years typical.  Often the battery's own 'self discharge' will kill it sooner than the chip it's backing up.

[StatFreak]

...
Often the battery's own 'self discharge' will kill it sooner than the chip it's backing up.

That certainly seems to be what happens with the S+.

Of course, when it comes to the Bally S5500...    

[stayouttadabunker]

I clipped this out from an old S+ manual Poppo!
I think it's really from one of those old ARMY machines like the one FrankA has
with the display screen, because the "12bb" would come up on the display-
which I'm pretty sure would mean "Bad Battery".
I personally would love to have one of those displays for my S+!!    
The error codes on that display are quite different from the normal "12" we usually see.  >>>

[poppo]

I clipped this out from an old S+ manual Poppo!

Interesting. I wonder if that message really means that you will get a 12 error at 2.9 (the comparator threshold) because the battery is going fast. Usually a battery will stay close to it's rated voltage and then drop relatively quick once it is going to die.

The funny thing is that once you get an error 12, you pretty much already lost the data since replacing the battery will dump the static ram unless you put an alternate power source on it while swapping the battery. The board I had that had error 12 had 0v on the battery, so the ram was already wiped.

[stayouttadabunker]

I never once even tried to save whatever what was on the cmos whenever I got a [12].
I just replaced the battery and started the accounting records anew.
It's usually considered a total loss of accounting records and only an educated
guess at what might have been in there based on past records.

I threw a multimeter on a [12] once and it showed me about .5v by the time I got to it.
I think it dropped from the threshold of 2.9volts to .5v in about 60 hours.
This is only a guess and is derived from memory from about a year ago....lol

[knagl]

I would presume that in a casino environment that if a 12 came up that the hard and soft meter readings would be documented on paper prior to anything else being done, and then the battery changed.

[StatFreak]

I would presume that in a casino environment that if a 12 came up that the hard and soft meter readings would be documented on paper prior to anything else being done, and then the battery changed.

But the catch-22 seems to be that once the 12 is displayed, the bookkeeping meters are no longer accessible. I've never had a dead battery on an S+ so I'll defer, but on the Bally 5500 that is certainly the case. I've often thought that this was a weakness of the design, considering the effort to which the manufacturers go to preserve data.

I wonder if the data are still readable through SAS when a 12 tilt code is displayed. If not, then the casino would lose all data following the last download or manual record. Even if that were only one or two shifts, it would compromise their financial records.

[poppo]

But the catch-22 seems to be that once the 12 is displayed, the bookkeeping meters are no longer accessible. I've never had a dead battery on an S+ so I'll defer, ...

If that is the case (data inaccessible), I would think the casino would have a test fixture that would apply power to the MPU while the batter is changed (as long as the battery has not gone too dead). I know that on my pinball machine, I can swap the backup battery while it's powered up to preserve the high score and settings. But the battery is easily accessible and in a holder. This is why I think the low battery error actually comes on well before the battery has gotten down to the minimum 2V the chip requires.

I wonder how many in-service S+ machines have ever had a 12 even pop up. One would think that back in the day of routine maintenance, batteries would be changed when the machine was taken out of service and all data was already recorded.

[knagl]

But the catch-22 seems to be that once the 12 is displayed, the bookkeeping meters are no longer accessible.

I wonder if the data are still readable through SAS when a 12 tilt code is displayed. If not, then the casino would lose all data following the last download or manual record. Even if that were only one or two shifts, it would compromise their financial records.

Point taken, but couldn't the hard (mechanical) meters be used to fill in the blanks?

[reho33]

Don't mean to "hijack" the thread but I have a request: Could someone post the history of the Military Slots? Now returning you to the topic at hand.....

[stayouttadabunker]

Don't mean to "hijack" the thread but I have a request: Could someone post the history of the Military Slots? Now returning you to the topic at hand.....

That's "Classified" information....