Bally E 1235-4 Not working - Code 100002 - no test switch found

[antichrist]

 
yeah , tomorrow I check the cpu socket and all (e)prom sockets again.
maybe I forget one pin ?

for checksum reasons you have to modify just only one byte of eprom.
With an 2650 disassembler you can check the routine for counting bytes and make some changes in the Bin file.

Did you find differences when You compare one Personal Binaryfile with another one ?

[ramegoom]

I compared two P1 chips which were very similar, but a 1% difference between the payout percentages. Chip part numbers 550287 and 550288, both checksum to 9056. They are used on E2000 machines and can be interchanged.

The chip's part number for all P1 chips can be found at hex location 0062 thru 0064. This data is displayed in the readout during the machine test procedure, i.e. 55 02 87

HEX Location             550287              550288
0064                               87                        88

0067                               33                        16
0120                               04                        05
0121                               06                        10
0123                               04                        05
0124                               00                        10     

These five bit changes apparently equate to the 1% payout difference.  Using a dis-asembler, you would likely be able to annotate the data.

[antichrist]

very interresting your investigations 
they change the checksum to the same value 9056.
did you change the bytes to a other payoutratio and burn a new prom with those and test it ?
maybe the checksum have to be corrected maybe someone has written a small calculator for that.

I wish I could test this by myself but I wait for the softwareupdate to read the 82s123 roms for my prommer to check if they are correct.
at the moment 100002 error is up and I clean and test all cpu , ram , rom sockets pin by pin everything is fine
next I change the 5101 with an 6264 sram because I don´t have these 5101 here..
the Adress/Datalines goes to socket p2 and this should be easy to connect

[ramegoom]

A quick look tells me that hex location 0067 is a flexible bit, not related to any equations. This data must be what was used to achieve the necessary checksum after the other locations were modified. Then, the percentage difference is likely modified by data in 0120, 0121, 0123 and 0124.

So, the hack would be to change these four bits of data to see if it significantly affects the payout percentage, but I'm not sure how you'd test the changes.

What we need here is someone experienced in assembly language to chime in, where this could be a classroom experiment....

[antichrist]

there are some eprom simulators (Pic processor with sram and outputbuffers) . unfortunately my one is very old with dos software and serial cable
I can´t remember where to find the disk (arround 1992) and have no dos computer anymore...
but there are 28c256 or something, you can burn with an eprom burner and plug in instead of P1 and then look what happend

correct 0067 is the checksum patch


0000-07ff m1
0800-0fff  m2
1000-17ff m2
1800-1fff  P1
2000-27ff P2
1e00-1eff RAM 5101 (lower and upper ram as 8bit device)

Interesting places in the P1 Chip:

Code: [Select]


Payout Table at P1:



18b0 00 00 00 20 00 20 00 20 00 50 01 00 00 00 00 00
18c0 00 00 00 00 00 02 00 05 00 20 00 10 00 14 00 18
18d0 00 00 00 00 00 00 01 00 04 08 10 20 01 02 04 08
18e0 10 20 01 02 04 08 00 00 01 02 04 08

X1868:
; Spule 0
; from bottom to top (glue is begining and End marking)
; Cherry     = 0a
; Plum        =12
; Bell        =13
; Bar         =24
; Orange    =11
; Seven     =85

    db    H'85', H'12'                ; REEL 1  Decoded bottom to top (85not used , start with plum /12 =2 = Plum)
    db    H'13', H'12', H'13', H'12'        ; 2,3,2,3,2,3,2,3,0,3,2,4,3,2,3,1..
    db    H'13', H'12', H'13', H'0A'        ;
    db    H'13', H'12', H'24', H'13'        ;  22 Symbols 85,13,0a,13,12,13,11,13..
    db    H'12', H'13', H'11', H'13'
    db    H'12', H'13', H'0A', H'13'
    db    H'85', H'12'                ; 85 Seven = END 12 = not on the wheel
X1880:
    db    H'85', H'11', H'24', H'13'        ; reel 2 1,4,3,4,2,4,1,4,0
    db    H'24', H'12', H'24', H'11'
    db    H'24', H'0A', H'24', H'11'
    db    H'13', H'24', H'11', H'24'
    db    H'11', H'24', H'0A', H'24'
    db    H'11', H'24', H'85', H'11'        ; 85 =seven und End
X1898:
    db    H'85', H'12', H'11', H'12'        ; reel 3 : 5,1,2,0,1,2,1,2,3,1
    db    H'11', H'0A', H'12', H'11'
    db    H'24', H'11', H'12', H'11'
    db    H'12', H'11', H'13', H'12'
    db    H'11', H'12', H'11', H'0A'
    db    H'12', H'11', H'85', H'12'        ; 85 = end 3
------

yes its a classproject , a lot of work. In germany at the humbold university they have a" introducing to microprocessing" at the media class with an signetics2650 Emulator, they write games for a Interton Game Console during the semester.

[antichrist]

not much at the moment as I´m waiting for my prom programmer...

[antichrist]

@ramegoom

This mistery 56 will be asked here:

Code: [Select]

00C1 0C 18 00 LODA R0 1800      56 00 00 00 00 00 00 20 42 01 00 00 00 00 05
00C4 1A 04    BCTR  N 00CA      56 00 00 00 00 00 00 20 42 01 00 00 00 00 05
00C6 E4 56    COMI R0 56        56 00 00 00 00 00 00 20 02 01 00 00 00 00 05
00C8 98 73    BCFR  Z 00BD      56 00 00 00 00 00 00 20 02 01 00 00 00 00 05
00CA 04 D6    LODI R0 D6        D6 00 00 00 00 00 00 20 82 01 00 00 00 00 05
00CC B0       WRTC R0           D6 00 00 00 00 00 00 20 82 01 00 00 00 00 05
00CD F0       WRTD R0           D6 00 00 00 00 00 00 20 82 01 00 00 00 00 06
00CE 3F 04 F6 BSTA UN 04F6      D6 00 00 00 00 00 00 21 82 01 00 00 00 00 06
04F6 20       EORZ R0           00 00 00 00 00 00 00 21 02 01 00 00 00 00 06
04F7 CC 1E 16 STRA R0 1E16      00 00 00 00 00 00 00 21 02 01 00 00 00 00 06
04FA CC 1E 17 STRA R0 1E17      00 00 00 00 00 00 00 21 02 01 00 00 00 00 06

you can clearly see, memory location c6 (m1 Prom) compare the Register 0 with with 56
and jumps when happen or not.
This codeline prevents you from alter a personal rom ....
so here you can change something and then you can begin to alter the perso prom and look whats happen ...

at the moment I try to build an prom adapter since I have no 82s123 proms ...

[antichrist]

I try to get an cc16 compatible coinmech (MC16) and I cannot see the 24 VDC Connector in my slot ?.
Can somebody please explain , where to find the Coinmech connector ?
Is there a wire loom or something ?


Where did you usually connect the coinmech ?
 


http://i.imgur.com/t0rv2js.jpg
http://i.imgur.com/vhsEWiY.jpg

b)
Is it true I have to modify the Hopper when I want other coins than Quarters or DM to spit out ?
if yes what part has to be modify and how ?

Cheers

..:: my slot is an bermudatriangle  ::..
One solution found many new problems came up

[ramegoom]

Don't know if this will help, but here's the connector for the comparitor on an E2000 machine. It simply supplies voltage to two of the six terminals, the red LED lights on the comparitor, and the coin passes thru. The logic in the CPU will either power it up when it's ready to accept a coin. Also, to use a different coin in the comparitor, you put a sample coin in the slot (brass token is in there now) so the comparitor will automatically acccept only that type of coin.

If you need to change coin types on the hopper, it involves a coin shelf and coin wheel change to match the particular size of coin.

[antichrist]

thank you ramegoom, you save my day !!

my is looking like that: http://i.imgur.com/F2WSlCY.jpg

Could you be so nice and tell me where did the red / Black wire going to ?
is the 2nd connector to the brass sensor comming with the comparitor or is this cable from the slot machine ??
cannot see this on the pic.
I don´t have this cable in my door 

Hopper: where can I get those wheel part for  ? want to use it with euro coins ...#
btw: interresting, brass is not magnetically working as I know 
maybe it changes AC frequency components.... hmmm

[ramegoom]

If you don't have the connector, you can probably fabricate one, wire it to 12 volts (assuming the comparitor is a 12 volt unit) and keep it on all the time that the machine is running. It will then allow the coin you are using to work for the machine. You might have to modify the coin slot to make the newer coin work.

For the coin wheels, look on Ebay. There should be Bally coin shelves and wheels there. What size is the coin you are trying to use? If it's close to the US quarter  (24,19mm, 0.950 in.) the hopper you have should work - you would need to adjust the coin knife and the payout switch to make it compatible.

[Mr. Dinoman]

You can use a mechanical coin mech. such as this one...Used on arcades and pinball machines and older slots...

[ramegoom]

^^What Mr. Dinoman said...yep that's the alternative. Wonder if it had a mechanical acceptor originally?

[Sunrise Side]

Pretty sure it had mechanical coin mech as it has the diverter mechanism in the holder. If you use a comparitor , you will not use the divert mechanism in the holder currently as the comparitor has it's own. you will use those diverter coil wires but I will have to take a look at which pins on the comparitor you use for the voltage that is supplied to that divert coil. The comparitor has multiple voltage choices.

[antichrist]

at the moment I validate which coinmech needed.

option a) the Comparitor mc16
this one needs 24 VDC supply but I don´t find any wires at this voltage in my slot ??
but the fist E1000/2000 slots should be runnung with 24vdc here  (cc-16 comparitor) if I´m right ?!

Terminals:
http://uk.farnell.com/jst-japan-solderless-terminals/xhp-6/steckergehaeuse-6kont/dp/1516269?CMP=KNC-GAT-FAT-GEN-KWL-OTH-JST
http://uk.farnell.com/jst-japan-solderless-terminals/bxh-001t-p0-6/crimp-terminal-xh-28-22awg/dp/1516301?MER=en-me-pd-r2-acce-con
http://uk.farnell.com/jst-japan-solderless-terminals/wc-110/hand-tool-crimp/dp/1516302?MER=en-me-pd-r2-acce-con

but unfortunately the costs for the crimp tool is higher that the whole slotmachine *sick*


option b) a cheapy mechanical one ...

the coins I want to use are euro coins since i live in europe.
1 Euro Diameter: 23,25 mm  (0,915354331 inches)
 thickness: 2,33 mm (0,0917322835 inches)
weight: 7,50 g = 0.26 oz

I don´t know if knives and wheels exists for that ...

[ramegoom]

The US quarter hopper will work for your coin. You would need to adjust the payout switch because the coin is slightly smaller in diameter, so the switch would need to actuate at the 11 o'clock position on the coin (where the coin passes thru the roller at that point). Otherwise that will work fine.

The coin mechanism wires can be soldered directly to the pins if you don't have the mating connector. Look for the voltage in the machine to power it up. Although, a mechanical acceptor would work, you might not be able to find one that will pass your coins properly, so the CC16 is more desirable since it senses the actual coin that you insert in the slide.

[antichrist]

thank you. thats nice , just modify the microswitch 

my progress on my pcb :

u17 and U19 are bad (proms) so I try an prom replacement socket:





at the moment I  because my cheap china chip programmer  cccp ( product TOP2011)  doesn´t burn the GAL chip correctly ..
very sad  If you follow me, don´t do it , its not worth the money ! get a real chip prommer !

[antichrist]

to ramegoom:

so if you wish to alter the Roms (m1,m2,m3) you have to switch off the ROMCHECK routine.
the easiest is to alter the M1 rom at adress 00ce from 3f 04 f6  to:

00ce c0
00cf c0
00d0 c0

you can do this with your eprom burner built in memory editor
after you put in the 3 c0 codes, no rommemory check (100006) again
so you can alter all Roms without tilting 

[ramegoom]

Good info! This makes hacking much easier, especially using a ROM emulator. I have an old Dataman S4 that works nicely for emulating on the fly.

Now if you can figure out how to increase the payout odds to say, 98% - that would be the icing on the cake.

[ramegoom]

Took a look at the M1 chip (E755-79) for an E1000. The data string 3f 04 F6 is at location 00F1

Looked at the M1 chip (E755-80)for an E2000, same data is found at 00ED.

By changing these three bytes to C0, does this bypass the checksum check for all program chips, or only the M1? EPROMs M1, M2, M3, and P2 all have checksums xx00, and all P1 checksums have xx56, so I'm thinking the bypass mod will ignore checksums from every EPROM?