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

[ramegoom]

Ah, good info. I just looked in my spreadsheet, and it appears the 585490 uses the 3701 P2 chip, which must be MIA on my machine. I'll have to take another look....!

Anyway, I have the following P1 chips, many of which I pulled of of a surplus lot of E2000 CPU's:
4A9483
1A9195
1A9091
1A6885
1A3090
655482
596394
585490
583784
568279
554190
550287
550288
540789
540090
485493
568177
450489
203279
183486
177589
162183
158483
155189
149085
147184
491986
455189
448390

Also have a number of P2 chips, seems the 3701 was most prevalent within the various machines I have.

Within this thread, there is a good deal of disassembly data starting to form. This could be a good info thread for understanding the programming.

[ramegoom]

Upon further checking, the E2090-253 does indeed have the P2 chip. And I did a meter check on the P1, it comes up as a 585490. And the sound does work, coin-in, reel spin, win, and door open. It might be a freak combination?

[GOS]

Ah, good info. I just looked in my spreadsheet, and it appears the 585490 uses the 3701 P2 chip, which must be MIA on my machine. I'll have to take another look....!

Anyway, I have the following P1 chips, many of which I pulled of of a surplus lot of E2000 CPU's:
4A9483
1A9195
1A9091
1A6885
1A3090
655482 225-371,2,3 3lp
596394  ?
585490
583784 239-136,7,8   es e2272-41
568279 234-321,2,3 3cm   numerous P2  double progressive
554190 231-339,40,41  239-753,4,5 2cm e2090-128
550287 234-138,9,40 3cm  e2090-120
550288 234-138,9,40 241-188,8,90 3cm e2090-119  es
540789 239-696,7,8 4coin  other strips e2296
540090 3cm triple bar numerous strip  es
485493  e2090-244
568177 234-318,9,20 3lp  p2 numerous progressive
450489 241-188,9,90 3lp 
203279
183486 ?
177589 ?
162183 239-765,6,7 5cm
158483 225-295,6,7,8 5cm
155189 234-183,4,5  3cm
149085 225-169,70,69,70 6cm p2 3701
147184 ?
491986 ?
455189 234-222,3,4 3cm
448390 234-219,220,221 more strips

the p1 with ?  either are modified p1 (renumbered by someone - i do the same when I modify one)  or later than my documentation.
the p1 with a letter imbedded are usually games with CREDIT meter. I have no doc on them.
the ones that are BOLD I could use copies for my use.  (if you have a programmer - you can create bin files).
best i can do

[antichrist]

so I change the bridge recitifer, the 5V cap and add 2x 100nF ceramic caps around the IC to help
minimize HF oscillations.
This is recommended by the manufacturer of this voltage regulator.

if this board is´nt running, the big blue cap will fly away next :-)

[antichrist]

Now I track down my problem. The CPU is constanly reseting (with mains frequency)

On Pin 15 of J1, I get 12V with an 2V Ripple.

on CR4 Zener I measure 12v with 2v ripple on the left side, on the right side i get 5V with 2Vripple (the ripple is 2V and rides on the 3V line)
so the Transistor U21 Pin9  switches on and off with mains frequency.
so the same for U21 pin 14 transistor switchs the whole CPU Reset unit on/off with 20ms period (mains)
and the same for Q1 the Delay driver Transistor goes on and off with mains frequency.

The CPU is constantly resetet so it can never pass the memorytest program because in the middle it resets and begins
again. stupid.

The delayrelais never click because its cleared via pin 4 J1 before it make contact.

So now I´m out of strategy. I tested the power supply, but the (sawtooth) ripple cames from the Big blue C1 cap (powersupply) and I don´t know why.

the diodes are new, the recifier is new, the cap c2 and q1 of the powersupply.
So its time to replace the big blue cap on the power supply.

What did you measure on Pin15/j1 (MPU Board) ??
or C9 if you prefer (+ side) ??

Did you have ac or dc 10v ??

Cheers

[ramegoom]

Did you check the two power supply regulators (and capacitors) on the MPU board? Those are responsible for the CPU and other components there.

[dhellis]

On my E2000 MPU board I measure about 8 volts on J1 pin 15 wit about 2 volts of ripple

CR4 I measure 8 volts on the cathode and 1-7 volts on the anode so the drop is 6.3 volts
which is correct for a 6.3 volt zener

[antichrist]

thank you dan.
the cap c1 didn´t help much only that now when I switch on the machine I get  ". . . . . . . . . . "
"0 . . . . . . . . . . "

and after 1 sec.

"100002"

as always. Now I check the power watch circuit around CR4.

thanks for checking this out !

@ramegoom
thank you for that hint. the regulators are fine and putting out 5v with 0.1v ripple.

Next I will check the groundings. I have some rust under the screws that holds the mpu pcb.

Cheers

[antichrist]

Found a lot cold solder joints arround IC CA3081 transistor array.

Now MPU boots, but the testswitch isn´t working due to a 0....... Error.
What does this code ?

When I disconnect the driver board, I get an 4........ instead the 0....... Error


Whats wrong here ?
Please help.

[ramegoom]

Odd display. Here are the memory or EPROM based codes.

[dhellis]

A lot of the errors that you are seeing seems to be consistent with memory errors. I would change U22
which is a CD4066. This chip is used to select the rams and I just fixed a board that was showing almost
the same error, mine was 1000006 (I think).

The 0..... may be caused by the same chip but could also mean the clock is not working so check
pin 5 of U31 for a 2 MHZ signal

[antichrist]

yes the 4066 is a crude method to tristate the Systemmemory from the MPU in case of powerloss or Reset during powerout or glitching via 555 monoflop circuit.
simple but working.

I piggyback the 4066 with an new one. usually with 74lsxxx chips this works but this is a analog ic. maybe its not so easy here.

So I think my proms are not OK and this is why it shows me a "4 . . . . ." but it isn´t.
 I figured out how the Powerup Selftest on my E2000 is working:

Code: [Select]

L0091:
    wrtd,r0                ; Port D (debugger?)
    stra,r0    X1FFF            ; an 1fff (load 00100000 in den shifter)
    stra,r0    *X0061            ; indirect adresse from 0061/62  -> (1e80 !!)
    stra,r0    X1FFF
    coma,r0    *X0061            ;readback
    bcfr,eq    L0088            ; no then endless LOOP 100002 ERROR showing
    coma,r0    X1E00            ; compare 1e00 (RAM Aera)
    bctr,eq    L0088            ; no ? then endless loop again.
    birr,r0    L0091            ; yes? increment r0++ and back to 0091 memorytestloop !
L00A7:
    lodi,r0    H'D3'
    wrtc,r0                ; 100003 to winmeter  "incorrect or no sencond rom" aka M-2 ROM
    wrtd,r0                ; 
    loda,r0    X0F45            ; Debugger?
    comi,r0    H'17'            ; 0f45 = 17 if yes then you have a correct *lol* Prom M2 (only 1 byte is tested!!)
    bcfr,eq    L00A7
L00B2:
    lodi,r0    H'D4'
    wrtc,r0                ; 100004 to Winmeter "incorrect or no third ROM"
    wrtd,r0
    loda,r0    X17FE            ; read 18 from memory location  17FE (3. rom) = 18 OK
    comi,r0    H'18'
    bcfr,eq    L00B2            ; no  ENDLesS LOOP !
L00BD:
    lodi,r0    H'D5'
    wrtc,r0                ; 100005 ans display "incorrect prom series, personality prom P1"
    wrtd,r0
    loda,r0    X1800   
    bctr,lt    L00CA
    comi,r0    H'56'            ; 1800 = 56 ? (usually  P1 ROM alaways 56 inside) *lol* one byte check
    bcfr,eq    L00BD            ; JUMP if not  to00bd (endless loop!!)


The Code 4..... isn´t coded in the computer.
So this should be a malfunction comming from

So I have to check the Interrupt system....

the odysea goes on ;-)

[antichrist]

Need help with the cc-16 coin comparitor.

which one to choose ? a 12vdc a 24VAC or 24VDC ? where did you get the voltage to supply ?

I don´t have the power connector in my slot so I have to take the power from the transformer ?
maybe the UR Pin from the power supply (11.8V AC) with four diodes and a cap to the cc16 12vdc.

*confused*

[antichrist]

Found the problem for the 10002 error.

after patching out the 10002 memory check program  and patching the Romcheck function (because I alter the M1 Rom so I have to avoid a romchecksum)
the mpu stop with 10003 error (the processor see no second Prom aka   M2 ROM in the slot )
exactly cpu looks 0f45 and looks for 0x17 if its written there, no 10003 error.
but in my case wrong value because adress line problems.

  either the adressbusdriver are damaged or the cpu can´t write to the adressline (cpu damaged) or simply ic socket problem (no reliabe contact) 
The adressbusdriver ics are checked last year so its seems the processor.
But with one million operations per second its really hard to see something ...
Lucky if you have an Logic Analyzer. I don´t have such fancy things in my drawer :-(

Best thing is to build an NOP Generator so you can see the adresslines dancing on the oszilloscope.
wow its going on....