Atari System 2 PCB Repair - Paperboy

[Purity]

Thanks Karl. Yes I agree it looks like the -5 comes from the -15v, so that should be covered. I will check I am getting -5 at the 7905 but I suspect I will be doing, and I have sound issues elsewhere

I definitely think we need some answers on what power options work and what doesn't work

Cheers

Ben

 

[karlcdoe]

Well as promised I successfully fired up a pcb (champ sprint converted from paperboy) on +5vdc and +12v dc only. I also shut off the power rail that was connected to the +15v line and then managed to get the pcb to boot back up on it when powered back up at voltages between 6.5vdc and 14vdc so while I'm sure that will vary a bit between boards it's hardly fussy, the only difference was the quality of the video output (as someone previously suggested), though it looks a bit rubbish on my monitor/g-nobbers converter anyway.

It's an old loom but looking at it I clearly never bothered wiring up the 10.3 volt mentioned or the -15v. Just tied all the GND and +5v sense wires to the relevant rails (I don't think the sense wires are even connected on later System 2 cabs such as 720/APB that use 1x SMPSU) then stuffed +12v down P15 pin 2 on the cpu board and pin 7 of the video pcb.

I should reiterate that I have no amp, speakers or control inputs wired up to it to test I suppose you may well need that -15v if you are working on the sound section or some random controller dac. As regards the 10.3 volt... anyone decided what that is for without pouring over the schematics, if anything. From memory some Atari games of that era used an unregulated voltage from somewhere to guess when the power had been switched off so it could write to the novram before the power supply collapsed.

PS game board has a running current of about 9 amps on +5vdc according to my psu.

Pictures below:

 

[karlcdoe]

As regards your actual issues with sound/video the short answer is none of those problems are ones I have seen/fixed in the past so not much use I know but:

Soundwise on the second round of self tests it seems to start by playing the pokey test sound when it says 'yamaha test' which suggests its a bit confused so if you've checked out sound cpu, sound ram and rom already my next stab in the dark, based purely on the schematics would be the address decoding at 4E and 1E.

GFX, not sure. Looking at the picture I get on a computer monitor + gonobes I'd find it hard to tell the difference between white and grey...

Missing roofs, I figured they must be predominantly grey but looking on tinterwebs maybe not. The most useful thing it says in the manual is that 'scrolling playfield test.... uses the entire playfield ram except four rows that make up black border of screen' rows of what? rows on screen? rows of ram on pcb? Does it run in game mode or test only? Be interesting to see what graphics are missing and if they have a common circuit but I assume there are multiple faults here.

karlcdoe2018-11-02 12:06:37

 

[Purity]

Thanks Karl, that's very useful!

Can I just ask you list everything you have connected in this format:

Video PCB:

P13

Pins 1,4 Ground

Pins 5,8 +5

etc, etc

The game does play when I take it out of test mode and similar things are missing. I will take a video of what I can see in-game

So far I have gone over the entire Playfield ram addressing sheet, and I think I went over the playfield rom pages. Also the colour ram sections with a logic probe, and found nothing wrong.

Was going to move onto the scrolling pages next

I've also accidentally killed by -15v on the PCB so I'm going to have problems with sound now, until I get this back up and working....school boy error :-(

 

[RaveN]

Guys, what are we trying to prove here, that the game "kinda" runs on certain voltages, or what the board actually requires to run properly?
You can get it to run on 12v/5v, but there's a fair amount that won't run properly, or will work but barely.

I wouldn't recommend running op-amps designed for dual polarity supplies on a single supply, or ICs designed for -5v on this for any length of time.

The AR3 delivers about 7amps @5v, the CPU switcher about 3 amps, I'd expect the total amps to be a bit less than that.

 

[Purity]

My own interest was whether you can get this working off a switcher without the need for the AR3 and Atari brick which takes up lots of space.

When I say "working" I mean good enough to have on a bench to fault find.

I've previously been told that without the +15v being bang on the game will watchdog/not boot.

From my testing tonight I've found the switcher is "good enough" for supplying the +5v and gnd to both boards, and the 12v on the switcher is also ok for both Power OK and the +15v. There doesn't seem to be a need to use a variable power supply for the +15v.

My board boots without issues off the switcher and looks the same as if powering off the AR3 and brick. Obviously as there is no -15v, I have no sound

Although it is not ideal to be doing this on a more permanent basis it looks to certainly be good enough for letting me look over the boards with a logic probe etc

Purity2018-11-02 20:07:12

 

[karlcdoe]

Irrespective of what anyone is/isn't trying to prove here these are the pinouts on my loom as asked:

(I never said running the pcb like this is advisable but equally I never toasted anything doing it in the past - use at your own risk, etc)

##Video PCB##

#J/P13#

1 GND
2 GND
3 Key
4 GND
5 +5
6 +5
7 +12
8 +5
9 +5
11 GND

##CPU PCB##

#J/P15#

2 +12
11 Key

#J/P20#

3 GND
4 GND
5 GND
6 +5
7 Key
8 +5
9 +5

 

[RaveN]

Your initial question was whether 12v (or 10.3v) was good enough for "Power Ok". I attempted to explain how the electronics for that worked.
I stated to ignore the 10.3v as it's probably a misprint (I'm obviously going against the manual here so there's a chance I'm wrong). I also requested you take a scope reading with that to see what it was doing. I was more than happy to be proven wrong and was intrigued to know the results to gain more knowledge on the matter.

I also explained the way the -5v originated/worked.

I'm not sure what else I could have done to help with your questions. If all you wanted to do is boot the board for logic testing and didn't care about proper gfx/sound, then you could have just used 5v and GND and shorted the Power Ok op-amp pin.

 

[Purity]

Andy I will level with you - my immedaite aim was to get rid of the AR3 and
brick. My stuff is current laid out on the dining room table and the
missus is not happy and wants it shifting asap. My previous attempts at
running this off a switcher resulted in being unable to get a stable
screen. I have tried the same method on my dining room table and it
seems fine now, so I must have made an error somewhere. I can now move
the stuff somewhere else where I can look over the board without being
nagged at

HOWEVER I am very much interested in the power requirements
as you are, and your explanations have been very useful in how the
electronics work for power ok and -5. The more info the better in my
opinion I do care about getting proper gfx/sound. I have also been told conflicting information about what is needed for these boards so it is useful to know for sure

I very much appreciate all the input from yourself and Karl and it has been very helpful.

I will take a scope reading as suggested and report back on results

Sorry, I should have explained myself better earlier.

 

[Purity]

Thanks Karl for the pinouts you were using for testing

As promised here's a video of in-game:

 

[Purity]

Pin 11 @ 5K on the scope

This is running at 10.3v DC from the AR3. It might actually be higher than this I will take a measurement

Purity2018-11-03 15:17:32

 

[Purity]

Bah, i have a communications failure now with the 6502

Swapped to anothercpu board and get the same thing

Any experience with this issue?

 

[strykr]

I have experience with this. Its exactly what my paperboy pcb does. Doesnt boot but can get into test and it scrolls through the diags and gfx testing screens and halts at 6502 comms errors. Ive swapped out three different 6502s now to no avail.....

 

[strykr]

I will say that a lot of the stuff mentioned in this thread is aboce my paygrade in terms of knowledge but its good to see a technical thread on vac with some of the clever chaps chiming in.

 

[GadgetFreak]

Bah, i have a communications failure now with the 6502
Swapped to anothercpu board and get the same thing
Any experience with this issue?

Never worked on a System 2 but I have had a System 1 where a faulty pokey caused the 6502 to freeze, could also be the ram for the 6502. Have you checked the data/address lines on the sound section rom/ram and also the control lines on the 6502 to see if it is running?

 

[Purity]

I've narrowed this down a bit

Interrupt logic, and adress decoders are good. I also checked the reset circuit and CPU and they are all good

Rams and roms are good. Pokeys are ok. 6502 is good

I can see some issues in the 6502 comms flag, at the 6502 itself and at its latches, so the problem is here somewhere

 

[Purity]

Not much further with this unfortunately. I've temporarily swapped CPU boards, and am back to where I was before my other CPU board started with the 6502 communications error. I've fixed some of the sprite corruption. This was a dodgy leg on one of the video roms.

I definately have a colour issue on the video board. I can see that there are no levels of colour on the test screens, and I think this is why I see no road or footpath etc. You can also tell that there is a problem because of the blinking Paperboy text.

I've looked through the output driver chips, the video intensity control and driver enables sheet, and the color ram sections with a logic probe. I see no obvious problems

Purity2018-11-05 11:55:20

 

[Ace`]

Have you checked everything on page 31 of this manual? http://arcarc.xmission.com/PDF_Arcade_Atari_Kee/Paperboy/Paperboy_SP-275_1st_Printing.pdf Looks like there are two versions of the manual. I forgot how good the Atari schematics are.

 

[Purity]

Have you checked everything on page 31 of this manual? http://arcarc.xmission.com/PDF_Arcade_Atari_Kee/Paperboy/Paperboy_SP-275_1st_Printing.pdf Looks like there are two versions of the manual. I forgot how good the Atari schematics are.

Yes dude

 

[Ace`]

What about the S157 @ 4E/F? 'Playfield' is pretty vague but maybe it's a layer missing rather than a colour issue?