Atari Pokey Tester

 Technical Info  Comments Off on Atari Pokey Tester
May 012017

I wanted to see if it was possible to test Atari Pokey chips using an Atari 7800 console. Here’s what happened:

Atari Pokey Tester

I would like to say a massive thanks to Michael “RevEng” Saarna, the author of Atari 7800 Basic, who helped to get the code working for the tester, and members of AtariAge forum who made suggestions on how I could move the project forward.

The program is available to download below should you wish to try this out:

Atari Pokey Tester (25 downloads)

Data I/O 29a Remote Mode

 Technical Info  Comments Off on Data I/O 29a Remote Mode
May 162016

Now that I had the Data I/O 29a in a working state I attempted to see if I could get it to communicate with a PC using the Promlink programming software. Unfortunately I fell at the first hurdle as my unit did not have a remote mode available. This can usually be accessed by using the command, “Select”, “F”, then “1” then “Start”, and “Start” again. On my unit this command was not even valid! Interestingly the manual does state that remote mode is “only in models equipped with this option”.

I then noticed that my device had a rom set labelled “324-0029”, whilst my friends unit, which could go into remote mode, had a rom set labelled “324-1024”. Both prom sets were exactly the same and these were labelled “324-1980 Rev A”. So I decided I would copy the 324-1024 roms and try them in my system. Luckily several collectors had some of these roms spare they could post to me (TMS2716’s) and I was pleased to see that once they were burnt, my unit now had the ability to go into remote mode aswell.

Next step was to build the serial port lead to make the connection. This required a 25 pin Dsub male connector and a 9 pin DB9 female connector. I found details of the pins used and connection settings at the following page

Once the lead was made I then had to find copies of the software. After a bit of searching I found a FTP site that had what I needed. The link for the site was

This site had various versions of the programs. Most notably 341-1991, and 61. What I discovered is that the unipak works best with 341-1991, and unipak2 or unipak2b works best with 61.

The site also had txt file lists of all the roms, and proms it could burn on each pak, so that was incredibly useful information.

I installed the software on an old laptop running Windows 95, and on my first attempt I was greeted with the message of “contact with programmer established”.

I wanted to try and get the unit running on a Windows 7 PC, soley for the reason that I didn’t want to use multiple PC’s or laptops to run all of my test equipment. I downloaded the software DOSbox and installed it on my Windows 7 box. The trick to get this working is to make sure that the serial configuration in the dosbox.conf file is configured correctly. I used the settings:

serial1=directserial realport:com1 irq:4 bytesize:8 stopbit:1 parity:E

Once that was configured I copied the promlink software to the root of my C:\ drive and launched DOSBox. Once the folder was mounted using mount c c:\promlink I was able to launch the Promlink software and establish a connection.

I’ve dumped both rom sets and prom set, and made them available below. If someone else comes across the same issues with remote mode it should be fairly easy to sort out with these files, as I could not find copies anywhere else:

324-1980 Rev A Proms (11 downloads)
324-0029 Roms (14 downloads)
324-1024 Roms (12 downloads)

Cook Race (BurgerTime Clone) PCB Fix

 Technical Info  Comments Off on Cook Race (BurgerTime Clone) PCB Fix
Feb 222015

I was generously given this PCB for nothing, and thought I’d try and get it working as I really like the game BurgerTime.

I started by building a Jamma adaptor for video, power and sound using the pinout below (Note the equivalent Jamma pins are in brackets):

On initial powerup I got the following:

I could not credit the game, nor could I hear any sound.

The PCB had quite a bit of corrosion on a lot of chips, so I cleaned up the legs or replaced the IC if it was really bad. I did end up replacing quite a number of chips. One chip, an LS138 @ 2L on the top board had also failed, so I replaced this with a working one. Also I could not find any schematics for this PCB, but it did seem similar to Burgertime in layout, so this would help.

After quite a bit of work there was much improvement

So the problems I still had were:

1) I couldn’t credit or start the game
2) The game has lines through it
3) There was no sound as far as I could tell without crediting the game
4) The game doesn’t consistently boot. When it doesn’t boot I just get a black screen with white lines:

So first thing I did was try and work out why the game wouldn’t credit. I had a look at the edge connector and found three broken tracks so fixed them up:

So I now got this and could credit a game. This confirmed I had no sound:

So I had a look at the sound section:

I swapped the 6502 for the 6502 on the main PCB – no difference. I also swapped the sound chips around and no difference.

Also I checked the larger capacitors in the sound section and they seemed ok. The amplifier also sounded like it is working. So I was stuck on problems 2, 3 and 4.

I did also try testing the larger caps on the main PCB, as I thought this might explain the inconsistent boot, but they also looked fine.

With the help of Philmurr from UKVac providing some schematics from BurgerTime we noted that the 6502 CPU and AY3-8910’s were doing very little in terms of activity on the sound address and data buses.

The next time I looked at the PCB I realised that the 6502 looked as if it had no power. Vss (ground) on pin 1 and 21 were low as I expected but Vcc (+5) on pin 8 was also low ?!?!!?!? Instead I measured pin 21 and pin 38, pin 1 and pin 38, both giving 4.79v, so it looked correct. I started to think that pin 8 was unused, but the datasheet said that it was a +5 pin. I was confused.

Phil noted some corrosion on the photo right next to pin 8, or thought there could be a “hidden” track underneath the socket of the 6502, so I removed it and took a picture of it so I could see where that corroded track went. Pin 1 of the LS74 and pin 2 of the 6502:

Phil pointed out that this was the LS74 which is 3F on the BurgerTime schematics…..which makes that heavy rail +5V! Clearly the track has been damaged to pin 8! So with this fixed I got the following:

I spent another hour on the PCB trying to fix the white line issue. Having done some reading around white lines/jail bars, I came to the conclusion it was probably a ROM issue. Several articles on different PCB repairs mentioned mask rom problems in relation to jail bars, so I thought I would start re-seating all the roms, despite not having any mask roms on the board. The PCB also had them one-sided pin sockets which are fairly poor, so I thought it wasn’t a bad place to start.

As I knew the sound data was held on the eprom labelled 6 I left that alone, and instead started by lifting out eproms numbered 5, 4, 3, 2, and 1 from the top board separately. Removing number 5 and powering up the board resulted in colour/sprites missing, but the lines were still visible. Damn! Lifting out number 4, and the same thing occured but with different colours/sprites as expected. However when I lifted out the number 3 eprom the lines disappeared!

So reseating eprom number 3 and booting up the PCB generated a game with no lines! However now there was no sound!!!!

So I thought I would lift eproms 4 and 5 out again, and see if that would make a difference as that was the only thing that I did, and I got this:

So it looks like the sockets are causing the issues. I’m also wondering if this is causing the game to fail to boot sporadically? I probably need to change them

Anyway I would say the game is fixed now, so thanks for those that provided help and guidance!