Monstermug
Active member
Got a Gravitar in the post today. Box arrived crushed but pcb miraculously was untouched.
Lovely condition PCB, complete with manual and schematics package. Just what I needed!
Before plugging it in I always make a habit of spending 5-10 mins visually inspecting the pcb.
I noticed that someone had attempted a repair on the MC1495L @C12 and the DAC @ F9. I noticed there were really bad soldering on the MC1495L with blobs of solder on the legs and the DAC @ F9 had soldered removed from the legs but still connected to the board.
Picture above showed half attempt at removing DAC @ F9. I think he decided not to pull this one out completely after messing about with the MC1495L and then just given up.
Rather than pulling it out completely, I just re-soldered it back in. If it turns out to be faulty I would need solder to pull it out anyway.
Apart from this there was no other visual damage to the board that I can see.
Plugging it in (using my space duel adapter with bench power supply, supplying the +/-22v) I got no picture. Turned off the bench power supply and used a probe to test pin 40 of 6502. It was pulsing rapidly.
First things first. Verify clock signal on pin 37. This was pulsing nicely. Double checked with probe and should get a square wave.
Then I switched to test mode to see if reset was still pulsing rapidly and sure enough it was. (This tells me that the CPU is unable to access the main program rom that runs test mode)
Next step is to verify roms. They all verified fine to set 2 Gravitar.
Next step is to verify rom sockets are good. Atari boards are notorious for bad sockets although Gravitar board set uses a "better" quality rom sockets than others.
Here is a trick that I have learnt from doing a repair on a Tempest board. If you hold the probe on pin 40 of the 6502 cpu and press down on the Rom sockets, if there is a bad socket the reset pin will go high. This is what I did and low and behold, pressing down on Rom @ H3 caused reset pin to go high.
Examined it visually and couldn't find anything bad. Must be internally loose. So I pulled it out and found most of the legs just broke off in half by itself.
Put another socket in and now the I can see the game working by checking reset pin 40 remained high (not watchdogging).
Turned power on to my bench power supply for +/-22v and still got no picture. I had guess this from the failed repairs on the dacs.
Time to probe the Analogue ouput section. First scoped x axis at the test point and got a nice signal.
then y axis test point and got nothing...
So we have a problem on the Y axis. Tracing back signal to TL082 @ D/E12 showed no input signals on pins 2/3/5/6. Traced back to MC1495L @ C12 and there was no output but there was input going in.
Piggy backed a fresh MC1495L on and got active signals coming out from X axis test point.
So time to change out the MC1495L.
This proved to be more difficult than I thought as I had found that previous owner had pulled of not 1, but 4 pads from trying to remove the chip.
Another trick I learnt to repair pulled traces are to remove a pad from a scrap circuit board and implant it onto the good pcb.
If you look closely you can see the tiny pad I stole from this old acid damaged A500+ ram module.
1 pad repaired. Another 3 more to go!
Here is the finished result. Not perfect but at least the solder has somewhere to stick too rather than just a blob at the end of the leg as it was before.
Fresh ceramic dip package installed.
Time to fire it up again!
Success!
Another classic from the grave.
Lovely condition PCB, complete with manual and schematics package. Just what I needed!
Before plugging it in I always make a habit of spending 5-10 mins visually inspecting the pcb.
I noticed that someone had attempted a repair on the MC1495L @C12 and the DAC @ F9. I noticed there were really bad soldering on the MC1495L with blobs of solder on the legs and the DAC @ F9 had soldered removed from the legs but still connected to the board.
Picture above showed half attempt at removing DAC @ F9. I think he decided not to pull this one out completely after messing about with the MC1495L and then just given up.
Rather than pulling it out completely, I just re-soldered it back in. If it turns out to be faulty I would need solder to pull it out anyway.
Apart from this there was no other visual damage to the board that I can see.
Plugging it in (using my space duel adapter with bench power supply, supplying the +/-22v) I got no picture. Turned off the bench power supply and used a probe to test pin 40 of 6502. It was pulsing rapidly.
First things first. Verify clock signal on pin 37. This was pulsing nicely. Double checked with probe and should get a square wave.
Then I switched to test mode to see if reset was still pulsing rapidly and sure enough it was. (This tells me that the CPU is unable to access the main program rom that runs test mode)
Next step is to verify roms. They all verified fine to set 2 Gravitar.
Next step is to verify rom sockets are good. Atari boards are notorious for bad sockets although Gravitar board set uses a "better" quality rom sockets than others.
Here is a trick that I have learnt from doing a repair on a Tempest board. If you hold the probe on pin 40 of the 6502 cpu and press down on the Rom sockets, if there is a bad socket the reset pin will go high. This is what I did and low and behold, pressing down on Rom @ H3 caused reset pin to go high.
Examined it visually and couldn't find anything bad. Must be internally loose. So I pulled it out and found most of the legs just broke off in half by itself.
Put another socket in and now the I can see the game working by checking reset pin 40 remained high (not watchdogging).
Turned power on to my bench power supply for +/-22v and still got no picture. I had guess this from the failed repairs on the dacs.
Time to probe the Analogue ouput section. First scoped x axis at the test point and got a nice signal.
then y axis test point and got nothing...
So we have a problem on the Y axis. Tracing back signal to TL082 @ D/E12 showed no input signals on pins 2/3/5/6. Traced back to MC1495L @ C12 and there was no output but there was input going in.
Piggy backed a fresh MC1495L on and got active signals coming out from X axis test point.
So time to change out the MC1495L.
This proved to be more difficult than I thought as I had found that previous owner had pulled of not 1, but 4 pads from trying to remove the chip.
Another trick I learnt to repair pulled traces are to remove a pad from a scrap circuit board and implant it onto the good pcb.
If you look closely you can see the tiny pad I stole from this old acid damaged A500+ ram module.
1 pad repaired. Another 3 more to go!
Here is the finished result. Not perfect but at least the solder has somewhere to stick too rather than just a blob at the end of the leg as it was before.
Fresh ceramic dip package installed.
Time to fire it up again!
Success!
Another classic from the grave.
