pac-man no sprites

JimD53

Active member

Donor 2024
Joined
Jan 19, 2024
Messages
221
Reaction score
40
I've been using the two bit diagnostic tool and have successfully fixed the colors and the character display. I'm now stuck trying to get the sprites to display. This is one of the diagnostic screens that should have sprites displayed in each box points to 3F or 3H. Both are new. When I boot using the ms pac-man eprom provided with the diagnostic kit the demo starts and I see dots disappearing but no sprites there either. I saw activity on 1F, 2F, and 1H which are downstream from 3F 7 3H. I checked all the possible causes in MikesArcade pac-man troubleshooting guide and everything looked good there. Any suggestions on what else to check would be greatly appreciated.

IMG_5241.jpg
 
Check that the 3E LS20 output at pin 6 is toggling, as well as inputs pin 1, 2, 4, & 5. That will indicate sprite data is at least being clocked out. Dead? Something much deeper on the board.
 
Check that the 3E LS20 output at pin 6 is toggling, as well as inputs pin 1, 2, 4, & 5. That will indicate sprite data is at least being clocked out. Dead? Something much deeper on the board.
Thanks for the info. I see signals on all pins you mentioned. I apologize that I'm a little tardy in updating this thread. Last night I got out the microscope I use for surface mount work and inspected the back side of the board for all of the ICs involved in displaying the sprites. I did find a very small bridge between 74LS161 (1E) pin 11 and the trace very close to it going to pin 15. Under the microscope I could see a very short trace width tab attached to the IC pin pad pointing toward the trace next to it. With the poor state of the green solder protector at that trace a small solder bridge was created but hard to see without magnification. That said I removed the extra solder and now I get ghosts but not in the correct locations. The bottom edge is animated. The test below indicates a failure may be caused by the ram at 3F & 3H. I swapped them with the ones at 2K & 2L. The sprites aren't in their respective boxes but they did shift to the locations shown below. Before they were near the top of the screen. I ordered some new ones so I wait unless someone can point me to a different failure that would cause the sprites to be dropped on the screen at incorrect locations. Thanks for all the help in getting this board working.


IMG_5250.jpg
 
If there was a solder bridge between 2 outputs of the 161's, might want to try piggybacking on them with another 161 to test.
 

User manual there ^ may help
 
If there was a solder bridge between 2 outputs of the 161's, might want to try piggybacking on them with another 161 to test.
Thanks , I did replace them. Same results.
 

User manual there ^ may help
Got it. I'll look at it again. The pac-man troubleshooting manual part two has a description of the sprite circuit but it will take some time to get through. For now I'd like to know how the sprite row/column information is encoded. Might be helpful to figuring out what bus signal is bad.
Thanks for the info. I see signals on all pins you mentioned. I apologize that I'm a little tardy in updating this thread. Last night I got out the microscope I use for surface mount work and inspected the back side of the board for all of the ICs involved in displaying the sprites. I did find a very small bridge between 74LS161 (1E) pin 11 and the trace very close to it going to pin 15. Under the microscope I could see a very short trace width tab attached to the IC pin pad pointing toward the trace next to it. With the poor state of the green solder protector at that trace a small solder bridge was created but hard to see without magnification. That said I removed the extra solder and now I get ghosts but not in the correct locations. The bottom edge is animated. The test below indicates a failure may be caused by the ram at 3F & 3H. I swapped them with the ones at 2K & 2L. The sprites aren't in their respective boxes but they did shift to the locations shown below. Before they were near the top of the screen. I ordered some new ones so I wait unless someone can point me to a different failure that would cause the sprites to be dropped on the screen at incorrect locations. Thanks for all the help in getting this board working.


View attachment 837918
I looked up this screen in the diagnostic kit manual. Red ghost should be in the upper left box, pink in the center box, and blue in the lower right box. I used my usb logic analyzer hooked to the attack ram at 2A (RA0-RA7) and I see what look to me as good signals. The data out, pin 7 on 2A-2D were all pulsing. For now the problem seems to be upstream of the 74LS161s at 1E & 2E. I have new ram on the way for 3F & 3H.

Here is what the logic analyzer shows.
pac-man sprite problem attack ram chip ra0-7.jpg
 
Following up - You're last at sprites in the wrong location; what about the motion test? What is the result there?
 
Following up - You're last at sprites in the wrong location; what about the motion test? What is the result there?
Thanks for checking. Actually I've regressed back to no sprites. Since I couldn't detect the error using the logic probe I decided to remove the chips touching the DR bus ( 3F, 3H, 1F, 2F, 1E, 2E) so I could check each one for short to gnd, +5, or each other using my multimeter. I also checked for a bad segment pullup at RM4. Finally I checked for continuity at all points along the bus. Everything tested good. I tested the RA bus between 1E & 2E and the ram at 2A-2D. No problems detected. Finally I looked at the data lines going from 2A-2D and 3C and 3D. Good all the way including the pull up at RM2. When everything was back in place I had no sprites. Doesn't make sense. I had ordered new ram for 3F, 3H, 2L, 2K. I was assured the previous new ram would work. I didn't match the doc as a replacement and it ran very hot. The new ram is 74S89 which is running at "normal" temperature. No sprites with either type of ram.

On a very old thread someone said that the 74LS75 at 3D could cause missing sprites. I have some new ones on the way. Since all sprites have disappeared I think that I either have a "stuck" ram at 2A-2D messing up the data on the bus with pull up RM2 or one of the chips to the right of it on the schematic. When I get back to the shop I also want to check /CntrlLd and /ObjOn which the troubleshooting guide V2 indicates are important to the attach ram process.

I'm ready to find my missing sprites so I can work on my missing sound.
 
I'll have to put the no sprites problem on hold for now. I still have no sprites and I now have a garbage screen filled with random letters. I got to this point by deciding to pull the diagnostic tool out and replace the Z80 on the board to see what the original game looked like. That operation did require removing and replacing the daughter card above the processor socket. When it booted it was out of both horizontal and vertical hold. Using the same technique I used to use on a old CRT TV - I hit it. I thumped the board, while laying flat on the table, and got the sync to behave but now I get this screen. I did discover that pressing at 3M I can cause the sync to lose lock but it doesn't help the screen data. This picture was taken with the diagnostic board connected back in place of the Z80 processor. I think part of my problem is a cracked trace. For now I won't be updating this thread until I can at least see the diagnostic screen. The 74LS75 mentioned in the previous post will be here tomorrow -- sigh.

IMG_5256.jpg
 
Back
Top Bottom