Haven't had chance to post any repair logs for ages but after an interesting discussion about getting some boards that aren't high value repaired I thought i'd inspire people to breathe life into some of those boards they chuck on the pile and just hunt down a replacement for. Hopefully this will jumpstart a few T&F and HS board repairs and maybe some of the junk closet might surface for sale or something - there's hardly ever any Hyper Sports boards for sale.
Firstly, i'd like to say a huge thank you to Jon (Porchy) and Shoestring for the diagnostic ROM for Track & Field, i'm hoping one of them will do a version for Hyper Sports or share the code and i'll write one and make it available.
x-x-x-x-x-x-x-x-x-x
Prof Peach's Track & Field
This came to me before xmas and was not booting - could see a screen of "0" and the picture just rolling like crazy.
Fault 1 - no sync - scope shows nothing coming out on the sync pin, traced it back on the board to the LS244 @ 1C on the lower board. Sync going in, nothing coming out, replaced chip, sync restored but still not starting up.
Probed around a bit and found all the signals going to all the various places - at least correctly looking.
Found that the 2 x LS244's @ 5C/5D had Vcc tied to 2G (pin 15) by a solder bodge blob - I looked at the schematic and this wasn't supposed to be there, sadly, they had been done quite badly that the chip legs crumbled so I had to replace these 2 chips anyway - at least the signals going to the address bus looked sensible now - how it worked previously is an unknown.
Still not booting. Decided to try out the diagnostic ROM here. I was in a rush and struggled to understand the page a little bit and tried putting it into 5A but this led me down a rabbit hole to try and figure out the problem - tried the diag chip in 1A and hey presto.
Now we're getting somewhere - 11B NG (no good I assume) - so what we learned here is that the board will continue to stall if it doesn't pass the WORK/OBJ RAM tests.
This took some finding - I looked at the LS245's on the data bus - no problems there - logically the diagnostic program is running so none of those can be stuck HI on output at least in the direction of the CPU.
Went around all the signals to the 10A,10B,11A,11B 2114 RAMs - all address lines moving, all data lines moving, all selects happening.
Found that it was the LS153 @ 9D in the address circuit to the 2114's - although the signals were moving, the functional logic output was not right. Comparator told me this.
This is another good example of where a logic probe won't cut it for everything but it'll help a lot of the time. Without the comparator i'm not sure how i'd have figured that out without removing each 153 in the circuit and testing out of circuit - I don't own an in-circuit tester like an ABI Boardmaster or even the mighty Fluke with some trickery wouldn't have helped.
Replaced the LS153 @ 9D and the game was back to normal and the test ROM told me so...
Tested in game mode, everything fine again now.
Frothmesiter's Track & Field
I forgot to take a picture of the symptom here but what was happening was that the background was just an off green colour, game would play in the background but you could only see the sprites.
Diagnostic ROM inserted and sure enough can't see anything on the screen but you hear the music play and the runner runs across the screen.
Started at the end of the circuit and worked backwards through the 157's and the 273.
Finding input to the 273 but nothing moving on the output I was poking with the scope when the graphics suddenly flickered back to life - note all signals TO the chip were present.
Reflow around chip and check traces for damage and backgrounds are restored.
Game has an intermittent halt/reset/watchdog issue still - every now and again the game will freeze and reset.
A little problematic to find this one - had to use the old "inspection eye" tool and look for problems - tapping the board brings it back to life.
Did preventative ROM pin cleaning and socket evacuation - gets better results.
Eventually I figured out by trial and error that the problem was on the sound board - reflowed a tonne of dry joints and found a small cap on the edge of the board twisted over so its legs kept touching - touch them together - game resets...... Spread them out neatly and all good again.
I booted the game and decided to have a quick go and discovered some of the speech missing - it would should "on your marks" but that was it.
A quick pull of the EPROMs on the sound board and a ROM ident via the online tool and find that C9 doesn't match anything - burned new EPROM and we have everything back to normal.
Not sure if this version of the diag rom has a speech test in or not but all sounds passed and sprites and backgrounds looked fine.
Frothmeister - Hyper Sports
Oh, you'll love this one...
Game sort of starts up - looks like a Commodore 64 loader or an Amiga decrunching a game or demo.
My first thought was that the horizontal clock was stuck doing 1H and thus repeating one pixel of the output - oh my no....
Pulled the board apart on the bench so I can poke about.. I have some nice 40-pin ribbon cables that work perfectly on these boards.
With the game on, I pushed on all the socketed IC's - firstly the custom 504 @1c on the lower board - no crunch but when pressed, backgrounds try to come back - remove custom chip, clean legs, clean socket, replace, backgrounds mostly back - squeeze socket - everything fine, replace socket, backgrounds return but with sprite issues.
(sorry about the sync bars)
You can see here that the sprites for the weight lifter are corrupt on another screen, the runner on the long board is fine but in mid air he's a blob of garbage.
I've seen this before a few times to its graphics rom region time. Quick look at the schematics and the objects are in the lower right corner of the bottom board labelled C11-C19 normally. This board has had the eproms compressed into 27128's.
Inspecting carefully I find the one in G17 to be inserted upside down - oh dear, well that's toast.
Now, I don't have 27128's on stock currently, only 27256 and 2764 - so I had to manufacture a 27256 with the images of the 2 roms it covers in the right order. This was a bit of guess work.
C15 and C16 are in one rom in G17 on this board.
On my eprom burner PC I downloaded the hyper sports ROM files and did:
And burned it - nope, sprites still garbled but clearer - it then occurred to me that while verifying the other ROMs that the G19 one which is labelled C18 I had to read as a 2764 so the C18 image is in the LOWER part of the ROM - ahhhh:
Re-burn and place in socket, game starts, weight lifter has jailbars but is clearer - again, having seen this on my other boards, squeeze the socket - ta-da, jailbars gone.
Replace socket at G17 with a nice new one and everything hunky dory.
Don't just cupboard these boards - they're worth repairing.
Hope this is useful info.
PS. These 2 games are a personal top favourite game for me, if anyone has junk boards they want rid of, i'm happy to have a go at turning them back into workers for the sake of preserving this awesome game.
RGP2016-04-11 22:53:35
Firstly, i'd like to say a huge thank you to Jon (Porchy) and Shoestring for the diagnostic ROM for Track & Field, i'm hoping one of them will do a version for Hyper Sports or share the code and i'll write one and make it available.
x-x-x-x-x-x-x-x-x-x
Prof Peach's Track & Field
This came to me before xmas and was not booting - could see a screen of "0" and the picture just rolling like crazy.
Fault 1 - no sync - scope shows nothing coming out on the sync pin, traced it back on the board to the LS244 @ 1C on the lower board. Sync going in, nothing coming out, replaced chip, sync restored but still not starting up.
Probed around a bit and found all the signals going to all the various places - at least correctly looking.
Found that the 2 x LS244's @ 5C/5D had Vcc tied to 2G (pin 15) by a solder bodge blob - I looked at the schematic and this wasn't supposed to be there, sadly, they had been done quite badly that the chip legs crumbled so I had to replace these 2 chips anyway - at least the signals going to the address bus looked sensible now - how it worked previously is an unknown.
Still not booting. Decided to try out the diagnostic ROM here. I was in a rush and struggled to understand the page a little bit and tried putting it into 5A but this led me down a rabbit hole to try and figure out the problem - tried the diag chip in 1A and hey presto.
Now we're getting somewhere - 11B NG (no good I assume) - so what we learned here is that the board will continue to stall if it doesn't pass the WORK/OBJ RAM tests.
This took some finding - I looked at the LS245's on the data bus - no problems there - logically the diagnostic program is running so none of those can be stuck HI on output at least in the direction of the CPU.
Went around all the signals to the 10A,10B,11A,11B 2114 RAMs - all address lines moving, all data lines moving, all selects happening.
Found that it was the LS153 @ 9D in the address circuit to the 2114's - although the signals were moving, the functional logic output was not right. Comparator told me this.
This is another good example of where a logic probe won't cut it for everything but it'll help a lot of the time. Without the comparator i'm not sure how i'd have figured that out without removing each 153 in the circuit and testing out of circuit - I don't own an in-circuit tester like an ABI Boardmaster or even the mighty Fluke with some trickery wouldn't have helped.
Replaced the LS153 @ 9D and the game was back to normal and the test ROM told me so...
Tested in game mode, everything fine again now.
Frothmesiter's Track & Field
I forgot to take a picture of the symptom here but what was happening was that the background was just an off green colour, game would play in the background but you could only see the sprites.
Diagnostic ROM inserted and sure enough can't see anything on the screen but you hear the music play and the runner runs across the screen.
Started at the end of the circuit and worked backwards through the 157's and the 273.
Finding input to the 273 but nothing moving on the output I was poking with the scope when the graphics suddenly flickered back to life - note all signals TO the chip were present.
Reflow around chip and check traces for damage and backgrounds are restored.
Game has an intermittent halt/reset/watchdog issue still - every now and again the game will freeze and reset.
A little problematic to find this one - had to use the old "inspection eye" tool and look for problems - tapping the board brings it back to life.
Did preventative ROM pin cleaning and socket evacuation - gets better results.
Eventually I figured out by trial and error that the problem was on the sound board - reflowed a tonne of dry joints and found a small cap on the edge of the board twisted over so its legs kept touching - touch them together - game resets...... Spread them out neatly and all good again.
I booted the game and decided to have a quick go and discovered some of the speech missing - it would should "on your marks" but that was it.
A quick pull of the EPROMs on the sound board and a ROM ident via the online tool and find that C9 doesn't match anything - burned new EPROM and we have everything back to normal.
Not sure if this version of the diag rom has a speech test in or not but all sounds passed and sprites and backgrounds looked fine.
Frothmeister - Hyper Sports
Oh, you'll love this one...
Game sort of starts up - looks like a Commodore 64 loader or an Amiga decrunching a game or demo.
My first thought was that the horizontal clock was stuck doing 1H and thus repeating one pixel of the output - oh my no....
Pulled the board apart on the bench so I can poke about.. I have some nice 40-pin ribbon cables that work perfectly on these boards.
With the game on, I pushed on all the socketed IC's - firstly the custom 504 @1c on the lower board - no crunch but when pressed, backgrounds try to come back - remove custom chip, clean legs, clean socket, replace, backgrounds mostly back - squeeze socket - everything fine, replace socket, backgrounds return but with sprite issues.
(sorry about the sync bars)
You can see here that the sprites for the weight lifter are corrupt on another screen, the runner on the long board is fine but in mid air he's a blob of garbage.
I've seen this before a few times to its graphics rom region time. Quick look at the schematics and the objects are in the lower right corner of the bottom board labelled C11-C19 normally. This board has had the eproms compressed into 27128's.
Inspecting carefully I find the one in G17 to be inserted upside down - oh dear, well that's toast.
Now, I don't have 27128's on stock currently, only 27256 and 2764 - so I had to manufacture a 27256 with the images of the 2 roms it covers in the right order. This was a bit of guess work.
C15 and C16 are in one rom in G17 on this board.
On my eprom burner PC I downloaded the hyper sports ROM files and did:
Code:
copy /b c15+c16+c15+c16 g17-2756.binAnd burned it - nope, sprites still garbled but clearer - it then occurred to me that while verifying the other ROMs that the G19 one which is labelled C18 I had to read as a 2764 so the C18 image is in the LOWER part of the ROM - ahhhh:
Code:
copy /b c16+c15+c16+c15 g17-27256.binRe-burn and place in socket, game starts, weight lifter has jailbars but is clearer - again, having seen this on my other boards, squeeze the socket - ta-da, jailbars gone.
Replace socket at G17 with a nice new one and everything hunky dory.
Don't just cupboard these boards - they're worth repairing.
Hope this is useful info.
PS. These 2 games are a personal top favourite game for me, if anyone has junk boards they want rid of, i'm happy to have a go at turning them back into workers for the sake of preserving this awesome game.
RGP2016-04-11 22:53:35
