Hyper Sports Pcb Repair

Del Griffith

Active member
vacBacker
Feedback
15 (100%)
Credits
805CR
Bought this dead Hyper Sports pcb off ebay a while back. The pcb was in excellent condition and better than my worker, I thought maybe something easy...... ha ha big mistake! I don't really log any of my repairs because i don't have the technical knowledge or skills to know what i'm talking about so most repairs are fluke or belt an brace jobs for now. I think this board deserves a mention tho. I don't have any pictures for the 1st stage as i didn't think it was going to be a major.

First i wanted to see which of the 2 boards were faulty using my worker and so after some swapping around i found both boards were faulty. Hmm i decided to start the bottom board first which was dead with a black screen and loud buzzing noise. I read all the roms and they all checked out ok, Cleaned all socketed custom chips and tried again with my working top board...... no change, Next i started poking around with the logic probe and found tons of floating pins on some of the Fujitsu chips. I've heard about them being likely candidates on bad boards and a good place to start so before trying to look over the horrible and virtually unreadable schematics for this game i started piggybacking some new chips over the ones with floating pins..... what do ya know one by one they started pulsing again, Found a total of 14 chips 12 fuji's and 2 toshiba 2114 ram chips now all piggybacked and pulsing away.

This was the result

pic_1.jpg


pic2.jpg


Board is booting and running partially with garbled graphics.

Thought at this point i better start replacing and socketing some of the suspected faulty chips and see if they really are all faulty. I'd say at this point you have to ask yourself how much do i like this game 14 chips an all!!! In my case alot as i love hyper sports. Taking it slowly to not damage the tracks or board i replaced them one at a time checking the results after each one......

All 14 replaced and board is running as pictured above. Time for a few days off it.

After a few days away from this board i came back turned it on and the colours were gone and game even more garbled!!! Stressed with it at that point i didn't take a picture of the fault but another failure somewhere was certainly making things even worse. I poked around again trying to read the schematics and piggybacking any chip that seamed to be acting oddly but no change at all, I was starting to lose faith with it a little, Thankfully a chance visit from the one and only Mitchell Gant armed with his scope got me back on track. Neil managed to locate a shady Fuji LS245 at location E9 which was causing the most recent problems (piggybacking this chip made no difference so i learned a good lesson there not to depend on that technique) Right excellent and back to running like the pictures above.

Here's how the board is looking so far

pic3.jpg


came back to the board 2 days later and this

pic_4.jpg


pic5.jpg


Aaaahhhhhh another failure without touching anything.... no sprites anymore and less backgrounds!!!

Thinking now if these chips are failing at this rate and all apart from ram have been Fuji's i might as well just swap them all out sein as i've come this far. I don't plan on getting rid of this board so the cost is not important plus the small ttl chips are cheap anyway.

Started swapping two at a time then checking the results,

H11 - LS244 & F12 - LS32 and this happened

pic_1.jpg


pic2.jpg


A few more with no change and then i swapped an LS245 at H1 and this

pic6.jpg


Much better, Just faded sprites and a slight wobble on the screen but really getting somewhere now.

Carried on swapping til all were done but this is as good as it was getting just swapping out the Fujitsu ic's so what next?? Everything seemed to be working fine when going round the board as far as i could tell. Read through a few repair logs of this game and decided to change the 2 toshiba 2114 rams i had not already done in the block of four at locations H8 and H9, Again piggybacking these made no change where as the two next to them did make a change earlier on in the repair so i left these 2 thinking they were fine, This was the result....

pic8.jpg


pic9.jpg


Working Perfect!!! The couple of glitches on the score is just because their's no battery in the top board.

Here's how the board looks (socket city!!!)

pic10.jpg


Roughly 40ish chips in the end, Not all faulty but how long would the rest have lasted after some play??? prob not long judging by all the other faulty ones.

Once i've fixed the top board i'll post an update on this thread, That's if i can get it sorted
smiley36.gif


Del Griffith2013-09-04 23:20:59
 

Mitchell Gant

Active member
vacBacker
Feedback
2 (100%)
Credits
884CR
Wow, well done Baz!!
smiley32.gif


Stubborn determination and you've got it running with just a logic probe, a soldering iron and solder sucker. Oh and a lot of days bashing away at it.

I bet you're a well practiced with the chip removal now!

I printed an A2 size schematic for you of this board, just hadn't dropped it in with you yet. Looks like you don't need it now
smiley2.gif
Was still crap quality because of the low res scan PDF though.
 

grobda

"Look at the size of that thing!"
vacBacker
Feedback
7 (100%)
Credits
1,061CR
result! :)

I had a pretty similar story with a time pilot '84. forget how many bad fujitsus were in it but it was more than 10.

Please tell me you didn't do all that with a manual solder sucker?
smiley36.gif
 

RGP

Meeter & Greeter
Feedback
5 (100%)
Credits
2,039CR
Awesome mate, well done, this is one of my favourite games too. I have a board with mega crashed sprites but its not the EEPROMS.

Wish the schemes were easier to decipher on this game.
 

cmjones01

Newbie
Credits
194CR
That's an impressive repair, well done. Without wishing to hijack the thread, do Fujitsu TTL chips have a bad reputation? I've had a lot of trouble with them on some arcade boards, but having only ever repaired a dozen or so, I didn't know how common it was. Other brands did seem more reliable.

Chris
 

Del Griffith

Active member
vacBacker
Feedback
15 (100%)
Credits
805CR
grobda said:
result! :)

I had a pretty similar story with a time pilot '84. forget how many bad fujitsus were in it but it was more than 10.

Please tell me you didn't do all that with a manual solder sucker?
smiley36.gif

Yeah unfortunately a manual sucker
smiley29.gif
Bought one of those Engineer Japanese made ones which are quite pricey at £20 but worth every penny, They really do speed things up.
 

Del Griffith

Active member
vacBacker
Feedback
15 (100%)
Credits
805CR
RGP said:
Awesome mate, well done, this is one of my favourite games too. I have a board with mega crashed sprites but its not the EEPROMS.

Wish the schemes were easier to decipher on this game.

The schematics really are terrible to try and read for this game. You probably know more about the board than me but I'm pretty sure it's the block of 8 2764 roms that hold the sprite data so maybe best place to look first is those 4 2114 ram chips and the LS244's & 245's between the ram and custom chip.
 

Del Griffith

Active member
vacBacker
Feedback
15 (100%)
Credits
805CR
I have only read about the Fujitsu chips being prone to failure myself, Have a few boards riddled with them but still all work fine so it maybe down to just certain batches that were made.

This board is a revision B but my other is revision C and has hardly any Fuji's on but instead has single legs lifted and cut off certain chips?? quite crude but seen others like it so it was prob factory done for some reason. The schematics are from the older revisions of the pcb tho. The rev C boards have Half the amount of roms and are Konami printed mask roms which for reference need to be read as 27128's and the files needed to burn new ones from mame are the Hyper Sports Bootleg roms ?? Strange i think but my mame is pretty old and i use winromident so that's possibly been changed on newer versions of mame
 

trm

Who loves you, and who do you love?
Feedback
2 (100%)
Credits
2,876CR
sh*tty/cheapo manufacturing it seems, a bit like the Atari flyback problem. It seems the bond pad wires corrode on Fujitsu within a year or so manufacturing range.
 

Del Griffith

Active member
vacBacker
Feedback
15 (100%)
Credits
805CR
I wonder how many were made in that manufacturing range!

Did think about over voltage with the amount of faulty ones at first but maybe other brands would also have failed if so. I'm just glad all the customs and proms are fine.
 

Muppz

Active member
Feedback
6 (100%)
Credits
362CR
Nice job, I've done dozens of Konami boards and its the same Fujitsu story. Good idea to shotgun, I would replace the faulty chip then days later another chip would pop on the board.

Even some bootleg capcoms have suffered the same problem.
 

Del Griffith

Active member
vacBacker
Feedback
15 (100%)
Credits
805CR
The top board partly worked, Cpu side was running fine but sound side was making an annoying buzzing noise that could be turned up and down with the pot but no game sounds. Amp and volume pot probably ok with the noise and also it being adjustable.

Probed around the sound section and every pin was stuck or dead, Only thing happening was a clock signal on the fm chip (I think it's the fm chip at location A13 ? rubbed out on the schems also ) The Z80 was also doing nothing. Shotgunning the Fujitsu's made no change so i started replacing them.

LS138 @ C18

LS02 @ D18

these 2 started the speech working and some activity on the Z80 - buzzing still the same

LS138 @ D17

2114 @ C16

2114 @ C17

LS138 @ C8

LS08 @ C9

LS367 @ A5

These 6 brought the sound back fully and cured the buzzing, all fuji's and toshiba.

1.jpg


Both boards fully working now and still in good nik even after all the work so well worth it i think.
 
Top