PCB Repair Logs – Page 67

Double Dragon repair log #8

PCB Repair Logs Comments Off

Mar 262016

Bought on Ebay this cheap untested Double Dragon original PCB:

Double_Dragon_PCB

Board came in not very good shape, CPU board was really dirty as you can see from picture above.I powered it up and got a solid black screen so I decided to washed it.After dried and removed the oxide from JAMMA connector :

Double_Dragon_PCB_washed ì

I powered the board again and this time it successfully booted but after started a game I noticed some sound FXs were wrong, replaced only by some noise :

Sound FXs are generated by two MSM5205 ADPCM chips which read data samples from two 27C512:

ADPCM_circuitry

The two MSM5205 were tested good on all pins and piggybacking them didn’t change nothing but when I went to test the two 27C512 I found that pin 7 (address line 4) of the one @IC95 was silent so it wasn’t being addressed.Schematics showed it connected to pin 3 of a 74LS393@IC61 :

27512@IC95_74LS393@IC61 :

Jumpering the two pins fixed the issue and all sound FXs were restored but it was an ugly solution so I followed the trace until the point of its break.This lead me under the socket of the 27C512 @IC95 where a pad was corroded and lost contact to pin 7 of this EPROM.I restored the connection with a piece of AWG30 wire after breaking the plastic of the socket in order to reveal the pad:

RSCN2560

Another arcade PCB preserved!

Bad Dudes Vs. DragonNinja repair log

PCB Repair Logs 1 Response »

Mar 252016

Some days ago I had on the bench this Bad Dudes PCB for a repair:

DSCN2505

Board played fine but clearly had color palette troubles which I narrowed in CPU board by swapped a good one:

DSCN2506

Putting my fingers on board some colors were changing so I could locate the palette RAMs in three TMM2018 chips:

DSCN2507

Probing the chips I found that pin 6 (address line A2 in common) of all of them was completely silent, this was confirmed also by my scope:

DSCN2508

According schematics this signal should come from pin 9 of a near 74LS157 @15H :

TMM2018@15K_address line_2

but I couldn’t measure any continuity with my multimeter.A closer inspection revealed some corrosion on pin 6 of the TMM2018 @IC78 :

RSCN2511 ac

I removed the chip and found that the pad was completely corroded and lost contact with its traces:

RSCN2521

I promptly patched them and this fixed the game completely:

DSCN2516

X-Men repair log #2

PCB Repair Logs Comments Off

Mar 242016

The board played completely blind.

With a logic probe I tested all the 74Ls07 which had fixed input and outputs as expected, therefore no activity at all.

Tested the clock (pin 11) on the 74LS273@9C and 10C and it was fixed high.

xmen-schems

Unfortunately the schematics of the game found in internet are not complete, so I had to trace back manually and found out that the output pin 14

of the 74LS04@17E was stuck, while the input was floating.

xmen

Replacing the chip with a good known one didn’t fix the problem!

So I looked on the back of the pcb and found that a pin of a chip was touching the line of the output it shorted it !

After lifting the pin, the pcb showed the game in colours but all the screen was blocky.

While searching for the source of the problems I shorted two pins of a custom chip by mistake and the pcb rebooted without any issues.

It was clear it was some kind of reset problem for the graphic chips.

The 68000 had it’s own circuit but the customs had a separate reset.

Looking back on the schematics, I found out that the system reset was given by the custom 051550 which I couldn’t find on the board….

Infact it was missing and I didn’t notice!

Soldered a new one from a scrap pcb and the game was fully restored!

xmen2

Pang repair log #3

PCB Repair Logs, Repair Logs Comments Off

Mar 232016

I desuicided this original PCB but after booting up I got this

Pang Pang2

After checking rams and buffer they were all good.

Just to avoid any doubts I dumped without convinction all the graphics eproms which normally are very reliable.

To my surprise when I lifted eprom PWE02 with the pcb in operation, the parts of the background corrupted turned to black revealing that at least the problem was connected to the circuit around the eprom.

Placing the eprom on my programmer I got this:

Foto 20-03-16 10 00 58

Turned out that the VCC pin of the eprom was making bad contact internally.

Burned a new eprom and the problem was fixed:

Pang_fix

A picture of the bastard before going to the bin:

eprom

Sunset Riders repair log #3

PCB Repair Logs Comments Off

Mar 162016

Got this Sunset Riders PCB from my friend Joachim:

Sunset_Riders_PCB

He told me PCB suddenly stopped working while he was playing at it.When I powered it up I got this :

Board was constantly resetting sign that the watchdog circuit was active.The main 68000 CPU was halted and its data/address bus in high impedance state.For first I disabled the watchdog by closing the jumper ‘JP’ near the JAMMA connector:

J_jumper

The /RESET signal was correctly generated by pin 10 of the custom SIL ‘051550’ but its clock input on pin1 was stuck high :

051550

This signal labeled ‘AFR’ is generated by pin11 of a 74LS138 @13G:

74LS138@13G

74LS138@13G_PIN15

Obviously, since the CPU was halted, I got no activity on inputs and outputs of this 74LS138 but I found that its pin 15 (an output) was almost shorted to ground (only 6.3 Ohm of resistance):

74LS138@13G_PIN15

This was clearly abnormal, two were the possible causes : either this output was shorted internally to ground, either something was tiying it externally.I removed the 74LS138 but it was good and pad of pin 15 was still almost shorted to ground.Part of schematics posted above shows pin 15 connected to pin 1 and 15 of some 74LS253 multiplexers.Using again my multimeter in continuity check I could find the point of less resistance in the area of the two 74LS253 @14B and 13B:

74LS253_13B_14B