Defender 4 light shenanigans

mecha

Well-known member
Joined
Apr 19, 2007
Messages
24,331
Reaction score
7,037
Location
MKE, Wisconsin
well, I too acquired a Defender recently. machine was super dirty, so my first step 2 weeks ago was to give the PCBs rubbing alcohol baths. I pulled all socketed chips. the ROMs had a lot of corrosion on the chips so I sanded those and reinstalled them accordingly. I used those commercial hand dryers in our bathrooms for drying the boards.

the triacs transistors on both the sound board and main board had either instances where the legs broke entirely, or they were a twisted/tangled mess. I bought replacement 2N4400s at Radio Shack (I researched this from another Defender thread on here) and ran across some 2N4403s along the way. I MAY have reinstalled the 4403s in wrong spots, I know all the 3 video ones are correct, I've seen on threads here where people are talking about parts locations on the boards, and mine have no identifications for what's what at all, unless I'm supposed to read the manual. :)

I get no startup sound, however you can hear the amplifier kick on, and all I have are 4 solid lights on the ROM board (they do not flash at all). there's a chance the color balance is jacked up on the monitor (a Sampo, if that helps) I get what appears to be a red screen

point of interest: I've noticed there's some oxidation on a few chips near where the battery holder was. I removed the old battery holder and picked up a CR2032 socket but haven't done that mod yet. I really just wanted to see if the game would even turn on.

I double checked all my wiring off this: http://www.cityofberwyn.com/schematics/williams/DEFlay.tiff

I had 5.20V at the switching power supply (mine's moded, the original PS board is missing).

thoughts anyone? did I fuck up when I cleansed the boards or is that a common problem if there's battery leakage? I never observed the green oxidation around the pins before cleaning.
 
The cleaning shouldn't have hurt anything. The sound board has a self test button, push it and you should hear a continuous sound test - you have to shut the machine off to silence it. Your machine isn't getting too far through the boot sequence, it's pretty early that the led's on the rom board are set and cleared (that first blink). I'd first recheck your connections (particularly rom board to cpu) then check the voltages at the 4 corners of one of the ram chips, then see if your cpu is getting a clock or reset signal.
 
If you have a logic probe check the Reset pin on the CPU. Very often when there is corrosion near the clock circuit the Reset pin is pulsing basically freezing the CPU. The other very common issue is the ribbon cable to the ROM board. Check to make sure the connector to the MPU board is tight all the way accross the top of the connector (where it squeezes the cable into the IDC connectors). If it is at all sloppy you may need to recrimp it. The following link shows the correct way to recrimp the connectors (ignore the part about cutting the ribbon cable unless you are going to replace it): http://www.arcadesolution.com/ribbon.html

ken
 
i dont have a probe but i have an oscilloscope, big ancient one lol. If it helps any i ran it for awhile last night and felt around the cpu board and nothing got hot. But could any of you tell me the orientation of the notched ends of the ICs? i think i have them installed the right way but most of my sockets dont have the U in them. and if so the shit would be fried anyway right?

im typing from my phone, not at work now.

can i hook a jamma game up to the Sampo monitor? i dont know what sync those run. Bitch is the tubes got scratches in it :( fires up fine. Probably needs caps. Vert lin by the looks might be off and my h-size(?) pot is broken
 
http://www.robotron-2084.co.uk/techwilliamshardwareid.html#defender

In the orientation of the picture, all of the notches are facing up. You can see them if you click on the picture and look very closely.

Jamma uses negative sync (williams games are positive sync). There should either be an additional header for negative sync or a switch or the monitor will autosense it.

ken
 
ok well, that big switch on the CPU board.... isn't that the Reset switch? what's the likelihood I damaged that by submerging it while cleaning? :)

is there a way I can pull that switch out just to see if it's the cause of the problem, and is there a suitable replacement from the new parts bin as opposed to chopping another board up?

I double-checked my triacs, there were some that the solder didn't flow through all the way to meet the traces on the component side, so I touched those up.

and I did the sound board test... it sang for me. so I'm inclined to believe everything's good, except maybe the board's stuck in Reset....
 
ok well, that big switch on the CPU board.... isn't that the Reset switch? what's the likelihood I damaged that by submerging it while cleaning? :) .......maybe the board's stuck in Reset....

the switch in the circuit is a single pole normally open switch. the ones i have seen are the f-type switch (see below for example) so any extra poles or terminals are not connected.

you can test it simply with the game off and with a multimeter on the continuity setting. usually the switch has three pins, centre is common, NO is one side, NC is the other. Check the PCB pads underneath it to confirm which two pins are used as the NO (normally open) RESET switch and then test continuity across them. if it is shorted without you pushing it in, then it may be full of fluid or shot. if it is shorted when you press it as well as when you dont press it, then definately shorted or shot.

if the switch behaves as it should (ie not shorted, then shorts when you press it) then you need to look more closely at Q9, Q10, Q11, Q12 and 6O (7411 AND gate) and associated resistors.

you could use a multimeter to test the RESET line state at pin 37 of 2I (the CPU). this line should be HIGH when inactive and pulse LOW when the reset switch is pressed. if it is neither 0v or 5v on the MM, it may be pulsing so fire up the oscilliscope, set the input volts range to 1v/divn, DC coupling, move the trace downward, then hookup the probe earth to ground, and the centre pin (on high impedance) on pin 37 of ic 2I (the RESET line.) theoretically, it should register 5v in normal use, no groundward pulsing. if it is pulsing, need to look at the inputs to IC 6O closely, and its output to check it out. if it is stuck on ground, check voltage across & resistance value of R15 (it should be conn to 5v at one end and RESET at the other) or 6O may be faulty

check those coupla items and report back!
 

Attachments

  • ftype sw.JPG
    ftype sw.JPG
    8.6 KB · Views: 0
Last edited:
thanks for the details sir. not sure if I'll get around to it this weekend though. (my busy point of the week as a tech at my family's arcade!)

actually had the cab thrown back in the storage garage for the time being.
 
Back
Top Bottom