Taito space invaders display showing duplicate elements

[kiyoukan]

I have a black and white 3 board stack space invaders game.
I have been able to rule out the PSU and monitor and driver board using external power supplies and monitor.
It worked once for a short bit after removing the monitor from the cabinet and putting it back in for inspection but only for a day.
I have checked on the board stack and I'm getting all my voltages as expected across the 3 boards.
The game is playable even with the messed-up display. the bases are still there you just cant see them and as the invaders make their way down the screen more show up and you can shoot them even off screen including the ufo.
I am comfortable with a soldering iron and scope.
I was hoping someone could give me an idea of what chips to look at and test.
I do have a schematic that is very close to my machine it has some extra chips but seems everything else lines up

https://arcarc.xmission.com/PDF_Arcade_Manuals_and_Schematics/Space_Invaders_Schematics_(Taito)_scan2.pdf

Looking at the above PDF I feel like my issue is in the bottom right page 8 looking at those chips and working backwards.

 

[wugly]

I think page 6 left side. page 8 can be bypassed with a braze kit

 

[wugly]

page 8 is rom chips fed by and chips which is used as a buffer that is fed from the cpu address bus. there isn't much back from there

 

[HudsonArcade]

Looks like the video counter that's muxed to make RAMA0 is stuck / floating.

Start checking at IC7 pin 2 (16H) on the middle board and work your way back to IC8 on the ROM board.

(Edit: Changed refdes to match the linked schematics.)

 

[kiyoukan]

Looks like the video counter that's muxed to make RAMA0 is stuck / floating.

Start checking at IC7 pin 2 (16H) on the middle board and work your way back to IC8 on the ROM board.

(Edit: Changed refdes to match the linked schematics.)

Awesome, i will start looking up the data sheets of the chips in the chain to understand what signals I should see.
Would I be correct that I should be able to compare the behavior of the chips vs its neighbors?
ics 7,6,5 are all the same on the middle board and on the rom board too I see that the chips in the chain are the same.

My only rub is this darn board setup makes it a real pain to work on being stacked and all with those short ribbon cabled between the boards.

 

[wugly]

its a mux

https://www.jameco.com/Jameco/Products/ProdDS/301612-DS01.pdf?srsltid=AfmBOopo2WIARF7zvqhaFXkk94aLFBmzqB1GFtsTQcHegbPZ43Ti087O

2 separate inputs that which input is passed along to the output depends on the selection on pin 1. if you see data going in changeing you should see data going out. consider though that since the game worked for awhile after you had it out that you might have a bad connection between those chips and one or more rams that was tweaked enough for awhile to make the game work and now one of those rams has a floating input

 

[HudsonArcade]

its a mux

https://www.jameco.com/Jameco/Products/ProdDS/301612-DS01.pdf?srsltid=AfmBOopo2WIARF7zvqhaFXkk94aLFBmzqB1GFtsTQcHegbPZ43Ti087O

2 separate inputs that which input is passed along to the output depends on the selection on pin 1. if you see data going in changeing you should see data going out. consider though that since the game worked for awhile after you had it out that you might have a bad connection between those chips and one or more rams that was tweaked enough for awhile to make the game work and now one of those rams has a floating input

If the RAM had a floating input, the game wouldn't play.
The problem is on the video side of the mux.

 

[kiyoukan]

Looks like the video counter that's muxed to make RAMA0 is stuck / floating.

Start checking at IC7 pin 2 (16H) on the middle board and work your way back to IC8 on the ROM board.

(Edit: Changed refdes to match the linked schematics.)

HudsonArcade​

Thanks for the advice on where to start looking and letting me know that because the game plays my roms are hopefully good.
I used my logic probe and made a truth table of sorts.
Interestingly, IC7 pin 2 and ic28 pin 9 are both stuck low as they are connected.

my question what the best way to tell what chip it is pulling it low?
The fastest way I can think of is to use micro flush cutters to cut the leg half way on ic28 pin 9 and bend it to not make contact and if IC7 pin 2 starts pulsing then I know ic28 is dead.
If it stays low then I know that ic7 pin 2 is dead and I can rebend the leg back on ic28 and solder bridge the cut line.

Obviously once I confirm the dead chip ill use my solder sucker to remove the chip completely and solder in a new one.

 

[kiyoukan]

just read the schematics it looks like ic 8 74161 (was late it meant ic18 74LS86) is dead
with 10 being high
pin 9 pulsing
output pin 8 should be pulsing but it is stuck low thus driving the above chips in the chain to be stuck low.

 

[mecha]

just read the schematics it looks like ic 8 74161 is dead
with 10 being high
pin 9 pulsing
output pin 8 should be pulsing but it is stuck low thus driving the above chips in the chain to be stuck low.

your pin 8 that's stuck low, try this with multimeter on diode test. touch the red probe to ground and then touch the legs of the chip with the black probe. given it's TTL (there's transistors inside) you should see junction drops (something like .450-.600~) but if it's reading close to 0 like it's shorted to ground then you know that gate of the chip is bad and it needs to be replaced.

there's exceptions to this, like if it's a chip where they're only using half of it, it's normal for the unused pins to be tied to ground (you should never leave them floating). 74161s are kind of a common failure item so this isn't surprising to me. anyone take into account SI is nearing 50 years of age?

 

[wugly]

regarding ic7 pin 2. that is an input. disconnecting ic28 pin 9 will cause it to float not pulse but you could look at ic28 pin 9 and see if it pulses. then again it could be the inputs on ic28 that drives the output of pin 9.

the 161 chip pin 8 is gnd the chip is supposed to be grounded there its part of what powers the chip

 

[kiyoukan]

it was late and i botched the above post.
The dead chip is 74LS86 ic18
Pin 9 input pulsing
pin 10 pulled high
So output pin 8 should also be pulsing but its stuck low.

Tempted to just cut the legs and whack a npn transistor across it with a few resistors.

 

[wugly]

wake up. I have no idea what you are talking about ic 18 is a 74ls08. pin 8 should mirror pin 9 if pin 10 is pulled high. question is if ic 24 which uses this address is pulling it low or is the output of pin 8 bad to begin with.

 

[kiyoukan]

the schematics are written in a confusing way.
i was referring to the ic number as per the PCB count of chips not as the BOM labeled on the chip as there are shared by multiples
SO on the PCB it is IC18 position but the bom list has it as IC14 but as you can see ic14 is listed for multiple chips as they are all the same type of ic.

 

[kiyoukan]

so i cut and bent leg 8 on ic18 and wired in a quick bodge npn transistor plus a few resistors to check how the display would look.
Seems I am on the right track.
Once the IC's arrive I will swap all of them out for good measure. 15, 16, 17, 18.
Hopefully, once I do that the bottom of the screen will also be good.

 

[HudsonArcade]

It's properly listed as 18 in the schematics.

 

[kiyoukan]

I know it was back in the day and this was drafted by hand but I am spoiled by modern computer-generated schematics.
I found it hard to read some of the text in places and then on data sheets they don't use the same letters such as Q is E.
I hope after this it will live a long and problem-free life in its new home.
I just hate the idea of someone throwing it away for parting it out on ebay.