Star Wars vector problem

davespicer

Member
Joined
Jun 26, 2009
Messages
45
Reaction score
3
Location
Cornwall, United Kingdom
After 10 years of procrastination, I finally got around to repairing the HV/deflection boards on a Star Wars cab that's been sitting dormant in my garage for 10 years. The monitor is working great, the AVG board not so well...

What appears to be happening is that any time the beam is moved while inactive, the path is offset upwards. In other words, a right horizontal jump becomes a diagonal up/right. For example, a line of text will have correctly rendered letters, but with the whole line sloping upwards. The BIP screen shows the boxes offset in the Y direction and winding YBIP to its limit doesn't bring them in nearly enough to correct the error.

Here's a quick movie showing the game in action:

http://www.sparcade.adsl24.co.uk/temp/star_wars_sample.avi

Note the sloped text. Given how jumbled the Star Wars logo is, it's possible the error manifests itself differently on larger jumps.

I've tried swapping AVG and vector ROM chips from spare (non-functional) AVG boards with no effect. The AVG output also looks bad on an oscilloscope, making me sure the deflection board is beyond reproach.

I'm about to have a long stare at the schematics and see if I can figure out what could be happening. In the meantime, I thought somebody on this board is bound to have seen something similar and will immediately say, "ahhhh, that's caused by xxxxxx". Does anyone have any ideas?

--
Dave
 
Last edited:
I had a VERY similiar issue but not as drastic as yours. I found on the AVG board there was a cold solder joint on one of the ceramic capacitors. There are so many components on that board it's easy to over look the culprit. Pull the board out and lightly attempt to wiggle the brown ceramic caps and see if you have one loose.
 
To me, that Y shift across X looks like a pincushion correction circuit error. Check the MC1495's (multipliers, 2 of them) and the opamps and discretes around them. One of the chips is probably dead. Hopefully it's just a stupid opamp. They're super cheap and easily available.

Dang, I just remembered that Star Wars and Major Havoc corrected pincushion, not on the mainboard, but by using a different yoke than games like Space Duel, Black Widow, Gravitar. Now I can't recall if the Star Wars even has MC1495's in it. Well, if that's the case, then your yoke is probably shorted somewhere internally.

Bill B.
 
Last edited:
I had THIS problem (picture) and it turned out that my YOKE was off about 1/8th of an inch. I loosened it a little and slightly turned it and the problem went away.
 

Attachments

  • screen.shot.02.jpg
    screen.shot.02.jpg
    49.8 KB · Views: 95
  • screen.shot.03.jpg
    screen.shot.03.jpg
    69.2 KB · Views: 91
To me, that Y shift across X looks like a pincushion correction circuit error. Check the MC1495's (multipliers, 2 of them) and the opamps and discretes around them. One of the chips is probably dead. Hopefully it's just a stupid opamp. They're super cheap and easily available.

Dang, I just remembered that Star Wars and Major Havoc corrected pincushion, not on the mainboard, but by using a different yoke than games like Space Duel, Black Widow, Gravitar. Now I can't recall if the Star Wars even has MC1495's in it. Well, if that's the case, then your yoke is probably shorted somewhere internally.

Bill B.

I think if I had a yoke problem, I would see the active lines bent too. Visible lines are rendered correctly, it's just the tracking when the beam is off that's screwed. Feels like an AVG chip problem, but I've already disproven that one.


--
Dave
 
Last edited:
I had a VERY similiar issue but not as drastic as yours. I found on the AVG board there was a cold solder joint on one of the ceramic capacitors. There are so many components on that board it's easy to over look the culprit. Pull the board out and lightly attempt to wiggle the brown ceramic caps and see if you have one loose.

I'd bet it is something annoying like this.

Efforts at diagnosis have now been thwarted by a more pressing issue. I kept losing focus at random so pulled the boot on the focus wire to check. Sure enough, the wire had been hanging by a thread and came away in my hand. One quick resolder job later and I now have no display at all. :-( Turning up the brightness on the HV shows the correct display being traced, so at least the deflection is working.

I seem to be going backwards.

--
Dave
 
One quick resolder job later and I now have no display at all. :-( Turning up the brightness on the HV shows the correct display being traced, so at least the deflection is working.

I don't even know how in the hell that's even possible? I think you need check the B+ voltage now. Ensure it's at or close to 180VDC. I really doubt this is this issue but having a dim display like that is typically a failing flyback. I don;t know what else to have you check :(
 
I don't even know how in the hell that's even possible? I think you need check the B+ voltage now. Ensure it's at or close to 180VDC. I really doubt this is this issue but having a dim display like that is typically a failing flyback. I don;t know what else to have you check :(

B+ reads 174V, anode voltage is around 18KV. The heater voltage is only measuring 0.8 VAC... shouldn't this be around the 6 or 7 mark? Since the Amp gets this from the transformer, it's hard to see how the heater voltage could be bad when the HV is okay.

--
Dave
 
B+ reads 174V, anode voltage is around 18KV. The heater voltage is only measuring 0.8 VAC... shouldn't this be around the 6 or 7 mark? Since the Amp gets this from the transformer, it's hard to see how the heater voltage could be bad when the HV is okay.

--
Dave

Having checked using a scope, the heater voltage is a 7V square wave @ 25KHz with a duty cycle of 25%. Is this normal? The output from the 555 is closer to a 50% duty cycle, which pretty much talies with the resistor values.

Just in case ZREF had gone on the AVG board, I probed that and the RGB outputs. All are active and peak at around 4.5V or so.

A possible clue is that the retrace lines when upping the brightness are now green, whereas IIRC they were white before.

--
Dave
 
Found it! Shorting Q26 caused the display to spring back into life. Effectively I have a permanently active spot killer, even though the LED goes on and off as expected. I'll have a hunt around for the cause after work. Should be easy enough. I expect there's a dry joint which was cracked when I pulled the boards out to fix the HV.

I never thought I'd be so glad to see those skewed, messed up vectors come back again. :)

Right, back to the original problem then. Thanks for everyone's thoughts on that so far. The cause is definately AVG-related because the vectors are skewed in the same way when viewed on a scope.

--
Dave
 
did you guys read the RECENT story on the dude that marathoned this machine at 52 hours straight! He's crazy! But...........has the world record. How many times would one blow up the death star in a 52 hour game??

Best of luck fixing your unit
 
did you guys read the RECENT story on the dude that marathoned this machine at 52 hours straight! He's crazy! But...........has the world record. How many times would one blow up the death star in a 52 hour game??

Best of luck fixing your unit

Hmmm, I'd need to add cushions and a minibar first. :)

--
Dave
 
I wonder how the cross-hatch screen looks ?

Here's a run through test mode:

www.sparcade.adsl24.co.uk/temp/sw_test.avi

I skipped the scale test too quickly. What's not shown is that the linear+binary test pauses periodically. Having never seen it work, I'm not sure if the behaviour is normal or not. The subsequent test completely freaks the monitor out.

Ignore the black blob on the intensity test by the way - that's just a burn in spot the size of a small moon.


--
Dave
 
Well from looking at the video you need to adjust your x and y bip pots on the game board. Use the bip test screen to align the green box so you have a one line box.
 
Well from looking at the video you need to adjust your x and y bip pots on the game board. Use the bip test screen to align the green box so you have a one line box.

See my first post. :)

Cranking YBIP doesn't provide nearly enough adjustment to correct the BIP screen. There may be a fault in that part of the board of course, although I've yet to find one.

--
Dave
 
Back
Top Bottom