Missile command ram issues

KazooBR

Well-known member

Donor 3 years: 2023-2025
Joined
Jun 9, 2010
Messages
1,887
Reaction score
1,555
Location
Kalamazoo, Michigan
I have two boards that are acting similarly and I'm stumped. Self test indicates ram issues (high beeps on certain chips) and I swap that ram out but the problems remain. High beep on chip 3, replace. Then high beep on 4, replace. Then high beep on 3 again.

Two boards are doing this. They both pass Fluke short ram tests.
 
Ah - yeah I'll pull mine off my machine and try that I suppose. Thanks for that reminder.
 
+1 for Braze kit

if replacing the rams isn't clearing the errors you'll need to trace backwards. I don't remember the anatomy of a Missile Command now.
 
yep. I tracked backward to a LS244 at P5 that the data goes through to get back to the CPU. It appeared OK but I pulled it anyway and it checks fine out of circuit. I put a new one in since I socketed that position, but no change.
 
I have two boards that are acting similarly and I'm stumped. Self test indicates ram issues (high beeps on certain chips) and I swap that ram out but the problems remain. High beep on chip 3, replace. Then high beep on 4, replace. Then high beep on 3 again.

Two boards are doing this. They both pass Fluke short ram tests.
Do they pass the Long RAM test?

What do the signals look like on the scope?

Anything look too low or too high?
 
Do they pass the Long RAM test?

What do the signals look like on the scope?

Anything look too low or too high?
^ this is the guy you need for this subject lol

I think you helped gamefixer and me with the Missile Commands we had come through. one of them after going back and forth over maybe a whole week I found that the poly caps near the rams had the legs super long and exposed and were touching the traces and that's what caused the game to crash when we ran it overnight.
 
Beep 3 = RAM@M4
Beep 4 = RAM@L4

They have this 74LS153 in common.. I'd be inclined to check it.. Not sure how the Fluke does it's tests..

1753280637374.png
 
Thanks all. I'll check this stuff when I can.

For now, I'll say it passes ram long but it's odd. First run was fairly short - 30 minutes or less. The second time kept going past 1 hour and I stopped it (workshop is also granddaughter bedroom and she gets priority at nap time 😄).

I'll have more time tomorrow on this.
 

Attachments

  • IMG_8789.jpeg
    IMG_8789.jpeg
    402.5 KB · Views: 10
You are checking RAM, but what about power? RAM will flag errors if your power isn't solid. Good 5 VDC, low AC ripple.

It's one thing that is common to all the chips (in addition to the 74LS153 noted above)
 
You are checking RAM, but what about power? RAM will flag errors if your power isn't solid. Good 5 VDC, low AC ripple.

It's one thing that is common to all the chips (in addition to the 74LS153 noted above)
I measured roughly +5.04, -5.4 and +11.78 at each chip.

And - using a workbench Atari power brick with new Big Blue cap. I haven't checked ripple since I installed this but it's working with everything else I'm testing.
 
I measured roughly +5.04, -5.4 and +11.78 at each chip.

And - using a workbench Atari power brick with new Big Blue cap. I haven't checked ripple since I installed this but it's working with everything else I'm testing.
Might as well check. It's the only way to be sure. Under load preferably.
 
Am I correct that my BugTrap won't help at all in this case since I'm dealing with chips on the address/data busses?

Or - can I use it to find problem lines with the understanding that the actual problem could be up or downstream of the chip I'm testing. I have yet to get comfortable with this tool and it seems like I'm mostly needing it when I get into these bus issues.
 
Beep 3 = RAM@M4
Beep 4 = RAM@L4

They have this 74LS153 in common.. I'd be inclined to check it.. Not sure how the Fluke does it's tests..

View attachment 834913

That mux is common to data in on those 2 chips, and MUSHROOM and MADSEL shouldn't be changing for the writes since he's ostensibly only doing 8b mode.

F/H/J/K3 muxes are common to the addresses on all of the DRAMs.
 
@HudsonArcade Missile Command is weird as you said..
I just read through some of my repairs.. Those 153's can take out one RAM at a time.. Good call..
On the FPGA CATBOX.. certain test patterns will pass and others will fail when the 153 is bad.. (I forget which - written elsewhere)

Now I'd also check the 153@J3 based on my notes..
 
Beep 3 = RAM@M4
Beep 4 = RAM@L4

They have this 74LS153 in common.. I'd be inclined to check it.. Not sure how the Fluke does it's tests..

View attachment 834913
yeah, I may have needed to replace a bad multiplexer. those projects were conveniently the ones I didn't take 20 pictures of what I did. it was also thousands of missions ago and my brain is running out of disk space to memorize every thing I do.
 
Back
Top Bottom