Space Invaders Deluxe

[orion3311]

Since we're dealing with digital electronics, you dont necessarily need to be concerned with voltages since most signals are either on (5v), off (0v) or pulsing (changing state). For that chip all you need to be concerned with is pulsing. If one output is not pulsing, THEN you figure out why. That last board i fixed, pin 15 was not pulsing because the input was stuck high on the chip generating it.

As far as knowing what to look for..you need the datasheet of the part in question. It helps to have a comp with internet access nearby.

What your describing almost sounds like what i was seeing while trying to figure out the rom mod, but im not sure where id start.

 

[KevinMullins]

For that chip all you need to be concerned with is pulsing. If one output is not pulsing, THEN you figure out why. That last board i fixed, pin 15 was not pulsing because the input was stuck high on the chip generating it.

That's where I was kinda having trouble interpreting my probe. On a couple pins the pulse light was lighting, but the tone was more solid than the others, so they certainly weren't identical. I forget now what I was getting on the inputs.
Went round and round on this thing so much I found myself forgetting where I was at on it.

But there has been some unexplained progress on it.....more to come.

 

[orion3311]

Ah you have audio too - mine doesn't have audio, just the 3 lights. It may sound different because it could be different clock frequencies. I wouldn't worry about it - since the reset line seemed to take effect I think you're ok for that chip.

 

[KevinMullins]

I just always look for alternative methods for double checking stuff. (with laymans tools) And one the the next steps involved "cutting out" legs on the 8216's to do further checks. (wasn't about to do that yet) Especially since I don't have an o-scope to check as they showed for examples.

 

[KevinMullins]

Ok, so here's what went down.
I messed with this thing for nearly 8 hours off and on finding nothing obvious.
Same exact stuck screen the entire time.
Pulled it out 2 dozen times, pushing prodding re-seating. No changes.
Had replaced E3 (74LS74) only because I had one handy and was mentioned in the manual. No changes.
Verified each and every socket for spot on continuity, doubled checked solder work and traces, everything looking happy. No changes.
All the logic probing appeared to be checking out best I cold tell at this point.
Finally about to give up for the evening and was going to shoot a short video to show the symptoms.....

Walked around front camera in hand, broken RAM lines. WTF!
Pulled all the ROMs back out, RAMs lines got a bit better.
Put the test ROM back in....bam! ... checking out fine.
Put the game ROMs all back in.... game came up just fine. 100% working, sounds and all.

I have NO idea what was going on, been running fine now for almost 24 hours.
Played several games on it. (yes, also verified that I suck at it)

I'm going to let it run some more and then maybe poke and stab at it to see if I can break it again. I don't like unanswered issues, they tend to come back up on ya when you least want them to.

Side note - I currently have S2 jumper removed. (had it removed the whole during testing) Going to let it run awhile and then put it back on the 5v line.

 

[orion3311]

Nice - at least its working again! Almost sounds like a cap type issue, doesn't it? Of coures there's only 8 trillion tantulum caps on there. Most of them just filter the +5v for each chip individually but I read somewhere where they can possibly go shorted, I dunno. It could possibly even be your power supply being noisy. Without a scope to check it thats always a possibility.

My third board is going to be a challenge, I think its in a lot worse shape than the other two. I also think it may have all rams bad, not sure.

Have fun with it though!

 

[KevinMullins]

I just hate phantom issues though, so yeah, glad it's working, still wish I had actually found something I could pinpoint though.
Since this will be a keeper I'll probably go through the power supply and whatnot after I run it awhile and give it some play time.

Hmmmm ... time to dig me up some spare boardsets.
I want to try my hand at that single ROM mod and maybe a Multi.

 

[orion3311]

You can make your own multi using the single rom mod as long as it's a Deluxe board, but overall the main benefit is that you only have to replace one rom socket, not all of them. If you google it you'll see its pretty much the same thing I listed but you add a switch to one of the address lines.

If you go the pre-made multi route, they put a board in place of the cpu so no actual rom socket is needed, and apparently it some some nice diagnostics too.

 

[Talon2000]

I'm still having issues with my SI.

Currently trying to use a double image 2732 and still having no luck.

Ohm'd out the pins.

2732 MB Pin
1 --->G3 11
2 --->G3 8
3 --->G3 3
4 --->G3 6
5 --->H3 8
6 --->H3 11
7 --->H3 6
8 --->H3 3
9 --->D2 4
10 -->D2 12
11 -->C2 4
12 -->Gnd
13 -->C2 12
14 -->B2 4
15 -->B2 12
16 -->A2 4
17 -->A2 12
18 -->E2 1
19 -->F2 8
20 -->Vpp +5
21 -->F2 11
22 -->F2 6
23 -->F2 3
24 -->Vcc +5

Still getting bars.. Checked all address lines, and Data lines with my logic probe and they are toggling.

New socket in H, New CPU socket. Verified CPU

Ugh..

What test rom image are you guys using? Perhaps mine is just screwy.

 

[Talon2000]

Ok I strapped my SID to match my SI, and tried the 2732 test rom. Works fine.

 

[KevinMullins]

So does your test rom now show a bad ram ?
Or do the grounding of pin 7 of each RAM and watch for a change on those bars, should be 16 nice equal lines, no separation, no random sparkly things, etc. (see pic)

These are the files I used for the Test ROM and all the game ROMs. (2716 versions)

 

[Talon2000]

Well on my sid, which has no problems the Test rom of course shows no errors. It's my SI that is having the problems.

It shows the bars like your picture, one fat line, and one skinny line however. I do notice that they shift positions. As in a upward downward movement every 10-15 secs, but they stay clean and crisp.

Ground pin 7 on ram chips appears to make no difference at all.

On a side note.

I did swap my SI, and SID sound cards onto the SID MB. Both sound cards work fine with the SID MB. There doesn't appear to be a part # on the boards, the only markings are Space Invaders, and then SID marking on the SID Sound board.

I though that kinda strange since I think it was stated somewhere that boards couldn't be interchanged.

 

[KevinMullins]

It's my SI that is having the problems.

Ah, I see now.... attached the SI rom files from the same place I found my SID rom files. These are the 2716 version still.

Also attached the test rom files for both 2716 and 2732.

 

[orion3311]

Those bars are basically what a (fairly) good board should be doing with no roms or test rom. If you put in the test rom into socket H and still get that, its not loading the program...either the strapping or the rom socket isn't functioning right, or the address decoder at E2 (I think - next to S6) could be bad/malfunctioning.

If you activate the slam switch on the coin door while the bars are there, it should shift around like you've seen. The reason it does it on its own, is (I believe) its the watchdog resetting the processor since there's no program running. Its doing the same thing as the slam switch. The nice thing about that is as long as it shifts around, it means the CPU is functional and the reset line isn't stuck or bad from the power supply.

As far as the sound cards, from what I understand, they're compatible on a hardware basis as long as you're running the program meant for the board. IE - you cannot run SID on a SI daughter board. Supposedly there's extra hardware for supporting SID and even on both boards. There's info out there on this that I found but nothing super concrete as to what's what.

In my testing I've successfully ran both SI and SID daughterboards on an SI board, and I believe vice-versa as the test rom probably isn't testing the extra SID sounds. Not 100% sure on that though.

Jeff

Well on my sid, which has no problems the Test rom of course shows no errors. It's my SI that is having the problems.

It shows the bars like your picture, one fat line, and one skinny line however. I do notice that they shift positions. As in a upward downward movement every 10-15 secs, but they stay clean and crisp.

Ground pin 7 on ram chips appears to make no difference at all.

On a side note.

I did swap my SI, and SID sound cards onto the SID MB. Both sound cards work fine with the SID MB. There doesn't appear to be a part # on the boards, the only markings are Space Invaders, and then SID marking on the SID Sound board.

I though that kinda strange since I think it was stated somewhere that boards couldn't be interchanged.

 

[orion3311]

Went back and read your other comments - Based on what I saw with my single rom mod using a 2764, try grounding pin 18 (chip enable) and see what happens. If that works, then its possible that the address decoder at E2 isn't working right or there's a solder bridge somewhere maybe?

Quick link to the 2716 datasheet which also has a side-by-side pinout of a 2732...
http://www.datasheetarchive.com/pdf-datasheets/Databooks-2/Book261-301.pdf

 

[hindered]

BTW Thank you all for the hard work and research put into this thread. This helped me get my board into booting shape.

 

[hindered]

PS - Just went back and looked your jumpers - COMPLETELY different from what I have!!!

This is the way my board came, with originally having 9316B roms, supposedly compatible with regular 2716's. I desoldered S2 to try un-grounding and putting 5v on it (I think it goes to pin 21 on rom H), but it turns out that didn't matter either! Once my ram sockets come in I'll solder it back together.

Also, other than your S2 going to ground and his going to +5v, these are strapped the same. Just different access points to the various rails. I was going over this last night while trying to strap my board and came to that conclusion... Mine is strapped the same as Kevin's but I chose a different point for the +5v.