Defender

Baiterhater

Active member
Credits
73CR
Ok so i could do with a little help, i have been pouring some serious love into a defender with a total restoration which ill post some pic up on here shortly its had a complete respray.

The very first game it repaid me by crapping out mid game at 100k, oh well thats 80's electronics.

I have traced the fault to the rom board, turn the game on and i get the start up sound and a continuous rug no led's on the rom.

I had a little probe around the rom and noticed that i had no outputs at all from IC 17 (74ls395) on pins 12-15 yet good pulsing inputs on pins 3 to 6, pin 10 clocks the 395 when low and this is held high so i guess this is why i have no outputs.

Tracing pin 10 back it comes through an or gate at IC14 and pin 9 of this is held high this comes from IC15 which is a 74139 decoder. inputs 2 & 3 are pulsing nicely and so is the enable at pin 1 yet 5 & 6 are both held high so my thoughts are suspect IC 15 ?

Im a little rusty been out for a while but still know my way around a meter and logic probe so any help muchley appreciated

Cheers

J

ic17.jpg
 

Mitchell Gant

Active member
vacBacker
Feedback
2 (100%)
Credits
888CR
Hi,

Yes you are correct in your deduction to suspect IC15, but there isn't enough information from a logic probe to be sure. It could be that IC15 is working perfectly, but the software program is not accessing address D000 and therefore not triggering the Y1 (pin 5) output.

You'd need a logic analyser or three channel 'scope to be sure, then you could check that when E1 input is low, if input A were high and input B were low all at the same time, then Y1 should indeed pulse low. You just can't do this with a logic probe I'm afraid.

My next course would be to piggyback a good 74LS139 (or even a 74HCT139) onto IC15 and see if Y1 then starts pulsing. Even piggybacking a good chip won't 100% prove it's the problem though. The original chip may be damaged in such a way that it prevents the piggyback chip from driving the pin properly. In my experience though it often does show up a bad chip, so well worth trying before having to desolder stuff.

Your worst case will be that IC15 really is fine and it's not accessing address D000. Then you've got a much more difficult hunt ahead to work out why that is not happening. Maybe the software waits for another input somewhere before trying to change ROM pages. Afraid I don't know Defender operation that well to advise.

My fingers are crossed for you that it's IC15 that is duff, and you were perfectly right first time!
 

Baiterhater

Active member
Credits
73CR
Thank you you're a diamond !

I have a 139 coming tomorrow so ill piggy back or swop it out, if its addressing issues I'm sending it out for repair I'm just not that good and don't have the kit
smiley36.gif


Thanks again ill update

j
 

Mitchell Gant

Active member
vacBacker
Feedback
2 (100%)
Credits
888CR
If you did want to delve deeper £10 gets you a cheap logic analyser with 8 channels:

http://www.ebay.co.uk/itm/Logic-Analyser-24MHz-8-Channel-Input-20cm-USB-Analyzer-Dupont-Flux-Workshop-/111674825674?epid=879778951&hash=item1a00569bca:g:pjwAAOSwBLlVWvHX

I've not got one like this, but a chap at work has and says they're pretty good for the price. You could check much more than a logic probe with this, easily prove your '139 function and more as long as you understand TTL logic. Your post shows you already do have a good grasp of that!

It's hard to keep this stuff running isn't it, and always seems to fail just when you think it's all going good.
smiley1.gif
 

Baiterhater

Active member
Credits
73CR
Thanks ill take a look at that !

I just couldn't believe it id had it in bits for months i replaced every molex connector, reflowed every header re built the power supply, cleaned everything, upgraded to 4164 ram and the very first game was perfect for 10 mins then all of a sudden it went into a rug and stayed there, couple of games in sure but i didn't even get to finish the first
smiley36.gif


Still the more you repair the better you get

j
 

philmurr

Active member
vacBacker
Feedback
48 (100%)
Credits
2,374CR
Good advice so far and definitely the first thing to do is to piggy back the 139.

But also as suggested, Defender doesn't access D000 until out of test mode, which includes checking the PIA

So if it's not the 139 I think you may need to replace the PIA as constant rug is a classic symptom of one that's failed
 

Baiterhater

Active member
Credits
73CR
Well i replaced the 139 but symptoms were still the same, also replaced the PIA but again still the same.

I have checked continuity on every socket on the roms and PIA.

so I'm guessing that although pulsing the data on pins 1 ,2 & 3 on IC 15 is incorrect, this comes from the address bus which looks like it comes from the roms, so I'm thinking i need to verify the rom images are correct ? anyone agree?

Cheers

Just
 

philmurr

Active member
vacBacker
Feedback
48 (100%)
Credits
2,374CR
It is worth checking the ROMs, unlikely for one of them to have failed mid-game, and the start-up ROM checks tend to work pretty well, but worth checking anyway

You say you've narrowed it down to the ROM board - is that by proving it with a working ROM board or some other method?

NVRAM failure on the CPU board can also cause the game to reset after the first rug pattern
 

Baiterhater

Active member
Credits
73CR
HI Phil

Yes i have a spare Rom board (blue roms) game works perfectly when installed so was able to narrow it to the rom board easily.

And yep it died mid game. I turn on now and get the start up sound with a constant rug, I'm just looking for a decent but not to expensive programmer to verify the rom images.

j
 

philmurr

Active member
vacBacker
Feedback
48 (100%)
Credits
2,374CR
Baiterhater said:
HI Phil?

Yes i have a spare Rom board (blue roms) game works perfectly when installed so was able to narrow it to the rom board easily.

Ah that's good to know, so ignore my comment about NVRAM on the CPU board!

If you want to send the board up to me I'm happy to check the ROMs for you
 

Baiterhater

Active member
Credits
73CR
Thank you a very kind offer but i have just found the problem it was IC 16 a 7432.

Id like to say i traced the fault in a real technical manner but i just piggy backed a new one over the top and bingo it came up.

Thanks to all who contributed anyone got any recommendations for an easy to use programmer that can burn 2716's and 2532's

thanks

j
 
Top