Amiga 1200 DRAM Signal Issues

[stormahead]

Hey guys,
Doing a Reamiga 1.5 board and having issues with DRAM.
Checked soldering, lines for shorts, bridges etc
Basically I've found that the entire DRD 1-15 across the top of Budgie has a ocassional pulse signal - like its waiting (pic.2)- same as DRD 16-31 before Diagram seems to test this area of memory. DRD 1-15 just never seems to become engaged like DRD 16-31 does with proper high low activity (pic1)

What would cause this? Some kind of address bus issue, bank select or something?

[stormahead]

I'm fairly certain now its an address line issue, perhaps Alice or Gayle.
A12 on Alice just looks weird compared to its neighbours. Checking the A20 line to see if its actually selecting the upper or lower bank of memory
Just hoping its not a dodgy Alice, bloody expensive chip

[pandy71]

IMHO it could be better to use logic analyzer than scope, activity on one of lines doesn't need to be suspicious especially if you use digital oscilloscope that perform probably undersampling...

[stormahead]

Any recommendations for a logic analyzer?
I've managed to burn diagrom 1.3 and run the better mem test. It confirms its not the address lines, but yes issues on DRD 0-15 regardless of which ram chip gets installed. Fastmem with accelator works perfectly.

[Lunda]

Originally Posted by stormahead

Any recommendations for a logic analyzer?
I've managed to burn diagrom 1.3 and run the better mem test. It confirms its not the address lines, but yes issues on DRD 0-15 regardless of which ram chip gets installed. Fastmem with accelator works perfectly.

Did you check the CAS signals?
Check if they are decoded correctly for byte, word, long word.

[stormahead]

I did and found a problem compared to a working 1200. The RAS/CAS on Alice seemed weird and I compared them. Image 1 from the non working, image 2 from working.
Thinking it must be a bad alice chip...

[giantclam]

...unless Budgie is pulling the signal down..

[stormahead]

Yes, possible Budgie pulling down on RAS/CAS inputs from Alice.
Lift pins on RAS/CAS input on Budgie and recheck?

Images of RAS/CAS input into Budgie.

[stormahead]

Just to go back and provide update for others..
It was a bad Paula connection causing the DRD signal issues which makes sense when DRD0:15 goes to Paula. Even though I had continuity and solder looked fine even under a micro- a full resolder did the job.

Next mission is to find why only 1024kb is being detected by Diagrom - $0000400 to $00FFFF only, nothing found higher. Address and data lines look ok on the scope.

[stormahead]

Resolved with a resolder of Budgie making painfully sure each pin was square on the pad and then a quick ultrasonic bath.
I did replace the Alice and Budgie in the end so its hard to pinpoint which was causing the addressing issues with whole chunks of mem missing.