DRAM Coreboard repair. Advice sought

[CheshireNoir]

Hi All,
I'm in the late stages of restoring a MB8342-3 DRAM Coreboard. It was a Starnet board and I'm converting it to be a floppy board. After sourcing all the parts, carefully soldering everything in, using sockets for all the new ICs, burning a new ROM (BN55 from the repository) and generally getting everything together, I turned it on today to get no picture. Not even synch.
What I did BEFORE I turned it on:
Made sure the baseboard works - Yep. I use it with my PC85b board without any problems.
Made sure there were no shorts between +5v and GND - Nope. Resistance reads at a healthy 2.3KOhm
Sanity checked polarity on all polarised components - All look good.
What I did AFTER I turned it on:
Made sure I was seeing 5V at the chips - Yes I am.
Soldered on a test point to the GND rail in unused IC 4 spot - Ground is good.

So that's the basics dealt with :-)

So, no picture. Time to break out my shiny new Oscilloscope. I'm even confident I'm using it right now! Good thing I have a nice accessible ground point.
Starting with IC1, the ROM, I probed around. The only things I can see that look wrong are _CE is being held high, (It may simply be that it's not reading from the ROM?) and D0 is being held low. All other pins are showing activity that looks "sane".

So, D0 held low. Is it shorted to GND? Err... Yes. Yes it is. Even if I unplug the core board from the main board. (which suggests the short has to be on the core board)

But how? I've inspected the top and bottom surface until my eyeballs ache. (Doesn't mean I haven't missed something)

So currently my working hypothesis is that it's an internal short on one of the components. The ROM should be fine as I was able to burn the new ROM image onto it, and it passed verification. The WDC2793 can be eliminated as it's in a socket. (Short persists with chip removed).
IC29 (74LS175), IC24 and IC 25(4164 RAM) are also hanging off D0. All are soldered onto the board. I have stock of the 4164 RAM and can easily get the 175, but before I start madly desoldering things, can anyone suggest anything I have missed, or if a particular chip is "Always the problem"? (It's always the PLA on the C64, for instance)

Morale is high, so far :-)

Cheers!

Chesh

[Ernest]

Hi All,
I'm in the late stages of restoring a MB8342-3 DRAM Coreboard. It was a Starnet board and I'm converting it to be a floppy board. After sourcing all the parts, carefully soldering everything in, using sockets for all the new ICs, burning a new ROM (BN55 from the repository) and generally getting everything together, I turned it on today to get no picture. Not even synch.
What I did BEFORE I turned it on:
Made sure the baseboard works - Yep. I use it with my PC85b board without any problems.
Made sure there were no shorts between +5v and GND - Nope. Resistance reads at a healthy 2.3KOhm
Sanity checked polarity on all polarised components - All look good.
What I did AFTER I turned it on:
Made sure I was seeing 5V at the chips - Yes I am.
Soldered on a test point to the GND rail in unused IC 4 spot - Ground is good.

So that's the basics dealt with :-)

So, no picture. Time to break out my shiny new Oscilloscope. I'm even confident I'm using it right now! Good thing I have a nice accessible ground point.
Starting with IC1, the ROM, I probed around. The only things I can see that look wrong are _CE is being held high, (It may simply be that it's not reading from the ROM?) and D0 is being held low. All other pins are showing activity that looks "sane".

So, D0 held low. Is it shorted to GND? Err... Yes. Yes it is. Even if I unplug the core board from the main board. (which suggests the short has to be on the core board)

But how? I've inspected the top and bottom surface until my eyeballs ache. (Doesn't mean I haven't missed something)

So currently my working hypothesis is that it's an internal short on one of the components. The ROM should be fine as I was able to burn the new ROM image onto it, and it passed verification. The WDC2793 can be eliminated as it's in a socket. (Short persists with chip removed).
IC29 (74LS175), IC24 and IC 25(4164 RAM) are also hanging off D0. All are soldered onto the board. I have stock of the 4164 RAM and can easily get the 175, but before I start madly desoldering things, can anyone suggest anything I have missed, or if a particular chip is "Always the problem"? (It's always the PLA on the C64, for instance)

Morale is high, so far :-)

Cheers!

Chesh

Hi Chesh,
Several things to start with where a short may exist or has been introduced.
1) The D0 is also present at IC 16 74LS174.
2) Although you have unplugged IC 29 & 35 the short maybe under or within the socket and not the IC.
3) DO is also present at Pin 1 of the 3rd row (20pin row of X2) where you would have fitted the 34Pin FDD socket. This row is used when you fit a HDD (or CF card) to the System.
4) Check for a short under the header if you replaced the X1 & X5 (25 Pin) that connects the Coreboard to the Mainboard.

Did you fire the board up prior to doing the upgrade?

Hope this helps.
Ernest

[CheshireNoir]

Thanks Ernest,
You got me looking in the right direction. It was the connector X2! A silly mistake on my behalf. The board had a bodge wire to connect the X2 connector to ground with a more sizeable wire, but the wire was way too long so I'd trimmed it and reattached. Of course I'd managed to short the pin 1 on the 40 way at the same time. 

Now I have some life! I get... a cursor :-)
   
No beep. No other life, but to get a cursor, it must be doing a certain amount of setup, right?

(To me, this suggests I'm using the wrong ROM? What ROM should I be using with a MB8342-3 in FDD mode?)

Also, no, I didn't test before I started the upgrade. With hindsight I probably should have.

Chesh

[someone]

Keep the 'M' key depressed during power up or during a keyboard reset.
The Monitor Title and prompt should appear.

---

BTW - Did you configure your baseboard to accommodate a DRAM coreboard?

[CheshireNoir]

Hi Someone,
The monitor title and prompt do indeed appear.
I have NOT configured my baseboard to accommodate a DRAM corebaord. PLease point this poor ignorant one at an appropriate guide and I will leap upon it with glee :-)

Chesh
(Who clearly has a LOT to learn)

[ChickenMan]

Chesh, have a read of Section B - ECN 5-60684-A in the microbee_hardware_notebook_with_updates.pdf. That manual is the Bible for Microbee updates and mods.

[CheshireNoir]

Cheers ChickenMan,
I have not. Will go and read now :-)

Chesh

[CheshireNoir]

a LOT of work later... (Board modded, drive controller tuned in)
   
Wow. It boots! I'm not completely out of the woods yet. I think I might have some bad RAM somewhere. Why? If I choose "Word Processor" I get:
     
and if I go into Help and back out I get
   
But so much progress!

Thanks to absolutely everyone who has helped me so far. This is one of my happiest moments with my hobby in a VERY long time :-D
Special shoutout to ChickenMan who is just super awesome and to Ernest and Someone for heaps of help.
(So, to do. work out what the error is causing the bad graphics. Possibly RAM? Fix the break key which has decided to no longer work. If I have to, I may end up kludgeing in something for now)

Chesh

[ChickenMan]

Fantastic, good to see its finally booting. Your booting up the final ver of Shell that came out on the 256TC, suggest to try something a little earlier.

[CheshireNoir]

Ah cheers!
Certainly haven't had any problems with the Arcade game disks yet. :-D

Chesh