exxos's Atari, Amiga & retro forum

This has sat for 12+ months waiting for me to find a moment. My plan of attack is the minimal number of ports, etc, to get this working to desktop.

Day 1

About three weeks ago I got started on some basics.

Monitor port:


Oscillator:


The SIMM socket. I don't consider myself a great solderer, so this filled me with the fear:


Once I got into it, it felt good. I don't think I bridged anything I wasn't supposed to.

Day 2 (three weeks later)

Time to get closer to running!

TOS jumper:


Reset switch:


Blob and keyboard pin 2 removal:


Populated SIMM + jumpers for RAM selection:


No flashy clock, so link pins:


Blob in the absence of RTC:


End of day 2 status

Board status with jumpers:


I feel like the above should be enough to boot to a desktop, so I wanted to move forwards. Time to plug in a keyboard and flip the switch...

... and the power LED lights up
... and my monitor senses activity and wakes up
... and ... I get a fully white screen. This is already better than I might have expected :D

If I bridge the TOS selector then I get a white border and a black screen. I don't have a TOS 2.06 decoder, and I'm not sure whether I should bridge the TOS selector or not!

I expect that some of my soldering on the SIMM socket is not good; it's a lot of pins, I'm sure I didn't get it fully right first time. So I figure I've got some checking to do on my soldering. Every continuity check I've done so far seems good, but it's not exhaustive.

As far as I've read, all my jumpers and other blobs are good to go. I've checked a couple of other things so far. I checked pins 17 and 18 on the processor, following this post. What I seem to get is that pin 18 sits at 5v, and drops low when I press reset. Pin 17 however, appears to always stay low. That definitely doesn't feel right.

That's where I have to leave it for today, but rather than stab completely in the dark, I'd appreciate best next-steps for this little project.

Also, can anybody tell me which way these jumpers should be set?

Looking at your jumpers in the pictures it looks like you are missing one on the BLITTER circuit.

There are two jumpers by the BLITTER and 68K socket they should both be closed if there is no BLITTER and both open if there is one.

I also suggest doing the resistor mod on the 32mhz osc. Without it, some machines don't boot up.

32Mhz OSC fix.

Icky wrote: 18 Sep 2022 08:09 Looking at your jumpers in the pictures it looks like you are missing one on the BLITTER circuit.

There are two jumpers by the BLITTER and 68K socket they should both be closed if there is no BLITTER and both open if there is one.

Ah! I was tired when I was finishing up and I pored over the board for any jumpers I'd missed... and still I missed this one! Sadly, closing the jumper doesn't bring my board to life yet.

exxos wrote: 18 Sep 2022 11:34 I also suggest doing the resistor mod on the 32mhz osc. Without it, some machines don't boot up.

32Mhz OSC fix.

Nice; interesting threads. I don't have any of those tiny resistors to hand, nor do I probably have the soldering skills. But for a rudimentary look at what's happening on this board:

Pin 2 on the shifter measures about 2.5V on my multimeter
Pin 39 on the shifter measures about 1.8V. Not sure if those values feel right or not.

If I quickly chain a couple of axial resistors to either ~430 Ohms or ~550 Ohms and wedge them across pins 2 and 20 much like in this post I see about 2.2V or 2.1V on pin 2 and about 1.78V on pin 39.

No changes, still a white screen with the TOS jumper open; white border with black screen with TOS jumper closed. (No 2.06 decoder.)

My gut says my soldering somewhere is bad, so I need to pore over those and check. Everything I've spot-checked so far looks okay, but I definitely haven't checked everything.

In my experience the TOS EPROM chip holder can be very flaky and with my H5 I had to pull the chip out, make sure all the legs are aligned (this might sound crazy, but sometimes the chip legs can become very tarnished and dark grey, in that instance I've actually carefully run a bit of sandpaper over the legs to clean them up) but yes, usually a white screen is a TOS issue.

Steve wrote: 19 Sep 2022 10:09 In my experience the TOS EPROM chip holder can be very flaky and with my H5 I had to pull the chip out, make sure all the legs are aligned (this might sound crazy, but sometimes the chip legs can become very tarnished and dark grey, in that instance I've actually carefully run a bit of sandpaper over the legs to clean them up) but yes, usually a white screen is a TOS issue.

Finally was able to grab a moment on this!

I have a cheap scope and it reads 16MHz on pin 39 of the shifter, so I figure the osc is good.

The chips that I can put into an old board definitely work (shifter, glue, MMU, floppy, DMA)!

I've pulled a cartridge port from a good board and dropped it into the H5 (I had wondered if the diagnostic cart would forego TOS, but perhaps not).

I've checked continuity between all SIMM pins and where they should go on the board: I don't believe I have any missed connections here.

I've pulled the ROM, jiggled the legs on the socket and the chip, and ran some de-oxit over the pins.

Still: white screen on boot. The hunt continues.

Have you worked though these steps ? https://exxosforum.co.uk/forum/viewtopi ... =78&t=2362

exxos wrote: 06 Nov 2022 04:00 Have you worked though these steps ? https://exxosforum.co.uk/forum/viewtopi ... =78&t=2362

Oh this is great. Today I got a solid run at this and it's been FASCINATING. And now I have a cheap oscilloscope, so I can check clock rates! I've learned way more about ST clock signals than I expected today.

Running through the tests:

• all chips get 5v
• reset signal: works as expected
• pull-ups: all pins (excepting GND, VCC, E, CLK) read 1K or 2.2K resistance EXCEPT pin 12 (which reads 1.5K; I figure this is okay)
• bus shorts: all pairs of pins look good EXCEPT between 20+21 (E and VPA?), which read 0 ohms.
• LS11 shorts: not applicable to H5?
• voltage test on the 68000 expansion ports: this is perhaps the real clue! I have voltages far below 5V on pins 2, 3, 4 (D1-3) and 23 (IPL2?).
• clock checks: all of my clock rates look absolutely spot-on.

I'm out of time for today though. D1-D3 are deeply suspicious to me: what to poke for those?

You need to do the first bus cycle test to diagnose. It's possible ROM or DMA is causing the low voltages so it might not actually be a fault.

exxos wrote: 07 Nov 2022 09:51 You need to do the first bus cycle test to diagnose. It's possible ROM or DMA is causing the low voltages so it might not actually be a fault.

Taking a look at this today: my cheap scope only has one probe (but I'll probably order a second anyway!), so I can't trigger on reset to measure each pin meaningfully.

So I'm trying this the alternative way: with the CPU removed, and with pins and wires connecting the address lines, AS, UDS, and LDS. It's VERY possible I've not landed this correctly, it's very janky.

At the moment all I get is 5V on all data lines, not the bit pattern from the other thread. Is that meaningful in some way, or does it nonsense?