exxos's Atari, Amiga & retro forum

ah pin 53 is gnd and should be highlighted on the pinout. I see its linked on the wires itself but not on the actual diagram , my bad :( will update it shortly..

So should be 4 pins "bottom right" connected.. not just 3.. So likely this was what was the problem was.. This is why I shouldn't write documents :lol: :roll:

exxos wrote: 12 Nov 2022 15:50 ah pin 53 is gnd and should be highlighted on the pinout. I see its linked on the wires itself but not on the actual diagram , my bad :( will update it shortly..

So I spent a LOT of time looking at that diagram assuming perhaps UDS or LDS was supposed to pull things low, because I was 99% sure I hadn't missed anything about GND :lol:

Janky socket back in place:


The outcome: it works!

I get a nice clean 0110 0000 0010 1110, i.e., 0x602e.

So, that's good at least. But ... unclear what now!

Pasting from an earlier post the other tests, for completeness:

• 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 think you need to properly do this fix viewtopic.php?p=90943#p90943

I mean really you should be boosting up. Probably try diagnostic cart over serial and see what you get.

exxos wrote: 13 Nov 2022 12:59 I think you need to properly do this fix viewtopic.php?p=90943#p90943

Okay. Since yesterday, I've had a couple hundred ohms wedged into the pins on the shifter, similar to this for now. I'll get the part to do it more permanently.

I mean really you should be boosting up. Probably try diagnostic cart over serial and see what you get.

That's kind of reassuring :) Mainly I'm still deeply dubious of my old soldering skills; I think it'll be a miracle if I got that SIMM socket right.

I'll look in the parts bin and/or go scavenging to get talking over serial. That'll take some time; will park this until I can make more progress!

200R might be to low.. Its not been tested. So you might be introducing a fault by not using the proper value.

Its possible you have a RAM fault. TOS is trying to boot so should be enough to get you into the diagnostic carts menu.

Okay, understood on the resistor. (I tried a few different values at the time. Now have 470 jammed in there.)

I soldered on a serial port just to see if it'd trigger something different, but I got nothing out. Serial settings work against other systems, so the cable and settings on the macbook I'm connected to aren't in question.

But now I've got a cheap scope, so I can go fishing elsewhere to get a better sense of what's going on.

I walked around the board to convince myself that my cartridge port and my RAM have good connections.

Cartridge: I tracked continuity from the pins on a cartridge to where they should go on the main board, and I spotted a couple of bad solder joints: now should be fixed.

SIMM: I put the scope on all pins of the SIMM (i.e., the exposed connections on the SIMM itself; no second-guessing). In all:

• ground and vcc all look good. The pins that aren't used for anything don't have any voltage on them.
• data lines: all sit at +5v
• MADDR lines: are all pretty active, transitioning high/low apart from pin 19 (MADDR10) which rests at +5v. But I can't tell from the schematics if MADDR10 is supposed to go anywhere; doesn't look like it goes to the MMU?

I haven't found any bridged pins on the SIMM or the cartridge. So that's all well and good, as far as I can see.

But I want to come back to the halt state. The system still doesn't get far in before stopping and I've been suspicious of the halt state since my first post in this thread, I just wasn't sure whether HALT should normally be high or low. Then I rediscovered this post.

I get the same behaviour: normally, HALT appears to sit low when reset is released, unless I remove the DMA in which case it sits high with reset released.

With the DMA in place, using the scope I can now also see that the halt line doesn't stay low, it bounces high then low when I hit reset:


Which corresponds to a brief moment where things seem to fly on some other lines, before stopping. So I guess mine's like Alienkid's, and I ought to go eyeballing the connections on the MMU, etc, sockets? I fear my eyes won't be good enough or well-trained enough to spot bad joints, but I can sure go fishing. I'd appreciate any educated guesses on the best things to look at, of course!

Have you done the clock checks in my troubleshooting thread ? I assume you have taken all chips out and checked for bent pins ?

exxos wrote: 27 Nov 2022 08:45 Have you done the clock checks in my troubleshooting thread ? I assume you have taken all chips out and checked for bent pins ?

Yeah, clock signals all look great.

I've inspected chips, and swapped in spares for at least the main ones (68hc000, glue, mmu, shifter, dma), and equally tested they all work in an STF board. I don't see anything wrong there, but I haven't futzed with the little legs inside the sockets themselves. Maybe that's something to do.

I only have one TOS chip to test, the one from your store that you can see in the first post, and at least we know that reads out the first bytes correctly.

I'm at a loss to be honest. You have correct bytes on the bus and its crashing as halt is going low. Does berr also go low ? The cartridge gets read very early in ROM. And doesn't seem your even getting that far.

I can only suggest you move your test wires and check address 2. So connect A2 to 5v and see what you get on the data bus. Try a few more addresses as well. They are listed in the diagnostic thread https://exxosforum.co.uk/forum/viewtopi ... 084#p79084

Does DTACK go low on address zero ?