After 10K

[Monotamp3]

Hi all, I require your expertise because my STF is killing me. I have an MB C070789-c1, and I installed the dual TOS with a decoder. It was working almost fine, so I decided to apply all the fixes suggested on the EXXOS forum. Everything was fine until the last one to solve the 10K nightmare.
Unfortunately, other than replacing all the SIL, I also removed the TOS sockets to lower the Dual Tos board. I wish I were clever enough to do one mod, test and then do the second, but I didn't. Now I find myself with a white screen when I turn the STF on. I checked the following:

On PSU exit, I have 5V and 12V
On the FDD power cable, I have 5V and 12V
On the video connector, I have: on pins 4 and 13 = 5v fluctuating between 4V and 5V.
on pins 9 and 13 = 15,79 Khz.
on pins 12 e 13 = 60hz that is weird because the TOS is UK.
No soldering seems cold, and on the bad-looking one, I checked the connection between the pin and the track.
I also used the EXXOS test cartridge, but still no video. And using a serial cable, on my terminal, only a weird character appears (see image).
I have attached a few images of the MB, the Test cartridge configuration and the terminal output.
I am not proud of my soldering skills, but my health conditions don't let me have very steady hands.
Only your skills can help me to try to solve the mystery.

20260926_142513.jpg (199.52 KiB) Viewed 187 times

20260926_132922.jpg (161.55 KiB) Viewed 187 times

20260926_132401.jpg (472.41 KiB) Viewed 187 times

20260926_132337.jpg (488.05 KiB) Viewed 187 times

Thank you all in advance for the help.

[rubber_jonnie]

A white screen generally indicates your ROMs aren't working.

Since you have a TOS 1.04-2.06 decoder be sure that you are not trying to use TOS 1.04 with the TOS 2.06 decode enabled as that won't work.

If it still isn't working than I'd suggest you pull out the just the DualTOS board and try with your original ROMs.

It could be that the 2.06 decoder isn't installed correctly and has bad contacts so you may need to consider that too.

[Monotamp3]

Hi, thanks for the tips. I'll give it a try.
Quick question: does the diagnostic cartridge replace the TOS when you insert it, or does it work with it?

[Monotamp3]

Hi, I swapped the dual TOS with the old original TOS, but it acts the same.
Out of curiosity, I tested the ripple of the PSU, and it is 25mV; I suppose that's more than acceptable.

[exxos]

It sounds to me the problem started when you removed the ROM sockets and soldered the ROM board in directly ? Most probably you have a broken track or via. So you would need to continuity check between the ROM pins and the CPU..

Also note when you are testing things on your scope, of course need to be testing on the chip in itself, also be aware that pressing down on the pin could actually make contact and give the appearance it is working.. When you remove the probes the temporary pressure you put on the pin is now gone and you can end up with a open circuit.. The need to be careful when taking measurements and continuity checking..

Also note Atari motherboards do not like manhandling at all and you could have completely other unrelated bad joints which have appeared on the CPU or other chips on the board while doing all the work.

Normally a good diagnostic step is to press down on various chips on various parts of the motherboard while continuing to press the reset button to see the springs into life. You can also hold the motherboard in the centre with your thumb, and gently bend upwards each corner of the board one at a time also are pressing reset to see if that makes anything spring into life. Generally doing those techniques you can probably work out a rough area the bad connection is assuming you have one which is intermittent..

[Monotamp3]

Thank you.
I will proceed with the tests suggested by you.

[Monotamp3]

Hi, looking at some schematics, I can understand, but I am not sure because I am not good at it, that the pins of the TOS chips U63 and U67 go to the CPU. And I found continuity between all of them (CPU AND TOS CHIPS), except for U67 pin 2, which has no connection to pin 2 of U63 and no connection to pin 41 of the CPU (A13 signal). It is a fault or it is me having misunderstood ?
Thanks

[exxos]

It will be a lot easier and clearer if you could post a image of the schematic with the tracks in question..

Thing is, a lot of motherboard and schematics are floating around the Internet and IC numbers can vary. We don't know what schematic you are following or even if the numbering matches your motherboard at this point. So it is always better just to post a snippet of the schematic you are asking questions of. Then you could even just paste arrows onto the things you are wanting to confirm and for other people to check.

[Monotamp3]

Hi, sure. In the pdf attached on page 4, you will see the schema of the 6 roms for the TOS. My configuration is with 2 chips Hi-0 and Lo-0 as it was before the mods to the dual TOS and it used to works.
My interpretation is :

1. Sheet 4 of the schematic
Next to each ROM, on the left, there are two columns of numbers. The outer number should be the bus address line (A1…A15), and the inner number should be the ROM pin. Inside the chip outline should be the ROM's own internal address name (A14, 13, 12… A0).

Each bus number is written on the slanted bus-entry stroke that leads into the row above it. So I didn't pair it with the pin number written on the same row. If I am right, this will read this way and get:

ROM pin ROM internal address Bus line
27 A14 A15
26 A13 A14
2 A12 A13
23 A11 A12
… … …
10 A0 A1

The 68000 has no A0 pin, because it reads 16-bit words, and the two ROMs (Hi and Lo) each supply half of the word. So ROM address A0 must go to CPU line A1, and every other line shifts up by one. If you paired the numbers on the same row instead, ROM A0 would connect to A2, and bus line A1 would have no pin at all, which makes no sense.

In the standard pinout for 28-pin ROMs/EPROMs (27256 type), pin 2 is A12, which matches what is written inside the chip outline on the schematic.
Considering the 68000 pinout that is the only pin out that I can be sure of, A1 is on pin 29, and the address lines continue in order: A2 = 30 … A12 = 40, A13 = 41, A14 = 42, up to A20 = 48.
Base on that thinking I made a table and I checked the continuity of all the pin of HI-0 and Lo-0 to the 68000 following that table and all beeped correctly exept for LO-0 pin 2 to cpu pin 41.

Here is the table tha I made:
PinROM (both) CPU
1 NC
28 +5V
14 GND
20 ?
22 ?
10 29
9 30
8 31
7 32
6 33
5 34
4 35
3 36
25 37
24 38
21 39
23 40
2 41
26 42
27 43
11Hi 61
11Lo 5
12Hi 60
12Lo 4
13Hi 59
13Lo 3
15Hi 58
15Lo 2
16Hi 57
16Lo 1
17Hi 56
17Lo 64
18Hi 55
18Lo 63
19Hi 54
19Lo 62

1040ST_Schematic_RevC.pdf

(1.71 MiB) Downloaded 4 times

!Message from: Moderator

@Monotamp3, we have placed your attachments in-line so they display correctly in your post. Here's the how-to guide.

[exxos]

Yes, I think you have found it. If U67 pin 2 has no continuity to U63 pin 2 or to CPU pin 41, then the low ROM has no A13. Half the ROM space will read garbage on the odd bytes, so TOS never gets off the ground. That's your white screen.

It also explains the 60Hz. The machine powers up at 60Hz and it's TOS which switches it to 50Hz, so if TOS isn't running it just sits at 60Hz. Same reason swapping back to the original ROMs made no difference: the ROMs are fine, the board isn't.

Your table looks correct to me. The address lines being shifted up by one is right, as the 68000 has no A0.

The fix is easy enough. Just run a short wire from U67 pin 2 to U63 pin 2 (or to CPU pin 41, whichever is easier to get at). Then recheck continuity from the legs of the chips themselves rather than the pads, so you know it's actually connected and not just making contact under the probe.

While you are there, I would also check pins 1, 20 and 22 on both ROMs. In the two ROM setup those are not unused pins. They go through the ROM config jumpers on the board, and as they were right next to the sockets you removed, they are just as likely to have taken a knock. Check them against sheet 4 of the schematic.

As for the diagnostic cartridge, it doesn't replace TOS. The CPU still has to read the start-up code from the ROMs first, and then TOS hands over to the cartridge. So with a broken ROM bus the cartridge won't run either, which is probably why you only got junk on the serial terminal.

Fingers crossed it's just the one track. Atari boards do like to keep a few surprises in reserve, like a bonus level nobody asked for.