33 bottles of beer on the wall - or 33 bombs on the screen

[jwd]

My original ST needs some love, i've put it off for years and have many other ST's/Falcon (which all work :) ) but I want this to be alive again, as well as learning something along the way in case my others start dying.

In a nutshell, 520STFM UK TOS 1.4, power on, 33 bombs. Major +point all the I.C.s are socketed, PSU re-capped and only 2 larger caps on board motherboard replaced (other caps another todo).

Diag cartridge works, following tests work fine :
Audio
RAM
ROM
Keyboard
(don't have harnesses/port pluggies so can only test with cart only)
So by the above passes is it safe to assume CPU, MMU, TOS, RAM etc is all hunky dory here?

Failure :
Timing - T6 1772 Bus Test - Diag Manual says faulty WD1772 replace, pretty obvious eh.
DMA - EB Bus Errors at FF8606 (DMA Mode/Status) + D0 DMA time-out - Not sure if this requires test harness plugged into DMA port?
Floppy EB Bus Errors at FF8606 (DMA Mode/Status) + E9 Bad Instruction Fetch (bus error)


My other ST's all have WD1772 soldered directly to board (reluctant to pull off at this stage) but have one donor board which i got the WD1772 off (Duratool de-solding station a saviour). Tried in broken ST still same results, thing is I can't be certain the donor board WD1772 works either (has many parts missing), also tried without WD1772 in place too, same results.

So my question (before i go buy WD1772 from Exxos store) is there anything else i can check/look at to confirm it is the WD1772 or another related IC?

I have an antique Oscilloscope so can do some probing :wink:

thanks in advance.

[rubber_jonnie]

IIRC the DMA bus tests do require the test harness, so that is likely a red herring.

In my experience the floppy usually fails for certain specific things:

1. Faulty DMA
2. Faulty 1772
3. Faulty cable
4. Faulty FDD

However 33 bombs means 'Trap Instruction Vectors' so not necessarily FDD related. Also, a TOS1.04 STFM should boot without a floppy, so have you tried booting with it disconnected?

Also, have you tried reseating the chips and cleaning the pins? Also, if you boot with the diag cart, but connect a PC via the serial port, do you see any errors? Are there any hot chips?

Just a few thoughts.

[jwd]

Thanks for the suggestions, additional steps tried so far :

Same = T6 1772 Bus test fail on startup with diag cartridge (same as T, timing test) or without diag cartridge - 33 bombs on startup.

With/without floppy connected - same
Reseat all chips and cleaned - same
Changed both WD1772 and DMA with another donor board (not 100% donor chips work so could be red herring here) - same
Different (new) PSU - same
Additional colour test on diag cartridge (found a colour monitor to use to check shifter) - passed.

Have done some probing with the oscilloscope with diag cartridge plugged in so ST not in bus error state :

Clock good to all chips
Reset circuit good - HALT + RESET working ok
All data lines pulsing away.

Observations :

DMA chip has all data lines + CLK pulsing away but nothing on the output side to the FDC (both DMA's tried same result) CD0 - CD7, should i see some observations while the ST is running (albeit in diag mode) ?

I'm leaning to get a new DMA and FDC but before i take the plunge any other checks/probing to suggest?

[PhilC]

Hi @jwd with your multimeter check that all pcb tracks to do with the connections between DMA and wd1772 are ok and then from wd1772 to the 74ls06 and then to the floppy port.

[jwd]

Thanks for the tip @PhilC all continuity between DMA and 1772 are good.

between 1772 and the corresponding 74LS06 are good only caveat is to slightly different pin numbers of the 74. The only decent schematics I could find look to be an older ST my motherboard revision is fairly new - C101314-001 Rev 1.1 :

2023-05-08_19h02_48.png

Continuity from all U9 links to U8 are good just to different pins on the U8 I presume because of board revision.

Having looked at the pictures of the C101314-001 on this site and over the internet the only difference i can see (as well as nearly ALL my chips being socketed??) is a diode on VCC pin of DMA :

2023-05-10_11h30_44.png

I presume this is ok?

Anyways in the meantime I have bought a load of replacement chips from @exxos just in case.

Any other suggestions/probing I can do/check while waiting on the parts?

[exxos]

Diode is normal for IMP DMA.

[PhilC]

@jwd change the 74LS06 as well if you can just to rule it out as they are rather cheap to buy.

[jwd]

Got my bag of tested chips from the store (nice and speedy so many thanks @exxos) and proceeded to replace the following :

DMA (you sent me a non IMP type so not used diode to VCC as mentioned @exxos)
WD1772
CPU
74LS06 (downstream from WD1772) as suggested @PhilC
TOS ROMS

Again re-seated all chips and sprayed with contact cleaner.

Unfortunately still same 33 bombs on startup or T6 1772 Bus Test fail when using diag cart (as mentioned previously audio/keyboard/RAM+ROM,Colour/HiRes tests have always passed fine so half the ST is surely working? ).

:(

Even though i'm still at the same place i have learned a hell of a lot about the board and used the oscilloscope to trace round the signals to see what's happening so still worth the effort.

Anyone have any more suggestions or things I could try now?

Thanks again.

[PhilC]

@jwd I've had some odd faults happen when a GLUE has decided to die, so might be worth swapping that and giving the socket it's in a good test and clean as well.

[exxos]

Can you post a image of your motherboard overall..