Looks like I'll be getting a Mega ST4 soon

[rubber_jonnie]

@exxos Thanks for all that, I will certainly give it a look, see what I can see, will hopefully get on it sometime today.

[rubber_jonnie]

A bit more troubleshooting done today, having a 2CH Scope is helpful for sure!

The reset circuit appears to be fine and works as described, and I can't find any bus shorts at all.

I started doing the first power up voltage checks as described for the H4, but it looks like some of the pins don't have 5v on them, so I will run the test on my working Mega to see how it compares.

I also started doing the clock signals, and that all looks a bit odd as the clock at the CPU was reading 10KHz, not Mhz, but I am still getting the hang of my scope, so it could be interpretation, so I will compare against known good again.

The clocks do look right in that 16Mhz and 32Mhz on the Shifter look to be 2x and 4x the frequency of 8Mhz on the CPU, FDC and DMA. That needs more work.

Interestingly, when I touch the 32mhz pin on the Shifter with my x1 probe, the white display goes blank, and when I remove the probe, it goes back to a white screen.

Hopefully I wll get the boot sequence/first bus cycle looked at/scoped tomorrow and maybe get my LA up and running, see what I can see.

Even if it never runs again, I guess it will be useful from the point of view of getting some experience with the tools and understanding the ST a bit better.

[rubber_jonnie]

Today's round of troubleshooting has been interesting.

As far as clock signals go,they are all exactly where they should be based on the H4 troubleshooting topic, plus I have cross checked with my working Mega and all the clock signals match there. I then moved on to what should be happening at boot time, and also on reset, and /AS never goes low. Ever. In addition, the /CE lines on the ROMs go high and stay there too.

I cross checked on my working machine and it looks like this:

IMG_20191231_154255.jpg

The yellow trace is the /AS pin on the 68K, the green trace is the /AS pin on the GLUE. I didn't bother to take a pic of the problem machine, because it was just two flat lines at 5v until power was turned off. Reset mad zero difference, they just stayed high at 5v.

So it seems that the CPU is held at some sort of halted state. /AS,/LDS,/UDS,/DTACK all stay high and don't go low on reset, and HALT is staying low permanently.

:sigh: :sigh: :sigh:

I suppose it's progress now I know this, but it's still a long way off being fixed :(

@exxos Any suggestions would be welcome, thanks.

[exxos]

do you have a pull-up resistance on the HALT pin ? Does the voltage ever change on HALT when pressing the reset button ? Is HALT shorted to 0v ?

[rubber_jonnie]

do you have a pull-up resistance on the HALT pin ? Does the voltage ever change on HALT when pressing the reset button ? Is HALT shorted to 0v ?

@exxos I'd need to check for a pull up, but the voltage never changes on HALT when reset is pressed. I'll check for a short to 0v too.

Will report back with results.

[PhilC]

Hi RJ ,

Have you verified that the reset circuit is working?

The reason I ask is that if you have the scope set up right you should see a very quick pulse high on reset and halt if the circuit is working. Might be worth checking with your LA on the two of them.

The reset circuit goes from the switch to the 556 and then through the inverter. Check that you get the reset signal at all those points but if it then stays low at the output of the inverter then you've got something else holding it off.

Probably something not sending a ready signal to the glue.

[rubber_jonnie]

@PhilC

I've been through the H4 guide regarding the reset circuit, and when I hit reset I see the reset pin go low as described. It goes low for about 1.5 seconds, then high again, as its supposed to.

I see no change on HALT at all.

I scoped both out at power on and reset, HALT goes low immediately.

When I had the machine running, reset was working OK, but it does seem like maybe it's somehow held in HALT.

The reset switch was replaced with a brand new one from exxos as it was not well.

[PhilC]

The pullups for halt and reset are the one just above the mega bus, so it might be worth checking continuity between there and the rest circuit.

It may be that your scope isn't quick enough to pick up the blip on halt after each reset, its 40mhz right? You could set to say 200ms, that way youd see the screen update slower or set to to trigger high with your probe on /halt?

Just a thought, have you got the 2ish mhz clock on the MFP? That will put it in halt if it's missing.

And if I remember correctly you said youd tested all the chips in an STFM?

What about Ram? Are ras and cas on both banks pulsing and you have good continuity on the mad lines? The ram used in a couple of my Megas had failed and also caused it to halt with just a white screen, not the 7sual white and black.

[rubber_jonnie]

Those pullups by the Mega bus have been replaced, but I will recheck continuity, the vias on this board are really not good, the board is terrible quality.

As for the scope speed, I checked on my working machine, and I do see blips on halt after each reset, so it's not a problem with the scope speed. They are definite blink and you'd miss them events if for some reason you don't get the time/div right.

The clock signal is there on the MFP, all the clock signals were checked as below and all were present and matched my good Mega:

Shifter should have 32MHz into pin 2 and output 16MHz on pin 39.

MMU should have 16MHz input on 5 and output 8MHz on pin 20, and 4MHz on pin 19.

GLUE should have 8MHz on pin 34. 2MHz on pin 54 and 0.5MHz (500KHz) on pin 43.

MFP 4MHz pin 35.

WD1772 8MHz pin 18.

BAD DMA 8MHz pin 39.

ACIA 500KHz pin 3&4.

CPU PLCC 8MHz pin 15, 68K sockets pin 15.

As for the RAM, I will check it, but based on what exxos suggested here I don't think it's even getting that far:

"If you never see /AS go low after reset then the CPU isn't running, like its held in HALT or RESET mode. If /AS does go low, the CPU is running, so the CPU will put all 0's on the address bus. if it doesn't, then GLUE won't decode and GLUE wont set ROM CE low."

I never see /AS go low, and I never see ROM CE go low. Reset correctly goes low at the CPU for about 1.5 seconds before it goes high again. HALT never changes, it just stays low all the time, and despite changing the time/div on the scope, there is no sign of even the smallest pulse.

Also, the RAM was working fine and had been run through multiple YAART tests when I had the machine running. I'm thinking there has to be a continuity problem between the CPU and the GLUE, so I will go back and recheck all the connections between them.

The only thing that changed with the board was I removed the CPU and fitted a socket. I've since removed the socket and fitted the CPU direct again, but no change, so if I eliminate the signals between the CPU and the GLUE next it will at least give me something to work on after that.

I will try and get the LA involved as well, I'll test it first on my working machine, I've just never used one before so I need to do some figuring out first.

Happy New Year by the way :) @exxos & @PhilC

[exxos]

I'd say lift the halt pin up on the CPU and wire link it to reset pin on the cpu and see what happens.

:thanksyellow: