1040STFM Halted CPU

Troubles with your machine? Just want to speak about the latest improvements? This is the place!

Moderators: Mug UK, Zorro 2, Greenious, spiny, Moderator Team

TeX38
Atari User
Atari User
Posts: 34
Joined: Fri Aug 19, 2016 11:49 am

1040STFM Halted CPU

Postby TeX38 » Mon May 08, 2017 3:54 am

If the MMU or GLUE were faulty, how do I tell?

The MMU is still producing the 4 and 8mhz clocks and the GLUE is producing the Hsync and Vsync as I have a stable image on the Composite line and on the screen.

But the CPU is Halted.

I assume if a ROM chip was faulty the CPU would halt?

Lynxman
Atariator
Atariator
Posts: 23
Joined: Sat Apr 04, 2009 9:39 pm

Re: 1040STFM Halted CPU

Postby Lynxman » Mon May 08, 2017 7:07 pm

Is the HALT-Pin of CPU low?

Have you already checked the reset circuit?

TeX38
Atari User
Atari User
Posts: 34
Joined: Fri Aug 19, 2016 11:49 am

Re: 1040STFM Halted CPU

Postby TeX38 » Tue May 09, 2017 3:07 am

Lynxman wrote:Is the HALT-Pin of CPU low?

Have you already checked the reset circuit?


Hi Lynxman,

Yes, I removed the LS05 to see if it may be the reset circuit at fault, the Halt remains low and the Reset circuit is working per expected.

The halt is Bi-Directional so the halt is being generated by the CPU.

There is no Data or Address activity.

I changed the RAM thinking that was the problem, my 1040 only has 512 for now but still no go.

When this thing originally died , I recall seeing a long line of bombs on the screen, upon reset the machine wouldn't restart.
Sadly I didn't think to count the bombs at the time.

Kubik
Obsessive compulsive Atari behavior
Obsessive compulsive Atari behavior
Posts: 137
Joined: Thu Jan 28, 2016 8:10 pm

Re: 1040STFM Halted CPU

Postby Kubik » Tue May 09, 2017 3:44 am

Could it be a ROM fault? Loose socket, perhaps?

TeX38
Atari User
Atari User
Posts: 34
Joined: Fri Aug 19, 2016 11:49 am

Re: 1040STFM Halted CPU

Postby TeX38 » Tue May 09, 2017 7:55 am

Kubik wrote:Could it be a ROM fault? Loose socket, perhaps?


Maybe, I am going to build a circuit to read each ROM using an Arduino to verify the chips.

I have an ST board here with French Roms, I plugged the English Roms into it but it wouldn't boot.

I also tried the French Roms in the 1040 but that failed as well....

Lynxman
Atariator
Atariator
Posts: 23
Joined: Sat Apr 04, 2009 9:39 pm

Re: 1040STFM Halted CPU

Postby Lynxman » Tue May 09, 2017 8:44 am

TeX38 wrote:Yes, I removed the LS05 to see if it may be the reset circuit at fault, the Halt remains low and the Reset circuit is working per expected.


OK

All Data and Address lines OK?
If CPU make Halt because of any trouble with the ROMs, maybe a connection is broken.

Did you made a cross check of Glue and MMU with known good ST?

Kubik
Obsessive compulsive Atari behavior
Obsessive compulsive Atari behavior
Posts: 137
Joined: Thu Jan 28, 2016 8:10 pm

Re: 1040STFM Halted CPU

Postby Kubik » Tue May 09, 2017 9:25 am

If you've got a logic probe, you could also check the behaviour of the address and data lines. I remember one case where I was able to find the hang cause with just observing the A/D signals :) Actually, that one was fairly simple - I have detected 4 ROM reads around address 0, which lead me to the conclusion it's a ROM fault.
With a good logic probe (preferrably the Superprobe I am promoting at every single occasion here :) ), you could check the ROMCS signals, count the number of reads before halt etc. That would give us better idea how far this actually gets.
I'd start with the ROMCS signals on TOS ROMs. If that's ticking, the next step would be ROMCS on the cartridge port, then RAM RAS and CAS signals.
BTW: did you check BERR too?

TeX38
Atari User
Atari User
Posts: 34
Joined: Fri Aug 19, 2016 11:49 am

Re: 1040STFM Halted CPU

Postby TeX38 » Tue May 09, 2017 11:14 pm

Did you made a cross check of Glue and MMU with known good ST?[/quote]

The MMU and GLUE are soldered to the board on my 1040, thinking to de-solder and replace with sockets.

The MMU and GLUE are socketed on my 512.

TeX38
Atari User
Atari User
Posts: 34
Joined: Fri Aug 19, 2016 11:49 am

Re: 1040STFM Halted CPU

Postby TeX38 » Tue May 16, 2017 8:38 am

Love my Oscilloscope..

I have found D3 D6 and D9 to be faulty on my 1040 Making me suspect a buffer at fault.

Next to change the 74LS244 and 74LS373 suspects.

User avatar
Greenious
Hardware Guru
Hardware Guru
Posts: 1003
Joined: Sat Apr 24, 2004 5:39 pm
Location: Sweden

Re: 1040STFM Halted CPU

Postby Greenious » Tue May 16, 2017 9:22 am

I assume you have the service manual?

http://dev-docs.atariforge.org/files/Me ... n-1991.pdf (actually contains all relevant service manuals)

Before going after GLUE/MMU, since they are soldered in, check their clocks, and it's probably easier to check MFP and DMA first. As per the service manual page 118 (1040STFM "troubleshooting a dead unit").

You've checked the reset circuit. And supposing the CPU is good, it halt's due to a double buserror (buserror within a buserror), which does suggest bad ram or rom.
Updated my guides as of june 28th, 2016. Check'em out and feedback!
viewtopic.php?t=5040

TeX38
Atari User
Atari User
Posts: 34
Joined: Fri Aug 19, 2016 11:49 am

Re: 1040STFM Halted CPU

Postby TeX38 » Wed May 17, 2017 7:15 am

Greenious wrote:I assume you have the service manual?

http://dev-docs.atariforge.org/files/Me ... n-1991.pdf (actually contains all relevant service manuals)

Before going after GLUE/MMU, since they are soldered in, check their clocks, and it's probably easier to check MFP and DMA first. As per the service manual page 118 (1040STFM "troubleshooting a dead unit").

You've checked the reset circuit. And supposing the CPU is good, it halt's due to a double buserror (buserror within a buserror), which does suggest bad ram or rom.


Hi,

All the clock are working ok, at their correct frequencies.

I do have the service manual, it could be very handy only I don't have a cartridge so my system checking is limited.

Will the cartridge even work in my situation?

I swapped the ram for another set but still no go, maybe it's ROM but I still suspect buffers.

Once I remove the buffer chips, I am expecting the Data lines will act normally for a moment or so on reset, if D3,6,9 are still faulty then it has to be CPU, pull-up resistors or maybe a capacitor.

TeX38
Atari User
Atari User
Posts: 34
Joined: Fri Aug 19, 2016 11:49 am

Re: 1040STFM Halted CPU

Postby TeX38 » Wed May 17, 2017 7:18 am

Question regarding the MMU,DMA and Glue chips.

Are these parts interchangeable between the 520 and 1040 systems?

User avatar
Arne
Captain Atari
Captain Atari
Posts: 393
Joined: Thu Nov 01, 2007 10:01 am

Re: 1040STFM Halted CPU

Postby Arne » Wed May 17, 2017 8:45 am

TeX38 wrote:Are these parts interchangeable between the 520 and 1040 systems?

In general: yes, all ST chipsets (260,520,1040,MegaST, STacey) are interchangeable. IIRC there do exist 80pin SMD MMU and/or Glue. But these are soldered to the PCB.
But better do not mix IMP and non-IMP manufactured ASICs.
Image

TeX38
Atari User
Atari User
Posts: 34
Joined: Fri Aug 19, 2016 11:49 am

Re: 1040STFM Halted CPU

Postby TeX38 » Thu May 18, 2017 12:47 am

Arne wrote:
TeX38 wrote:Are these parts interchangeable between the 520 and 1040 systems?

In general: yes, all ST chipsets (260,520,1040,MegaST, STacey) are interchangeable. IIRC there do exist 80pin SMD MMU and/or Glue. But these are soldered to the PCB.
But better do not mix IMP and non-IMP manufactured ASICs.


Thanks Arne,

The 2 Ataris I have have different part numbers for the MMU, GLUE etc but the correct pin count so I am hoping I can use them in my non working 1040.

TeX38
Atari User
Atari User
Posts: 34
Joined: Fri Aug 19, 2016 11:49 am

Re: 1040STFM Halted CPU

Postby TeX38 » Thu May 18, 2017 8:26 am

Kubik wrote:Could it be a ROM fault? Loose socket, perhaps?


I removed the ROMs and all data lines came good aside from D6 which is still stuck high, it pulses low by a few millivolts on reset so it is almost a bad Rom problem, not sure about D6 though, probably a buffer.

May explain why my working french STF wouldnt boot when I tried the Roms from my 1040 in it and with the crap D6, why the french roms wouldnt work in the 1040.

TeX38
Atari User
Atari User
Posts: 34
Joined: Fri Aug 19, 2016 11:49 am

Re: 1040STFM Halted CPU

Postby TeX38 » Sun May 21, 2017 5:44 am

TeX38 wrote:
Kubik wrote:Could it be a ROM fault? Loose socket, perhaps?


I removed the ROMs and all data lines came good aside from D6 which is still stuck high, it pulses low by a few millivolts on reset so it is almost a bad Rom problem, not sure about D6 though, probably a buffer.

May explain why my working french STF wouldnt boot when I tried the Roms from my 1040 in it and with the crap D6, why the french roms wouldnt work in the 1040.


Attached 2 pics, 1 of D6 other of an example of good data.

I doubt it to be a buffer problem as it is the same on input and output for D6.


The measurements were taken on the Shifter ic.

The data is just high at the cpu until reset.

Bad CPU maybe?

Something is causing D6 to do this, any ideas?
You do not have the required permissions to view the files attached to this post.

Kubik
Obsessive compulsive Atari behavior
Obsessive compulsive Atari behavior
Posts: 137
Joined: Thu Jan 28, 2016 8:10 pm

Re: 1040STFM Halted CPU

Postby Kubik » Sun May 21, 2017 10:18 am

Hmm, if I understand it right, it looks more like something pulling the D6 high. Did you try to remove all non-important socketed chips (AY, FDC, DMA...)?

TeX38
Atari User
Atari User
Posts: 34
Joined: Fri Aug 19, 2016 11:49 am

Re: 1040STFM Halted CPU

Postby TeX38 » Sun May 21, 2017 10:54 am

Kubik wrote:Hmm, if I understand it right, it looks more like something pulling the D6 high. Did you try to remove all non-important socketed chips (AY, FDC, DMA...)?


Thanks Kubik,

Sadly most of the chips on the board are not socketed, I have removed ROM, DMA and RAM.

Nothing happens at all when I pull the ram.

I expect that when I put the ram back in, if D6 comes good and one of the other D lines goes bad then it has to be ram at fault holding high.
ie. the ram chip order will have changed. (unless I fluke it :)
Lucky for me they are the surface mounted type so only 4 to de-solder at a time.

Exciting times here I tell you.

Kubik
Obsessive compulsive Atari behavior
Obsessive compulsive Atari behavior
Posts: 137
Joined: Thu Jan 28, 2016 8:10 pm

Re: 1040STFM Halted CPU

Postby Kubik » Sun May 21, 2017 5:07 pm

If I am not mistaken, the RAM data signals go through LS244 bus drivers. You'd better check the D6 on both sides of the corresponding bus driver before spending time and effort on removing the SMD chips :)

TeX38
Atari User
Atari User
Posts: 34
Joined: Fri Aug 19, 2016 11:49 am

Re: 1040STFM Halted CPU

Postby TeX38 » Mon May 22, 2017 4:11 am

Kubik wrote:If I am not mistaken, the RAM data signals go through LS244 bus drivers. You'd better check the D6 on both sides of the corresponding bus driver before spending time and effort on removing the SMD chips :)


Yep, did that earlier, the waveform on D6 is the same on both sides of the LS244.
Enters on Pin 17, exits on Pin 3.

If there is no ram, shouldn't the CPU still be trying to send data over the Data bus anyway?

The processor generates both data and address does it not?

hmm..

Kubik
Obsessive compulsive Atari behavior
Obsessive compulsive Atari behavior
Posts: 137
Joined: Thu Jan 28, 2016 8:10 pm

Re: 1040STFM Halted CPU

Postby Kubik » Mon May 22, 2017 7:04 am

The first cycles after reset will be CPU reading content of ROM. You should see CPU reading addresses 0 - 7 and then jumping somewhere to area FC0000 - FEFFFF. In that case, CPU generates the addresses, while ROM provides the data. I'm not sure what's happening next, but I am guessing presence of cartridge will be tested.
I am wondering if a diag cartridge would help in this case. When I had problems with memory on one system, I was able to pinpoint it based on the diagnostics... However, you seem to be on a good track already.

TeX38
Atari User
Atari User
Posts: 34
Joined: Fri Aug 19, 2016 11:49 am

Re: 1040STFM Halted CPU

Postby TeX38 » Tue May 23, 2017 8:23 am

Kubik wrote:The first cycles after reset will be CPU reading content of ROM. You should see CPU reading addresses 0 - 7 and then jumping somewhere to area FC0000 - FEFFFF. In that case, CPU generates the addresses, while ROM provides the data. I'm not sure what's happening next, but I am guessing presence of cartridge will be tested.
I am wondering if a diag cartridge would help in this case. When I had problems with memory on one system, I was able to pinpoint it based on the diagnostics... However, you seem to be on a good track already.


Yes, a cartridge would be very handy but I don't have one.

In fact, if anyone reading this subject has one hanging around unused I would love to hear from them.
I would be interested in just the bare board if available.

Kubik
Obsessive compulsive Atari behavior
Obsessive compulsive Atari behavior
Posts: 137
Joined: Thu Jan 28, 2016 8:10 pm

Re: 1040STFM Halted CPU

Postby Kubik » Tue May 23, 2017 8:33 am

TeX38 wrote:In fact, if anyone reading this subject has one hanging around unused I would love to hear from them.
I would be interested in just the bare board if available.

I still might have one spare PCB. It needs minor modification, but it works relatively well :) Send me a PM if you're interested.

TeX38
Atari User
Atari User
Posts: 34
Joined: Fri Aug 19, 2016 11:49 am

Re: 1040STFM Halted CPU

Postby TeX38 » Thu May 25, 2017 10:59 pm

Kubik wrote:
TeX38 wrote:In fact, if anyone reading this subject has one hanging around unused I would love to hear from them.
I would be interested in just the bare board if available.

I still might have one spare PCB. It needs minor modification, but it works relatively well :) Send me a PM if you're interested.


Hi Kubik, I PMd you.

TeX38
Atari User
Atari User
Posts: 34
Joined: Fri Aug 19, 2016 11:49 am

Re: 1040STFM Halted CPU

Postby TeX38 » Thu May 25, 2017 11:32 pm

I am thinking part of the problem with my 1040 is the Glue, there are no Rom strobes at all.
Strange, it has been there in my early testing, I guess the Glue has been doing strange things
and just maybe has now given up the ghost.

A/ The Glue in my 1040 is C070714-001.
B/ The Glue in my 520 is C025915-38A.

Can I use B in place of A?

You may ask, "Why don't i just use the 520 and forget the 1040?"

The 1040 is an STFM and the 520 is STF, I can't get the 520 to show a picture except in Hi-Res, I have tried everything to get Low/Med but nothing works, It is a French machine with a French TOS and I suspect it may be SECAM and the reason why I can't get a display.
Last edited by TeX38 on Fri May 26, 2017 8:26 am, edited 1 time in total.


Social Media

     

Return to “Hardware”

Who is online

Users browsing this forum: No registered users and 1 guest