[Beetle]

Hi folks,


I have a 1200XL and did Bob Wooley's PBI mod to it.
I want to run my CSS BlackBox with Floppyboard on it. The BB is okay and running on my
1500XL Laptop; if you remember, it has a 600XL inside.
On the 1200XL, however, it refuses to work. No booting from HD, no BB menu coming up
when menu button is pressed. Other tahn that, 1200XL works as before.

800XL OS and MMU it already had working a long time.

In the text linked above, i think i found one error when PBI pin 40 should be wired to U14, Pin 14 (the MMU) .
PBI 40 is refresh output according to this wiki page.
Only PBI pin 43 'Math Pack Disable' should go to that pin 14 of U14, not both.
PBI pin 40 i wired to ANTIC pin 8 (same as MMU pin 11).

Am i right with that?

I didn't have a 74HC08 handy, but tried both 74F08 and 74LS08 without luck.

I rechecked my wiring with ohm-meter but didn't find any mistakes.
I added a 22pf cap to phi2 as i was suggested on IRC.
I also added a resistor to the CSS BlackBox i want ro run on that machine. That tip i got from Bob Wooley's stabilizer mods..

Quote

MODIFICATION #3

This is actually a Black Box modification. Again, due to varying phase two clock signals, a timing circuit on the
Black Box MAY need to be modified. The D1FF latch uses a R/C delay to insure the latching occurs while the bus is valid.
A late 02 signal can skew this delay, causing the latch to grab random values at times.

The fix is simple. First, locate the resistor and capacitor combo that is just below the BB's SCSI port. Look at the color
bands of the resistor. It should be brown-black-brown. Now look on the bottom side of the BB, and see if there is another
resistor soldered in the same place. If so, then no modification is needed. If you see no resistor on the bottom
IN THAT LOCATION, then solder a 220 ohm (red-red-brown) resistor across the two resistor pads.

But with no luck again. The 1200XL runs as usual, not recognizing the BB at all.
It won't either boot from the (already configured!) Harddisk nor bring up the BB menu.

When doing my Google search, i found this text on the net:

Quote

I got into the Atari when the 1200xl was released. That was
my first 'real' computer. The maligned orphan 1200xl happened to be the
best 8-bit machine Atari ever made. I've got one still and have done the
800xl rom mod, and the internal basic mod. On this particular one I tried
Bob Wooley's 1200xl PBI mod. ALLL THOSE CONNECTIONS. I double checked,
rechecked etc. I even redid the mod completely with a new set of chips and
cable. Nothing. Not 'no video'. The machine works perfectly when
connected to my Black Box or not. It just doesn't SEE the Black Box.

I haven't seen any updates to the 1200xl PBI mod and I don't think it
quite works. I think the PBI is not being selected or something. One
more line and it would work I swear....

So, that guy ran into same situation as me. I'd really
love to get that BB running, because the 1200XL board i work on is
that one in my 1250XLD, the 1450XLD remake. A Floppy on PBI plus
a SCSI HD inside would make my remake a bit closer to what Atari wanted
the 1450XLD to become. The Flagship of the XL line.

Is there someone who can give me some hints or, even better, did solve
the same situation?

Thanks for reading,
Beetle

This post has been edited by Beetle: Mon Jun 2, 2008 3:53 PM

[Guitarman]

Hopefully Bob will read this and answer himself. I also have a 1200XL with PBI mod I got off of Ebay a while back. When Bob saw a picture of it, he thought it might have been one of his old ones. I have not tried mine with a MIO yet as the connector on the end of the ribbon is a edge slot connector, like the one on the MIO. I need to make a passthrough to connect them!!

[bob1200xl]

As it says in the article, I used this PBI mod on a Supra MicroPort. I am pretty sure that it also worked on the MIO but I have never tried it on a BB - don't have one. You are correct about the REFRESH line on pin 40 - it should go to MMU pin 11, not 14. You might try manually selecting the BB with an ED/ASM cart in BUG. Store a bit in $D1FF and see if the BB disables the Math Pack at $D800-$DFFF. (I think you'll just crash - display $D800 and compare it to what's in the OS ROM) One or more bits should bring up MPD on the MMU. You can also look in $0247 after you boot with the BB connected. It will show what PBI 'devices' were found and initialized.

Some people have said that the BB doesn't follow PBI protocol. I don't know what that refers to but the hardware in the 1200XL is essentially the same as an 800XL.

Bob

Beetle, on Thu May 29, 2008 4:52 AM, said:

Hi folks,


I have a 1200XL and did Bob Wooley's PBI mod to it.
I want to run my CSS BlackBox with Floppyboard on it. The BB is okay and running on my
1500XL Laptop; if you remember, it has a 600XL inside.
On the 1200XL, however, it refuses to work. No booting from HD, no BB menu coming up
when menu button is pressed. Other tahn that, 1200XL works as before.

800XL OS and MMU it already had working a long time.

In the text linked above, i think i found one error when PBI pin 40 should be wired to U14, Pin 14 (the MMU) .
PBI 40 is refresh output according to this wiki page.
Only PBI pin 43 'Math Pack Disable' should go to that pin 14 of U14, not both.
PBI pin 40 i wired to ANTIC pin 8 (same as MMU pin 11).

Am i right with that?

I didn't have a 74HC08 handy, but tried both 74F08 and 74LS08 without luck.

I rechecked my wiring with ohm-meter but didn't find any mistakes.
I added a 22pf cap to phi2 as i was suggested on IRC.
I also added a resistor to the CSS BlackBox i want ro run on that machine. That tip i got from Bob Wooley's stabilizer mods..

Quote

MODIFICATION #3

This is actually a Black Box modification. Again, due to varying phase two clock signals, a timing circuit on the
Black Box MAY need to be modified. The D1FF latch uses a R/C delay to insure the latching occurs while the bus is valid.
A late 02 signal can skew this delay, causing the latch to grab random values at times.

The fix is simple. First, locate the resistor and capacitor combo that is just below the BB's SCSI port. Look at the color
bands of the resistor. It should be brown-black-brown. Now look on the bottom side of the BB, and see if there is another
resistor soldered in the same place. If so, then no modification is needed. If you see no resistor on the bottom
IN THAT LOCATION, then solder a 220 ohm (red-red-brown) resistor across the two resistor pads.

But with no luck again. The 1200XL runs as usual, not recognizing the BB at all.
It won't either boot from the (already configured!) Harddisk nor bring up the BB menu.

When doing my Google search, i found this text on the net:

Quote

I got into the Atari when the 1200xl was released. That was
my first 'real' computer. The maligned orphan 1200xl happened to be the
best 8-bit machine Atari ever made. I've got one still and have done the
800xl rom mod, and the internal basic mod. On this particular one I tried
Bob Wooley's 1200xl PBI mod. ALLL THOSE CONNECTIONS. I double checked,
rechecked etc. I even redid the mod completely with a new set of chips and
cable. Nothing. Not 'no video'. The machine works perfectly when
connected to my Black Box or not. It just doesn't SEE the Black Box.

I haven't seen any updates to the 1200xl PBI mod and I don't think it
quite works. I think the PBI is not being selected or something. One
more line and it would work I swear....

So, that guy ran into same situation as me. I'd really
love to get that BB running, because the 1200XL board i work on is
that one in my 1250XLD, the 1450XLD remake. A Floppy on PBI plus
a SCSI HD inside would make my remake a bit closer to what Atari wanted
the 1450XLD to become. The Flagship of the XL line.

Is there someone who can give me some hints or, even better, did solve
the same situation?

Thanks for reading,
Beetle

[mos6507]

bob1200xl, on Thu May 29, 2008 7:33 AM, said:

Some people have said that the BB doesn't follow PBI protocol. I don't know what that refers to but the hardware in the 1200XL is essentially the same as an 800XL.

Bob Puff told me it doesn't follow the PBI guidelines. The original concept of the PBI was to enable a card cage. So each PBI device was supposed to have a unique ID, kind of like an IRQ in ISA slots. But the Black Box takes over all the IDs and won't coexist with any other PBI devices. The only time that's a problem is with the 1400 prototypes because they have internal parts that interface through the hardware as if they were PBI devices. So the only way to make those work is to swap out the ROMs which disables the speech, the modem, and the parallel floppies.

[Beetle]

Thanks for your answers so far.

The BB will be the only PBI device in my 1250XLD, so if it takes all ID its no problem to me.
In the meantime i tried with the Phi0+Phi2 mod according to the stabilizer mod, looking at the
different pin numbers of the 1200XL, of course. Didn't do the trick, anyway.

:-(

I took Bobs suggestion and got my 1200XL with BB the ASM/ED cart.

DEBUG
DD800
D800 20 A1 DB 20 BB DB B0 39

DEBUG
CD1FF<FF

DEBUG
DD800
D800 20 A1 DB 20 BB DB B0 39

DEBUG
D0247
247 00
248 00 00 00 00 00 00 00

So, what do you think?

[dmlloyd]

I've been helping Beetle with this (#atari on irc.newnet.net) - it seems to me like the latch isn't being activated, though without a scope or a schematic for the BB it's hard to say for sure. But we've tried adjusting phi2 in about 3-4 different ways, no luck...

[bob1200xl]

When you write to $D1FF the BB should lower MPD, an input to the MMU. This kills the ROM at $D800-$DFFF. If nothing else, it will enable RAM in that space. Remove any PBI devices and try grounding MPD. You should see RAM at $D800. (the ASM/ED cart will probably crash) If MPD does its job, then you need to see why the BB isn't decoding $D1FF. Write a little routine to write $FF into all locations of $D1xx.

Do you have anything else (like an MIO) that you can try?

Bob

Beetle, on Thu May 29, 2008 12:07 PM, said:

Thanks for your answers so far.

The BB will be the only PBI device in my 1250XLD, so if it takes all ID its no problem to me.
In the meantime i tried with the Phi0+Phi2 mod according to the stabilizer mod, looking at the
different pin numbers of the 1200XL, of course. Didn't do the trick, anyway.

:-(

I took Bobs suggestion and got my 1200XL with BB the ASM/ED cart.

DEBUG
DD800
D800 20 A1 DB 20 BB DB B0 39

DEBUG
CD1FF<FF

DEBUG
DD800
D800 20 A1 DB 20 BB DB B0 39

DEBUG
D0247
247 00
248 00 00 00 00 00 00 00

So, what do you think?

[bob1200xl]

Hey... you built that 1250XLD? I took a look at the thread about its introduction. Wow - I'm impressed!

You have a lot of extra stuff on that motherboard. That could be a problem. Or, not.

Let me know how you make out with MPD and all.

Bob

Oh, here's the one I made...





So, that guy ran into same situation as me. I'd really
love to get that BB running, because the 1200XL board i work on is
that one in my 1250XLD, the 1450XLD remake. A Floppy on PBI plus
a SCSI HD inside would make my remake a bit closer to what Atari wanted
the 1450XLD to become. The Flagship of the XL line.

Is there someone who can give me some hints or, even better, did solve
the same situation?

Thanks for reading,
Beetle
[/quote]

[sup8pdct]

Beetle, on Thu May 29, 2008 10:52 PM, said:

Hi folks,


I have a 1200XL and did Bob Wooley's PBI mod to it.
I want to run my CSS BlackBox with Floppyboard on it. The BB is okay and running on my
1500XL Laptop; if you remember, it has a 600XL inside.
On the 1200XL, however, it refuses to work. No booting from HD, no BB menu coming up
when menu button is pressed. Other tahn that, 1200XL works as before.

800XL OS and MMU it already had working a long time.

In the text linked above, i think i found one error when PBI pin 40 should be wired to U14, Pin 14 (the MMU) .
PBI 40 is refresh output according to this wiki page.
Only PBI pin 43 'Math Pack Disable' should go to that pin 14 of U14, not both.
PBI pin 40 i wired to ANTIC pin 8 (same as MMU pin 11).

Am i right with that?

I didn't have a 74HC08 handy, but tried both 74F08 and 74LS08 without luck.

I rechecked my wiring with ohm-meter but didn't find any mistakes.
I added a 22pf cap to phi2 as i was suggested on IRC.
I also added a resistor to the CSS BlackBox i want ro run on that machine. That tip i got from Bob Wooley's stabilizer mods..

Quote

MODIFICATION #3

This is actually a Black Box modification. Again, due to varying phase two clock signals, a timing circuit on the
Black Box MAY need to be modified. The D1FF latch uses a R/C delay to insure the latching occurs while the bus is valid.
A late 02 signal can skew this delay, causing the latch to grab random values at times.

The fix is simple. First, locate the resistor and capacitor combo that is just below the BB's SCSI port. Look at the color
bands of the resistor. It should be brown-black-brown. Now look on the bottom side of the BB, and see if there is another
resistor soldered in the same place. If so, then no modification is needed. If you see no resistor on the bottom
IN THAT LOCATION, then solder a 220 ohm (red-red-brown) resistor across the two resistor pads.

But with no luck again. The 1200XL runs as usual, not recognizing the BB at all.
It won't either boot from the (already configured!) Harddisk nor bring up the BB menu.

When doing my Google search, i found this text on the net:

Quote

I got into the Atari when the 1200xl was released. That was
my first 'real' computer. The maligned orphan 1200xl happened to be the
best 8-bit machine Atari ever made. I've got one still and have done the
800xl rom mod, and the internal basic mod. On this particular one I tried
Bob Wooley's 1200xl PBI mod. ALLL THOSE CONNECTIONS. I double checked,
rechecked etc. I even redid the mod completely with a new set of chips and
cable. Nothing. Not 'no video'. The machine works perfectly when
connected to my Black Box or not. It just doesn't SEE the Black Box.

I haven't seen any updates to the 1200xl PBI mod and I don't think it
quite works. I think the PBI is not being selected or something. One
more line and it would work I swear....

So, that guy ran into same situation as me. I'd really
love to get that BB running, because the 1200XL board i work on is
that one in my 1250XLD, the 1450XLD remake. A Floppy on PBI plus
a SCSI HD inside would make my remake a bit closer to what Atari wanted
the 1450XLD to become. The Flagship of the XL line.

Is there someone who can give me some hints or, even better, did solve
the same situation?

Thanks for reading,
Beetle

have you checked the external select line (pin 2) on the pbi and it does do what it is supposed to? (it disables ram access regardless of address) another thing to check is all the address lines are the right way round. also the datalines. both BB and MIO have there own decoding of the address lines for registers etc.
One trick is to continuity test the 1200xl cart port and the one on the BB pin for pin. There will be a couple of signals not present but it will test the data and most address lines at least.

James
From memory, both have MPD and external select line connected on the device.

[MEtalGuy66]

USe an 800XL/130XE motherboard.. (actually, given space considerations, a 600XL motherboard with the Mega-HZ SRAM upgrade would be even better.) You can easily build a cart-slot reciever PCB to attatch to the 1200XL case's big cartridge-slot "guide-boss" piece, and then connect it with a ribbon-cable.. Same goes for the joystick ports.. The 1200XL keyboard is easy to convert to work with the 800XL/130XE (Ive done 4 of them so far) The function keys even work, since they dont use any additional decoder lines.. I'd hook L1 and L2 to the two black-box LED circuits If I was you.. If you need the cable that goes from the 800XL PBI to the Black box, I'll make you one for $10.00US, you pay shipping. Just let me know how long you want it to be (in centimetres)..

[Beetle]

Well Ken,

Quote

USe an 800XL/130XE motherboard..

that would in fact be my last resort and would give me a clear picture, too.
Although not bad, the 1200XL's PAL conversed picture is not as sharp as my 800XL.


Bob,

With a little work here and there your 1450XLD can shine! Tell us the story of it.
A remake? Original prototype found while dumpster diving?

I will go, check that MPD line...

[Mathy]

Hallo Beetle

There's a better way to make Phi2 work as it shoud be. Forget Bob Puff's mod. Contact Hias, Guus Assmann or BigBen. I guess HardwareDoc might be able to help you too.

Greetings

Mathy

PS after they tell you how it's done EXACTLY, please tell us (read: me). I know it involves a resistor and a capacitor, but I can't get them to tell me which one and where.

[Beetle]

Hi Bob1200XL

Quote

When you write to $D1FF the BB should lower MPD, an input to the MMU. This kills the ROM at $D800-$DFFF. If nothing else, it will enable RAM in that space. Remove any PBI devices and try grounding MPD. You should see RAM at $D800. (the ASM/ED cart will probably crash)

ASM/ED cart and also BASIC crash when i put MPD to ground.
After testing that i connected BB and did a BASIC loop which poked 255 to 53759 all the time, but MPD stays high.

Quote

If MPD does its job, then you need to see why the BB isn't decoding $D1FF. Write a little routine to write $FF into all locations of $D1xx.

I did that routine, too. After a few seconds it said 'READY'. No crash, MPD stays high.

Quote

Do you have anything else (like an MIO) that you can try?

Bob

No, i don't. But the BB works on my Laptop (600XL based). So its okay.

I should now look why MPD doesn't go low, right? But how to?

This post has been edited by Beetle: Fri May 30, 2008 6:00 AM

[Beetle]

Quote

have you checked the external select line (pin 2) on the pbi and it does do what it is supposed to? (it disables ram access regardless of address) another thing to check is all the address lines are the right way round. also the datalines. both BB and MIO have there own decoding of the address lines for registers etc.
One trick is to continuity test the 1200xl cart port and the one on the BB pin for pin. There will be a couple of signals not present but it will test the data and most address lines at least.

James
From memory, both have MPD and external select line connected on the device.

Hi James,

Thanks for your tip as well. the 'external select' signal is the one that has to be generated on the 1200XL PBI mod.
The 1200XL doesn't have that signal on board. I made it using a 74F08 ( i dont't have HCT ones and tried with an LS, too)
according to the PBI mod. Too bad i don't have an oszilloscope. All i can say is, the extsel line stays always high. BB never
makes it low.

For the wiring, its checked twice according the mod's numbering.
Then dmlloyd checked with me pin for pin if each PBI pin reaches the target pin on Sally, MMU or whatever.
We encountered no mistakes.

Greetings,
Beetle

[bob1200xl]

It would be helpful to know whether this is a BB specific problem or a PBI generic failure. That would at least tell us that the address and data lines are correct.

MPD is latched in the BB by decoding $D1FF and write. If that is not happening then it doesn't matter if EXTSEL, ROM or anything else is working. It doesn't appear to be functional. Can you take a detailed picture of the top and bottom of the BB board? I should be able to trace the circuits from them.

I would meter from the CPU out to the connector pins on the BB - verify that D0 goes to D0, D1 goes to D1, etc. Pay particular attention to the address lines because a crossed wire (by me or you) will probably ruin the decode.

This is going to be tough without a scope... you are in Europe?

Bob

Beetle, on Fri May 30, 2008 2:23 AM, said:

Hi Bob1200XL

Quote

When you write to $D1FF the BB should lower MPD, an input to the MMU. This kills the ROM at $D800-$DFFF. If nothing else, it will enable RAM in that space. Remove any PBI devices and try grounding MPD. You should see RAM at $D800. (the ASM/ED cart will probably crash)

ASM/ED cart and also BASIC crash when i put MPD to ground.
After testing that i connected BB and did a BASIC loop which poked 255 to 53759 all the time, but MPD stays high.

Quote

If MPD does its job, then you need to see why the BB isn't decoding $D1FF. Write a little routine to write $FF into all locations of $D1xx.

I did that routine, too. After a few seconds it said 'READY'. No crash, MPD stays high.

Quote

Do you have anything else (like an MIO) that you can try?

Bob

No, i don't. But the BB works on my Laptop (600XL based). So its okay.

I should now look why MPD doesn't go low, right? But how to?

[classics]

This may not be extremely helpful, but I did one of these PBIs once a long time ago and I remember that one of the connections listed was incorrect, listed as a duplicate or something.

Steve

[Beetle]

Hi Bob,

Quote

It would be helpful to know whether this is a BB specific problem or a PBI generic failure. That would at least tell us that the address and data lines are correct.

Well, BB runs on two other XL maschines i have.

Quote

MPD is latched in the BB by decoding $D1FF and write. If that is not happening then it doesn't matter if EXTSEL, ROM or anything else is working. It doesn't appear to be functional. Can you take a detailed picture of the top and bottom of the BB board? I should be able to trace the circuits from them.

I'll take some pics. Reload this page later, i'll ad the pics here.

Quote

I would meter from the CPU out to the connector pins on the BB - verify that D0 goes to D0, D1 goes to D1, etc. Pay particular attention to the address lines because a crossed wire (by me or you) will probably ruin the decode.

Thats what we did yesterday. It is and was okay.

Quote

This is going to be tough without a scope... you are in Europe?

Yes, i'm in Germany. I'm trying to get a scope borrowed, but no luck yet.


Hi Steve,

Quote

This may not be extremely helpful, but I did one of these PBIs once a long time ago and I remember that one of the connections listed was incorrect, listed as a duplicate or something.

That one we ironed out already:
Only PBI pin 43 'Math Pack Disable' should go to that pin 14 of MMU, not 43 and 40.
PBI pin 40 is to be wired to ANTIC pin 8 (same as MMU pin 11).

This post has been edited by Beetle: Fri May 30, 2008 10:14 AM

[classics]

Beetle, on Fri May 30, 2008 11:07 AM, said:

That one we ironed out already:
Only PBI pin 43 'Math Pack Disable' should go to that pin 14 of MMU, not 43 and 40.
PBI pin 40 is to be wired to ANTIC pin 8 (same as MMU pin 11).

Ok, I found the notebook and it was indeed this connection. Hope you figure it out.

Steve

[bob1200xl]

Well, BB runs on two other XL maschines i have.

Yes the BB is OK but would anything work on the PBI?

Quote

I would meter from the CPU out to the connector pins on the BB - verify that D0 goes to D0, D1 goes to D1, etc. Pay particular attention to the address lines because a crossed wire (by me or you) will probably ruin the decode.

Thats what we did yesterday. It is and was okay.

Did you measure from the CPU pins out into the BB?

Bob

[Mathy]

Hello guys

As the BB doesn't really stick to PBI rules, it might not be the prime candidate to test a faulty PBI.

@Beetle: The TurboFreezer (new version) should be PBI compliant. But I'm not sure if you have one (I don't).

Greetings

Mathy

[mos6507]

Beetle, on Thu May 29, 2008 12:07 PM, said:

Thanks for your answers so far.

The BB will be the only PBI device in my 1250XLD, so if it takes all ID its no problem to me.
In the meantime i tried with the Phi0+Phi2 mod according to the stabilizer mod, looking at the
different pin numbers of the 1200XL, of course. Didn't do the trick, anyway.

Cool project.

This post has been edited by mos6507: Fri May 30, 2008 11:43 AM

[Beetle]

Mathy, on Fri May 30, 2008 7:25 PM, said:

Hello guys

As the BB doesn't really stick to PBI rules, it might not be the prime candidate to test a faulty PBI.

@Beetle: The TurboFreezer (new version) should be PBI compliant. But I'm not sure if you have one (I don't).

My Turbofreezer is soldered onto my 800XL board. Inside the case :-)
And all the BB doesn't do PBI compliant is taking all PBI ID numbers instead of one,
making it not compatible with other PBI devices on the same bus.



Anyway, here are the pics:

Blackbox top view


Blackbox bottom view


Blackbox plus floppyboard on 1200XL mainboard


ugly bottom of 1200XL board

Click pics to get the giant ones. Everything should be visible.

[sup8pdct]

only thing I can think of now is you may have a week cpu. the ones made in mexico I believe are the ones that are not real good. Do you have a logic probe? I do most of my trouble shooting with one of these.

James

[dmlloyd]

Just to summarize what's already been tried:

* Weak CPU - definitely not the prob; I believe Beelte tried a CPU from another system that was able to drive the BB without a problem.
* EXTSEL - checked, rechecked, and triple-checked. All wiring for EXTENB/EXTSEL is correct.
* MPD - also checked thoroughly.
* A0-15 and D0-7 - checked one by one at least two different ways.
* I think we also checked IRQ a few times, but we'll check it once more.

So the current primary symptom is that /MPD and /EXTSEL are never being asserted - my best guess is that the latch on the BB is failing to fire at the right time due to possibly a timing problem in relation to the address bus. Beetle is messing with the RC circuit on the BB now.

Beetle has found two errors now - one is the aforementioned MPD error, and the second one is that phi2 was going to a ??? pin on SALLY rather than the actual phi2 pin.

[dmlloyd]

Hooking up phi2 properly seems to have improved matters. Now writing values to $D1FF seems to freeze the system fairly reliably.

However the menu button still doesn't work. Checking /IRQ now - phi2 timing might still not be perfect.