Powertran Cortex Issues

[Jim Hearne]

One drive out of the pair shouldn't have the terminator resistors fitted.
That drive probably wont work by itself without you swapping the resistors (probably a resistor pack) over.
Just a thought.

 

[plagued]

yes, I've tried swapping the resistor DIP package over to the other drive and setting them up the same, and no joy. I did also have a look at the signals without termination, and while not recommended, it does look like on a short cable run like I have for internal drives, there's no discernable difference the resistors on the floppy end of the chain make over just the ones on the cortex side.
I've not tried to diagnose the other floppy, and realistically I'd be better off getting an 80 track one that I can directly try a known good standard disk on.... although it would be good to get it fixed just so I don't have a broken floppy sat on the shelf

 

[Jim Hearne]

Just a thought.
If you want more drives, those suitable for the BBC Model B worked on my Cortex with no changes.
If it wasn't all packed away i would be trying to help solve your problem.
It sounds like you have Stuart on the case though, i'm sure he will figure it out.

 

[Stuart]

I've done a bit more feretting ...

The BOOT command essentially does the following:

(1) Loads the 1st sector (just one sector) of drive 0, side 0, track 00 into the BASIC scratchpad area. The first 32 bytes of this sector are a boot header.
(2) Verifies the checksum calculated over the boot header bytes.
(3) Gets the load address specified in the boot header.
(4) Loads track 00 (the whole track) of drive 0, side 0 into memory starting at that load address (>6100 in our case).

Up to this point, if there are any problems (no disk, unformatted disk, checksum error, ...) then the BOOT code is called again - it will just loop and loop. If you see an error message then everything above must have worked.

It then runs the code just loaded starting at another address specified in the boot header (>6120 in our case) which:

(5) Checks the FDC primary status register. If the FDC reports an error then you get the Fatal System Error message and it hangs. I'm not sure of the conditions in which it is able to load track 00 in order to have some code to run, yet the FDC reports an error.
(6) Applies some patches to the standard BASIC.
(7) Loads SYSTEM$ from the disk. I think it must be at this point that the Controller Error is displayed, but having trouble unravelling that code.
(8) Performs a warm restart, which displays the BASIC prompt.

What it might be useful for Plagued to now do is find out what is causing the Fatal System Error. I think we can do that by patching the error routine to instead jump to the Monitor. So do a reset, then MON, then:

M 1DAA<return>06A0<space>0AF8<return>

Then G<return> to return to BASIC.

Then BOOT from the floppy.

It should then jump to the Monitor. Do:

M F140<return> and what number does it give you? Hopefully not FFxx this time.

 

[plagued]

awesome, so if we do that I get F140=EFC7
not my forte, but is that suggesting error code 31 "Drive 1 not RDY after 1.25 S"
I did have a quick go at connecting the second somewhat faulty floppy drive into the loop so it would have a drive 1 and it made no difference.

 

[Stuart]

awesome, so if we do that I get F140=EFC7
not my forte, but is that suggesting error code 31 "Drive 1 not RDY after 1.25 S"
I did have a quick go at connecting the second somewhat faulty floppy drive into the loop so it would have a drive 1 and it made no difference.

I think it's a different error. The value we're interested in is the >EF, in binary 1110 1111. The error code is in bits 2 - 7 which is 10 1111 = >2F, which is Data CRC error.

Page 10 of the 9909 manual suggests it could be caused by the DMA Controller not responding quickly enough. Or other problems, I assume.

Wondering if there could be a problem with the data separator circuit, which is formed by those components shown below the FDC in the Cortex manual schematic. There are some notes in the manual and news letters about changing the value of R70 from 18K to 12K - just checked my board and I've got 12K. Might be something to try.

There's a letter in one of the news letters about someone also getting the Controller Error problem, but no reply ... ;-(

 

[Jim Hearne]

My R70 is 12K as well.

 

[fixit9660]

My R70 is 12K as well.

My R16 (R70) value is 12k too. The schematic shows the original as 18k. I've attached a corrected schematic together with the Data Manuals for the main suspects involved, if they're of use to anyone.

 

[plagued]

cheers, I did actually swap out my R70, and used a pot to adjust the resistance until I got a 2u wide signal on pin 5 of IC70, which was about 10k.
But I've just gone and swapped it to 12k to see if that made any difference.
I've also just picked up an old BBC micro 5.25 80 track drive, I'm not 100% it's working OK, but I'll have a play to see, but as it is I get the exact same issues with the fatal system error, so I'm starting to think I've caused this by poking about. I'll have to re check over everything.

 

[plagued]

hmm we could be poking in the right spot. I've just gone back to step 1 and checked the signal on pin 5 of IC70 and it's much narrower than 2us. I've attached a 47k pot in place of R70 as I did before and this time it makes no difference to the signal on pin 5 other than when you drop it to 0ohm it pulls the line high and kills the signal all together. Not sure what I've done to cause this as this is likely related to the newer CRC related system error rather than the original boot issue.

EDIT.... No scrub that .... maybe I was on the wrong pin or something but I've just checked again and I'm fully able to adjust the width. Odd ... I've left the pot in place, but it's now almost exactly 2us according to my scope.

 

[plagued]

ok ...... so I went from:

to:

so I think the reason why I couldn't get this far was 100% a floppy drive/disk image issue......
However, the reason why I was having so much trouble is I do seem to have a floppy write issue, but not a read issue.
What I've done (other than going back over everything I had already gone over looking for issues I couldn't find)
Was I got an 80 track BBC micro drive back up and working. I then loaded the Recovery Utils 80SSSD disk image from the PC util into HxCFloppyEmulator and made a raw image template (with just basic FM settings) to then convert the image to a hfe raw image that I then wrote to a floppy using the greaseweazle on my laptop.... and much to my surprise I have a booting CDOS. Also because I now have a working hfe image, I poppet it on the gotek and that boots fine too! But I still cannot format with saformat or write a disk using the cortex and diskimg. But I'm quite excited to actually boot from a real floppy
If I had the knew how to make (or had) the gotek disk image from the start I could have saved a lot of poking about ... so here it is.
I'll have a go at making some others when I get a chance.

 

[Stuart]

ok ...... so I went from:

to:

so I think the reason why I couldn't get this far was 100% a floppy drive/disk image issue......
However, the reason why I was having so much trouble is I do seem to have a floppy write issue, but not a read issue.
What I've done (other than going back over everything I had already gone over looking for issues I couldn't find)
Was I got an 80 track BBC micro drive back up and working. I then loaded the Recovery Utils 80SSSD disk image from the PC util into HxCFloppyEmulator and made a raw image template (with just basic FM settings) to then convert the image to a hfe raw image that I then wrote to a floppy using the greaseweazle on my laptop.... and much to my surprise I have a booting CDOS. Also because I now have a working hfe image, I poppet it on the gotek and that boots fine too! But I still cannot format with saformat or write a disk using the cortex and diskimg. But I'm quite excited to actually boot from a real floppy
If I had the knew how to make (or had) the gotek disk image from the start I could have saved a lot of poking about ... so here it is.
I'll have a go at making some others when I get a chance.

So out of interest do you still get an error booting from the disk that fixit9660 sent you, or is that OK now?

Have you checked that the WRITE GATE signal is getting from the FDC to the floppy drive?

 

[plagued]

So out of interest do you still get an error booting from the disk that fixit9660 sent you, or is that OK now?

Have you checked that the WRITE GATE signal is getting from the FDC to the floppy drive?

so yes I'm still getting a boot error on the other disk, but it was written as a 40 track disk, and I'm not sure if I can force the 80 track drive I have to double step. And it also doesn't read on the 40 track drive......
But you were close with the write gate .... it was WRTPROT and my fault
I'd got so caught up in trying to get it to read, that I'd ignored most of the write circuit. I had traced the route between all the floppy pins and the FDC but I'd missed the fact that I had a bridge holding WRTPROT low ..... I did even consider it at one point, but I was sure that SAFORMAT would have popped up a message like "take the write protect sticker off you twit" rather than format aborted....
It wasn't until I had sat down to test all the write circuit, got the scope hooked up and thought "what should I use to trigger a write command ... I'll use the RENAME app to try and rename a file", and that immediately said ... the disk is write protected
So that didn't really need a scope to see it was held low and then find the bridge.
So picky drives aside .. I think I'm done ... it's all working. I do want to upgrade the 9929 with something that can do RGB or VGA like that RP2040 project as the PAL output is terrible even with the video "fixes" applied.
Anyway I'm going to go get a stiff drink and dare to put the other 4 screws in to hold the motherboard down.

Thanks everyone for the help!!!!! I'm sure I'll have some more questions ... like what's the best game?

I'll post some glamour shots when it's back together.

 

[fixit9660]

So glad you got this sorted!!! I've only just seen this series of posts; I've been laid low with a really lousy Cold for weeks and unable to move far or think straight.
Hoping to get back to it shortly.