Another C1P build

[Thomas]

I see that the data rate reported by OSIHFE is 250 (KHz)

Shouldn't the bit rate it be 125 (KHz) for a 5¼" floppy?

Hmmmmm?

Congrats to your progress. The actual data bit rate is 125kHz bit FM. Data is coming together with a clock signal which results in signal of a 250kHz if a data bit is present and 125k, if not.
With all the progress, you should see data coming into the ACIA on the 610, if “D” was pressed.
Also the Gotek should turn on and present the INDEX pulse and data on the DATA Out pin.
In my Gotek firmware, I can see if the drive is selected and what the actual track is and if side 1 or 2 is selected.
The rest is measuring and tweaking.

[Thomas]

I see that the data rate reported by OSIHFE is 250 (KHz)

Shouldn't the bit rate it be 125 (KHz) for a 5¼" floppy?

Hmmmmm?

Congrats to your progress. The actual data bit rate is 125kHz bit FM. Data is coming together with a clock signal which results in signal of a 250kHz if a data bit is present and 125k, if not.
With all the progress, you should see data coming into the ACIA on the 610, if “D” was pressed.
Also the Gotek should turn on and present the INDEX pulse and data on the DATA Out pin.
In my Gotek firmware, I can see if the drive is selected and what the actual track is and if side 1 or 2 is selected.
The rest is measuring and tweaking.

[HOWARDHB]

Thanks Thomas!

Now that I've verified that the "external" hardware does in fact work correctly, I can "scope" the signals from the 600D & 610...

And, the Gotek OLED display does in fact display the head and track numbers, as it is being "read" by the Greaseweasle.

H.

[HOWARDHB]

Hi Thomas.

Could you report here what the jumper settings are on your Gotek, specifically JC, S0, S1 and MO / MOR
Just trying to eliminate basic connection and setup errors...

I can't find any definative info online, with respect to OSI & Data separator and Gotek.

I have the drive sellect jumper on S1.
(if S0 is jumpered, the green "drive sellected" led is lit permanently)

My FF.CG config file on the USB stick looks like this:
interface = shugart
index-suppression = yes
pin02 = nc
pin34 = rdy
motor-delay = 400

H.

[davisgw]

The S0 or S1 jumpers will determine if the Gotek is drive A or drive B. And if you have the "twist" in the flat cable it should be set to S1 for it to connect as drive A. MO is not jumpered.
The FF.CFG settings that are working for me are:
interface = shugart
host = unspecified
pin02 = nc
pin34 = nc
write-protect = no
max-cyl = 40
side-select-glitch-filter = 2
index-suppression = yes
head-settle-ms = 12
motor-delay = 450
indexed-prefix = "OSI" - this is the prefix for the images on the flash drive
I have not tried changing each of these values to see the effect.

[HOWARDHB]

Thank-you !

I will edit my FF.cfg with your settings.
(I have read that the new FF.cfg settings will only be "read" and saved into flash memory after first "resetting" the Gotek with no USB stick)

I'm using a "straight" 34way ribon cable, so I'll jumper "S0" and remove "MO"


H

[HOWARDHB]

This is a very strange coincedence, or, it is not a coincedence at all:

With NO activity on the floppy data and control signsls, upon pressing "D" after a reset, I was hunting for the cause...

I found a cut track, a very specific cut. Not a manufacturing fault, clearly a CUT
It is the "buffered" 02 clock, B02 (U2 pin 3) without which no floppy control or data signals can be generated...
I bought this 610 PCB, un-populated, from a seller on Ebay....

B02 track CUT.jpg (333.73 KiB) Viewed 1969 times

Very strange indeed....

Anyhow, I repaired the cut, and now, upon pressing "D", I see the Track number (displayed on the Gotek) toggle briefly from 0 to 1, and back to 0.
But nothing further happens after that... Why, I as yet, don't know...

So, I decided to again connect up the Shugart A400, 35 track, drive and see if the head would move...
(The drive motor spins permanently, on application of power, which is a quirk of the OSI floppy system)

I inserted a disk (that I created using the Greasweasle), with a 35 Track version of Picodos, that bxdanny made for me.
Upon hittng "D" after a reset, the head will position to TRACK 0 (if not already there) and the head-load solenoid engages...

Once, only once, I saw "SYNTAX ERROR" appear on the monitor screen after a few rotations of the disk...
This is progress, me thinks!

Now that I have a full 32K RAM installed (8k+24k), I' should write C1-OS65D.HFE to a new floppy and commence scoping using the Gotek.

However, @ 2am I retired to bed to cogitate...

H.

[ronin47]

How in the heck did it get cut?

[HOWARDHB]

looking at the cut closely, it seems to be certainly deliberate....
(the rest of the board is "pristine", and is Gold flash plated)

[bxdanny]

Well, it's good that you found that cut. Based on your photo, it looks like it would be hard to notice.

The symptom you describe of going to track 0, then track 1, and back to track 0 sounds a lot like what happened when I first got my 5.25" disk drive and 610 board. It happened because I only had 12k installed, and the disk I was trying to boot (which was the standard version of Pico-Dos) actually needed 13k. But you have the full 32k, so that shouldn't be the issue. And you have booted disk images from the Gotek unit successfully, Still, if you haven't run a thorough RAM test, perhaps it would be a good idea.

You can check if track 0 is in fact loading by going into the Monitor and entering .FC06G. That should cause track 0 to be loaded, and the CPU then return to the Monitor program. You can then check the RAM at $2200 to see if what is there is what should be there. The first five bytes at $2200, with either 65D or Pico-Dos, should be A9 01 8D 5E 26.

BTW, Pico-Dos has no directory structure; the program descriptions you saw were extracted by Mark's software from REMark statements within the programs themselves.

I'm pretty sure that any 40-track .hfe file can be converted to a 35-track file by simply changing the byte at offset 9 (tenth byte of the file) from 28 (40 decimal) to 23 (35 decimal), and that the Gotek and Greaseweasel software will both then treat the image as containing only 35 tracks. But I can't be positive. You could help verify (or refute) this by trying it on some other images.