Dual XF-551 clone firmware

[+mytek]

1 hour ago, Dropcheck said:

Curious.  There's a SF826 transistor listed as Q1.  I'm not finding that particular version on Mouser or Digikey.  What NPN transistor would be a good substitute?

That's just being used as a digital switch so any general purpose NPN transistor should work in its place (e.g., 2N3904, 2N2222). And I believe the pin-out is a match for either one of those.

 

[BillC]

On 8/17/2020 at 6:48 AM, manterola said:

Thanks for the answers... My understanding is that at least the floppy cable provides two motor and two drive select (DS) signals for managing two drives. On the other hand, it looks like the Wd1772 controller has only one MO(motor on) signal, which drives the DS0 and MOTOR input of the mech. So it looks like additional logic is needed to redirect this signal to DS0 or DS1 accordingly. So it is not only a matter of firmware change but also adding some electronic components.

Also thanks Mathy for the idea, I might need to figure out with Sascha how to login in ABBUC forums and contact him there. 

It looks like the simple solution is assembling two boards   , which mainly means two WD1772 and everything, assuming I can fit two inside the case... I'll check this possibility, as well.

The way Atari implemented DS0 on the XF551 was to tie it to MOTOR via a trace, since the IBM method of using a twisted cable re-assigned the DS0 pin as the additional MOTOR signal both MOTOR signals are always active at the same time. 

 

The CSS dual-drive mod still used a straight 34-conductor cable, with DS1 for the 5-1/4" and DS2 for the 3-1/2 " drives, because of the way the MOTOR signals are tied together I believe both drives should work configured for DS1 when used with a twisted floppy cable. 

 

I believe the primary reason for the potted module of the CSS mods was copy protection. If the need for the module was eliminated from the firmware I think that 2 of the 8040 outputs could be used as DS1/DS2 for the dual drive version.

[Simius]

16 hours ago, manterola said:

That is my question... It look like the schematics for 601 is the same as 602, because of the memory and the wiring to DS1, etc. Is the ROM dump also good for xfd602?

Dropcheck: Good to see you in this topic, I was wondering why you weren't in the discussion. 

Schematic is the same, software different. There were also a few changes in later production - 10MHz crystal, RPM switched to 360 and apropriate software.

 

 

[Dropcheck]

I've lost confidence in the schematic.   Simius,  do you have a more accurate schematic?

 

 

[Simius]

43 minutes ago, Dropcheck said:

I've lost confidence in the schematic.   Simius,  do you have a more accurate schematic?

It's an irrelevant, obvious clerical error. 

 

[Dropcheck]

 I also do not see connections for U7 pins 1,2 and 3.  U3 pin 26 is not shown.  I can assume a lot of things, but not necessarily the right one. 

 

Sobola did a good job in the schematics, but everybody makes mistakes.   ?

[manterola]

7 hours ago, Dropcheck said:

 I also do not see connections for U7 pins 1,2 and 3.  U3 pin 26 is not shown.  I can assume a lot of things, but not necessarily the right one. 

 

Sobola did a good job in the schematics, but everybody makes mistakes.   ?

I believe pin 26 of U3 is VDD in the 8050 and ~stop in 80c50 so it should be set to 5V.. right?

Now regarding the U1 , U7 and U8: U7 is a ls38 so pins 1, 2 and 3 are U7A  and those are connected (the error is that U7A is LS38 not an LS00). U1 is LS38 as well. U8 is LS00 and U8A looks like no used.

Edit: I confused u1 and u8 , now the above statement is correct

Edited August 19, 2020 by manterola

[Dropcheck]

The problem is that the schematic is supposed to eliminate assumptions and guesses and tell you exactly WHAT goes to WHAT and HOW.  And serve as a template for trace routing in pcb routing software.  I know I sound anal, but I hate wasting time and money on wrong assumptions and guesses.  ?

 

I'm willing to assume that U7A LS500 should read as U7A LS38, but then the question shifts to U8A and what is connected or not to pins 1,2 and 3.    Do I assume that all three pins are NC? Or that Sobola just missed tracing those pins?   Grrrrr.....   ☹️?

 

And what is D1?  The schematic symbol indicates a zener diode with a voltage of 3.9V if I read the schematic right.   Would a 1N5228 series work?

[manterola]

Yep, I totally hear you. And assuming everything is right and you solved the hardware side of the puzzle, then it comes down to finding the firmware that supports two drives. 

[manterola]

For know I will check if I could fit two sf551 boards in the enclosure. The dual drive with single board seems like a long shot, moreover when the chances for me modifying some existing firmware are slim (very limited knowledge of assembler). Still the changes are not zero.

[Dropcheck]

I'll be working on a proto board based on this schematic in the next few months.  Not sure what kind of enclosure I'll be using.  Maybe a Hammond plastic case if I can find one that will fit two 3.5" drives.  Maybe a repurposed 1050 case. 

 

Right now I've got a few other irons in the fire.  ?

[manterola]

That sounds great. I look forward to hear news from that iron. From my part I'll put together a couple of sf551 boards, which I could  use for this dual drive or later for other sf314 transformation I have in the backlog. 

You will get my order soon. 

Mauricio