WDC FDC 2797 -- Pin 25?

[+Larry]

I was working on a some dead 1050's today and I found two drives that had pin 25 missing on the 2797. But one of the two drives works perfectly. Then I found another working 1050 that had a 2797, but with pin 25 intact. Does anyone know why the pin 25 might have been was removed. I don't believe it is just a coincidence that pin 25 was missing on two separate drives. (?)

 

Also, another 1050 will not always "seek" when powered up, but if I bump the drive slightly, then it works fine. Removed, inspected and replaced all socketed IC's, and it still does the same thing. Has anyone run into this before, or can hazard a guess what might be going on?

 

-Larry

[rdhays82604]

I was working on a some dead 1050's today and I found two drives that had pin 25 missing on the 2797. But one of the two drives works perfectly. Then I found another working 1050 that had a 2797, but with pin 25 intact. Does anyone know why the pin 25 might have been was removed. I don't believe it is just a coincidence that pin 25 was missing on two separate drives. (?)

 

Also, another 1050 will not always "seek" when powered up, but if I bump the drive slightly, then it works fine. Removed, inspected and replaced all socketed IC's, and it still does the same thing. Has anyone run into this before, or can hazard a guess what might be going on?

 

-Larry

 

Sounds to me like you have a loose solder joint. Give them a check.

[Bryan]

I've seen many 1050's with a pin cut off the controller. I looked it up at one time and I'm pretty sure it's for a feature Atari isn't using. Many 1050's have 2793's and the PCB may have been designed with that chip in mind.

Edited March 14, 2010 by Bryan

[ijor]

I was working on a some dead 1050's today and I found two drives that had pin 25 missing on the 2797.

 

The pin is cut off because this pin has a different purpose on the 2797 (than the 2793), and the drive board is the same. On the 2793 it is an input, on the 2797 it is an output.

 

I'm not quite sure about what would happen if the pin is connected. It might not work, or it might work with just some extra power consumption.

[mikey]

I'm not quite sure about what would happen if the pin is connected. It might not work, or it might work with just some extra power consumption.

 

I don't know either but may I suggest that there's a pullup and a HLT on the same wire.

Wouldn't putting SSO output on this wire disrupt the HLT signal?

[re-atari]

I was working on a some dead 1050's today and I found two drives that had pin 25 missing on the 2797. But one of the two drives works perfectly. Then I found another working 1050 that had a 2797, but with pin 25 intact. Does anyone know why the pin 25 might have been was removed. I don't believe it is just a coincidence that pin 25 was missing on two separate drives. (?)

' Back then' a friend of mine had a 1050 with a 2797, pin 25 was snippped off there as well. IIRC (can't find the datasheet right now) this pin controls side select on drives with 2 RW-heads.

 

re-atari

[+Larry]

Ok. I seem to remember this coming up somewhere before. Found a datasheet:

http://www.retrotechnology.com/herbs_stuff/fdc_datasheet.pdf

This shows the SSO (2797 FDC), and ENMF (NOT) (2793 FDC).

However, I'm not knowledgeable enough to interpret what that difference in pin use really means. Can someone "interpret?"

-Larry

 

I'm not quite sure about what would happen if the pin is connected. It might not work, or it might work with just some extra power consumption.

 

I don't know either but may I suggest that there's a pullup and a HLT on the same wire.

Wouldn't putting SSO output on this wire disrupt the HLT signal?

[Bryan]

This shows the SSO (2797 FDC), and ENMF (NOT) (2793 FDC).

However, I'm not knowledgeable enough to interpret what that difference in pin use really means. Can someone "interpret?"

 

2797 - SSO = Side Select (output). If you select the top or bottom head using the 2797 registers, then you connect this pin to the drive.

2793 - ENMF = Enable Mini Floppy (input). Normally, the clock is divided by 2 to produce the correct bit-rate for 5.25" floppies. Pulling this pin low causes the data rate to double.

 

The data sheet says ENMF is for 8" drives, although I wonder if this controller would support 3.5" HD floppies in fast mode.

 

EDIT: The datasheet is confusing regarding !ENMF. It says 2MHz is used for 8" media and 1MHz for 5.25" media (mini-floppies) [see CLOCK on page 2]. Since !ENMF is designated as a low-enable, you would think that pulling it low would put you in 1MHz mode, but on page 3 [ENABLE MINI-FLOPPY] it says the pin is pulled high (logic 1) to enable mini-floppies. It looks like they have indicated the wrong logic level on the pin name.

Edited March 14, 2010 by Bryan

[+bob1200xl]

It is confusing, isn't it? But, everything stated is correct. The -ENMF line, when pulled low, divides the input clock by 2. So, if you are running both HD 8" drives and DD 5.25s, you feed a 2MHZ chip clock into pin 24 and, when you are talking to an HD drive, you leave -ENMF high for a 2MHZ data rate. To switch to a 5.25 DD drive, you pull -ENMF low for a 1MHZ data clock. If you are only talking to 5.25 DD drives, you can feed a 1MHZ clock into pin 24 and keep -ENMF high. (which is what the 1050 does) Could you input a 2MHZ clock and use -ENMF to run either kind of drive? Certainly. SD, DD and HD are all available.

 

Don't know how you could run with the SSO pin connected. It defaults to a "0" on power up.

 

Bob

 

 

This shows the SSO (2797 FDC), and ENMF (NOT) (2793 FDC).

However, I'm not knowledgeable enough to interpret what that difference in pin use really means. Can someone "interpret?"

 

2797 - SSO = Side Select (output). If you select the top or bottom head using the 2797 registers, then you connect this pin to the drive.

2793 - ENMF = Enable Mini Floppy (input). Normally, the clock is divided by 2 to produce the correct bit-rate for 5.25" floppies. Pulling this pin low causes the data rate to double.

 

The data sheet says ENMF is for 8" drives, although I wonder if this controller would support 3.5" HD floppies in fast mode.

 

EDIT: The datasheet is confusing regarding !ENMF. It says 2MHz is used for 8" media and 1MHz for 5.25" media (mini-floppies) [see CLOCK on page 2]. Since !ENMF is designated as a low-enable, you would think that pulling it low would put you in 1MHz mode, but on page 3 [ENABLE MINI-FLOPPY] it says the pin is pulled high (logic 1) to enable mini-floppies. It looks like they have indicated the wrong logic level on the pin name.

[+Larry]

All good info about pin 25 -- sounds to me like its safer to just snip it.

 

And thanks, rdhays82604, it does sound like a bad connection, so I'll start checking/re-flowing the solder joints.

 

-Larry

 

Sounds to me like you have a loose solder joint. Give them a check.

[+bob1200xl]

No, don't cut it off! Just bend it out a little so it doesn't enter the socket.

 

You may need that SSO output at some later date...

 

 

Bob

 

 

All good info about pin 25 -- sounds to me like its safer to just snip it.

 

And thanks, rdhays82604, it does sound like a bad connection, so I'll start checking/re-flowing the solder joints.

 

-Larry

 

Sounds to me like you have a loose solder joint. Give them a check.

[Mathy]

Hello Bob

 

You may need that SSO output at some later date...

Do tell!!!

 

greetings

 

Mathy

[Bryan]

The good news is that you can create your own side select if you have any extra I/O in your circuit, and then you too can use either controller.

[Rybags]

I wonder... with a custom firmware and a bit of extra hardware, could you mod the 1050 electonics to control two drives using side-select to choose the active drive.

 

Chuck in some extra RAM and build in a warp-copy mode too.

[Bryan]

I wonder... with a custom firmware and a bit of extra hardware, could you mod the 1050 electonics to control two drives using side-select to choose the active drive.

 

Chuck in some extra RAM and build in a warp-copy mode too.

The most complicated part would be switching all the raw signals since there's no controller on the drive. There are some nice high-speed, low-resistance bilateral switch chips out there that could route the signals to the heads. Now all you need is a double-high 1050 case mold to make the XL version of an 815.

[+bob1200xl]

You would be better off using standard 34 pin drives, I think.

 

Bob

 

 

 

I wonder... with a custom firmware and a bit of extra hardware, could you mod the 1050 electonics to control two drives using side-select to choose the active drive.

 

Chuck in some extra RAM and build in a warp-copy mode too.

The most complicated part would be switching all the raw signals since there's no controller on the drive. There are some nice high-speed, low-resistance bilateral switch chips out there that could route the signals to the heads. Now all you need is a double-high 1050 case mold to make the XL version of an 815.

[Fox-1 / mnx]

I wonder... with a custom firmware and a bit of extra hardware, could you mod the 1050 electonics to control two drives using side-select to choose the active drive.

I was wondering how long it would take until someone came up with such a question and I'm a bit disappointed. Why did it take so long? :-)

 

 

 

 

(note: this is a positive reply)

Edited March 16, 2010 by Fox-1 / mnx