8" drive support on PC's

Randy McLaughlin cctalk at randy482.com
Thu Mar 17 16:04:36 CST 2005

From: "Dwight K. Elvey" <dwight.elvey at amd.com>
Sent: Thursday, March 17, 2005 2:06 PM
> >From: "Randy McLaughlin" <cctalk at randy482.com>
>>On the Nec 765 (Intel 8272) the direction pin is also use for reduced 
>>current (TG43).
>>When the step line pulses the direction pin controls the direction.  All
>>other times it is high if the track counter is 43 or less, low if the 
>>counter is greater than 43.
>>The current disk controllers are based on the 765.
>>Does anyone know if the direction pin does the same on the newer clone
>>chips?  I have started looking at the datasheets but so far I have not 
>>any reference to that function on "modern" parts.
>>If it's true then the 34 to 50 pin cable I use (and others) can be made to
>>include TG43 without using an adapter like FDADAP.
>>It may turn out some do and some don't but since everyone is basically 
>>the same licensed core I would expect all newer chips to act the same. 
>>question is was TG43 support kept from the original 765?
> Hi Randy
> See web page:
> http://www.dbit.com/fdadap.html
> You'll note that it mentions that a normal PC does not
> supply the TG43 signal. Their adapter includes what looks like
> a uP to keep track of the current track to generate this
> signal for writing to 8 inch disk.
> If you were doing software to write 8 inch disk, I would guess that
> you could double up this function on the step direction wire.
> There is otherwise no signal to do this for you.
> Don Maslin's connection was primarily to be used for reading
> 9 inch disk. It may also be assuming that the drive has a built
> in track 43 sensor.
> I think this is were the confusion over pin 2 comes in. This
> is usually uses as a speed control on PC's but historically
> it was a write current control.
> Dwight

There is confusion over pin 2 use at least in my head.  The first use was 
with the SA800 interface for reduced write current.  Today on the 34 pin 
connector that is what it is now used for even for 3.5" drives. 
Traditionally the write current line was used completely differently from 
how it is used today.  Traditionally it changed according to where on the 
disk the data is being written, for today's drives it is for the entire 

I finally found in one of the Micro-Cornucopia magazines more references to 
the 5.25" HD drives stating that they were designed to be two speed specific 
brands are listed.  That with low speed they acted as quad density drives 
and with high speed they were similar to 8" drives.  Twenty years after the 
fact it is difficult to find the truth about something no longer greatly 

That is a side issue to my question about 765 TG43 controls.

On a real 765 it generates the reduced write signal and outputs it on the 
same controller pin as the direction signal.

Today the 765 is emulated via VHDL code and I am curious as to whether the 
TG43 logic was kept.  I can and will experiment on my systems but I would 
like to know if others are familiar with the issue.

Obviously all PC's that use a real 765/8272 does generate the signal and 
does bring it out to the 34 pin connector, it just happened to be hidden on 
pin 18.  I question if it is still there?

The FDADAP states that "This signal is not normally provided by PC FDCs" 
obviously at the very least many did maybe all.  The FDADAP does more than 
just generate TG43 it also greatly simplifies wiring.


More information about the cctech mailing list