Firing up the pdt11

[Chris Zach]

Curiouser and curiouser....

On a lark I got a 47 ohm (actual reading 50.1 ohms) resistor from the 
box, put it on the anode supply line of the sector pulse LED on the 
drive that is in the PDT11 (not the one with the blown LED) and hooked 
it up to the power supply.

With the resistor in I fired up the supply, set the voltage to 3v, and 
checked the current. Now the LED only draws about 30ma measured at the 
supply instead of over 120ma without the resistor. So, different.

Then I put this assembly into the actual PDT11 circuit: Now when I check 
voltage at the resistor point I see 1.9v across the diode instead of 5v. 
And when I put a disk in the PDT11 does drop the solenoid down and tries 
to "read" the disk (but fails, but it is 8 slow failures, and not 8 fast 
ones). So I think the LED is now "lit".

Then I looked closely at the wires on the actual floppy drive to board 
interface. I see on the *other* plug (the one that runs the track zero 
detection) that there *is* a small resistor in line on the harness 
itself. Haven't checked the value but I'm guessing it's 68 ohms or so. 
But there is no similar resistor on the line to the sector pulse LED. I 
just checked the broken drive, same deal.

Maybe they "forgot" the resistor on the sector pulse LED? They didn't 
put it on the board or I wouldn't see +5 with the LED lit. More 
interestingly maybe the actual disk assembly for a PDT11 is 
*deliberately different* from an RX01 so you would have to order the 
"right" specific part. I don't know; I haven't taken apart my RX02 to 
see if that resistor is in line on the disk drive itself for either of 
these.

It could also have been an oversight, I'll check the top disk to see if 
the resistor is visible. Maybe the LED works at full +5v brightness for 
awhile, then opens at +5 but is still closed at lower voltages and lights.

This brings up another issue: As these are soft-sectored, does the 
amount of light the LED produces matter? Maybe my LED+resistor is bright 
enough to trigger sector pulses, but they are now not where the data is 
anymore? How critical is sector pulse alignment for an 8 inch floppy drive.

This is progress, and this is interesting. I wonder how deep this rabbit 
hole goes...

CZ



On 10/12/2020 11:11 AM, Tony Duell wrote:
> On Mon, Oct 12, 2020 at 4:04 PM Chris Zach <cz at alembic.crystel.com> wrote:
>>
>>> In a 'real' RX01, each sensor LED (track 0 and index) has the cathode
>>> grounded and the anode connected to +5V via a 68ohm series resistor.
>>> If you measure the voltage across the pins with the LED disconnected
>>> (or if the LED is open-circuit) with any reasonable meter, it'll read
>>> 5V. The meter will not draw enough current for the resistor to drop a
>>> noticeable voltage.
>>
>> Ok, that makes sense. Is the resistor on the board or on the drive wire
>> harness somewhere?
> 
> I'm pretty sure it's on the PCB.
> 
>>
>> I'll jumper the disk 0 sensor into my test board here (actually just the
>> extender board from another PDT11) and see what the voltage looks like,
>> then compare it to the same spot on the second drive.
> 
> I'd look at the voltage on each pin of the LED in the 3 other sensors
> (index and track 0 for both drives)
> 
> -tony
>