Interfacing PDP 11/05 to VT 50

[Joseph Lang]

The current loop is actually "proper" .
There are 3 parts to current loop 
1) transmitter (switch)
2) receiver (opto coupler in dec stuff)
3) current source

You will have problems if things don't match. An active transmitter has to connect to a passive receiver. 
Passive transmitter to active receiver.
Active means it has  a current source,passive no current source. 

The 11/05 is active tx active rx. So expects contacts on the keyboard and the selector magnet in the printer. 
Most industrial current loop converters are not this way out of the box.

Joe

> On Dec 1, 2015, at 1:56 PM, Paul Koning <paulkoning at comcast.net> wrote:
> 
> 
>> On Dec 1, 2015, at 1:22 PM, william degnan <billdegnan at gmail.com> wrote:
>> 
>> ...
>> Sorry about the wording of my question.  Thanks for the replies.  I was
>> only able to get the VT50 to receive, I could not send.  So I decided to
>> research the problem.  I found the link above, the author of the page says
>> in effect 20mA did not work (for him) as desired.  So I was wondering if
>> anyone here has been successful to attach a M9970 to a VT50 or VT52.  I
>> spend more time on it, but I was curious if it was even worth it given the
>> hardware.
>> 
>> The author writes:
>> 
>> -start quote--
>> 
>> I first tried to connect a PC with the 11/05 over a industrial
>> RS232-to-20mA converter, but this failed.
>> 
>> A 20mA interface works by one side providing a 20mA current, which drives
>> receiver and transmitter of the closed loops for Transmit and Receive.
>> 
>> But the 20mA interface of the PDP-11/05 is not a proper one: the receiver
>> is more like a low impedance voltage sensor, while the transmitter simply
>> switches voltage at the levels +3.5V to -15 V.  At best you call the
>> PDP-11/05 serial interface a "TTY interface": it is well suited to read
>> data generated by mechanical switches to GND, and driving solenoids for
>> transmit data.
>> -end quote--
> 
> Hm.  I wonder by what definition it is not "proper".  A current loop transmitter, in an electronic terminal as opposed to a mechanical one, would be a switching transistor that turns the current on and off.  The only issues I can think of are the voltage rating used, and the polarity.  For example, in earlier 60 mA loops (for the Model 15 Teletype and machines of that era) you might find loop voltages around 100 volts or so, with a big series resistor.  The purpose was to reduce the distortion from the inductance in the loop (in the receive solenoid).  In the later 20 mA loops I would expect lower voltages to be used.  But the specification is pretty wide open -- "20 mA" is really about all there is.  You can build conforming devices with optocouplers, switching transistors, solenoids and cam operated switches, etc.
> 
> I don't know what "simply switches voltage at the levels +3.5V to -15 V" means.  The voltages at the pins aren't directly relevant, only the voltage difference between the two transmit pins -- or more precisely, the resulting current in the loop.  I could easily imagine that in the off state both pins are at -15 while in the on state one goes to +3.5, or something like that, driving through a resistor of a bit under 1000 ohms.  
> 
> The fact that the "industrial converter" didn't work does not imply that the VT50 would have troubles.  Just the opposite, actually; I would assume that DEC would make sure that its 20 mA terminals work with its 20 mA host interfaces. 
> 
> Did you check the voltage and current in the two loops when you hooked it up?  I wonder if you might have the polarity backwards on one of the circuits, assuming that the devices involved care.
> 
>    paul
>