Urgent Current Loop Question - ASR 33 / PT08 - test-drb@ccmp.vtda.org

17 Oct 1999

That's it Tony, and yes... I'm serious. They connected the ASR-33 across an
inverter and shorted it out when the relay transmits. The R107 is RTL (with
a real transistor) so it can't damage it.

I made the change and...... the PDP 8/s is now running DECUS tic tac
toe!!!!!!!!!!!!!!!!!!!!!!

I got the docs finally on the PT08 and found a -3V reference supply was in
the wrong slot... That would account for the problem I had initially.

Otherwise the entire system is now happy.

BTW: I also picked up the ASR 33 schematics and the "Teletype Service
Manual - ASR 33" as well.

Thanks for the help.

john

-----Original Message-----
From: Tony Duell &lt;ard(a)p850ug1.demon.co.uk&gt;
To: Discussion re-collecting of classic computers
&lt;classiccmp(a)u.washington.edu&gt;
Date: Saturday, October 16, 1999 2:33 PM
Subject: Re: Urgent Current Loop Question - ASR 33 / PT08

...
 >
> Before I attempt to figure this one out I would like to know if anyone else
...
   can help me.

I know the ASR33 pretty well, but I've never seen an 8/s. But maybe I can
help. I'm going to assume you have the sort of ASR33 that's normally used
with computers -- that's to say one with a current loop interface.
There's s 3 position knob on the front of the machine (line/off/local)
and no other controls on the 'call control unit' (electronics module).

>
> I hooked up an ASR-33 (non DEC) to my PDP-8/s. I ran a maindec program and
...
 > the PDP could talk to the teletype no problem,
never an error.
>
> After spending 4 1/2 hours cleaning up bad solder joints on the receiver I
...
 > got it up and running.... *kind of*
>
> The problem is this: The ASR-33 is interfaced to the PDP currently by tying
...
 > one receive line to the input of a R107
(transistor inverter) and the other
...
   receiver line
to the output of the same inverter. Then, the signal is
 *conditioned* (inverted) a few times before it is passed onto shift
 registers,,, etc.. 
Are you sure this is right. It sounds crazy to me (and now I'll get a
pile of replies pointing out how DEC were being clever here)...

The ASR33 transmit loop is electrically a switch. Actually it's a complex
array of switches. but in the end, it either shorts the 2 Tx terminals or
opens them. There is no voltage on either terminal (in a 'stock ASR33')
from the ASR33 electronics. There are no other components involved.

Connecting a switch between the input and output of an inverter is not
normally a good way to get a clean logic signal ;-)

The normal circuit (used in all sorts of machines) is to connect one of
the tty Tx wires to one supply rail, possibly through a small resistor
to limit the current in the event of a short somewhere). The other tty Tx
wire to the other supply rail through a (larger) resistor. And then to
monitor the voltage at the 'top' of that resistor, say by connecting it
to the input of a logic gate.

>
> If I hook up the teletype the WAY the circuit exists now I get a 0-1V data
...
 > signal (notice positive) out of the first inverter
which of course is
> totally incompatible with the others and the signal never passes through
> beyond the first inverter.
>
> If I disconnect the FIRST inverter from the second one and LEAVE the one
> receive line in the input of the first inverter ("floating") and hook up
the
...
 > SECOND receive line to the input of the now
*disconnected* second inverter I
...
   get data. Most
of the time good, but some bad characters get through.

 This is obviously not a working current loop solution.

 If the teletype is passive should I just connect the receive line on the
 input of the first inverter to ground or -3V??? 
Well, I could understand connecting the tty Tx loop between the input of
one inverter and the output of another one, especially if the input of
the latter was tied to a known logic level (so the output is just a
constant voltage source, etc). This makes an assumption about how
floating inputs behave, which may be OK on transistorised logic like this.

You are _sure_ that you've connected to the right pins on the R107
module, and that in fact that's the right module to be using?

> #1) As I don't have docs nearby on the ASR-33 and have never had a VERY
> simple interface like this CAN I short the one receive line to ground or -3
...
   volts????

Sure. Just make sure the TTY contacts don't have to carry high currents
(like don't connect the loop across the PSU) unless you like repairing
burnt/melted switches...

>
> I just want to make sure the ASR 33 does not have any active part that will
...
   eat my PT08
for breakfast. 
Not that I know of...

-tony