17 Oct
2003
17 Oct
'03
3:56 a.m.
Message: 16
Date: Sat, 11 Oct 2003 09:04:11 -0400
From: Ian Primus <ian_primus(a)yahoo.com>
Subject: Re: ASR33 Teletype interfacing
To: "General Discussion: On-Topic and Off-Topic Posts"
<cctalk(a)classiccmp.org>
Message-ID: <67FFB9E5-FBEB-11D7-86CD-000393D7845A(a)yahoo.com>
Content-Type: text/plain; charset=US-ASCII; format=flowed
I am late in an response to this ASR33 discussion but the comments here
definitely hit some memory banks buried deep.
In the early 70's, I recall that 120VDC was the source for 60ma current
loop to teletype, and it seems to me that "full-duplex" had the send
side sourced from the teletype and the receive side sourced from the
connected transmission equipment. There was something special that was
done to "echo" at the local printout any typed keys, It seems that it
was jumpers at the transmission equipment between the incoming "xmit"
current and the outgoing "Rcvd" current, again it must have put them in
series. It also rings in my head that the interfaces changed when 20ma
came out and they could not be intermixed. A machine would be either 60ma
or 20ma and required an overhaul to change it. 20ma was a result of reduced
physical part requirements for robustness and EMF splatter reduction.
Test equipment of the time put out "RY's", and that was almost a perfect
square wave at 120Vdc, 60ma as I remember. I have conveniently forgotten
most of what we did to repair these units, but I do remember a tank where
the units were given a bath in something??? If air couldn't get things
cleared and freed up, it got a bath then a complete relube???!!!!
The comments to "running open were correct. That is the sign that no
current was available to the unit. The current kept it "held".
On Thursday, October 9, 2003, at 04:52 PM, Tom Jennings wrote:
Somehow I'm not surprised. Something told me that it was too simple to
work properly.
I have been
attempting to get my ASR33 teletype connected to something
and communicating, but so far I have not been successful. I have built
the interface here :
http://www.daedalus.co.nz/~don/computing/20mahack.html
It wont work, sorry...
Teletypes are inductive loads. Though they only want
20 mils, the
voltage needs to be high to get the initial magnet pull-in (basic RL
theory). ASR33 loops were generally run at 100V or so, but I run my
Model 28 at 14V, with non-perfect error rate, and I don't use the
keyboard.
I take it that the voltage isn't that crucial, just the current?
The keyboard and printer are IN SERIES. If you hit
keys while it's
printing you foul it up. Normal.
What about on a full duplex machine? Is it the same, or are they
separate?
Because it's inductive, it makes a spike when yuo
turn the voltage off.
You need to suppress this with a diode, a resistor and capacitor, for
example.
They're not subtle interfaces, and weren't meant to be.
If you just want to print, you can rig up a power transistor, two
resistors, a diode, and a high-voltage DC power supply to do the trick,
and drive it from the serial port.
If you want to receieve also, you can use another transistor and
resistor to pick off the change in loop current that happens when you
press keys which open the loop, and drive the serial port.
I've done one of these fairly recently, and if poked with a
not-too-sharp stick, I'll scan the schematic and pu on my website.
*poke*
<grin>
If it's not too much trouble, I'd like to see it. From what I have
heard, there are lots of ways to do this, and i would be very
interested to see how you did it.
Thanks!
Ian Primus
ian_primus(a)yahoo.com
---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.528 / Virus Database: 324 - Release Date: 10/16/2003