[GreenKeys] Associated Press Special Dataspeed Receivers

[Don Robert House]

Jim,

Thanks for the history on the DRPE.  We had good results with the  
DRPE.  We had many of them in service.
The worst problem was the operator failing to empty the chad bag.  It  
would back up into the punch.
I still have my punch cleaners for the different punches.  In the data  
test center I installed both a Type 5
sender and receiver and a Type 2 sender and receiver.  We also had  
installed the discrete generator
and recognizer that you are famous for designing.  When testing to one  
of those Dataspeeds we kept
a package full of those round discs to cut for the different company  
confidential codes.

The Type 4 with error detection and correction was a sight to behold.   
At the TTY Corp school on Wells
Street I attended the installation and maintenance class on the  
beasts.  The maintenance kits just for
the sender cost over $2500.  I got in trouble for ordering two of them  
for the two of us trained on the
one we had in service.  Penn Precision Products was our customer.   
They made precision parts for
NASA.  The majority of the Type 4 receivers were in Chicago and the  
technicians there spent much time
with the receivers.

Watching the machines was like a Rube Goldberg machine on steroids.   
Someone should have made
a video tape or movie of them.  The machines in the school sat next to  
each other and used 400 type
data sets.  They had the darn things running at 2000 bps.  The  
instructor would open the cable between
the machines and you know what happened.  The temporary open would  
spoil the generated parity
check words which is how the two machines sent blocks of data with  
both serial and circular parity.
So the receiver would turn off the sender, which in turn would back  
the tape up exactly the number of
blocks that were errored.  That was amazing, but the receiver would  
then back it's tape up through
the punch block and over punch the errored blocks with Rubouts.  Then  
the receiver would turn the
sender back on again. They made the most unusual sound when  
operating.  The use of the block
by block check words would interrupt the punch ever so slightly so the  
sound was high and low
like a metronome.  Verrr rumm Verr rumm Verr rumm almost hypnotic.

Getting back to the BRPE and the GRPE.  Both used the same punch and  
the same motor.  The
GRPE was different only in the mounting.  The BRPE was higher than it  
was wide where the GRPE
was laid out horizontally with the motor to the side of the punch with  
a belt drive.  Someone once
told me it was designed that way to be mounted in an aircraft.  The  
standard BRPE apparently did
not fit in the equipment racks on USAF and NOAA aircraft.

Now that I have put everyone to sleep I will sign back off.

Don


On Fri, 9 Jan 2009, Don Robert House wrote:

> Six level advanced feed hole Type 2 Dataspeed receivers.
>
> I designed the circuit and station equipment for the last installation
> at Paddock Publications in Arlington Heights, IL in the 70's
> Clever subset allowed for switch settings depending on the type of
> news to be received.  I have always hoped that at least one would
> still be around.
>
> ...  USAF and the Atomic
> Energy Commission also used the BRPE AND GRPE.
> My favorite was the DRPE.  A really fast punch.  More than 1000 words
> a minute.

GRPE?  That's one I don't know.

The DRPE was in somewhat the same category as the Inktronic, a kluge
(too clever for its own good).  The idea was that the electromagnet
holds up a metal reed. Cutting off the current releases the reed,
which swings down, drives the punch pin through the tape, and swings
up and the magnet grabs it again.  One problem was that the reeds
need a lot of magnetism to pull them up when power is first applied
and they are in the neutral position far from the magnet pole pieces.
But it seems and looks like such a simple thing, just a magnet and
reed for each punch pin and the only motor is a little one used to
feed the tape.

The worst problem was eddy current generation in the solid metal reeds.
They would get red hot.  The solution was to grind out a depression
in the reed and fill it with a ceramic magnetic material.  At first
these inserts were glued in with epoxy, but the bond tended to fail,
so they had to be held in with pins.

Another problem was the acoustic noise level, much worse than any
previous product.  I remember when the punch was being developed in
the old Wrightwood building in 1959 - sounded like somebody was
running an (old model, poorly muffled) power lawnmower in the lab.
It took lots and lots of padding and muffling to get the thing to
be tolerable for use in an office.  The padding was probably that
same stuff that turns to crumbs when it ages.

There were three generations of drivers for the DRPE.  The first, used
only in a demonstration model, used conventional high voltage high
resistance, just like in a selector magnet loop, to get fast
response.  It was tremendously wasteful of power.  The second design
used some big inductors, and a scheme of shuttling energy between the
punch magnet and the inductor, for a considerable saving in energy.
This was used in the FAA Service B switching system, which I believe
was the first production use of the DRPE, and in the high speed ASR
set for Long Lines Project 176, another government system.  The third
design used a brief pulse of high voltage to pull the reed toward the
magnet and then a lower voltage to hold the reed attracted.  This was
used in Dataspeed Type 5, and later in Type 4.

Oh, yeah, the advanced feed holes of 6 level Teletypesetter operation.
What that means is, in 5 through 8 level paper tape the centers of the
small tape feed holes are on a line that passes through the centers
of the code holes.  But in 6 level Teletypesetter the front edges of
the feed holes are tangent to a line that is tangent to the front
edges of the code holes, so the feed holes are "advanced" in comparison
to the other way.  I have no idea how it got to be this way, but this
is the way it was.  So we had in mind a reader and punch with user-
adjustable tape guides to handle any tape width from 5 to 8 levels;
except it wouldn't work for 6 levels because of the requirement for
advanced feed holes.

The BRPE punch was popular in a lot of applications, such as computers,
to the extent they used paper tape at all.  Tally Register Co. was a
competitor of Teletype for this kind of product.