HP2748A papertape reader : connector ?

From: Bob Shannon <bshannon_at_tiac.net>
Date: Mon Jul 2 19:13:52 2001

I've done some digging, and I think I have some new info...

Jay West wrote:

> Bob wrote...
> > "Incorrect" may have been a poor choice of words....however I have
> documentation
> >
> > that shows a HP2100 shipped by HP with the older reader interface board.
> In
> > fact
> > one of the reader boards I have was removed from a basket-case 2100
> system.
> >
> > I wounder when HP switched over, and why?
>
> No Kidding!? That's truely interesting. You have my curiosity up, I'd love
> to know what the deal is for my own future reference.

Apparently HP switched from the dedicated tape reader board sometime in the
early 1970's. I located a guide to HP interface cards that describes the 8 bit
duplex
board as being used with the HP2748A and B, as well as the tape punch.

The older 'Tape Reader' board could never drive a punch, its only an 8 bit
transistor
drive input port, with flag and control bits.

My current guess is that sometime during the transition from the 2100 to the M
series
machines, when logic had gotten cheap, they dropped the tape reader board and
used the 8 bit duplex board as both the reader replacement interface, and also
as the
punch interface.

But I gather that if you had a punch and reader on a M-series box, you had 2 8
bit duplex
boards...is this correct?

The old-style tape reader board also interfaced to the HP2847 readers, which had
the
same ITT-Cannon connector as the '48's.

> I do have a lot of sales literature and marketing collateral for these
> systems, and all of them show the 2748B paper tape reader as using an 8-bit
> duplex register board. Hummmmmmm.... You know what - is it possible that the
> ones that use this "tape reader interface board" that you've seen are the
> 2748A reader and the B version uses the 8-bit duplex register board? All the
> literature and docs I have shows the 2748B, I have nothing that relates to
> the 2748A which is why I'm conjecturing this. One wouldn't think that HP
> would release a board with the same electrical specs as the 8-bit duplex
> board, so wouldn't you think the electronics in the reader would have to be
> different for the two boards? Most interesting... could you check and see if
> your reader is the A one that uses this board Bob? Or maybe show us a pinout
> of the old style reader interface vs. the new style?

I can post the pin-out for the HP Tape Reader board, its totally different than
the
8 bit duplex board. There are 4 bits of data on the top left, and 4 bits on the
bottom
left of the I/O finger connector. Of course the duplex register has all 8 bits
of each
direction on one side of the board, etc.

As for readers themselves, I've had too many problems with the clutches, and
some
of my paper tapes are a bit fragile, so I copied them onto eproms and built an
emulator.
Frankly, I'm not sure I still have any physical HP readers. The last actual
reader I used
is an older Decitek...

At least one old HP reader was canabilized to make my first reader emulator...

> > Umm, how is the 2114 not an HP machine? (The 2114 was developed by HP.)
> >
> > Your thinking of the 2116 perhaps, which was originally developed by a
> small
> > firm that
> > HP later absorbed?
>
> I knew the first 21xx machine was already designed and put together by
> another company, and HP just absorbed the company. When they saw how
> sucessfull it was, they then came out with the other two. I was thinking
> this first one was the 2114, and the HP made ones were the 2115 and 2116. I
> stand humbly corrected - must be the 2116 that was absorbed and they made
> the other two. The 2100 did come later. Interesting tidbit - one of HP's
> largest users of the 2100 system was a company called "MeasureX". This
> company had it's own engineering staff and redesigned the 2100 to be a
> single cpu card. They sold this design back to HP and it became the 21MX
> line. The MX in 21MX stands for MeasureX. I got this directly from a phone
> call with the former VP of engineering of MeasureX, so I suspect it is
> accurate historically.

What we know today as the HP2116 machines were developed by Dymec Instruments.
This is the same firm that developed the HP2401 digital volt meter (with nixi
tubes!).

Dymec was absorbed into HP, and so began HP's computer systems division.

I may be mistaken, but I think a friend actually has a re-badged Dymec 2116!

>
> > at least 1 2100A shipped using the older style interface board
>
> I'm just wild guessing here, but is it possible that the 2100A that you got
> had the tape reader (old style) added later and wasn't actually sold that
> way from HP? The reason I'm wondering this is I have the sales/ordering
> guide for the 2100A and the 2100S, and both only specify the 8-bit duplex
> register. Most Strange! My earlier comments were based on the HP
> sales/ordering guide, and in retrospect you're quite right, those guides
> were almost certainly revised from time to time and earlier ones than mine
> might indeed use the old style reader board. Again, I'm wondering if it's
> not the A reader that used it?

Hmmm, that was a long time ago. I cannot say if the 2100A used the A or B
reader,
but the old-style reader board workes all the way back to the 2747 reader, so I
suspect
that the interface never changed.

The 8 bit duplex register board does use the same logic levels, assuming your
have
an inverted duplex register.

As for the 2100A I found with a old-style interface, that machine had come out
of MIT
so its an easy bet people had been screwing around with the configuration.

If your super-curious, I have the old-style board and cable. I suspect it will
plug into your
2748B and run perfectly.

> > What is your definition of a '2100 machine'?
> >
> > 2114, 2115, and 2116 machines run HP 2100 software, and were the original
> > processors
> > used in HP2000 time share basic systems. They originated the HP board
> interface
> > used
> > all the way up to the MX and E series. The software on the HP2100 archive
> site
> > is that
> > which originally shipped with 2114 thru 2116 systems.
> >
> > If these classic machines are not 2100's, then what are they?
>
> >From the sales/ordering guides that I have (but perhaps I'm reading them
> wrong or drawing incorrect conclusions): The 2100 machine is a 2100 machine.
> 2100 was not a family or class of machines, it was a specific machine (2100
> with option pack A or S). There is no difference in the A and S versions
> except the S version included options as standard which were also available
> from the outset on the A. From what I can gather, the 2114/5/6 machines were
> not referred to as "2100's", they were referred to as 2114/5/6. Part of my
> basis for this is logical - they wouldn't name the 2100 (which came later) a
> 2100 if that was the generic family name for the 2114/5/6 machines I'm
> guessing. The early nomenclature of HP's mini products is *SO* confusing...
> cpu's that were called 21MX's were later redesignated 1000 machines, etc.
> Plus the whole 2000 thing that doesn't refer to a computer, but rather a
> specific configuration of both hardware and software (for TSB). What a mess.
> What's your take on this Bob?
>
> Regards,
>
> Jay West

Hmmm, you could get a HP time share basic system with a 2116 (original version)
or
a 2100A, so to my eyes, the 2100 is simply another model in the same line as the
2114
through 2116.

But the 2100's look and feel, and style of construction is totally different
from the earlier
machines. I just don't know.
Received on Mon Jul 02 2001 - 19:13:52 BST