• Please review our updated Terms and Rules here

Pertec T8640A Tape Transport

Mal

Experienced Member
Joined
Apr 3, 2008
Messages
91
Just wondering if anyone here can help with some high-level information about hooking up Pertec T8640A Tape Transports to a QBUS (or UNIBUS) PDP-11. I have searched these Forums without much success.

I have 3 x T8640A Tape Transports. Each one has 3 x 36-pin edge connectors on them.

One T8640A has an Emulex cable kit attached to it. This connects the 3 x 36-pin connectors through to 2 x 40-pin (not 50-pin) header connectors via two lengths of twisted-pair ribbon cable. But it's not clear to me how these two 40-pin headers are supposed to connect to a QBUS system. I assume this Tape Transport was once attached to a QBUS system, as it is mounted in an otherwise almost complete QBUS rack.

I have another T8640A that has 3 grey ribbon cables leading from the 36-way connectors to a small PCB that has a 100-pin (S-100 style) female edge connector. This presumably connected to an external Pertec Tape Formatter (such as an F8x9 series, which I also have). But this Tape Formatter then has another 100-pin edge connector that connects to the "Tape Controller". I am not sure whether there was a QBUS or UNIBUS board that worked with these drives.

Any information you have as to how to connect either of these T8640A's to a PDP-11 system would be appreciated.

Here's a link to a page with more photos and information about this system -> Pertec T8640A Tape Transport

t8640a-front-cropped.jpg
 

Al Kossow

Documentation Wizard
Joined
Sep 1, 2006
Messages
2,874
Location
Silicon Valley
Just wondering if anyone here can help with some high-level information about hooking up Pertec T8640A Tape Transports to a QBUS (or UNIBUS) PDP-11. I have searched these Forums without much success.

I have 3 x T8640A Tape Transports. Each one has 3 x 36-pin edge connectors on them.

One T8640A has an Emulex cable kit attached to it. This connects the 3 x 36-pin connectors through to 2 x 40-pin (not 50-pin) header connectors via two lengths of twisted-pair ribbon cable. But it's not clear to me how these two 40-pin headers are supposed to connect to a QBUS system. I assume this Tape Transport was once attached to a QBUS system, as it is mounted in an otherwise almost complete QBUS rack.

I have another T8640A that has 3 grey ribbon cables leading from the 36-way connectors to a small PCB that has a 100-pin (S-100 style) female edge connector. This presumably connected to an external Pertec Tape Formatter (such as an F8x9 series, which I also have). But this Tape Formatter then has another 100-pin edge connector that connects to the "Tape Controller". I am not sure whether there was a QBUS or UNIBUS board that worked with these drives.

Any information you have as to how to connect either of these T8640A's to a PDP-11 system would be appreciated.

Here's a link to a page with more photos and information about this system -> Pertec T8640A Tape Transport

View attachment 35194

the connectors on the drives are unformatted, read, write, and control. They would either go into one of the connectors on the pertec formatter, or go directly to a qbus interface that accepted an unformatted drive input. the most common type of qbus interface has 'pertec formated' inputs.. the back end of a formatter or tape transport with two connectors coming out of it. to use the formatter you have, you have to break out the other 100 pin edge connector to two ribbon cables, the pinouts of which should be in any manual for a qbus formatted tape coupler. there are dozens of manuals on bitsavers.org to choose from.

or.. find an unformatted tape input qbus controller, which aren't very common.
it probably was used with an Emulex TC01 controller. manual is TC0151001-D_TC01_Oct82.pdf on bitsavers.
the connections to the TU11 cables are on page 4-08

also, the drives are NRZI only (no 1600bpi phase-encoding support), and the controller only supports 18 bit addressing

copies of the schematics would be nice to archive, my copy of the manual was incomplete
unfortunately, they are difficult to copy because they are so long
 
Last edited:

Mal

Experienced Member
Joined
Apr 3, 2008
Messages
91
Al, that is very helpful information, thank you. I've managed to dig up a QBUS 2-board set that consists of an Emulex TC01 and the optional PE board for it. The T8640A also seems to have a PE option board fitted to it, so maybe 1600bpi PE will be possible. I'll post an update here once I've had a chance to check out all the connections and cable it up.

I will also work on scanning those long-form schematics. I had good results recently scanning and stitching together long-form schematics for an electro-mechanical pinball machine. I'll try the same technique.
 

Doug G

Veteran Member
Joined
Jun 16, 2012
Messages
541
Location
SoCal
Our group at CDC maintained a PDP-11/34a with a Pertec tape drive that looked like the one in your picture, but we weren't responsible for the tape subsystem. My recollection is the unibus tape controller was Pertec manufactured.
 

Qbus

Veteran Member
Joined
Feb 23, 2011
Messages
921
Location
Salisbury Maryland
My 11/23 has a Pertec controller for my CDC tape drive but it’s a standard dual fifty pin connector setup. Runs without issues on the Q bus system running RT-11 ver 5 and responds to the alias of MS0 The interesting thing is that the tape drive and controller were once part of a MicroVax 2 system and required almost no work to get them to work with the Q Bus.
 

Mal

Experienced Member
Joined
Apr 3, 2008
Messages
91
Well, I have the Pertec tape transport almost working on RT-11. Lots of trial and error so far as tapes were concerned. Many original DEC or DECUS tapes would give a partial directory read, leaving behind residue on the heads (which needed to be cleaned after each tape pass). On other DEC or DECUS tapes I seemed to be able to read the entire directory, though I suppose I can never be sure whether the controller/drive gave up before reaching the true EOT as I don't have a listing of what was supposed to be on each tape.

I then tried writing to some blank tapes. I chose two that were still in sealed plastic bags.

On each tape, when I first copied an entire RT-11 distribution (COPY/SYSTEM DD0: MT0:/POSITION:-1) to the tape, both tapes seemed to be written out successfully and both yielded a full directory with the expected number of files, when read back with a DIR command. I did one tape at 800 BPI and the other at 1600 BPI.

But things have since gone down hill somewhat from there....

When I try to write over the same tapes (after doing an INIT MT0: first), I have no trouble writing out file-at-a-time from the RT-11 command line. But if I try to write an entire RT-11 distribution again, the Pertec transport gives up after about 9-12 files. It lets go of the tape tension and goes offline. RT-11 hangs until you press CTRL-C twice.

I'm at a bit of a loss to understand what's going on here. Perhaps the Pertec transport isn't sufficiently erasing the old data, before writing the new data? Or should I perhaps be bulk erasing tapes before attempting to re-use them?

I've tried setting this up in 3 different QBUS backplanes (two of which are known to be reliable), with the same results. I've also tried adding heavy earth straps between the QBUS chassis and Pertec chassis, and also running all components from the one power distribution panel. No noticeable improvement in reliability.

Any suggestions?
 
Top