Logica VTS 2300 (Kennet)

 Home > Browse Our Collection > Computers > Logica > Logica VTS 2300 (Kennet)

[Neal Crook writes: I was hired as a graduate engineer by Logica VTS and worked there for just over a year before being made redundant. I was not involved in the original development of the Kennet A but did a lot of work on the machine as it moved into production. Here are my memories of this machine. Stuff marked ??thus?? indicates where my memory is hazy]

Logica VTS 2300 (Kennet A)

Logica Video Terminal Systems (Logica VTS) was a division of Logica plc, and was based in Drakes Way, Swindon. This site contained the hardware and software design and development teams, software validation/QA, manufacturing, drawing office, purchasing, canteen and other functions.

The Kennet was designed as a dedicated word-processor and a successor to the 2200 "Whirlwind" machine. It came to the market just at the time that the IBM PC became a credible threat. Although the Kennet was based on the Intel 8086 and could (and later did) run MS-DOS, it did not provide any degree of software compatibility with the IBM PC (at the time, the crucial measure of compatibility was: could you run Microsoft flight simulator. The Kennet could not).

There were 2 chassis variants: Kennet A (2300) which had 1 expansion slot and Kennet B which had ??4??

The Kennet B looked like a taller version of the Kennet A. Internally, it had a larger/higher capacity power supply.

There were 2 motherboard variants: a motherboard providing a monochrome display, and a motherboard providing a colour display.

The original and most common model was the monochrome Kennet A. It comprised a base unit, a monitor and a keyboard. The monitor had an amber phosphor and attached to the base unit with a cable and D-connector. The cable provided 12V power and the display signals. The keyboard was attached to the base unit by a variant of the BT modular jack (??or did it attach to the display??)

The case was plastic. After removing 4 screws (2 on each side) from the bottom tray, the cover could be slid forward (to clear the disk drive facias) and lifted off. There may have been a reinforcing metal plate attached to the inside of the cover. The cover was 4 separate mouldings, screwed together: a lid, a front facia and two side panels (one on each side).

The Kennet B used the same bottom tray and top, it used a different/taller front facia and stacked multiple side panels on each side to get the additional height.

The cream/grey colour scheme was used for Logica-branded machines. Machines badge-engineered for BT and ICL used a chocolate-brown facia around the monitor instead of the grey. All were manufactured in Swindon.

On power-up, the machine attempted to boot from disk. Usually, it would boot straight into the word processor, which had integrated file manager and print spooler. This software was entirely proprietary, and written in-house.

Internally, the unit had a switch-mode power supply in an open-frame metal chassis running the full length of the unit along the right-hand side. The IEC inlet on the back and the power rocker-switch on the front were both mounted in this chassis. Later units had a "nomex" (a kind on resin-impregnated cardboard) shield covering the high-voltage areas of the power supply.

To the left of the power supply, a backplane ran across the front of the machine and the rest of the internals of the Kennet A were a 4-layer sandwich. The bottom layer was the expansion slot. The next layer was the motherboard. The next layer was the Option Bus and the top layer was the drive tray.

The Kennet B was identical except that there were multiple expansion slots below the motherboard.

The drive tray could accommodate 5.25" floppy drives and Rodime 5MByte hard drives. There was space for 2 floppies, or for 1 floppy and 1 or 2 hard drives. The drives were powered by flying leads from the power supply. The data cables for the floppies connected down onto the motherboard below. When hard drives were fitted, there was a 3rd-party (Adaptec??) controller board mounted on the drive tray and connected down onto the motherboard (maybe via an O-bus card?)

The motherboard could be slid out from the rear of the unit by removing the metal back panel, but its removal also required the cover to be removed so that the disk drive ribbon cables could be disconnected from the motherboard.

A set of small daughter-cards could be mounted along the back edge of the motherboard. These were connected electrically through 0.1"-pitch 2xN connector pins that were fitted to the motherboard and fixed mechanically by click-in nylon stand-offs. The electrical connection was called the "option bus" orO-bus. The O-bus cards were not mechanically interchangeable; there were 3 (??or4??) different positions supporting cards of different sizes. The cards were ~4"x4" (some were wider, some narrower). Connectors mounted on the back edge of the card aligned with cutouts in the metal back panel of the base unit.

There was an O-bus card that provided an ARCnet LAN interface, and another that provided a daisy-wheel printer interface. There were probably others that I don't recall.

The backplane used 2 DIN 41612 connectors per slot. For Kennet A there were 2 slots: the top slot for the motherboard and one slot for expansion. For Kennet B there were ??5?? slots: the top slot for the motherboard and the remaining ??4?? for expansion.

Each expansion slot was divided into a slot about 4" wide and a slot about 8" wide; there was 1 DIN 41612 connector for each section. The 4" slot was used for memory expansion. Memory expansion cards were 4" square with a male DIN 41612 connector on one edge and a female DIN 41612 connector on the opposite edge. Up to 3 could be daisy-chained in the slot extending the full depth of the unit.

The 8" slot was used for adding additional functionality. The only board that I remember was a dual screen/keyboard expansion. This provided connectors for 2 more monitors and 2 more keyboards; it contained 2 duplications of the video/keyboard circuitry on the motherboard. The monitor could not be powered from the Kennet; a visually identical "self-powered" monitor was required. With this expansion board, a Kennet A could serve 3 separate users, all running independent word-processing tasks and sharing the single processor, the file storage and any other peripherals.

The motherboard was about 13" square and was packed with through-hole components and connectors. There were many MMI PAL devices but no other "custom" components.

The PCB layout was subcontracted. There was so much circuitry to fit that a 6-layer board was needed (2 outer signal layers, 2 inner signal layers, ground layer, 5V power layer). Even then it did not fit. Design rework had to be performed to cram additional logic into PALs to reduce the component area.

The colour motherboard was the same size but had even more logic; in this case, higher-density 24-pin PALs were used.

The motherboard design and timing analysis was done by hand, and the different sheets of the A0 schematics showed different handwriting and drawing styles depending upon the designer who had been responsible for that part of the design. After the design had been completed and debugged a separate project was undertaken, using 2 Daisy CAD workstations, to schematic-capture the entire design and to perform a logical simulation and timing simulation.

The motherboard contained an 8086 processor, boot EPROMs, floppy disk controller, DRAM, keyboard controller, video sub-system, battery-backed real-time clock, interrupt controller (and probably more).

I thought that the video sub-system was character mapped with attribute bits for underline, bold etc. but I read elsewhere that it was a bit-mapped display so I must be remembering that wrongly.

The motherboard was designed in-house. The design of the VDU, the power supply and the keyboard were subcontracted. I think that the keyboard manufacture was also subcontracted. The VDU, motherboard, power supply and mechanical components were all assembled in-house in the Swindon factory.

A walled-off area within the shop floor contained the (noisy) pick-and-place machines for inserting ICs, passive components and the Harwin pins that made up the connectors. There was a flow-soldering machine, a bed-of-nails tester and a commissioning area where technicians attempted to diagnose faulty boards. The rest of the space were sets of conveyer belts where staff performed their steps of placing hand-inserted components or assembly of the sub-systems into complete units.

The design offices and labs were effectively a "slice" of space carved off the shop floor, and there was a door from the lab straight onto the shop floor, which could be used as a short-cut to the canteen.

The working environment was friendly and relaxed. There were good working relationships between the engineering staff and the manufacturing staff, and a good mix of men/women in all of the job functions (apart from design, which was predominantly but not exclusively men).

Kindly Donated by Satherley Design Associates who were involved in the industrial design of the machine.

Manufacturer: Logica
Date: 1st January 1985

Comment on This Page

Other Systems Related To Logica VTS 2300 (Kennet):

Item Manufacturer Date
Logica VTS 2200 Logica 1st January 1984
Logica VTS 2200 Logica 1st January 1984

This exhibit has a reference ID of CH55474. Please quote this reference ID in any communication with the Centre for Computing History.


 Logica VTS 2300 (Kennet)

Click on the Image(s) For Detail


Help support the museum by buying from the museum shop

View all items

Founding Sponsors
redgate Google ARM Real VNC Microsoft Research
Heritage Lottery Funded
Heritage Lottery Fund
Accredited Museum