Introduction
Data General Dasher 200 terminal

The first "Dasher 200" terminal came with the first system that I picked up, the NOVA 3. I got a second D200 terminal from Geert. The Dasher 200 terminal has a very typical, recognizable shape, and of course the "Data General" blue color. One keyboard has a few keys of which the spring action is gone, so those keys are always "pressed". Definitely repairable.

The Dasher D200 terminal displays standard ASCII characters but has a cursor control and other special feature code set different than other vendors. Later Data General Dasher terminals did have ANSI 1964 support along with the "Data General mode" support.

The keyboard cable connects to the rear side of the base unit (with the CRT) using a rather simple header. The separate wires are visible. A less fragile connection, for example, a sub-D connector would have been a better choice.

Description
The Dasher 200 terminal consists of a separate keyboard and a base unit. In the base unit is the screen, electronics and the power supply. Pretty standard. Not so standard is the way to open the base unit. To open the base unit, you first remove the off/on-brightness knob, by pulling it out. After removing 2 screws at the bottom you can swing the top part of the housing up until it locks in a held-open position.

The terminal has the following connections at the rear side.

9-pin single-in-line header for keyboard connection
DB-25 female connector for EIA / current loop connection to the computer or modem
8-position DIP switch package for terminal communication settings

On the front is just one knob. You pull the knob to switch the terminal on, and rotate the knob the adjust the brightness of the screen. Push the knob in to switch off the terminal. If the friction of the knob on the shaft is low you end up with the knob in your hand when you switch on the terminal.

Keyboard connection D200 keyboard connection cable
The keyboard is connected to the base unit with a simple 9-pin header. This is probably the minimalistic solution to connect something. No strain relief, no shielding, no proper hood to handle the cable when you plug it in. OK, at least there is one pin that serves as a key to prevent wrong plugging it in.
On the other hand, in a normal environment you put the base unit on the desk, put the keyboard in front of it and plug in the cable.

Perhaps we nowadays would have used a tie-wrap as strain relief and protection against not-intended disconnection and bent pins when a typist gets "too enthousiastic", and gives the keyboard a firm whack.

Terminal connection D200 terminal connection cable
The terminal connects to the computer (or modem) with a standard DB-25 female connector. On this connector are pins for EIA ("RS-232") and for 20mA current loop.

DIP switch baudrate

1 2 3 4

0 0 0 0 50

0 0 0 1 50

0 0 1 0 75

0 0 1 1 110

0 1 0 0 134.5

0 1 0 1 150

0 1 1 0 300

0 1 1 1 600

1 0 0 0 1200

1 0 0 1 1800

1 0 1 0 2400

1 0 1 1 3600

1 1 0 0 4800

1 1 0 1 7200

1 1 1 0 9600

1 1 1 1 19200

Left, next to the terminal connection are 8 DIP switches. The left four DIP switches (numbered 1 to 4) set the baudrate. As you can see, there are a few baudrates that would be considered "weird" today.
DIP switch 5 and 6 set the parity.

DIP switch parity description

5 6

0 0 none There is no parity bit.

0 1 odd The state of the parity bit is set such that there are an odd number of ones in the transmitted character.

1 0 mark The parity bit is always "1".

0 1 even The state of the parity bit is set such that there are an even number of ones in the transmitted character.

DIP switch 7 and 8 are not used ("reserved").

Some first-hand information
Sometimes, writing a web page has very nice "side effects". Just a few weeks after I added this page about the Dasher 200 terminal, I received an email from Peter Simpson. Peter really knows the D200 ... he was on the team that designed the D200! This is what Peter told me.

Working on the design of the D200 was his first job out of college. He wrote some of the code (see below), and Peter was responsible for the keyboard electronic design and test. The Data General internal code name was "PEGASUS".
The overriding goal was LOW COST. Data General's existing terminal, the 6053, was big, heavy, and expensive to build. At the time DG was competing with things like the Lear Siegler ADM-3, which was only a few hundred dollars. The goal of the design team was to design a full featured CRT terminal for less than the cost of an ADM-3. For that time, the team did some very innovative stuff.

The D200 used a generic microprocessor (Motorola 6802)
The team wrote their own code in assembly
Communication speeds up to 19.2kbps (this was not available on all terminals)
All components placed on a single PCB (caused LOTS of issues, more later)
The team designed their own switching power supply -- these were leading edge at the time
The team designed their own CRT subsystem: sweep, power, the works (most terminals had purchased subassy's)
The team designed their own capacitive keyboard (expensive molds, but very cheap keyboard to produce)
A simple, two-piece moulded housing
The D200 was designed for volume production

You can see the complete schematic diagrams for the system here (it is on the last pages) :
Data General Dasher Display terminals Models D110-D200 User's Reference Series.

Due to some of the design choices, the D200 had a few problems.

The CRT High Voltage was also generated on the logic PCB!
The display drive circuit could burn out if the microprocessor locked up!
To make it worse, the microprocessor locked up regularly, because the CRT (High Voltage) anode connection was directly above the CPU!
A shield was added later to improve this.

Peter's story continues.
So yes, there were some things that we might have been able to improve. But we sold lots of them, and they were cheap to build.
My neat contribution:
You'll notice that the capacitive keyboard is susceptible to noise pickup. Place the keyboard on top of any CRT and it will start generating random characters ... Put the keyboard on top of its own CRT -- no garbage! I phase locked the keyboard scan to the CRT horizontal sync. The keyboard data is not scanned when the CRT flyback pulse is generated, so no interference!

That was a clever idea (IMHO).

One more tidbit.
There is a not-so-secret command to download a Motorola S1/S9 encoded program to the video RAM of the terminal. Check the back of the manual in the link mentioned above.
Enter the command sequence Enter Remote Test followed by the S1/S9 hex code to download and run programs in the terminal memory.
Today, this would be considered a huge security hole, but back then, it was just a fun thing to include. The team wanted to include a PACMAN game as an easter egg, but it turned out that there was not enough RAM for it (for some reason Peter can't remember) - the optional code had to go into RAM rather than be in masked ROM. There are four (?) empty positions on the front right corner of the PCB. These are for "extended RAM" - 2102 or 2104 static RAM, but here Peter's memory is not 100% clear ... Those were the days!

back to top