Data Acquisition Corp. DAC-512
The Data Acquisition Corp. DAC-512 "Digital Computer"
Image Courtesy Charles Falconer
DAC-512 in use. Note electronics package containing the logic behind the keyboard unit.
Image Courtesy Charles Falconer
The Data Acquisition Corp. DAC-512 is an early
programmable desktop electronic calculator. Although the company
called the machine a "Digital Computer", it was most definitely a calculator*,
albeit a very capable programmable calculator. At the time it was introduced
(Spring of 1966), it was arguably the most powerful desktop electronic
calculating machine available, and though eclipsed historically by the
Olivetti Programma 101,
the DAC-512 remained one of the most
powerful programmable electronic calculators
available until the introduction of the Hewlett Packard 9100A in early 1968.
The DAC-512 was designed by the founder of Data Acquisisition Corp., Charles
B. Falconer (9/13/1931-6/4/2012). Falconer was born in Geneva, Switzerland
but grew up in Eastern Canada. As a youngster, he had a keen interest
in math and science. He attended McGill University in Montreal, Quebec,
where he studied physics and mathematics, with a particular interest in
nuclear physics. After college, he moved to Ontario, Canada, and took a
Chalk River Nuclear, where he became deeply involved in the development
of custom hardware and computer software for
control and managment of nuclear power facilities. He became well-known
for his skills through his work at Chalk River, which resulted in Falconer
spending a lot of time traveling around the world as an expert in the field
of applying computers to the process automation for controlling nuclear power
facilities. In 1958, he moved from Canada to Boston, MA, a hub of technology
enterprise, with the intention of starting his own business
developing custom hardware and software
solutions for the nuclear power and other industries requiring precise and
fail-safe real-time process control systems. He realized his dream of starting
his own company, which he founded not long after arriving in the US.
His company was called Data Acquisition Corp., which was located
in Hamden, CT. Sometime in 1964, Falconer, having been deeply
involved with the use of expensive minicomputer systems
for data acquisition and control, believed he could design a much lower-cost
machine that could do the job. Mini computers, while powerful and fast, were
complex and could be difficult to program and use. Falconer thought that a
somewhat less-powerful, but very user-friendly machine could be a valuable
tool for data acquisition, analysis, and even process control. Falconer did
most all of design himself, and by mid-1965, a prototype of his vision
was up and running.
The prototype machine was a standlone programmable calculator, without
I/O interfacing capabilities. By the time the DAC-512 became a production
product in the spring of 1966, Falconer had developed the
I/O interfacing capabilities for the machine that allowed it to "talk" to
external equipment that could allow the DAC-512 to become the basis for
a data acquisition and control
system. At introduction, the retail price for the DAC-512 was $9,500.
This was a lot of money for a calculator, but given that a typical
mini computer system of the time outfitted for data acquisition and
process control sold for in excess of $30,000, the DAC-512 was quite a value.
The DAC-512 consists of two parts, an
operator's console that provides the user interface for the machine, including
a keyboard with keys that would light to show available operations to the
operator, a Nixie tube based display to show calculation results and
program steps, as well as other operational controls. The other
component of the system was a sizable (and heavy -- approx.
60 pound) electronics package containing the calculating logic, magnetic
core memory, and power supply.
The two units are connected by a cable that carries digital signals and power
The electronics package of the DAC-512 was built with
discrete transistorized solid-state logic (approx. 1000 transistors and 3000
diodes), and an 8192 (8K) bit magnetic core memory
(with a cycle time of 10 microseconds) which stored the working registers,
program steps, and memory registers. There were about 40 circuit boards
inside the electronics package, with each board densely packed with components.
The DAC-512 used a microcoded architecture, but not in the traditional sense of
a having a read-only memory serving as the microcode store that contains
the microinstructions that govern the operation of the machine.
Instead, the microcode of the DAC-512 was wired into specific
sections of the logic of the calculator. This
provided some functional flexibility, but required some reworking
of circuit boards in order to change the functionality of the machine,
but didn't require a complete re-design as hard-wired logic architectures
This scheme was chosen as a reasonable compromise between cost and flexibility,
as read-only memory technolgy at the time was generally quite expensive.
The DAC-512 operated with full algebraic logic (which had only been
implemented by Mathatronics with their
Mathatron calculators up
to that time), observing the classic order of precedence of operations,
as well as allowing for overriding order of precedence using
parentheses. The machine processes the basic four math
functions, with approximate average calculating speeds quoted by Mr. Falconer
of addition/subtraction in 10mS (0.01 second), multiplication in 50mS,
and division in 70mS -- quite fast for the time. The DAC-512 utilized a binary
coded decimal floating point numeric representation with nine significant
digits and a two-digit exponent ranging from +50 to -49. The machine provided
120 memory storage registers and 512 (with this being the "512" in the name
of the machine) steps of program memory, which was
far more than anything else on the market at the time. Program memory was
divided into eight areas of 64 steps each. Each area could be used as a
standalone program or a subroutine. Eight keys at the left side of
the keyboard could cause immediate
execution of any of the eight program segments. Conditional instructions
allowed decision making within a program, and program looping for iterative
algorithms, as well as subroutine nesting capability. Every DAC-512
came with a comprehensive library of advanced math functions such as logarithms,
square roots, raising numbers to powers, statistical functions, and much more.
Due to the machine's flexible I/O architecture, an interface was developed
providing a connection a
Teletype Model 33-ASR data terminal to the calculator, providing
the ability to load and punch programs to paper tape, as well as accepting
numerical input and printing numerical output from/to the terminal.
Picker Nuclear-badged DAC-512, circa 1967.
Due to extremely competitive market conditions both in the process
automation and electronic calculator marketplace, Data Acquisition
Corp. ran into financial difficulties, and was purchased in 1967 by
Picker Nuclear, a division of Picker Corporation. Picker Nuclear
was a producer of electronic measurement instruments for nuclear-related
systems, and later, of medical imaging (XRay, Gamma Camera, Ultrasound)
equipment. The DAC-512 calculator was retained as a Picker Nuclear
product, and was re-branded and sold under the Picker Nuclear badge.
However, with intense competition in the calculator market, it soon
became clear that the company's investment in the calculator was not going
to be profitable, and production and sales of the calculator ceased.
Total production of the DAC-512 was somewhere around 100 units.
The Old Calculator Museum is looking for any examples of the machine,
literature, or memories from anyone who may have had experience with the
DAC-512, either at Data Acquisition Corp., Picker Nuclear, or perhaps were
involved with the DAC-512 as an end-user.
If you have anything to contribute, please click the EMail button at the
top of this page to contact the curator.
* The Old Calculator Museum considers a calculator different from a
computer based the criteria that a calculator operates fundamentally on
numeric values. The manipulation of strings of alphanumeric
characters (e.g., A-Z and special characters other than + and -) is
a characteristic of a computer. Some later programmable calculators could
display or print out alphanumeric character strings, but, in general could
not accept alphanumeric strings as input, nor could they process character
strings, such as sorting, searching, or modifying character strings. There
are admittedly fuzzy boundaries to this distinction, with machines such the
Hewlett Packard 9830. The 9830
could input, output, and process character strings. However, most of this
class of machines (inclduing the 9830) would operate as a calculator by simply
entering a math expression and pressing a key to process the expression,
which would evaluate the expression and display
and/or print the answer, without any special programming.
Other machines, such as the
Wang Laboratories 700-Series, and
Tektronix Model 31,
could input and output alphanumeric character strings to
attached peripheral devices, but had very limited ability to manipulate
character strings. These machines are considered
calculators because the majority of their function centers around
performing operations on numeric values, although with their capabilities,
they certainly blur the line between calculator and computer.