Page images
PDF
EPUB
[merged small][merged small][merged small][merged small][ocr errors][merged small][merged small][ocr errors][merged small][merged small][ocr errors][ocr errors][merged small][ocr errors][merged small][merged small][merged small][merged small][ocr errors][merged small][merged small][merged small][ocr errors][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][ocr errors][merged small][merged small][merged small][ocr errors][merged small][merged small][merged small][merged small][merged small][ocr errors][merged small][merged small][merged small][merged small][merged small]

1.

[ocr errors]
[ocr errors]

Controller slot

...

24

List of Figures

MIDAS System consisting of a crate, power supply and a
number of single- and double-width modules

2. Simplest Stand-alone MIDAS System

3.

[ocr errors]

Stand-alone MIDAS System controlled by Programmer
or Computer Module .

vi

6

7

Page

4.

MIDAS System Controlled by Remote Time-shared Computer 5. MIDAS System Controlled by Local Minicomputer

[merged small][ocr errors][merged small]
[ocr errors]
[merged small][merged small][merged small][ocr errors][merged small][merged small][merged small][ocr errors][merged small][ocr errors][merged small][merged small][merged small][merged small][merged small]
[graphic]

Figure 1.

MIDAS System consisting of a crate, power supply and a number of single- and double-width modules. The system controller occupies the rightmost slot position.

MIDAS

Modular Interactive Data Acquisition System
Description and Specification

Charles H. Popenoe and Mack S. Campbell*

The task of interfacing experiments to computers
and data-logging systems should be made as painless as
possible for the scientist. With this intent, MIDAS,
a user-oriented, modular digital interface system based
on CAMAC hardware and USASCII-bus data communication
has been developed. MIDAS modules enable the experi-
menter to set up, program, modify and operate automated
or computer-controlled experiments independently of the
experts. Salient features of the concept are described
and operating configurations discussed both with and
without computer control. System interface requirements
are specified in sufficient detail to enable one skilled
in the art to design and construct modules operable with-
in a MIDAS system.

Key words:

Computer-controlled experiment; computer interfacing; data acquisition system; digital interface; instrumentation; laboratory automation; MIDAS; program

mable controller.

1. Introduction

Historically, laboratory automation has followed two patterns either a hard-wired dedicated system has been designed to accomplish a specific function efficiently, or a digital computer has been interfaced to laboratory instruments and then laboriously programmed in assembly language to perform somewhat more flexibly one or a number of jobs. The first approach unquestionably leads to an efficient but very inflexible system which performs the specified task well. However, requirements in research laboratories continually change, which sometimes leads to scrapping very expensive systems that will no longer satisfy current needs. The computer approach on the other hand, is a marvelous one for a well-financed, well-staffed establishment with the necessary expertise in minicomputer programming and digital interfacing techniques. Lacking this expertise, the experimenter is forced to either divert time from his research functions to gain education in these fields or alternatively, to employ the services of experts who will supply him with the system that he thinks he wants at the time of

*Compumetrics Division, Tri-Com, Inc., 12216 Parklawn Drive, Rockville, Maryland 20852.

[ocr errors]

specification. The latter leads to a situation where the experimenter does not really understand the system and is therefore not capable of making small changes without again bringing in the experts. Neither of these approaches supplies the user with what he really needs a readyto-run universal interface to computers and instruments from any manufacturer which he may plug together quickly himself as easily as he might plug probes and plug-ins into his familiar oscilloscope; and which he may then program, run, modify and reprogram on the spot until his experiment is running to his complete satisfaction.

The decreasing cost and size of integrated circuit logic has recently made possible data acquisition and control systems which may operate independently of a computer, yet have many of the attributes of computer-based systems. These relatively new "programmable" systems are generally modular in concept and programmable by "software" techniques, employing easily changed program storage such as paper tape, magnetic tape or punched cards. Such systems must necessarily sacrifice some of the extreme flexibility and high speed attainable with a machine-language programmed computer in exchange for simplicity of operation and ease of programming. This exchange is effected by increasing the complexity of hardware logic to assume functions once accomplished by sophisticated computer programming. Hardware costs have, however, been shown to be small or insignificant when compared to the labor involved in "bringingup" a complex system. We can well afford to spend a few extra dollars on logic hardware if by doing so we may save a few minutes of an inexperienced user's time in programming or using the resulting system.

We have developed a new system with the acronym MIDAS, which we believe will alleviate many of the present experiment-automation difficulties. MIDAS is a user-oriented, modular, programmable digital interface system based on CAMAC hardware and on USASCII-bus communication between modules. It has been designed with the idea of making the task of interfacing instruments, experiments, computers and data-recording devices as painless as possible for the scientist-user.

The MIDAS concept is based on two strong beliefs. The first is that a good universal automation system should have enough capability to handle perhaps 80% of experimental situations without reliance upon a computer. Computers should properly be reserved for functions that computers do best -- high-speed computation and decision-making. To use a computer for data acquisition and sequential control is "over-kill". When these computer attributes are required, however, the same system which has been running stand-alone must be able to be quickly plugged into and indeed be controlled completely by a computer, whether it be a local minicomputer or a remotely accessed time-shared facility. It is important that the systems be upgradable in a rational manner through a series of easy transitions ranging from the simplest stand-alone data recording system up through a full parallel multi-instrument complex operating at computer speeds.

« PreviousContinue »