lane@dalcs.UUCP (12/01/86)
I'm looking for any info on Process Control and Measurement and Automated
Data Aquisition... Books & articles on the subject, systems and software
for doing PCM... particularly in the low-budget/PC-based areas of the
marketplace. Main interest is a survey of what's available (and good)
but also to get an idea on what there is an unfilled need for.
A tall order I realize... Please mail and I'll post a summary (promises,
promises :-). Thanks in advance for any offerings.
To add some salt to the topic, We have an HP 9816 system with a 3421A data
aquistion unit. Any ideas on what we could do with it? (heh, heh...)
--
John Wright
--------------------------------------------------------------------------------
Post: c/o Dr Pat Lane, Biology Dept, Dalhousie U, Halifax, NS, Canada, B3H-1J4
Phone: 902-424-3805 or 902-424-6527
Ean: lane@cs.dal.cdn
Uucp: ...!{utcsri,garfield,dartvax}!dalcs!lane or ...!lane@dalcs.uucp
Bitnet: lane@cs.dal.cdn or JW@dal.bitnet (goes to Dal Cyber)
--------------------------------------------------------------------------------ornitz@kodak.UUCP (barry ornitz) (12/04/86)
In article <2082@dalcs.UUCP> lane@dalcs.UUCP writes: >I'm looking for any info on Process Control and Measurement and Automated >Data Aquisition... Books & articles on the subject, systems and software >for doing PCM... particularly in the low-budget/PC-based areas of the >marketplace. Main interest is a survey of what's available (and good) >but also to get an idea on what there is an unfilled need for. > >To add some salt to the topic, We have an HP 9816 system with a 3421A data >aquistion unit. Any ideas on what we could do with it? (heh, heh...) > >John Wright Welcome to the wonderful world of process measurement and control. You may be surprised to learn that micros have been used for this purpose far longer than there were home computers. In fact the minicomputer was in common use for this purpose long before it found its way into offices. The micro- processor was readily accepted into the measurement and control area almost from the day of its introduction - I can remember when the Intel 4004 4-bit micro cost over $400 for the chip alone, and 2K of ROM and 256 bytes of RAM made up a system that cost over $10K but was easily sold for control systems. As for books and articles, I would suggest the following three magazines: Control Engineering, Intech (Journal of the Instrument Society of America), and Measurement and Control News. Three books come to mind: "Microproces- sors for Measurement and Control" by Auslander & Sagues, "Transducer Interfacing Handbook" by Analog Devices (ed. Dan Sheingold), and "Micro- processor Systems Handbook" by Burton & Dexter (also Analog Devices, one of the major suppliers for A/D/A converters and analog I/O products). A good book on the basics of process control is that by Pradeep Deshpande. As for finding unfilled needs, there are plenty out there but the market is filled with some very heavyweight competition. You need to find a very specific need if you want any chance of making a successful business in this market. I have seen a number of fairly large companies fail in this business, but I have also seen small ones form and prosper. Your system would be considered rather small and underpowered by typical standards. It is very slow on the analog input, it has no analog output, and it is quite expensive for what it does. H-P does, however, make extremely reliable equipment, and their HP-1000 series has been a respected process measurement and control computer for years. Having said all this, I'll turn up the flame on IBM. The editorial of the November 1986 Control Engineering begins: "Personal computers, or PCs, and especially clones and repackagings of the IBM PC based on the PC bus, are just about omnipresent these days in almost any booth in any industrial control show." The editorial claims that most of this is "show-biz glitz" and that "none of these ... provide a real basis for any conclusion that these file-processor-oriented personal computers are likely to make profound changes or inroads in real-time control." I can add to this my experience at an instrumentation and control forum at a national Instrument Society of America meeting. Well over thirty major companies using industrial process measurement and control computer systems were represented at this forum. When the use of personal computers for control was brought up, the unamimous conclusion was that no computer system based on the IBM PC (including hardened PCs) was yet available that could provide the reliability needed for industrial process control. Most of the experts present felt that the personal computer was useful on the engineer's desk for design, word processing, etc. and also good as an inexpensive monitor or console for the actual control computer; they also were confortable using the PC for noncritical laboratory data collection systems. For process control, where safety and profitability were concerned, they wanted a "real" system. This ISA meeting was held late last spring but I haven't seen any major changes in the way of better and more reliable PCs. You should note that many of the same microprocessor chips are used by the makers of industrial control systems; they just don't take the shortcuts when designing their systems. As an example, we have an STD bus system monitoring a fiber dryer in our plant. It is in a sealed NEMA box with no cooling air and the box is too hot to keep your hand on - yet it has run without a single failure continuously for over four years. I know the heat will ruin it eventually but the production people will not let us move it until it fails. OK IBM, let's see you do that with your 7552 repackaged AT! Final IBM flame -- On October 8, 1986 "IBM today announced its first plant floow computer that monitors and controls manufacturing operations without an operator in attendance." What about the IBM 1800 of the mid-sixties - one of the finest discrete transistor process control computers ever made! IBM should look to its past to see the folly of PC/DOS and floppy disks on the plant floor! John, I rambled on far too many lines and I apologize to you and the net. I wish you success in your venture and I also wish someone would come up with reliable hardware with a real-time multitasking operating system that I could buy as cheaply as a clone. Maybe someone reading this....hmmmm! By the way, the comments are my own and should not be taken to represent the official policy of this company in any way. Barry L. Ornitz Process Instrumentation Research Laboratory Eastman Kodak Company Eastman Chemicals Division Kingsport, TN 37662