|
 |
Wednesday, December 19, 2007 |
It was light blogging Monday and Tuesday because I drove up to Chicago to visit Molex and Omron. There have been changes at both companies, and I have not had an opportunity to share the editorial focus and vision of Automation World. Two great meetings with companies with similar challenges.
You may not recognize "Molex," but you will by the end of 2008 if their plan succeeds (and I write enough). Most of you would recognize names such as Woodhead, Brad Harrison, SST. Well, Molex wanted to expand deeper into the industrial space and decided the best way was to buy someone, and Woodhead was the best choice. The integration of the companies is well underway, and it appears to be going well. You'll be seeing phrases such as "Brad from Molex" with Woodhead, SST, Brad remaining alive as product names. The new division is Automation and Electrical Products Division combining some previous Molex businesses with the former Woodhead business. So the marketing people face the challenge of "re-branding," something that is never easy.
There are people in the industry who love to go around and call us a "systems" magazine--as if that's bad. Some of you may even wonder why I'd talk to a company that does cables and connectors. Well, first off, nothing in automation is going to work if it's not connected. Second, systems doesn't mean just computers, PLCs and software. An automation system includes all the stuff to make it work--sensors, motors, controllers, connectivity stuff, software--plus the strategies that put it all together to make manufacturing profitable and an integral part of the enterprise. The new Molex innovation center with displays of various solutions is a great reminder of the engineering and innovation that goes into these products and how cool they are.
Then I visited with Omron Electronics. It was my first visit with the new marketing team. They also face a daunting challenge. Through a couple of changes of management at the North American headquarters in Schaumburg there have been a couple of significant changes in marketing direction. Now it seems that the company is once again focusing in its core automation strengths and adding new products like crazy. New vision products were the topic of the day -- making vision easier for people who aren't vision, or even automation, specialists. Then the mechatronics term was broached. Nice to hear. I always considered Omron a viable competitor to the usual PLC suspects -- Rockwell, Siemens, Mitsubishi, Schneider, GE Fanuc, AutomationDirect -- but it seemed to fade away the past 2-3 years. Good competition is a good thing, so I'm glad to see Omron showing signs of shaking off the dust and dusting off some competition (couldn't resist that one).
9:18:13 PM 
|
|
OK, this is marketing, but it's interesting. Servers have become as important in manufacturing applications as in IT, and keeping them up is a big deal. Stratus Technologies developed fault-tolerant servers at a fraction of the going price several years ago. So, I open the mail today and pull out a "Dummies" book (I'm expecting one on programming for Dummies any day now :-} ). Actually it's more of a booklet size, but entirely in the "Dummies" format -- "Fault Tolerance for Dummies." You can get yours here. It's an interesting read.
8:47:38 PM
|
|
A couple of days ago I had a tip about a pending MTL acquisition. Well, it happened. And I'd have never guessed Cooper Industries. Fuses, lights and instrumentation. I guess when you're a conglomerate, "synergies" don't enter the equation. The executives try to make a fit, but it looks as if MTL grew so much that it needed more capital to grow some more.
From the press release:
- The boards of Cooper Controls (U.K.) Limited ("Cooper UK") and The
MTL Instruments Group plc ("MTL" or the "Company") are pleased to
announce that they have reached agreement on the terms of a recommended
cash offer to be made by Cooper UK, an indirect wholly-owned subsidiary
of Cooper Industries, Ltd. ("Cooper"), for the entire issued and to be
issued share capital of MTL (the "Offer"). It is intended that the
Offer will be implemented by way of a public offer under the UK City
Code on Takeovers and Mergers (the "Code" or "City Code").
- Under
the terms of the Offer, MTL Shareholders will receive 708.5 pence in
cash for each MTL Share. A Loan Note Alternative will be made available
to all MTL Shareholders (other than Restricted Overseas Persons).
8:23:30 PM
|
|
Ah, just got another comment on programming as I was posting my last update. According to Rob (his comment is below), he's disappointed in my knocking ladder diagram. It's like knocking English. Well...actually English is easy to knock, too. Back to ladder, I sincerely hope that the technology providers continue to improve their programming tools so that ladder can be hidden. That is where the greatest strides will be taken to improve coding and enforce standards. It may be useful to realize that ladder is big in the US, not so much elsewhere. Eventually it will go away for a variety of reasons. Maybe when a better graphical method (say something like NI's LabView?) gains traction? Or, check out this podcast on something called Boku, a programming "game" for children. MacLaurin has interesting ideas.
As for English. I was just listening to a podcast where one person said the great thing about English is that it has no grammar. It's all usage based. However people use words and phrases is what it means. I've also heard it said that one reason English is becoming the universal second language is because you can speak it poorly and no one knows. I find those thoughts horrible and offensive. I suppose in everyday conversation you can use slang and partial thoughts and sort of communicate. But language is a vehicle to help us convey meaning. Unstructured use of language usually follows unstructured thinking which follows cluelessness.
Back to programming. I just commissioned an article on IDEs and what's happening in the programming tools area. It was a good start, but a little shy of what I was hoping for (January Automation World). I'll try again later. (Not the writer's fault. We have occasional difficulties getting good information during our research.) I'm trying to come up with an article definition tying things like UML and structured thinking and programming tools. Any ideas of sources welcome. Maybe I can get it done by summer.
Anyway sorry, but I guess I think that ladder, like Basic, if left to the lowest common denominator encourages unstructured thinking. I need tools to help me structure my thinking (since I don't use them to blog, the blogs sometimes ramble). I use OmniOutliner and the outliner in Word a lot. Anything the suppliers do to help will be surely welcome.
From: Rob Holl
Hi Gary, I often read your column and am very surprised / disapointed by your comments. Why slam Ladder Logic? Especially when you are a voice for our industry. That's ! like slamming English as a language. Ladder or any booleon solver is the brain of manufacturing, and will continue even when design tools are fully integrated into the digital factory. You are really pointing to problems with people and process. Possibly an article on the advances that many people have made with standard code structures and automation to generate PLC code or ladder based on a proven structure would help you regain some credibility.
12:45:44 PM
|
|
I ran into Bryan Singer at Automation Fair last month. He used to be at Rockwell, now he's with Wurldtech--the security services company in Vancouver. He's also still with the ISA SP99 cybersecurity committee. I don't have time for it, but I was doing a little surfing this morning and ran across his (actually the company's) blog. Good one to check out.
Nice rundown of current affairs in the world of OPC on Eric Murphy's blog.
Want to know why I recommend selling from your strength and not from your analysis of your competition's weaknesses? Check out Carl Henning logically dissecting some comments posted on the PTOBlog. Stick to explaining what's good about your solution. It's less painful. [I'm working on getting a podcast interview with some Rockwell Automation people to discuss what they mean by "standard, unmodified" Ethernet. More next year ;-) ]
I like to throw out comments and see what they stir up. On my quick posting about programming the other day, I got two really good comments. By the way, David great to hear from you again. It's been too long. Oh, and I looovvvve sequential function charts. When I learned about them years ago, I was fascinated. Another student and I sat in the classroom quite late figuring out different ways we could use the method. Then I got a book on Petri Nets and Grafcet to study more. Also got an intro to UML. UML concepts are behind some of the standards thinking going on today. There is no single nirvana, but I'm a believer in thinking about the whole thing before diving in (for smaller projects obviously, those of you who are part of huge teams each person contributing a part can use ideas such as Extreme Programming).
I agree strongly that much ladder logic is programmed in a style that
is called spagetti logic. For PLC's I prefer the term combinational. I
believe that spagetti code refers to code that jumps from one part of
the code to another, and back again. So that to try to map out the
logic it resembles a pile of tangled spagetti. In ladder logic, I tend
not to see the logic jumping from one location to another, but that a
given output is controlled by a combination of logic and or logic
statements to resolve it's state. There are ways to program PLC's that
demystify this, in particular to use Step logic. Step Logic is the
genisis for Sequencial Function Charts (SFC's). The use of SFC's helps
to contain the order that a machine operates in, and makes it very
logical, and simple to trouble shoot. It still requies some glue logic,
and a portion of logic I call "Mode Control". Mode control is
responsible for determining when automatic is allowed, single Step,
jog, single cycle, cycle stop, stop, and how to handle and recover from
Faults. One of the nicities is that once mode control and the glue
logic for step sequences is developed, it can be reused, only the
actual steps for each sequence, and the individual Inputs and outputs
delt with. This gives the machines very predictable behaviour, and
makes them easy to trouble shoot and build. david Coulter âo¢ 12/18/07; 9:32:09 AM #
I've always thought flowcharts were a waste of time -- some quite good
developers agree with that, too. Nobody uses them today - for one, they
can't express object oriented code well, let alone more advanced
techniques.
Supposedly the whole point of my blog is applying modern "best
practices" to automation software - there's a lot more to it than
programming. For example, anyone who does not use version control is
not a professional developer, especially since excellent, solid, well
documented version control is available for free.
I'm partial to the Extreme Programming approach - changing requirements
is part of life, and it's the first approach to take change seriously.
But it's important to know various approaches, and to know which is
appropriate for what part of a system. For example, I can see combining
BDUF (big design up front)/MDA (model drive architecture) at the lower
levels (e.g. safety critical) with XP (extreme programming) at higher
levels.
A good programmer programs well in any language (although the results
will be better in a well designed language). A bad programmer will
create spaghetti code in any language - I've seen too much of that, in
ladder, VB, C++, VB.NET, etc.
State machines (or state charts if you like UML-talk) are an excellent
design approach for a lot of automation, but aren't easy to do in
ladder logic (I've done them easily in IEC 61131 structured text). For
most machines, it's a much better approach than combinatorial ladder
logic, and I think is pretty similar to David's approach. Tony âo¢ 12/18/07; 7:13:17 PM #
11:40:33 AM
|
|
So, last night driving home from a quick trip to Chicago, I'm listening to my podcast library and heard an interview with Gigi Sage about relationships, men/women differences and how women can understand how they react and then use it to their advantage in business. The book is "Hello, Tarzan."
There must be some sort of meme going here. This morning I go directly to my favorite part of the daily newspaper--the comics--and see "Zits." The aptly named Pierce giving his fellow high school buddy some life coaching tips asks this quintessential Zen question, "If a dude says something and his girl-friend doesn't hear him, is he still wrong?"
11:21:25 AM
|
|
© Copyright 2008 Gary Mintchell.
|
|
|
