Systems Vendors / Integrators


Thread Starter

Ken Brown

I would like to know your thoughts on the following observation. Is it just me or is there a dramatic increase in competition from automation product manufacturers for systems work? We have three very good corporate customers who have each signed or been approached to sign agreements with Rockwell to have them provide 100% content, application engineering, fully integrated control panels, startup and on-going systems support for a variety of coordinated drive and motion applications. (whew that was a long sentence . . .) Ken Brown Applied Motion Systems, Inc.

Kirk S. Hegwood

Yes, I've seen the same. Even suppliers becoming integrators. Its fun having to buy from your competition. Good luck, Kirk S. Hegwood President Signing for Hegwood Electric Service, Inc. [email protected]

Guy H. Looney

Same thing being seen here in the Southeast. Rockwell is definitely working the market. Rexroth Indramat also does the whole shibang. Guy
I don't really see a future with someone like Rockwell providing full control packages to their customers. Most customers are not only interested in the control side of their applications, but rather in the full turnkey services that a supplier can provide to them. My company is in the business of providing complete engineering, manufacturing, electrical, and software development. Out biggest problem as of lately has been the large amounts of smaller companies coming into the market and underbidding everyone because of low overhead (contract engineering, contract manufacturing, contract electrical and contract software development). But they are proving most times that they are not able to deliver what a full turnkey company can provide.. We hang in there, and keep providing systems and services that the other companies (big or small) can't provide. Good luck! Eric Nagy Indelec Automation Inc.

R. K. Jagasia

Here in India, we have Rockwell Distributor taking up Systems assignments. How do SI compete in such situations?
Ken Brown <[email protected]> asked : >Is it just me or is there a dramatic increase in competition from automation >product manufacturers for systems work? Jim Pinto response : Yes, all the Automation majors are not showing any sales growth - and indeed their sales are declining. As a result, they are moving into Systems Integration, and thereby competing with their own SI customers. See my article : Why is Industrial Automation Declining? And other related article : Changing Face of Automation : Cheers: jim ---------------/ Jim Pinto email: [email protected] Web: San Diego, CA. USA ---------------/

Hullsiek, William

Do it faster, cheaper, and with a lower cost of ownership. What do your clients value ???
Automation Listers : The discussion on Technology Stocks is related closely to the other thread on Systems Vendors/Integrators. Walt Boyes <[email protected]> commented : >The jury is still out on Rockwell et al. moving into the integrator space. Jim Pinto responds : Rockwell is doing it intentionally, and will fail - because it is competing with its customers. R. K. Jagasia <[email protected]> complained : >Here in India, we have Rockwell Distributor taking up Systems assignments. Jim : Exactly - Rockwell SIs are confused and irritated. Jagasia : >How do SI compete in such situations? Jim : By offering better value in the complete installation. And, perhaps, offering alternatives (other PLCs). Eric <[email protected]> predicts : >I don't really see a future with someone like Rockwell providing full >control packages to their customers. Most customers are not only >interested in the control side of their applications, but rather in the >full turnkey services that a supplier can provide to them. Jim : Right! Eric continues : >My company is in the business of providing complete engineering, >manufacturing, electrical, and software development. Our biggest problem >as of lately has been the large amounts of smaller companies coming into >the market and underbidding everyone because of low overhead <clip> Jim : the problem is that PLC Systems integration has become a "commodity". The "knowledge" has spread to a lot of people. With "commodities" - low-price wins! Eric : >We hang in there, and keep providing systems and services that the other >companies (big or small) can't provide. Jim : Good strategy! But, to grow, you'll need to migrate to more "knowledge intensive" systems businesses - eg: networks, fieldbus, data-base analysis, wireless-networks, etc. Cheers: jim ----------/ Jim Pinto email : [email protected] web: San Diego, CA., USA

Anthony Kerstens

The problem of PLC SI becoming a commodity is that quality is a variable in commodity pricing. Any monkey can build a box with fiddly bits and make a machine run half-cocked. (Which I've had to clean-up on numerous occasions.) A quality system takes careful thoughtful design to make it easy to maintain and extend in the future. To the customer, a system is a black box that either runs or doesn't, and they may not understand how or why, or even care to understand. To buy something of quality requires knowledge on the part of the consumer, so part of the successful SI's job is marketing and consumer education. Otherwise, the low price will win. Anthony Kerstens P.Eng.
Hi, Re: Networks and fieldbus-I see this as a "sleeper" right now. For the past six months I've been intensely involved with Ethernet, Linux, TCP/IP protocols, and the different software packages and hardware needed to make all this stuff work. I've also been (too) deeply involved with the OTHER different device-level protocols since 1993. It may be that anyone will be able to plug 'n play Ethernet I/Os when they become available, but that isn't all there is to setting up the network. And the field will need people who understand not only the Internet protocols etc., but also how to set up, maintain, and implement such networks to each unique customer specification, in an industrial setting. The actual functioning of the network for control will be a small part of the total demands for a network's operation. IMO, a systems integrator who wants to get a jump on where things are going and make sure they have a path for growth should start learning about this stuff NOW. There is big learning curve, and the knowledge required is broad as well as deep. So here's a recommendation: Get an old Pentium 75 or other "clunker" and install Linux. My personal favorite is SuSE. Make sure you have a 10 gig HD, 'cuz the full install including source takes 7 gigs. Set up a small network in you house or business just for fooling around. Start learning how DNS works, how to configure the router, how to use tcpdump, web servers, Samba, DHCP....the list is huge. This will not cost you very much money, maybe $300. There is so much to learn and digest that it's impossible to do in a short period of time; but if you start now and do it this way, you will be ready for what's next by the time it gets here. And there will definitely be a vacuum (at least for several years) while people grapple with all the new issues that will arise from using this technology in this new setting. As systems integrators for industrial controls, you already have a unique and specialized knowledge base that needs to have added to it the idiosyncrasies of IP and Ethernet. Any SI will become much more valuable if they not only know the possibilities of adding Internet accessibility to control systems, but can actually implement them in a timely, secure, and safe manner. Hope this sparks a few people! Regards, Willy Smith Numatics, Inc. Costa Rica &lt;snip> >Eric : > >>We hang in there, and keep providing systems and services that the other >>companies (big or small) can't provide. > >Jim : > >Good strategy! >But, to grow, you'll need to migrate to more "knowledge intensive" >systems businesses - eg: networks, fieldbus, data-base analysis, >wireless-networks, etc. >

Adolfo Jimmy Saldivias

Willy: I have started this way a few months ago with a couple of books written by Mark Minasi. I have bought Mastering Windows NT 4.0 Server and Mastering Local Area Networks. Besides I have bought a couple more of the Dummie's book series on the same subject. I found the last ones the difficult to read, since they try to simplify so much, that I just get lost. Anyway, my next question is: 1. Could you, as a knowledgeable person, reccomend additional literature for the subjects you think we should be preparing 2. You were talking about spending $300, and I think I have surpassed that amount of money already with the books I have acquired so far, so maybe I am choosing the wrong sources of information, so I will be glad if you could be more specific about titles and authors. If you think this subject is not adequate to put in the list (I don't really see a reason why, but then again, what do I know?), I will be glad if you could e-mail me directly. (Moderator's note: Yes, certainly, post them here!! --JP) Thanks MBA Ing. Jimmy Saldivias TECSIM Phone: 591-4-523438 Fax: 591-4-523413
Literature is one thing. If you don't actually have an NT 4.0 Server set up for you to play with that you can mess around with, and have never actually installed Windows NT or used it at all, well, I think that you've still got some work to do. Books are *wonderful* for showing you everything that can be done. For example, O'Reilly's "Running Linux" book is WONDERFUL for showing you all the nifty things that you can do with Linux. It covers, well, pretty much everything that you can do in Linux in 730 pages. You'll need more indepth books when you start doing specific things, but this'll give your the feel of what you can do. But when it comes down to push and shove, you need hardware. This is one of the great things about Linux -- you can download it for free and install it on old hardware. Just go to and download the CD images. If you don't want to (or can't) download 2 650 images, go to your local computer store and buy it. For less that the price of a Microsoft operating system alone, you get the Linux equivalent of the operating system, Back Office, and Office. It can't be beat on price. But don't take my word for it. Get it, install it, play with it, screw up your system, reinstall it, download things, install them, whatever. Then you'll start figuring out how you can make it work for you. Alex Pavloff Software Engineer Eason Technology
¡Hola Jimmy! (Press delete now if you don't like long, boring, personal posts) Ok, I will reveal my secret "guerilla" tactics for learning about this technology on a shoestring ;^) First, my $300 figure was for a junk PC and Linux, and you'll have to adjust that amount according to how much you can wangle an old PC for. Even though I live in Costa Rica, I managed to scrape together a Pentium system for about $250. I just bought the pieces from different computer dealers who were selling old stuff cheap. It took several months. It would be a lot easier in a more developed country, or if you have enough money to go buy a PC. The thing is, you need a dedicated PC that can be your Linux box to play with. If you live in the US or another country where you have access to E-bay, maybe you can get one off that site. So, I didn't count any books in the price. Since we're on that subject, E-bay is a great place to get technical books. I just got a sealed 3 volume set of the IEEE 1999 Microwave Symposium Digest for $20 ($330 on the IEEE site). I've also bought other used books over the net. They seem to be one of the easiest things to get through customs with little or no duty, and the special rates make international shipments practical. I just checked, there's a copy of Mastering Windows NT Server 4 Third Edition on E-bay "buy-it-now" for $14.99. It's been up there since Saturday at $5.99, and has had no bids. Although technical books like these fetch a lot of money new, their value quickly drops in the used market. Don't spend a lot of money on books. Anyway, I haven't spent much money on books for this subject, because it's all on the web. The main thing is to get Linux and start messing with it. By learning Linux, you will automatically be forced to learn more than you ever wanted to know about the Internet and its protocols. I tried some different flavors of Linux, but I like SuSE the best because of the depth of the installation and support. You can check out their websites: for German for Spanish for English The German language part has more stuff on it; if you can't find what you need in your own language, pretend you can read German and use the German section search engine. In general, don't be afraid to use sites in other languages. You'll be able to see either the file you're looking for, or the words you need to know since they're usually transliterations of English; I've gotten some cool stuff from Hungarian sites, even though I don't know a word of Hungarian (except igen). A lot of these smaller countries have really smart people with no money. Odds are they're running Linux. Once you get Linux installed, you have to start somewhere, and choosing where to start can be difficult! You now have a box that boots Linux--Big Deal. Since I am lazy and have teenagers, I made them learn some of it with me. I find that kids are very useful to help me learn. They ask wonderful questions and they think of things that you don't. Plus they don't think anything is impossible, they just go ahead and try to do it. I didn't have to convince them much, though; I just told them that if we got a network going in the houses, we could all share MP3 (Napster) files and play them on anyone's computer. That piqued their interest. So I taught them how to make Cat5 cables, we bought $9.00 NICs for all the computers, and we started with our Linux experience by getting Samba to run. This will illustrate how to find out about how to make it work--you use Google to find FAQs, mailing list discussions, and other documentation for the immediate problem. It took a few evenings, but we got it working. Next, we wanted to set up the server to allow us to have simultaneous internet access. So we had to get the router daemon going. That was tough. In fact, we ended up having to reinstall Linux about three times after we messed it up so bad it wouldn't do anything. We were still so green at that point that it was just easier to start over. Again, we downloaded a bunch of pages off the web and printed them out, trying different things until it worked. In this process, we were forced to learn how DNS works, more about TCP/IP, and how to use tcpdump. Tcpdump saved us. We kept looking at all those packets, figuring out where they came from and where they were going. Eventually we understood, but this was the biggest hurdle. It took weeks of beating the head against the wall. Next came the Apache server. Edict from Father: "Everyone has to make their own website, telling about yourself and your interests". I confess that at this point, I had to buy a bigger hard drive. We extended our network and included the neighbor's house on our intranet. One kid has 80 pages already on his site. We had to learn about virtual domains and a bunch of other stuff to get that working. The web sites are still only on our intranet, but I'm hoping we can get a fixed public IP address in the next few months so they can be out on the Internet, too. At this point, we still don't have the e-mail stuff working, but that's next. We got a junk modem and configured ppp and mgetty so we can dial in and access the network. We'll eventually have to figure out how to make the firewall work, and also we want to get the DHCP server working. I didn't want to use DHCP right away because I wanted everybody to have to understand IP addresses and netmasks, and to be able to configure their own computer in Windows. What happens when you do things this way? All the details (like ports, script syntax, devices, *n*x directory structures, and other myriad details) begin to sink in without you really noticing. There are hours, days, and weeks of frustration; but you don't have a production line down - - the worst thing that happens is your daughter can't listen to the Dixie Chicks for a few days. So, you give it a rest, and in a day or two come back, do a little more searching on the web, and it eventually works. You add some hubs and switches, learn how to make crossover cables, in a comfortable time frame. All these things will be incredibly useful to me when I start having to answer questions about Industrial Ethernet applications. Sure, during my day job I'm designing and debugging Ethernet hardware, writing some device drivers, and other low-level stuff. But I felt totally lost about "the big picture" until I started this Linux Server projet last July. I don't really *feel* any smarter, but I know by how quickly my co-workers eyes glaze over that I have at least learned a BUNCH of new acronyms. I'm not posting any of the links we used, mainly because I don't have them all in a file. But anyway, there's so much out there on this technology, you'll do just fine with just a search engine. As SuSE says on boot-up, "Have a lot of fun!" Regards, Willy Smith Numatics, Inc. Costa Rica PS Just got an e-mail from e-bay, got 20 10baseT NICs for $3 each. If anybody's coming to Costa Rica, wanna mule 'em in here for me?

Curt Wuollet

Hi Jimmy, Willy I would certainly second Willy's advice on using Linux as an educational vehicle for networking, programming, and software studies in general. For a free download you have world class networking, at least a dozen programming languages, a remarkable collection of FAQs, HOWTOs and ready examples of source code for practically any endeavor. It is all free and intended for just that purpose. There is simply no way to obtain this amount of information on any proprietary OS and the tools alone would be several thousand dollars at a minimum. What's more, If you want to know how a router or a firewall really work you can get world class examples that are the standards in use everyday. You can even learn a lot more for a lot less about Windows file sharing and networking from Samba than MS will ever tell you. Once you get started in using Linux and hit a few of the sites you would be amazed at the information available. With me, it's gotten so I would have no use whatsoever for an OS that was shipped bare, without even a compiler, where everything (except items that further lock-in) was an expensive add on. Why just for the tools I use everyday, I'd spend a mint and then get nickle and dimed to death to get the dozens of utilities I need. I can often create a useful mew program with 90% done by those who share and a few dozens lines I add. I've been doing it since you put together your own distribution and am still constantly amazed at how powerful this can be and how few people know about it or use the leverage it provides. While most people are reinventing the wheel you can be building a Porsche. You might start with the Linux Documentation Project and see if you can't find exactly what you're looking for. Regards cww
Pardon a non-cynical view from an instrumentation supplier.
Instrumentation is becoming more sophisticated and more complex applications are being addressed.
For the instrument supplier this means building a specialised skills/knowledge set to support the instruments in the application. If the instrument manufacturer remains a product only supplier then two things can happen: the instruments are not optimally installed or the skills and knowledge are supplied as "free" consultation to the SI or contractor who reaps the financial benefits.
Speaking from personal experience of viscosity measurement we can either offer an instrument guarantee or we can offer a performance guarantee of a fully designed and fabricated system. This has proven particularly critical in hydrocarbon applications where base viscosity determination is especially difficult to entrust to a third party. Invariably we have found the thrid party will listen to all the "free" advise and guidance and then compromise the design in what they think is a non-critical way but which results in a serious departure from optimum performance.
So at whatever level you want to consider it there is a case for the manufacturer tocapitalise on the skills and knowledge they have acquired and deliver a genuinely better performance for no more cost to the end user.
(Jon Watson; Solartron Mobrey)