PLC Brand Name USA Market Share

[Chris Elston 79]

So does any one have any idea what market share of installation say Allen Bradley has in the USA market? Or Omron? Or GE? etc?

[Duffanator 1]

The last time our Mitsubishi rep was in and we were talking with him he said Allen-Bradley has somewhere between 90-95% of the market share in the US. That doesn't suprise me, Allen-Bradley is everywhere. I don't know why though, I think there are much better solutions out there than Allen-Bradley but that's just my personal opinion. He did, however, say that Mitsubishi is doing a much better job of marketing in the US now. There are two car plants, I think Honda and Toyota, in the US that are 100% Mitsubishi driven and I think GM is looking into building plants with Mitsubishi control as well. Our plant is food production (pork products) and we have always used Mitsubishi, but any new equipment that comes in is either Allen-Bradley, Siemens or Mitsubishi control. I would think it goes in that order as far as popularity, in the food industry anyway. We never see Omron, GE, Idec or any other brand of PLC in any equipment we buy.

[HandledException 1]

Seconded. --HandledException

[PLCMentor.com 60]

Ok, I will disagree. I think AB processors are ahead of the competition. Frankly I think the rest of the PLC world is scrambling to catch up with the logix. Now prelogix processors you may have a point, but they still had some things that they did well such as not having a flat memory space for their data types. The way they handle data types was always better than the flat memory structure. On the other question, I think I had heard from my rep that their market share was somewhere in the 60-70% range. Siemens is second and all the others I believe have pretty small market share. They claim that they are making strides in the international market share area. Who really knows?

[Crossbow 295]

I haven't seen a report in like 5 years, but AB used to be 40%, Siemens about 15%, Modicon about 30%, and the rest filled what was left. AB has NEVER has 95% market share....

[paulengr 44]

The PLC was a joint project between AB & GM. Together, they invented the PLC. So for a very short period of time at least the market share stood at 100%. I'm not sure how long it was before the second PLC brand name appeared on the market though.

[Eddie Willers 2]

In the USA, Rockwell/A-B has about 40 to 45% of the controller market. It's much higher in Canada; some say as much as 60 %. Siemens has between 20 and 25 % of the market in North America. It's the other way around in Europe, of course, and in Germany proper Siemens completely dominates the market. You'll see Omron and Mitsubishi as the principal controllers in Japan. The rest of the market is "everybody else". Depending on your industry, you might see Telemechanique/Schneider/Modicon have almost 10% of the market, or Omron, or Mitsubishi. Despite their popularity on this Forum, Automation Direct has a very small market share, about 2% last time I read a report. I get these numbers from an Automation Research Council (ARC) report; most marketing organizations of major vendors have access to these. I don't read the control-oriented magazines anymore because I'm sick of hearing how magical Ethernet or Wireless or Mesh or Zigbee are.......

[NeTree 0]

I am of course familiar and fond of Siemens, however I can think the country of origin of machine maker dictates this popularity.

[JesperMP 2]

Sorry, I cant let such a statement pass unchallenged. That is correct as fas as that AB created the term "PLC". But of course, the Modular Digital Controller was there first, created by Chris Morley. If his company had been slighly more succesful, we would all be programming "MDC's" in stead of "PLC's".Now, back to the original topic. Worldwide, I believe that Siemens has approximately 30-35%, AB/Rockwell 15-20%, Mitsu 15%, Omron 10%. It sounds realistic to me if AB/Rockwell has approx 40% in North America.

[paulengr 44]

But, but, but...I just got a quote to run a wireless signal to display 3 limit switches from a "remote" building about 100 meters away for $10,000 (Elpro). It's so much easier, cheaper, and less reliable than just pulling two wires, right? Oh, and same outfit ran sensors for roughly half the inputs all over a kiln in the same area to pick up roughly 20 inputs for a mere $200K. Of course I just duplicated that plus all the outputs too with a CLX system for about $150K for the hardware and installation alone, including controller, IO, and programming for about $300K. Wireless is clearly the way to go! Wireless has its place. I also have remote pumping stations that are kilometers from the nearest network and running fiber or copper to them would be 5 times more expensive than wireless radios. I've also run wireless to displays mounted in cabs on bridge cranes where wired networks are much more difficult in terms of reliability. I'm just not sure what the point of the various "mesh" or Zigbee systems is. Every time I check into it, the pricing is outrageous and the ranges and speeds are vastly underwhelming. In terms of Ethernet, I disagree in part. The trade magazines seem to be decidedly one-sided with regards to it but at least they point out some of the very obvious issues. Ethernet has an almost completely flexible topology and speeds that are so high that network congestion on a controls network is ALMOST a non-issue. I haven't been very impressed overall with the motion control implementations over Ethernet. From a programmer/troubleshooter's point of view Ethernet CAN be much easier and simple to work with. However, there are plenty of places where Ethernet is a very, very bad idea. Do NOT carry office traffic across I/O networks for instance unless you've walled up everything and have bandwidth controls in place (and can maintain it). In fact Ethernet can get you into trouble very quickly precisely BECAUSE network design and planning is placed in your hands and not dictated automatically by the topology and network interfaces. Terminations are critical and often buggered up. Depending on the equipment and application, network interfaces can be costly. The very annoying (and documented) "feature" of Cisco switches to arbitrarily lock out ports at the slightest hint of noise that requires administrative access to correct makes them (and the AB branded ones) junk in an industrial environment, but Allen Bradley insists on using them. In a conveyor application where bus networks and the lowest density I/O available (single network-capable sensors such as ASI or DeviceNet) make the most sense from a simplicity point of view, Ethernet is the clear loser compared with more traditional network protocols.

[aggattapauer 2]

the PLC, as far as I know, was invented by Morley in 1968 (Modicon 084 and next 184).

[Chris Elston 79]

Mr. Willers, Thank you very much for this direction, the ARC report. That's what I was looking for, where to get the information or what orgainzation has that data. Chris

[Crossbow 295]

Dick Morley, not Chris. And he invented the PLC, contrary to what the AB guys want you to believe. He worked with the automotive industry (GM and others) to design a controller to facilitate line changes. AB did not invent everything, contrary to their belief.

[HandledException 1]

Sorry, I should clarify that I was referring to their software offerings. Coming from a software development background I am appalled at some of the crap our customers have purchased, or insist that we purchase, from them (at extremely high prices to boot). In my opinion they are so far behind, and so far off track, that they will never be able to catch up. I can't comment much about their hardware offerings, not having any sort of electrical background, aside from their plc programming software (Logix 5000) isn't too bad. Being able to program in a text-based language instead of ladder logic has been a lifesaver for me. Now all I need is for Rockwell to have AOIs to be truly object-oriented by allowing them to have multiple methods (routines), and I know I am reaching here...inheritance. --HandledException

[NeTree 0]

Ladder logic is not very much difficult if you see "mentally" addition BASIC commands as learning such as I did: See... -||--|/|--() as... IF -||- AND -|/|- THEN -()-.

[Crossbow 295]

Handled, Ladder is not designed for PC programmers. It's designed for electricians, as it mimics an electrical schematic. If your background is PC programming, then you would find structured text very similar to C++ programming. That's the plus of IEC style programming, is different languages for a different class of user. But beware as you write structured text and then hand it over to a shop floor full of 20-year ladder logic users. One response I got was 'What the ... is this ...., and how do I convert it to ladder?'

[PLCMentor.com 60]

Well thats a little different... I dont know if they are way behind or way off track, but they definately have some work to do. Ok a lot of work to do. russell

[NeTree 0]

HA! I would be one of them!

[paulengr 44]

I cut my teeth doing PC programming before there were even IBM PC's, in 1978. At that time, the number of canned software packages available could be counted on one hand. Other than their BASIC interpreter, Microsoft was a complete unknown. I think C and Unix were still operating on PDP-11's. Fast forward. I didn't work on my first PLC program (I didn't even know anything about them) until 1997, almost 20 years later. So PC programming languages are "native" for me, and PLC programming is a "second" language. Perhaps third if you count stack-based languages (HP calculator, Forth, and Postscript). Fast forward. I've worked on debugging countless programs on both PC's and PLC's written by others. It is much, much, much easier to debug programs written in ladder logic or even function block languages than Algol-derived languages such as Structured Text, C, Pascal, or Fortran. It is certainly possible but actually much harder to write unreadable PLC code. I have never for instance heard of an "obfuscated code contest" for PLC's. Here's a wikipedia entry talking about the most famous one for C: http://en.wikipedia.org/wiki/International..._C_Code_Contest If there is one, then I will certainly have to nominate code from GE environmental to win the prize. They have managed to produce some of the most goofy crap I've ever seen. They go out of their way to make the lives of themselves and their customers difficult.

[Arun Opto 22 0]

"Dick Morley", actually, not Chris Morley. Dick and Modicon (Schneider Electric) credits themselves with the invention of the first PLC...the Model 084 (84th attempt).

[Armadillo852 25]

We can all "guess" at the share, but unless someone comes up with an article published by a reliable source, we will never know. I have worked at many GM, Ford, Chrysler, BMW, Toyota, and Tier 1 and Tier 2 automotive plants. I can say almost for certain that GM will not go plant wide with any other PLC than AB. I say this for the same reason Ford doesn't want to stop using the PLC-5's. The cost to "train" union Employees on a new pice of technology is way too much investment for the return. Hence, "no bang for the buck." I have worked with a lot of PLC's and they all have a place. If you want motion control built in, AB for sure. If you are looking function blocks , then Siemens is the way to go. If you want high speed, than Beckhoff is the way to go. Again, I think it really comes down to who you like as to who would get the major share in the market. Just my 2 cents worth.

[Rodney 21]

I know this topic is for the U.S market share. However it is not the here in the U.K. I would say Siemens with "Simotion Software" is making big gains. It also makes a difference on where the O.E.M is located again Germany is a big influence on major automation prgects. My 2 Pence Rodney

[Crossbow 295]

Not true on the GM thing. I was in 2 plants in NY where the spec was Modicon for all controllers in a GM Powertrain plant. 984 and Quantum. A Tier 1 supplier in NY was Modicon too. I've been in a couple of Tier 1 suppliers for Toyota, and Mitsubishi was used. Used to see quite a bit of Omron in automotive plants too, though my line of work has changed and I don't get on site as much.

[Mike Lamond 1]

One Tier 1 supplier (ex-Chrysler owned) in NY that I've been in had Modicon so old that it pre-dated Gould, but the new equipment was A-B PLC-5 or SLC-500 for small stuff. There were four different brands of robots. When GM built the Saturn auto plant in Spring Hill, Tennesee, they chose GE as the controls vendor. In 1988, I was still fairly new at my first job when we built four control panels for an equipment supplier using Series 6+ PLCs and IDT Panelmate 1 or 2 (pre Cutler-Hammer!) HMIs. My second job picked up the pieces after the first company went bankrupt and in 1994 we built an identical fifth panel. When a brand new (old stock) memory board failed, the only replacement available came from the plant spare parts supply. My understanding later is that the plant utilities were built using the old Series 1 thru 6 hardware, while the production line equipment used the new 90-30 and 90-70 PLC families. Mike