Why do you pay for PLC programming software?

I think that if we're looking for a system for my plant I have to consider all the options and prices, of course, but also the learning curve and how easy is to programe the PLC for me and the maintenance people.

AB is more expensive than Omron and others, but it's very easy to create a project because the software do a lot of things for us in the background and everything is quite integrated. AB products are very easy to communicate between them.

Siemens or Modicon are as much as expensive as AB, but their software is more difficult to use and the communications aren't as fast to implement.

Depends on your plant's philosophy, you might want to spend time (money) doing some job that with other software could be faster (cheaper) but with a cheap software OR spend little time in learning and developing but more money in software.

The first example is cheaper at medium term, but the latter is more cheaper at long term... it depends on you...
 
B
The real answer is that most places don't need all that many software licenses, and the cost kind of fades into the general overhead. Its just not as big a cost as the cost of using some software that is less expensive, and not having the engineers and techs up to speed and loosing production when
something goes wrong.

The hardware cost of all the brands is very close, especially at the OEM and larger user level.
 
I agree that the initial cost of the software licenses do not have much impact. But, when an OEM like AB publishes 17 versions of RSLogix5000 that have to be in 'lock-step' with the controller's firmware, you run into some problems. If you're into making stuff work, being on the phone all day with the purchasing department, the distributor, and the OEM to sort out your licensing and support agreements is not pleasant.
 
D
I agree with you that this can be painful but it goes for most of the vendors now. When you have flash able ram in all of these controllers, all of the vendors are starting to "lock step" you.

One thing on the Logix side, after version 10, you can have multiple "versions" of the software on the computer so that you can talk to all regardless of the version.

Dave
 
D
I agree with a lot of what you said but I have a little different take on the RSLogix 5k and ControlLogix.

CLogix supports the 61131-3 (4 of the 5 languages).

The languages supported are Ladder, SFC, Function Block, and Structured Text.

None of these are really IT languages but structured text comes closest. Structured text programming like languages have been around in PLC's forever IE S5 Siemens. Now if you say in AMERICA ladder for electricians is "normal", I would agree. But SFC has been in PLC 5 forever, Ladder of course goes without saying and Function Block is not there for the IT folk, it is there for the Process people (DCS folks) who are moving over to the PAC (Programmable Automation Controller) world.

I think that the inclusion of these languages is not per say for an IT function but for flexibility of things like motion control, Process Control, batch control structure, etc. I can still do all of those things in PLC 5 but I find Clogix far easier to use. As with anything, if you do not comment and write good structured code, no one will be able to troubleshoot, not the Electrician, Engineer not the IT guy. I have been a part of some European programs that are extremely efficient from a "programming" point of view, bit shifting, etc., but noone can fix it when or if it breaks. That has nothing to do with the PLC but the programmer.

As always I claim it is the Painter and not the Paint set.

Dave Ferguson
 
D

Dave Ferguson

Sorry for the late reply was on vacation...

No there is no issue that I am aware of, I also checked with someone else if they had any timer issues inside a custom FB. There are none we are
aware of, they work fine, I use them for my fault logic for custom motor and valve control and they time fine...

My friend uses them in simulation blocks he has built and they work fine.

Dave Ferguson
 
D
Just thought I would put my two cents in:

Upon changing jobs three years ago I was introduced to Beckhoff Automation's PC control platform,and their programming software, Twincat. After using their software/hardware for the past three years I would much rather implement a new control system using this than I would any other PLC/PC control manufacturer. The only reserve I have to making that statement is that I have never used S5 or S7 but from my discussions with the OEM's (who previous used S7 but changed due to flexibility and price) who installed our equipment they are somewhat similar in functionality but Twincat is more open, flexible, and overall more user friendly. Another plus is their software cost. When you purchase one of their industrial pc's or embedded industrial pc's (much less than an AB Control logix processor) you get one licence for their software which will reside on that pc, which will obviously controlling your equipment. For your maintenance and engineering personnel the Twincat development software is free of charge you only are required to purchase a license for the runtime ( so only one license required per PC actually running equipment) I think this is fair. Hardware costs are also what I feel as "reasonable", definitely cheaper than AB or Siemens. I am not trying to make a big sales pitch (although it sounds like it) but if anyone is interested in a more open, cost effective, and still reliable platform then I would suggest at least giving them a look. You can download a full version of twincat including runtime for free from their website. http://www.beckhoff.com
 
Liability? Every user license I've ever read for software places the liability on the user and no fault can be be placed or implied on IDE software.

In fact a lot of software costs appear to be lawyer fees in writing the user license in such a way as to absolve them of all liabilities.
 
Beckhoff Incorporates a diagnostic tool in there software called Twincat Scopeview. It has a configurable update rate down to 1ms. You can also set up an I/O task and link it directly to your scope and even trigger a pre-configured scope to start recording automatically within your PLC logic. Very handy.
 
C
The problem with both is that they only work with one brand. And bear in mind you can get a Mixed signal scope with a 16 channel logic analyzer and two analog channels with 512k samples memory depth and 400MSamples/sec for about $600 and use it on anything. And you can believe what it tells you. Makes big bucks for a one trick pony seem like kind of a waste. A DSO with flexible triggering and deep memory makes a really good time machine for what happened before the once every two week failure. Digital power is cheap and becoming cheaper. Google Rigol DSO.

Regards
cww
 
K

Ken Emmons Jr.

Of course you can use hardware analyzers. We use an HP 2 channel DSO scope with 16 digital inputs. For complex triggering and viewing data on internal variables, you can't beat a scan based softlogic analyzer within the PLC. That and pulling in variable names and things automatically is just icing on the cake.

KEJR
 
M

Michael Griffin

In reply to Curt Wuollet: The software I/O timing chart (AB had this years ago) is really intended for something a bit different than a real logic analyser or storage scope. The logic timing chart is intended to help find bugs in your program logic, while a logic analyser or storage scope tells you what is really going on, including hardware problems.

I've used a scope to diagnose machine problems a number of times, but I only ever used a PLC programming software based software timing chart once (more for the sake of trying it out, than anything else). It's one of those things that sounds nice in theory, but is not really that useful of a feature in practice. It's definitely not a substitute for a scope.
 
C
I've found them pretty much useless especially for timing problems since they depend on the system that isn't working. And many factors, some of them very subtle, can cause them to "lie".  Many  glitches  wouldn't be glitches if they occurred during the detection windows.

The glitch is there for the same reason that the system can't see it.  But looking at the real waveforms is not for the faint of heart, you sometimes wonder how the thing works rather than why it  occasionally doesn't.  Brute force filtering covers many sins. Most "doesn't work" problems with encoders for example are immediately obvious with a long record.

Regards
cww
 
Hi All.

Nice debate.
Well, I have used Omron (Old DOS based), AB (500 & 5000), Siemens (S5&S7), Opto22, GE Fanuc and Beckhoff.

I currently suggest Beckhoff for every replacement not only because the software is free, but also because they support almost every fieldbus you can think of. It's simple. It has 7 languages which are IEC 61131-3. The support is great. The documentation I would say needs a little bit of a touch up, but pick up your phone and there is always someone to help. All there components are well priced as well. On top of that their new embedded systems offer the power of PCs and the robustness of Standard PLCs. The software is easy to use, quick, you can monitor as many pages as you want, upload, download, online changes, hundreds of FBs to help. (You can also create your own FBs and functions. Also you don't have to pay extra fees for extra functions in the software. And the entire range of Beckhoff only uses one software. Not like RS500 or S5 where now you have to change your hardware and software and pay for it because they're discontinued. I have to say when I first used it like most things that are new I was hesitant to use it, but the Beckhoff guys helped me in my first project with them and I have never looked back. In fact when I install other systems I feel really sorry for all the SIs who have to use other products and have to worry every time they buy from a company if the firmware is going to be okay.

That's my 2 cents. Let me know if anyone needs any details.

Waz
 
Ok, here is my 2 cents. A&B's software is way over-priced. Here's why I think this. A&B's software runs on Windows. A&B charges $1100+ for software. Microsoft charges $200 to $400 for a complete OS, that takes years to develop and test? So A&B is telling me since their software is $1100+ that it is more complex than the operating system that it runs on. "Ya right". Also the programming language that they are using costs less than their software.

This is right off Microsoft's web page:
Visual Studio 2005 Professional Edition - $799

So how can A&B justify their cost
if Microsoft can sell a complicated OS for under $500 and a complete programming package for under $1000? There is no way that A&B can tell me that their software is worth the price you are paying for a name. "The software should be part of the support for their product at no cost. There are lots of companies that offer their software for free and that is the way it should be."

I would never pay over $1000 for any software.
 
N
That argument has little basis. Microsoft sells orders of magnitude more money worth of software than AB - it's a matter of economy of scale. In nitch markets vendors sell lower quantities of software, typically at higher prices. If someone could profitably sell a better product for such industries at a more favorable price they probably would.

To determine if the software value is worth it for you, simply weigh the cost versus benefits considering alternatives.

I don't see how you can reasonably expect a top notch piece of software to be included, updated, and supported for free in this specific industry. I am a fan of the idea and open source. Too bad such things don't currently exist for us.

----
Nathan Boeger
http://notanotherindustrialblog.blogspot.com
"Design Simplicity Cures Engineered Complexity"
 
M

Michael Batchelor

I certainly don't want to get into a fight defending either Microsoft or AB, but your analogy is flawed.

In a commodity market, like 4-40 screws, the materials and production cost are significant to the cost of goods sold, while the design and engineering of the "product" are secondary to the cost of the design of the tooling to manufacture the product.

In the software world the materials and production costs are lost in the noise of the system compared to the design and engineering of the "product." (In these examples Windows and RSLogix) The vendor has to recover the costs *SOMEWHERE* in the market scheme, either in the price of the hardware - as for companies that give the software away for "free" as an inducement to get you to buy the hardware, or by charging for the software directly. Obviously since Microsoft doesn't really sell "hardware' (game boxes and keyboards aside) they can't recover the cost of Windows, Office, or Visual Studio in the sales of the hardware so they must charge for the software directly.

Let's work with round numbers on a scale we can all think about instead of trying to look at real sales figures. Never mind the fact that neither Microsoft or AB are going to give me a quick peek at their books.

Assume that MS Windows is magically exactly 1000 times more complicated than RSLogix, and takes exactly 1000 times the production effort to produce. Also assume that somehow both MS and AB magically have identical overhead.

Now assume that the market size for the MS products are 100,000,000 units, while the market size for RSLogix is 1,000 units. (Certainly not exact numbers, but I think it's a credible wild guess.)

Now, if you scale the production costs (based on effort - remember we pretended Windows was exactly 1,000 time more complex than Logix) and then spread them across the pool of customers to amortize the costs, you see that Logix costs 100 time more per unit sale to produce. Wow, maybe Logix is a bargain!

NOW FOR THE KILLER REVELATION

Fudge these numbers around any way you want, but the truth is that the invisible hand of the market sets the price. If everyone stopped buying Logix, the price would plummet, just like the companies who "give it away for free." If everyone stopped buying Windows the price would plummet. For both products, once the return crossed the threshold where the sales no longer offsets the cost to produce the product would vanish.

The market, and only the market, controls these prices. Arguments about the production costs are moot. They might make you feel better, or make you angry. And while it's true that MS does enjoy some ability to manipulate prices because of a monopoly position, the market willingly gave them that monopoly because they met customer needs adequately. Perhaps they didn't do a stellar job of meeting customer needs, but they did an adequate job in the face of other reasonable contenders, and people voted with their dollars.

AB is the same thing. They do have a lion's share of the US market, and they got there because people voted with their dollars. I have never once had an AB goon squad show up at my building and break out the windows simply because I got P.O. at them and didn't pay the support bribe that year. And one of my biggest customers is making a concerted effort to rip out all of the AB and replace them with Seimens across all their plants in the US and Mexico, so AB isn't holding on to the market. But guess what? The Siemens software costs money, too. Does that mean they are secretly in cahoots with AB to rape the customer for software? Not on your life. The Seimens guys I've met would like to cut AB's throat and hang the dead PLC carcases on pikes in front as a warning in every plant they can. But the marketing model stands because the customers find it adequate and pay the price, either willingly or grudgingly. But the invisible hand of the market rules the roost.

MB
--
Michael Batchelor
www.IndustrialInformatics.com
 
Surely it's all about the value of the return.

Someone once told me they happily paid £1000 for an early calculator, because it had a square root button, and it saved them hours of design time every day.

And way back in the 80s, we once wrote some software for an engineering company for doing their gear calculations. It allowed them to do a design in five minutes when it previously took them a week at the drawing board. Not unsurprisingly, they were very happy to pay over £1000 for it.

You have to view the programming software as a means of improving the productivity of the end user's automation - and it's either worth buying it or it ain't.

Don't forget that before programming software, we had to buy programming terminals - and they cost a lot more than $1000 (as well as weighing a ton - anyone remember the T50?!). ICOM started out as a two-man company; they reverse-engineered the protocols and program structure and developed a far superior (and much cheaper) programming tool compared to these dedicated terminals. The reason it caught on with automation companies is because it very quickly paid for itself many times over. A similar thing happened here in the UK with Siemens S5 in the 80s.

When AB realised their terminals couldn't compete with PC-based programming tools, they tried developing their own and when they failed horribly (APS!), they simply bought ICOM - and succeeded (at least for a while) in keeping the excellent team together. My bet is if anyone thought it would be profitable to "do another ICOM" and produce an independent, commercially viable programming tool, it would have happened by now.

It hasn't, because developing and supporting progamming software IS expensive. The reason MS can sell Windows for $400 is because they sell MILLIONS of copies. PLC users don't present that kind of market, so the costs have to be spread among fewer buyers.

However, one underhand trick the PLC manufacturers do pull is to keep their comms protocols secret these days. For example, back in the 80's, the DF1 protocol was published openly - anyone could design a program for any device to read and write PLC data through a plain old serial port. For the more ambitious, ICOM used to sell a C library so your DOS applications could talk via a KT card. No more. You HAVE to buy RSLinx. This is the kind of profiteering that the PLC manufacturers should be made to feel ashamed of: ethernet hardware costs pennies and TCP/IP protocols are open to the world - unless PLCs are involved. Then you are back to buying hardware and software costing thousands.

An interesting debate indeed.
 
Top