Why do you pay for PLC programming software?

J

James Ingraham

Anonymous said:
"AB may be better but it is not 5 to 10 times better as the cost would suggest."

I have NOT found Allen-Bradley to be 5 to 10 times as expensive as other brands. Siemens is every bit as expensive as A-B. We haven't used Modicon in a long time, but when we did there was no significant difference in price compared to A-B. Mitsubishi and GE Fanuc are also similar in cost. In fact, I've never run into ANY PLC that apples-to-apples is one-tenth the price of A-B.

Granted, there is a premium for A-B. That premium is very similar to the one for other "Tier 1" vendors like Siemens. AutomationDirect.com does have $99 PLC's, while a full-blown ControlLogix could run $10,000 or so. That's a hundred times more expensive! But it's not like comparing a Kia to a Cadillac; it's comparing a motor cycle to an 18-wheeler. A-B also has a $99 PLC.

-James Ingraham
Sage Automation, Inc.
 
C
Just about the only thing that proprietary software and the policies that accompany it do better than FOSS is to transfer money from your pocket to theirs. Unfortunately, this overriding goal tends to be in direct conflict with the users typical goal of knowing enough about the product to productively use it. This gets particularly bad for old products (often just a year or two) when they won't even sell you the information you need. That's why I like FOSS. I get the source, I get the information, and they are both mine for as long as I need them. No one can make them obsolete or bend you over the barrel. I'm not sure what's so whacked about that, it simply meets my goals rather than theirs.

Regards
cww
 
P
Great I understand PLC companies have to make returns, however as many posts indicate there is a real posibility for PLC companies to make a lot more money by making more sales. At present, costs would be the decider in what could gain PLC sales, the major portion of control system sales, not just for mining and the big players as it is presently structured toward. I worked in the mining industry as a tech, however trying to justify PLC costs in local industry even over BMS can't be considered. I am one of many, imagine the money you big guys are missing out on.

Just a thought.

Peter
 
S

someone who knows it's all true

This is so so true!!

Basically RA have 2 faces, and they can both be ugly. Even while there are a few good people there, their customer service is in effect, distributor integrator, stockist, etc. and it does not pay for you to give them any information.

I must point out though their products are top drawer, and there are inexpensive products in the range, ML 1100 on ethernet with a web server €360 cable €40 and prog sware €400.

Keep up the Oviscating,
 
E

Edward Mulder

A thread that runs for almost 6 years... Must be something that keeps us busy. Well the facts are these: if you want to use a PLC, you should prepare to pay anywhere between $400 to $1800 for the software. If you want to use an operator panel, another $30 to $400 for software.

Then the hardware. Somewhere between $1000-$3500, depending on your wishlist. Some vendors have $99 PLC's if you want something REALLY cheap.

Why these prices? Some people - vendor related - claim that you get what you pay for. Well, if we look at the microcontroller business, all the players (Microchip, Atmel, Luminance, Freescale, Zilog, etc.) offer their software for free. That's right: for nothing! And you get an integrated development environment, and that means: editor, assembler, in-circuit-debugger support, programmer support, and best of all: a simulator. All incorporated in one software package.

Their policy is to sell hardware, and make it as easy as possible to use that hardware. A microcontroller sells for $5-10, so you must sell a lot to justify that, and so they do.

Robustness? Well, that is only for the HARDWARE, because ladder-language is quite simple to implement. Open your PLC and you will see what that means: opto couplers, varistors, snubbers, DC-DC converters and RFI end EMI filters. I/O modules are normally coupled to the bus by means of 573 or compatible bus-buffers.

Why does my last client force me to use Siemens? Well, he paid a lot of money on licenses, programming cables, converters and so on, so he feels obliged to keep on using it. My project files measure 1.8Mb for a 40 rung program. The project file for the operator panel (12 text lines with 8 variables) is... 13.6Mb. Sorry guys, but this is ridiculous.

For a former project I used AutomationDirect. 80 PLCs plus operator panels. Everything worked fine. No hang-ups, crashes or other troubles. Has been functioning for 8 years now. Project files are 360Kb for a 3000 instructions program.

By the way, AutomationDirect - formerly Koyo - used to build PLCs that where relabelled as General Electric (you can mix them, I tried it) and... Siemens!!! So robustness is no excuse.

Does anybody know why the unifying IE11000 language did not provoke cheaper universal programming platforms? The industry does not want to give up such a profitable business?

Greetings,

Edward
 
C
In the past few years, I've had the chance to use many PLC brands, which is somewhat unusual as most places tend to stick to one. But I don't see at all the differences that justify the disparity in cost. I've been "auditioning" small PLCs for individual machines and integrating them into cells and I'm moving towards AD as well. The reasons might surprise fans of the high priced brands. One is that, for what I am doing, the "advanced" lines are way overkill. Another is that AD generally understands that you wouldn't be buying it if you didn't need it and they have a track record of getting stuff shipped and on site when you expect it. Their other gear is pretty carefully selected not just for cost, but to do the job reliably. I have AB support locally and that's about it. I find AD's support more useful and I don't get shaken down for it. Of course for other people, who have different needs, the big bucks may be worth it. But the biggest point is that I have found AD's small PLCs to be _much_ more reliable than the Micrologix series. Yes, the AD software isn't as kewl as RSLogix, but I have far less investment in supporting their entire line than AB or Siemens. For some reason the spendy guys can't seem to ever use the same setup twice, and when you support several ages and types, you need cabinets full of expensive adapters and crap. A couple of simple cables, which they give you the diagrams for, and which use standard and readily available connectors, fit all the AD gear I work on. That's the way it should be, as the other crap offers _no_ advantage that I can see. In general, if the AD PLCs won't do it, there's a much better way to do it than with a PLC. So it is the total package and the thinking behind it, that has to align with what I am trying to do. The software is important, but what good is the most convenient software when you need it all the time to recover and replace failed units (AB)? Or if using the software is a lot bigger project than the job you are trying to accomplish (Siemens)? I like the balance struck by AD and the value. The spendy brands should examine their value proposition in comparison.

Regards

cww
 
D

Dave Ferguson

I have 98 AB processors all either PLC-5 or Controllogix in about a 2-1
ratio with the PLC-5's being slowly phased out. They are all networked
together and have yet to have the following Curt quote happen...........

" but what good is the most
convenient software when you need it
all the time to recover and replace failed
units (AB)?"

I have had one processor failure in 19 years that was not recoverable and
this was the only "non-human" related failure I can think of in 19 years.
It was a recall chip failure that was a firmware issue.

But then again I do not use the "cheap" stuff. Let's not talk about
INITIAL cost of stuff and talk about lifecycle cost and productivity
costs. I pay more, but then again I have a single vendor training and
spare parts and backward compatibility. I cannot afford to find the
Siemens expert, the AD expert, the guy who can hook up the Brand - X skid
and on and on. We choose to run 24/7/365 with roughly 2-4 days long downs
a year. We have had a total mill down roughly 8 times in my 19 years there
where you could effect the total mill processors.

But then again this is a Paper Mill that cannot afford minutes let alone
hours of downtime........................................

Write back in 19 years and tell be about costs and failures, until
then................more FUD. YOU GET WHAT YOU PAY FOR........LONG TERM.

Dave Ferguson
 
M

Michael Griffin

In reply to Dave Ferguson: The Koyo PLCs have been around for a long time, and they've seemed pretty reliable in my experience. Their smaller PLCs (which are their main sellers) are also pretty simple and quite conventional. I doubt anyone would need an expert to figure them out. The smaller Siemens PLC (S7-200) series are similar in that respect (although the S7-300/400 can be quite complex). If someone has trouble understanding either of these, I wouldn't let them near an AB Micrologix either.
 
C
Hi Dave,

> I have 98 AB processors all either PLC-5 or Controllogix in about a 2-1
ratio with the PLC-5's being slowly phased out. They are all networked
together and have yet to have the following Curt quote happen...........
>
> " but what good is the most
convenient software when you need it
all the time to recover and replace failed
units (AB)?"
>
> I have had one processor failure in 19 years that was not recoverable and
this was the only "non-human" related failure I can think of in 19 years.
It was a recall chip failure that was a firmware issue. <

I haven't had any such problems with the larger
AB PLCs either, But I have installed and replaced
at least 8 Micrologix in 3 years. Just recently I've started replacing them with Mitsubishi FX PLCs because some of the replacements were acting up. Your recollection on processor failure is as it should be, it just shouldn't happen. I'm not especially fond of the FX or it's software, but I can't justify buying a product with that kind of reliability issues. And these are small point count applications that aren't really economic with a SLC even it I had the room. I had the local support team look at it and tried the suggested filter and grounding, etc. I wish the AD in this class came with sensor power, they have better availability than the Mitsubishi, I can never get the lowest point count models from Mitsubishi.

> But then again I do not use the "cheap" stuff. Let's not talk about
INITIAL cost of stuff and talk about lifecycle cost and productivity
costs. I pay more, but then again I have a single vendor training and
spare parts and backward compatibility. I cannot afford to find the
Siemens expert, the AD expert, the guy who can hook up the Brand - X skid
and on and on. We choose to run 24/7/365 with roughly 2-4 days long downs
a year. We have had a total mill down roughly 8 times in my 19 years there
where you could effect the total mill processors. <

Yes, but when you are building a compressed air
economizer or some other low point count app you won't get the work if you propose a $2k PLC.

> But then again this is a Paper Mill that cannot afford minutes let alone hours of downtime........................................
>
> Write back in 19 years and tell be about costs and failures, until
then... more FUD. YOU GET WHAT YOU PAY FOR... LONG TERM. <

No FUD, I can get you the AB case #s.

Regards

cww
 
D

DAVE FERGUSON

I had an interesting conversation with a Rockwell person, who had been around for a long time.

He commented that it used to be that they (and others) used to build stuff with XXXX years of mean time between failures and that the components were of shall we say "A much higher class and lifespan than today".

He commented how now all of the vendors and electronic components are in his word "Off the shelf stuff" that will not last like the old stuff.

We were discussing how I had equipment that was 20+ years old and still running and that we had no plans of removing some of it. He said that people today are still assuming that the stuff they buy now is made the same. He said that a lot of stuff is built with the expectation that it will be swapped out in 10 years or less. I think there are a lot of people in for a big surprise down the road in obsolescence.

PS: He pointed out that it was everyone's stuff not just theirs...

Dave
 
The Koyo is way harder to understand than AB. To work with Directlogic you have to have a clear understanding of numbering systems, i.e. octal, hexidecimal, just to be able to figure out the inputs. At least with AB all you have to really understand is an integer and you could get by. I have never touched a Siemens but come on... how hard can it really be?
 
M

Michael Griffin

In reply to DAVE FERGUSON: I've never personally had any complaints about the quality of AB's hardware. But then Mr. Wuollet's problems seem to be just with their latest smaller PLCs.

As for stuff lasting 10 years or less, that is a reasonable expectation for smaller PLCs in a lot of industries. The machines they go in end up being scrapped or rebuilt in about that time frame because that is how long the product they are manufacturing is in production. The problems I see with some PLCs (without naming any brands) is when the hardware doesn't even last a few months or falls about when being installed. And it isn't necessarily the "cheap" brands where this problem is the worst.

Even if the design is good, a lot of automation companies (perhaps all now) have outsourced production to whatever subcontractor bids the lowest. And sometimes the cheapest is pretty shoddy.

There is a big market in counterfeit electronic components in certain countries that assemble a lot of electronic goods. The counterfeit components are sometimes completely fake. Other times they come from the genuine factory, but are defective scrap parts that were salvaged and repackaged or lower spec parts that were relabelled. In some markets, up to 30% of the electronic components are salvaged scrap or relabelled.

The purchasing people handing out contracts to subcontractors know what is going on, and they know why they are getting a lower price. So do the people they work for, all the way up to the top. They just cover their eyes and pretend they don't know what is going on and collect a nice bonus for finding a cheaper supplier. When the problems come up later, they get blamed on "the supplier" or "its QC's fault for letting it happen". It's never the fault of the people who actually made the decision.
 
J

Jeremy Pollard

Dave - not that I know much about it but Rhos will also make a big difference. Solder without lead, is proving to be a big deal breaker in the contracts of longevity!

As with everything else we are in a throw-away society!!

Cheers from: Jeremy Pollard, CET The Caring Canuckian!
www[.]tsuonline.com

Control Design www[.]controldesign.com
Manufacturing Automation www[.]automationmag.com
 
C
It depends on where you are coming from. Having a computer deal with decimal integers is actually quite a high level function. Octal is a little strange for those late to the party. But you have to understand hexadecimal to get far in AB as well. Siemens organization, with all the more bizarre bits of indirection and OO is far more complex, even for trivial programs. Not necessarily bad in itself, but far more complicated than it needs to be for most PLC fare. It's like assembly programming. Programming a Z80 with a flat map is just a lot more fun than dealing with x86 segmented addressing, etc., etc. The burden on the coder is much less.

Regards

cww
 
M

Michael Batchelor

I'm sorry, but I have to disagree with this. Neither the Koyo nor any of the AB line is "intuitive" in the sense of a infant suckling. They are learned skills, not intuitive understandings.

I will freely admit that there is a cultural componant to what we find easier to learn, i.e. it's easier to work with something that "looks like" something else you already know. If you grew up in the TI world then the Koyos look like a home run king.

How hard can a Siemens be? No harder than a Koyo, or an AB, or a Mitsubishi, or an ABB, ya-da, ya-da, ya-da. They're all different, and they all do some things well and other things poorly.

MB
--
Michael Batchelor
www.IndustrialInformatics.com
 
M

Michael Griffin

In reply to brian: AB used to use octal addressing. That didn't stop AB fans from claiming that the way those PLCs worked was only right way of doing things.

The Koyo PLCs are pretty typical Japanese style PLCs. They were also sold at various times under the GE, TI, and Siemens labels. The present Siemens S7-200 series are pretty similar to the Koyo, except they use IEC style addressing and dropped the older BCD style instructions (Koyo has both natural binary and BCD instruction sets). All of these are reasonably simple. As to whose is the "best" is a matter of splitting hairs.

The Siemens S7-300/400 is very different from the Siemens S7-200. The S7-300/400 *are* very complicated, and the programming software is very large and complex (I believe this is the software Mr. Wuollet was referring to as requiring more work to use the software than to write the PLC program). If you've never done anything with an S7-300/400 or an S5, then you haven't seen a complicated PLC.
 
C
Part of it depends on how GUIfied you are also. But there are good and bad of all genre. Step 5 and Omron's old stuff, most would agree are kinda nasty and don't flow logically.

RSLogix is about the best Windows implementation I've seen, they take the best advantage of features for prompting and flow. Many of the others show more signs of being Windowfied DOS programs. Some of the old DOS programs are easier to work with than some of the newer Windows offerings IMHO.

I prefer Logicmaster to _any_ of the new GEFanuc offerings which share some traits of Step 7. Since I am usually interested in the ladder, I like software that is written to deliver you there.

I don't like the program broken up into dozens or even hundreds of files. None of these can prevent me from using the software, just liking it.

Again, it's like other programming. I could do everything in assembler, given enough time and patience. I greatly prefer C as it's much less fatiguing. And the higher level languages gloss over too much detail. I've thought about this quite a bit lately as I wonder what the next step in automation programming will be. Perhaps a "machine description language" where you state what you want to do and standard block are put in place. Who knows?

Regards
cww
 
D
The point to the argument is that no matter which PLC you use, there is a learning and maintaining curve. Some info is transferable, and if you are just adding a "rung" an input and a coil, most should be "fairly" straight forward.

But At 3 AM, with what you think is the proper laptop, proper software version and copy of the program and comments and the proper cable and write-up, having 10 different types of skid controllers can and does become a nightmare (been there done that.)

So sometimes having a lot of "cheap" controllers isn't so cheap. We have chosen to use a couple of standard controllers from a single or small list of manufacturers. And although there are a couple of instances where we have Porsche's for getting groceries (way overkill), the benefits of spare parts and the tax consequences of the spare parts, training, experience, software, cables, and long range obsolescence and UPTIME far outweigh the money we save short term by getting the "deal of the week."

Just the way we have chosen to do things. We also standardize on one industrial switch, try to standardize on PC's (harder) and laptops (still harder) and DCS, discrete valves, control valves, transmitters etc. Sometimes we fight with vendors and pay much more to get it "our way" but again UPTIME is paramount for us and we cannot afford very often to have even a little skid go out on us as it may feed something highly needed to run.

Do we see the short sighted "bean counters" all the time trying to fight us, you bet. But we have had at least for a long time, a management that recognizes that sometimes the short term budget gain, may burn you later. Usually with the projects we do, the control system is such a small, small part of the overall project cost that it really is peanuts to us (most of the time.)

I have experienced the short sighted "low bidder" mentality, and I will take what we do over those 3 am or all day Saturday nightmares that the "low bidder" bean-counter at all costs, short term thinking create. I also will take off-the shelf programming when it makes sense over "custom" programming in some obscure language by the one guy we have stuff as inevitably, he leaves, moves to a different position, had it on another computer etc. By sticking with major vendors, we can have expertise, but can also find 200 guys on Monster, at some Integrator or down the street.

If we go with 50 one-offs, good luck.................

Dave Ferguson
 
C
Yes, Step 7/S7-400 is the epitome of abandoning the virtues of the classic PLC.  It isn't much simpler than using an assembler and a good macro
library on the same hardware.  It's pretty complex when you are doing the programming and can be really tedious figuring out what others
have written, especially since fans of the beast like to alternate between ladder and instruction list. A large program may have a whole screenful of blocks and you jump around a lot.

Finding a particular bit of logic and rounding up it's data can take a long time, especially if you don't use the stuff every day.

Regards
cww
 
Top