Beckhoff vs. Allen Bradley... who is better?

Thanks Archie. Great post.

I've been an A-B guy from way back and a CodeSys (less Beckhoff, more others) over the past 3 years.

RSLogix is a friendlier environment than either CodeSys 2 or 3 BUT when it comes to ease of interface, as a trained programmer, CodeSys stole my heart. The ability to write protocols over RS-485 or Ethernet, having a PLC that can "talk" to MS-SQL and is an FTP server, the ability to send secure emails, the built in web-server capabilities, external and open libraries such as OSCAT, etc, etc, convinced me that, while RSLogix is great for most common tasks, when you need that extra something, CodeSys provides it (generally) and provides it at no cost (generally.)

I almost forgot that CodeSys works on multiple platforms by multiple manufacturers.

CodeSys controllers tend to be more apt to stopping the controller than A-B, Siemens, or the others, which basically means that they fit a discrete manufacturing better than a process control.

I don't agree entirely with the learning curve being steep even though there will still be a learning curve. There are some definite benefits to CodeSys ladder such as the ease of integration of fb into Ladder that shouldn't be overlooked. And Archie is spot on regarding the other languages. It seems to me that CodeSys was designed with ST as it's primary programming language.

Good luck,

Yosi
 
I have no opinion about Beckhoff vs AB, but I do have an opinion about the need to change the PLC of a standard machine to standardise in a particular plant.

Sure it is an advantage to have the same PLC everywhere, but for the machine supplier it is exactly the same, to stick to their standard design and not change anything.
Changing the PLC is even harder for a machine supplier than it is for a plant.
Chances are that the programs (both PLC and HMI) are much more complicated than any program that the plant maintenance are designing by themselves, maybe the machine supplier has invested man-years in developing a program for a particular machine. If the machine supplier has to change the PLC they may have to start all over again. This means for the customer: Longer delivery time of the new machine, higher cost for the machine, probably a series of issues that will only be fixed after some timer after delivery, much worse or non-existing support from the supplier.
What the plant customer gains is that spare parts are in line with what they have already, and the ability to "troubleshoot" the program.
I put the "troubleshoot" in quotes because I see this as a false advantage. If a machine PLC program is well designed and has been finetuned for many years, there will be zero reason for an end-customer to troubleshoot the machine by looking into the PLC program.

Statements like this makes me cringe:
bjh said:
How easy is it to find a service technician to hire in the US that can at least futz around with RSLogix 500/5000 enough to get online and troubleshoot a bad sensor?
To have to go online with the PLC to "troubleshoot a sensor" is a massive fail. No need to explain why !
ASF said:
or because the night shift guy didn't get trained yet and has to google how to get online with the PLC
How about contacting the 24/7 support of the original machine manufacturer instead ?
 
I see this stuff all the time. Beckhoff was a big learning curve. AB was a big learning curve. EtherCAT is fabulously fast - do you need it unless you are doing motion control? I normally use Omron and that is my preference. They have EtherCAT if you do motion control even with EtherCAT ring so you can detect loss of a section of the ring. It is all relative. Ladder in Beckhoff is woeful! AB is better but can be painful as well.m Faster machine? Rubbish! All PLCs are quite fast these days. I usually use Omron's remote I/O - it is a CIP remote I/O system.1000 I/O updated in 1 millisecond - the processor is not that fast! I can understand using EtherCAT for motion - small packets - extremely efficient but do you need it? Have also used GE, Schneider, Siemens, Hitachi and quite a few others as well. Sticking where I am to be honest.
 
I have no opinion about Beckhoff vs AB, but I do have an opinion about the need to change the PLC of a standard machine to standardise in a particular plant.

Changing the PLC is even harder for a machine supplier than it is for a plant.

If the machine supplier has to change the PLC they may have to start all over again.


+1
All of the machine builders that I've seen have a favourite PLC, HMI, etc.
They design the machine with that, first.

If you really want that machine with a different HMI or PLC ... take a look at Jesper's post again.
 
Chances are that the programs (both PLC and HMI) are much more complicated... probably a series of issues that will only be fixed after some timer after delivery, much worse or non-existing support from the supplier.
What the plant customer gains is that spare parts are in line with what they have already, and the ability to "troubleshoot" the program.

Amen. Our plant has standardized on A-B hardware. We've purchased a handful of machines from builders that don't normally use A-B. The end result is a machine that mostly works, but was built and programmed very inefficiently because the builder was not used to A-B. I've spent more time debugging "standard" machines than I should have for this reason.
 
I have no opinion about Beckhoff vs AB, but I do have an opinion about the need to change the PLC of a standard machine to standardise in a particular plant.

Sure it is an advantage to have the same PLC everywhere, but for the machine supplier it is exactly the same, to stick to their standard design and not change anything.
Changing the PLC is even harder for a machine supplier than it is for a plant.
Chances are that the programs (both PLC and HMI) are much more complicated than any program that the plant maintenance are designing by themselves, maybe the machine supplier has invested man-years in developing a program for a particular machine. If the machine supplier has to change the PLC they may have to start all over again.

OK, a bit OT here. I would also question this... We have a supplier that built a number of machines for our process, that all used AB ML1200s. They had a standard program library that had sections that could be called depending on which options were purchased with the machine, and were built with AB hardware for everything from the switches and pilot lights to the terminal blocks. We had all the necessary spares to prevent issues if something failed. About 5 years ago, the latest two machines they built for us came in with Siemens EVERYTHING. Now we needed new spares, new software, and new learning curve, because they obviously redid their programming. Did they save money, because we certainly did not. [/rant off]
 
Have mostly been using AB over the years, but from a programmers point of view I'll pick Beckhoff any day.
You're not tied down by what functionallity the PLC maker thinks you need, as you can program yourself out of almost anything with Beckhoff.
If they made redundant controllers I would pick them whenever I got the chance.
AB is way too expensive for what you get, and om top of that you have to pay for support and software. I personally don't think RA would be in business for long if they weren't already so big in plants like yours.
This probably won't help you much though as your PLC has already been programmed when you receive it, but might help shine a light on why your machine builder chose them.


Just my $0.02
 
Last edited:
To make machine faster, OEM may change to use more servo axis to replace mechanical system. For more axis, motion controller need faster scan speed for control synchronizing axes. At this point, Beckhoff may have advantage if TwinCAT run in powerful PC or embedded PC with high speed CPU. You can use even multicore processor to separate task between motion and other PLC tasks.

Not only motion controller can make machine faster but servo drive system with fast network speed, accurate position control will be factor for large synchronizing axes. EtherCAT network from Beckhoff is high performance network with many members which you can check from EtherCAT.org
 
Unless your building machines and shipping them out then I dont think it much matters... the last large manufacturing plant I worked at we had a couple hundred PLC's and we had just about everyone made, even a few custom made (roll your own) we had a lot of these http://www.robust.de/en/products/roll-cutting-and-winding-machines and at that time they had a micro processor also a S7 running them, it was a nightmare troubleshooting but as long as you have the tools to get in then it will not be a issue and should be fun learning something new
 
Thanks Archie for the great post....

I am reading these posts with mixed emotions, after working with only Allen Bradley for so long it would be exciting to get on to a whole new line of hardware and programming software....

But... I am not too excited about the possibility of sweating bullets at 3am with the line down, and not 100% sure of the fix.
 
I am reading these posts with mixed emotions, after working with only Allen Bradley for so long it would be exciting to get on to a whole new line of hardware and programming software....

But... I am not too excited about the possibility of sweating bullets at 3am with the line down, and not 100% sure of the fix.
When I did my first Beckhoff TwinCAT project, I dove in head first to a big project. My mistake was going into it with an AB programming mentality. That created some not so well written code for the TwinCAT/CodeSys world. To get the most out of TwinCAT, you have to think in a different paradigm.

To give a very simple example of what I mean, I will use the common task of creating a toggle. Using a single input to switch a single output on/off on the rising edge of the input. Most AB programmers (including myself) will tackle this by thinking of how to create the ladder code with contacts, coils, sets, resets, one shots, etc.

If I were to tackle a toggle for the first time in twinCAT, I start with the thought process of "this is an elementary element and a common task, so how do I create a re-usable function block that is easy for others to understand." This is what I would want it to look like in the program:
attachment.php


Since I plan to encapsulate the function into something that most people should be able to figure out what it does without the need to see the actual implementation code, that gives me the flexibility to use any language that makes it easiest for me. My choice would be structured text. I would then create the function block and use this code:
Code:
(* If the input was previously flase and now true (rising edge), then invert the output *)
If In AND NOT LastState THEN
     Out:=NOT Out;
END_IF

LastState:=In;

What I now have is an elementary function block that is re-usable as many times as I want. From the perspective of next programmer or person troubleshooting the machine, the function block encapsulates the complexity of the implementation which should make it easier to read the program of interest (the ladder code) and not be bogged down by the toggle implentation. For my future efforts, I do not have to copy and paste the code to implement the toggle function, I just import my Toggle function block into my next program or make more instances in my current program. The key is to plan them well and try to start with elementary functions. As time goes on you will build yourself a re-usable library and you will become faster with creating new programs using these.

If you plan your functions and function blocks carefully, they can be re-used to build more complex function blocks. Think of it as building a house. Cinder blocks, plywood, studs, etc are your function blocks that have already been created. You now use many studs to build a wall which can now become a higher level function block of a wall. Assembling many instances of all these pieces will result in a house. Using the exact same elementary pieces and you can build a completely different house.

ToggleModel.png
 
Last edited:

Similar Topics

Hi guys, I'm looking to design and program a robot pick and place cell. I'm in between using a Beckhoff or Allen Bradley. I like the Beckhoff...
Replies
2
Views
833
Hi, I would like to know what are the basic things we should consider before start migrate from the AB PLC to Beckhoff Controller? Is ther any...
Replies
1
Views
2,002
Hi everyone, This is my first time posting, so please forgive any omissions or mistakes. I am attempting to control the velocity of a stepper...
Replies
18
Views
754
Hello sameone have Beckhoff PLC Siemens Sinamics V90 configuration example?
Replies
0
Views
85
hello, I am using Beckhoff with TwinCAT3 and when I change or add some new hardware or for any reason, there is a mismatch in the real hardware vs...
Replies
1
Views
111
Back
Top Bottom