Hi,

I am returning to this list after a break of around a year and I wondered if much is happening in the area of ethernet as a fieldbus ? A year ago there were lots of posts about eg Modbus/ip , Profinet etc etc but now there doesn't seem to be so much. Are people using ethernet in factories atm or is this still very much a future technology ?

Mike Davies
Senior Software Engineer

 

Hi Mike

I fear adoption has been put off again by the race to proprietize and destroy all virtue of Ethernet in automation. Once folks get wind of the
fact that Ethernet is going to be simply more of the same, the interest dies rapidly. We'll have a dozen protos on Ethernet with none of them meeting expectations for interoperability. When people hear Ethernet, they think of the compatible, universal, interoperable networking we
enjoy in the general computing world. What is being delivered is exactly the same Tower of Babel on a different type of cable. From the general lack of interest, I'd say very few folks are fooled this time around. It's a pity no one "gets" what the promise of Ethernet was or what it could do to promote fieldbus in general. Those who don't study history are doomed to repeat it, I guess. It'll be interesting to watch how long they wait for it to take off before they decide to work together and attempt to provide people with a reason to adopt. Perhaps studying why Ethernet is ubiquitous elswhere might provide a clue.

Regards

cww
--
Free Tools!
Machine Automation Tools (LinuxPLC) Free, Truly Open & Publicly Owned
Industrial Automation Software For Linux. mat.sourceforge.net.
Day Job: Heartland Engineering, Automation & ATE for Automotive
Rebuilders.
Consultancy: Wide Open Technologies: Moving Business & Automation to
Linux.

 

Please remember that the Ethernet specification only defines sufficient levels of the protocol stack so that different higher-level protocols can
share the same physical medium and the same or different computers - just as the people next to you can be speaking in French while you and a
friend are speaking English.
If anything is being hijacked it is the name "Ethernet"

I think that everyone will admit that Ethernet has been one of the great standardization successes of the last decade or two!
Peter

Peter Clout
Vista Control Systems, Inc.
176 Central Park Square
Los Alamos, NM 87544-4031
(505) 662-2484
FAX (505) 662-3956
[email protected]
http://www.vista-control.com/

 

Yes Mike look at "www.ab.com":http://www.ab.com for Ethernet IP is now in the field with some success.

 

> What is being delivered is exactly
> the same Tower of Babel on a different type of cable. From the general
> lack of interest, I'd say very few folks are fooled this time around.
> It's a pity no one "gets" what the promise of Ethernet was or what it
> could do to promote fieldbus in general.

What about the standard data formats like XML and protocols like SOAP ? I see the OPC Foundation ( "http://www.opcfoundation.org/":http://www.opcfoundation.org/ ) is doing something
along these lines, does anyone see any devices supporting these protocols out in the factories and plants yet ?

ISTM that swapping Profibus, Devicenet, Modbus etc for their ethernet equivalent doesn't buy much advantage. Until the application layer data
format is universally spoken then it's hard to see how the advantages of things like self-reporting devices a la Universal PNP can be achieved.

I have an interest in this because I am trying to make a system with a user interface that can be understood by any MIS or controller. As far as I can see there is just no one interface that can confidently be expected to be universally understandable at some point in the future. Unless someone here knows different ?

Mike Davies

 

Hi Mike

"Davies, Mike" wrote:
>
> > What is being delivered
> > is exactly
> > the same Tower of Babel on a different type of cable. From the general
> > lack of interest, I'd say very few folks are fooled this time around.
> > It's a pity no one "gets" what the promise of Ethernet was or what it
> > could do to promote fieldbus in general.
>
> What about the standard data formats like XML and protocols like SOAP ? I
> see the OPC Foundation (http://www.opcfoundation.org/) is doing something
> along these lines, does anyone see any devices supporting these protocols
> out in the factories and plants yet ?

Well, they are standards, kinda. They don't get you all the way there either and they are vast overkill for pushing bits around. I personally
hope the trend towards using office technology doesn't continue. In the case of OPC and close cousins, they were adopted because that's the only
facility provided by the controlling monopoly and all must pay homage. It's been kludged and hacked enough to work but that hardly makes it desirable if you have a choice. XML and to a lesser extent SOAP are being used as if they were Open Standards, but I see compatibility with what
Microsoft wants to do as the primary reason for adopting them. Except for that, what do they offer for automation? Practically anything else you could mention would be more specific to our purposes. That's why I think it would be a better solution to Open up protocols that were designed
for automation rather than just accepting what Redmond is willing to let people do. There are many better means already in existence (and we
definitely don't need more) the only thing preventing more than token uptake is the IP and license issues that effectively prevent others from adopting them. This long standing paradox is antithetical to standardization on efficient protocols purpose built for automation. If even a small portion of the efforts expended on creating duplicate functionality were put towards standardization we'd have been there long ago. Instead R&D dollars are burned in the vain hope that even if the first X closed proprietary schemes achieved only mediocrity, X+1 will take the world by storm and deliver the market to them. I find very little basis for that improbable dream and the waste of resources is
appalling.

> ISTM that swapping Profibus, Devicenet, Modbus etc for their ethernet
> equivalent doesn't buy much advantage. Until the application layer data
> format is universally spoken then it's hard to see how the advantages of
> things like self-reporting devices a la Universal PNP can be achieved.
>
> I have an interest in this because I am trying to make a system with a user
> interface that can be understood by any MIS or controller. As far as I can
> see there is just no one interface that can confidently be expected to be
> universally understandable at some point in the future. Unless someone here
> knows different ?

That's very much by design. You have now gazed upon the Tower of Babel.

Regards

cww
--
Free Tools!
Machine Automation Tools (LinuxPLC) Free, Truly Open & Publicly Owned
Industrial Automation Software For Linux. mat.sourceforge.net.
Day Job: Heartland Engineering, Automation & ATE for Automotive
Rebuilders.
Consultancy: Wide Open Technologies: Moving Business & Automation to
Linux.

 

> What about the standard data formats like XML and protocols like SOAP ? <

XML just adds another layer, it doesn't actually help much. The actual meat of the protocol is still very much up to the vendors - tag names, whether values are in attributes or containers, how they're represented (unscaled? scaled? what units?), how queries or subscriptions are done, config, etc.


Jiri
--
Jiri Baum <[email protected]> http://www.csse.monash.edu.au/~jirib
MAT LinuxPLC project --- http://mat.sf.net --- Machine Automation Tools

 

Mike,
Take a look at "www.transparentfactory.com":http://www.transparentfactory.com .

First product was released in 98!... There are thousands of devices embedding Modbus/TCP in the field!

Regards,
Marc Leconte