Industry 4.0 and the programmer of today...

No soft PLCs for me! Useless. I see that the IIOT **** talks about predicting failures - bloody magic stuff it it can predict the failure of a prox sensor! Give it to me!

I never understood the appeal of a soft PLC when PLC families are kept for decades and anything related to a computer will last about 5 years in the market place. Why would someone embark on that is really beyond me.

The failure prediction is complete and utter poppycock... even yesterday I had a visit from Endress+Hauser where they were raving about a flowmeter with a wireless webserver in it and it created its own calibration reports. I asked how can that be if nothing is running through it? How can those calibration reports hold against a known and maintained standard? He then came out with a TUV certification (that wasn't readable and therefore likely to not answer my question).

Then he came out with a self calibrating temperature probe that is going to disrupt the market because at a specific point of 130 degrees it checks that the temperature is correct.
I didn't have the time to tell him the uselessness and stupidity of that claim as there is no such thing as a single point calibration.

Instrumentation and automation manufacturers are coming up with all this to maybe try and impress the young uns starting now (I'm 37, by the way) with all the buzzwords and new-ish technologies.

The reality however, is that with time and proper procedures it is possible to estimate fairly accurately the overall impact of failures in a system, but impossible to determine which. Meaning that over the course of a year, it is possible to know that X amount of dollars is going to be spent in fixing stuff, but exactly what is to be fixed is up to the gods or good technicians that inspect equipment.
Obviously, you can install diagnostic equipment in certain machines (like engines and so on) and get a good indication... but not at the proximity switch level I think.
 
I never understood the appeal of a soft PLC when PLC families are kept for decades and anything related to a computer will last about 5 years in the market place. Why would someone embark on that is really beyond me.

The failure prediction is complete and utter poppycock... even yesterday I had a visit from Endress+Hauser where they were raving about a flowmeter with a wireless webserver in it and it created its own calibration reports. I asked how can that be if nothing is running through it? How can those calibration reports hold against a known and maintained standard? He then came out with a TUV certification (that wasn't readable and therefore likely to not answer my question).

Then he came out with a self calibrating temperature probe that is going to disrupt the market because at a specific point of 130 degrees it checks that the temperature is correct.
I didn't have the time to tell him the uselessness and stupidity of that claim as there is no such thing as a single point calibration.

Instrumentation and automation manufacturers are coming up with all this to maybe try and impress the young uns starting now (I'm 37, by the way) with all the buzzwords and new-ish technologies.

The reality however, is that with time and proper procedures it is possible to estimate fairly accurately the overall impact of failures in a system, but impossible to determine which. Meaning that over the course of a year, it is possible to know that X amount of dollars is going to be spent in fixing stuff, but exactly what is to be fixed is up to the gods or good technicians that inspect equipment.
Obviously, you can install diagnostic equipment in certain machines (like engines and so on) and get a good indication... but not at the proximity switch level I think.
We have used Rockwell Prox switches with built in diagnostics for years. This is nothing new. The thing is that the value of this is not that great in practice and hardly anybody bothers to check the data that is available.
 
No soft PLCs for me! Useless. I see that the IIOT **** talks about predicting failures - bloody magic stuff it it can predict the failure of a prox sensor! Give it to me!
IIOT shouldn't have a problem predicting failures. It will be the cause of most of them. (n)
 
Microsoft is joining the OPC UA band wagon.
Link

B&R come with both classic and UA but I have not used either yet.
Rockwell is also releasing or has released a sort of “soft plc” that runs windows and Linux. I don’t see any information on their site but that is what our vendor mentioned a few days ago. I want to say it would be in-chassis like a Contrologix rack.
Ya, I was at Microsoft campus for a OPC UA seminar last month. Over 1/2 of the people there were developers. They also showed us their IIOT lab where they help product developer getting their product to market. Classic OPC got a ton of product due to its origin, I think we can all agree on that and we do need something to fill that role.

Another point is that I think most control folks have a natural tendency to push back on the concept but I think that's shortsighted. IIOT, or whatever one call it will happen. Instead of taking the attitude that it's not needed, plan for it. Otherwise you will get IT people running wild in your space and as we know, that's usually not a good thing.


/2 cent
 
What's your guys take on OPC UA? I been hearing it for years but haven't yet to see field implementation. The concept seems fine though and I'm wishing for something better than classic OPC. I'm in the middle of troubleshooting an intermitten classic OPC problem and it's frustrating.

Big fan of OPC UA. Don't need to mess with DCOM settings, can run on any OS, not just windows. It almost always "just works" unless you're trying to do secure communication (then you need to mess with certificates, which is always a pain).

I've used it in an S7-1500 before. Minimal config, you pretty much check a box and make sure you have the right license. You browse to it in your OPC UA client (need to know device name/IP and port number) and then you can see all the tags available.

Here's a very good read, that everyone should read!

That is a very good read. Excellent vendor neutral summary of some big trends.

I think its important to add, thought, that MQTT isn't really competing with protocols like EIP/Profinet/OPC UA. It's more like TCP or UDP (built on top of TCP), where it defines how data is collected and transferred. I see application protocols like OPC UA adding options in their specs to start utilizing MQTT as an option for data transfer.

I dont like soft PLC and cant see them getting big appeal.

To me, soft PLCs are a niche product. Most soft PLC vendors have good controls in place to protect against things like a Windows crash affecting the PLC (virtualization, multi core CPUs, etc). The big advantage of PC based automation is the ability to offload some processing to windows. If you have a really advanced algorithm that needs to churn on some data for a while, or maybe you want to run your vision processing on the same box as your PLC, it can be a great tool.

Probably doesn't replace a traditional PLC for 90+% of applications, though. Product lifecycle issues of PCs (even if you get long lifecycle industrial PCs) mean that for support purposes you'll need to replace it 2-4x before you would a PLC.
 
Microsoft is joining the OPC UA band wagon.
Link

B&R come with both classic and UA but I have not used either yet.
Rockwell is also releasing or has released a sort of “soft plc” that runs windows and Linux. I don’t see any information on their site but that is what our vendor mentioned a few days ago. I want to say it would be in-chassis like a Contrologix rack.

CompactLogix 5480

Here is the controller you are talking about, I believe.
 

Similar Topics

Hello colleagues Anybody knows, what could be the best or a better option to bring with you when you are in service at factory to connect via...
Replies
1
Views
242
This is a great group that helps the Energy people to stay informed about Automating the Energy Industry. You are welcome to join...
Replies
0
Views
711
Hello Collegues ! I am here today looking for help, i recently started a magister in controls and automation. I was thinking in doing an automated...
Replies
7
Views
1,965
Hello all, I recently got a new job working as a controls technician. I have been doing it for 3 months, coming up on 4 now, and the company was...
Replies
2
Views
1,194
... yesterday I removed the plastic liner on the tail-gate of my Kia Soul to get at the broken reversing camera... It has 12 of those plastic...
Replies
17
Views
7,221
Back
Top Bottom