Sign in to follow this  
Followers 0
robh

ControlLogix I/O

45 posts in this topic

Well, I do have the age (40+ years) and the experience (15+ years as a controls engineer) and we seem to have 50 machines located in your neck of the woods in the last 5 years alone. (Must be doing something right there, chap! ) Try explaining RSNetworx to a new maintenance guy or the task of scheduled versus unscheduled data. With the large manufacturing turn-around, this is a never ending training session. I site, as one example, TMMNA. They have half of their plant set up with ControlNet....the newer side is set with Ethernet communications. Their new spec' allows ethernet only processors! They must enjoy those new toys, too! They are gathering quite the market share on the domestic manufacturers. I relate such to the automotive market. GM and Chrysler are stuck with OLD designs and they seem to be losing grip to the import market. Ford, thankfully they have the new Mustang design to keep them afloat. I used Toyota as an example because they are probably in your neck of the woods. Too, there is FNOK, MPP, GECOM, etc. I would love to carry this debate on further with you. It is always nice to re-visit the past.

Share this post


Link to post
Share on other sites
Good then you also recall when AB was preaching use CNET not ENET and spreading the "horror" story of the Ethernet Paper mill which self destructed all because the IT guy ran a network who. We can debate pro's and con's of ethernet vs. controlnet till the cows come home, but no one platform fits every situation or the others would be out of existence. Suffice it to say design well and you'll have fewer headaches at runtime.

Share this post


Link to post
Share on other sites
Hi, I have read the lively discussion unfolding here. I have recently tested out 1794-AENT modules. They seem to work fine and I/O update time seems good. I have a couple of concerns. I will definitely keep the control subnet different from the IT subnet. But I will have to put the HMI on the same subnet. Unless I use different ENET cards for I/O and HMI. Even on the same subnet, HMI does not seem to pose a problem, with scan rates of at least 1 sec. Also messaging volume should not eat into bandwidth. My main concern is at time of program upload / download. While at download, the I/Os will not be in critical control frame and even if they lose com for some time, it is ok, since the plant will be in safe state during download. But at upload, is the volume of traffic enough to eat up bandwidth and make the I/Os lose com for a time > 100ms?? If so, should I use different switches for I/O and RSLogix5000 PC? Generally ENET is at 10/100 MBPS. What speed is there for ENET/IP? Is it 10MBPS or 100 MBPS? What about the Gigabit ENET network? Are their any efforts to put I/O at 1GBPS + speed in the future?

Share this post


Link to post
Share on other sites
We currently have all our 1734-AENTs set at "auto negotiate port speed and duplex". With our configurations, the AENT reports 100 Mpbs and Full duplex. Correct, we utilize separate switches and have found the best results with the MOXA EDS series switches. There were some issues with the Hirschman and Pheonix switches in prior testing. (Too, another debate could rear it's ugly head as to managed or unmanaged. No need to go there at this time.) My recommendation is additional switches. I believe that is what you were asking?!? Edited by 5150

Share this post


Link to post
Share on other sites
As long as the controls engineer has unlimited control of the ethernet, and it is properly planned stay away from ethernet for I/O. I will reference my experience with our corporate governed IT "upgrading" our ethernet switches with new switches adding 3 VPN's, subnetting, etc so that the CEO and CFO who could really care less what goes on at our level can have pretty Web Enabled displays of the shop floor. We couldn't get our 80+ PLC's to talk, I couldn't imagine if I had ethernet I/O to deal with. I'll take RIO anyday it scares IT.

Share this post


Link to post
Share on other sites
Yeah but the PLC's wouldn't talk because those "dumb" maintenance guys messed up the new enet switches, just ask the IT Gods. I'd LOL if it weren't so sad that this is true.

Share this post


Link to post
Share on other sites
Funny how that works when asked how our network performed before the upgrade my respone was "We didn't know we had a network till you guys upgraded for us"

Share this post


Link to post
Share on other sites
Why would they not talk? Different addresses? Common class A/B/C/D addresses? I do not understand this statement. Please elaborate.

Share this post


Link to post
Share on other sites
HUH? Did you mean unless the controls engineer has unlimited control? Edited by robh

Share this post


Link to post
Share on other sites
Bad choice of words, we couldn't keep them talking. Reasons: 1. Initially computers that never were on "our" network now were, and they were all set up by the local IT guy who well we'll just say my 12 year old has a better understanding of TCP/IP. 2. Shortly after we were subject to hey don't open this email because............. too late Bob already did. Enter virus 1 then 2 worms. 3. Ed decided it was time to play with the macros in his email program. Well that garbage he broadcast shut us down for an hour. 4. My all time favorite one of our consultants feeling good after a cup of coffee an good solid hours worth of work decided it was time for a well deserved break. Pulled out his laptop and started playing Flight simulator with his budy in Texas. It took our corporate IT 2 hours to track it down because we (engineering group) are too stupid to have access to the sniffers we were staring at, in our branch. Best part he still does work for us everyday. The list goes on, my point too much you don't control.

Share this post


Link to post
Share on other sites
Well, I understand the point of the dude with the tape on his glasses between the lenses and pocket protector, but not all IT/IS individuals are that bad. (Shows a reflection of the HR and IT manager's idocy.) There should be a machine level switch so that when the network DOES go down or wig out during a lunch hour with the Flight Simulator marathon, the machine is going to function. Once again, the IP addresses should be set in accordance with the standard (class A/B/C/D---noting E is reserved--- to prevent such issues.) Some devices are smart enough to detect, such as the Omron ETN which knows if the IT/IS guy initially set up a system with a class A IP address and a class B subnet, DOH! I will close by stating for the hard core ControlNet/DeviceNet individuals, this comes to mind.... "Two Caterpillars were chewing on a cabbage leaf when one looked up and saw a butterfly flutter by. Turning to the other Caterpillar he said " Look at that! You will never get me up in one of those things." Can't stay cocooned forever.....ethernet is going to be THE majority means of communication in industrial controls momentarily....tick-tock. Edited by 5150

Share this post


Link to post
Share on other sites
Not until a couple generations of IT and Controls People retire. You'll also need a much better educated IT and controls community. Not to mention that Ethernet does not do redundant controllers or I/O right now.

Share this post


Link to post
Share on other sites
Refuse to open that cocoon, do ya'. (I see that you are a refugee from the RSLogix forum and 99.5% A-B.) As I have stated, across the pond beckons! The big three are dwindling away and the foreign three are heavily invading. True, there are OTHER industries, but they usually follow suit with the automotive industry....after all, the first PLC was developed for use where? "Two Caterpillars were chewing on a cabbage leaf when one looked up and saw a butterfly flutter by. Turning to the other Caterpillar he said " Look at that! You will never get me up in one of those things."

Share this post


Link to post
Share on other sites
...and you seem to be BEHIND the times on the redundant ethernet issue.

Share this post


Link to post
Share on other sites
Before giving response like this read the original statment. Bob is talking about redundand Controller setup and not redundand Ethernet. Bob is absolutely correct - redundand controller system has no Ethernet I/O capability, ControlNet is the only option. ControlNet is and will be for a long time the BEST option for I/O. Pub 1756-RM094 Edited by Contr_Conn

Share this post


Link to post
Share on other sites
It appears that YOU shoot from the hip, so put down that (A-B) six shooter before you hurt yourself youngster. You site a 1794 (A-B) publication. I did not state A-B only devices. I site alternate examples....as in Wonderware, Intellution and Cimplicity coupled with, for example the Most Logic System Hybrid controller. REDUNDANT ETHERNET CONTROLLER SYSTEM. Life exists outside of Rockwell, let go of the coat tails. Edited by 5150

Share this post


Link to post
Share on other sites
You should stop talking like this with person older than you are - yongster - I am older than you! ****************************** <---eh admin edit Wonderware has nothing to do with redundand ControlLogix Controller and I/O. Question is about Controllogix I/O Edited by chakorules

Share this post


Link to post
Share on other sites
Yes, that last statement shows real maturity!

Share this post


Link to post
Share on other sites
I vote 5150 get 86ed from this forum.

Share this post


Link to post
Share on other sites
It's been handled on BOTH ends. 5150 and Contr_Conn....

Share this post


Link to post
Share on other sites
Guest
This topic is now closed to further replies.
Sign in to follow this  
Followers 0