Point IO vs Flex IO: Do you have a preference?

[dcooper33]

I've read a few threads on this topic here on the forum, and it seems that the prevailing general opinion is that Flex is better for RIO applications, while Point is better for Ethernet applications. These were fairly old discussions, though.

Myself, I've had a fair amount of experience with Point IO integration into Controllogix platforms in the last couple years, for the most part positive. I'm not as experienced with Flex IO.

Can anyone offer any insight as to durability/reliability of the two platforms? Is there any concern about the age of the Flex IO platform, becoming incompatible with newer Contrologix hardware? Any types of modules that anyone have seen that work much better on either platform?

Thanks in advance!

Dustin

 

[Steve Etter]

I'm in much the same boat you are with a fair amount of experience with Point IO and just a little with Flex. What I have seen, though, is several module failures with my Flex IO and none so far (that I know about) on the Point IO. Everything has been on Ethernet, so I can't comment on the RIO platform.

Not much help, I know, but at least it's something.

Steve

 

[Archie]

I have worked on machines that have both FlexIO and PointIO over Ethernet. There is a very noticeable delay in the FlexIO connecting after power up. In the order of 30-45 seconds. I prefer the PointIO because of the density. Another thing to note about FlexIO is that it has a limitation of 8 modules per coupler.

 

[Geospark]

I've read a few threads on this topic here on the forum, and it seems that the prevailing general opinion is that Flex is better for RIO applications, while Point is better for Ethernet applications. These were fairly old discussions, though...

...Can anyone offer any insight as to durability/reliability of the two platforms? Is there any concern about the age of the Flex IO platform, becoming incompatible with newer Contrologix hardware? Any types of modules that anyone have seen that work much better on either platform?

Dustin,

While I cannot really call it a "discussion", I did recently post some details on the main differences between these two I/O platforms. They are not wholly comparable feature-wise. They are from different generations platform-wise. Point I/O is a more complementary and newer option than Flex I/O, but it cannot replace it in every way. Flex I/O, by it's design and nature, is slower than Point I/O in many respects, but still it's a perfectly good option, if it suits the application.

I always say these selection decisions should come down to the application and what suits best, but I will say that Point I/O is probably more widely selected nowadays where the feature differences do not come in to play...

http://www.plctalk.net/qanda/showthread.php?t=88331

Regards,
George

 

[dcooper33]

Thanks for the replies, fellas. Exactly the type of information I was looking for.

I have worked on machines that have both FlexIO and PointIO over Ethernet. There is a very noticeable delay in the FlexIO connecting after power up. In the order of 30-45 seconds. I prefer the PointIO because of the density. Another thing to note about FlexIO is that it has a limitation of 8 modules per coupler.

This is the sort of thing that worries me about selecting Flex for any new jobs here where the push is to control everything over ethernet. I had read that about the 8 module limit on the Flex. I worry about Flex adaptors ability to keep up with demand as network traffic becomes more and more sophisticated .

I'm in much the same boat you are with a fair amount of experience with Point IO and just a little with Flex. What I have seen, though, is several module failures with my Flex IO and none so far (that I know about) on the Point IO.

I've not seen any outright failures of any of our Flex modules, but have had issues with a couple of DC output cards that were way more susceptible to noise than they had any right to be.
I've heard a few disparaging comments from guys here looking at a Point IO rig saying that the modules themselves are cheap-looking, not industrial, etc. While I understand the impression, in a way (the cards do kind of look like the old NES cartridges), I don't have any reason to believe the hardware is prone to failure. Like I said before, I've got probably 200 or so modules on 40 or so 1734-AENT in my plant without a single failure, but the oldest installation is barely a couple of years old.

We're looking to standardize hardware platforms on everything, and one of the hot topics is remote io. There are a lot of voices for Flex IO, but I think that's partly because of its age, and more guys being familiar with it. The fact that some of the Flex modules are actually silver-seried is very unsettling for me when considering using the platform for a controllogix remote io expansion over Ethernet.

There have been a couple of recent jobs where Flex IO was selected for upgrades on sprawling old machines with PLC--5 controllers on DH+ and RIO network, and nearly all 120v control circuits. In those cases, Flex made perfect sense. I guess the thing to take away is that a standard is great, but every job is different, and you have to be able to select the right tool for any given job intelligently.

George,
I'm not sure how I missed your posts on the subject in my forum searches. Just careless, I guess. Thanks for the great info, as always The different types of built-in electronic fusing is another great feature of Point IO, IMO

Flex I/O, by it's design and nature, is slower than Point I/O in many respects

Can you elaborate on this? By slower, do you mean that the Point I/O modules, in general, are built to handle lower RPI's? Or that the AENT adaptors are faster?

 

[rdrast]

Here, we have completely switched from PointIO to FlexIO on all new equipment and upgrades, simply because the PointIO has been much more reliable, and is much smarter then the Flex.

 

[JesperMP]

Here, we have completely switched from PointIO to FlexIO on all new equipment and upgrades, simply because the PointIO has been much more reliable, and is much smarter then the Flex.

I totally get this.
The Flex io is so unreliable that your job is safe forever.


I think that there has been no new Flex products for many years, whereas there are new Point products, and Point is integrated into some models of Compactlogix. So point seems more futureproof.

I have had 1 Flex module go bad on me, but that is statistically insignificant to say anything. It has been 7-8 years since I made the last project with Flex. I wouldnt choose Flex today.

 

[kamenges]

The one thing holding up Point I/O is AB doesn't appear to have a Point equivalent to the 1797 family. That would be nice to have. But outside of that I've switched over to Point exclusively and haven't had any issues with it.

Keith

 

[JonHopper]

I much prefer Point IO to Flex IO. Regardless of communication type.

I've integrated probably close to 30 Flex IO modules in my four years of experience, and around the same for POINT IO modules.

I lost track of how many Flex scanner modules I've had fail on me. I've had NEW modules fail, old modules fail, and slightly used modules fail. I've had them be fully tested and operational, only to pull the FlexIO system apart, put it back together, and the module doesn't work.

I've also had countless numbers of DC modules on FlexIO fail completely or have bad outputs / inputs. I've also had a lot of base modules fail.

I have had about four module failures on POINT IO so far.

MUCH better success rate. Not to mention the diagnostic abilities are far better. I wouldn't recommend FlexIO on any new machine when you have the option of using either Flex or Point.

Another thing to mention is the fragility of Flex IO modules vs. Point IO. They are extremely easy to break, the pins are easily bent, whereas the Point modules are pretty durable and there are no pins to bend.

 

[Paullys50]

Good timing on this thread. We had Rockwell come in to discuss some various items, we asked this very question.

There are no plans for a 1797 equivalent in PointIO

They continue to sell more FlexIO than PointIO, a "next generation" FlexIO system will be released (would not give a time frame) and will eliminate some of the "limitations" of FlexIO, again they would not get into details. No signs of FlexIO going into obsolescence anytime soon.

If you are using PointIO, make sure you are using the 1734-AENT Series B this has a new CPU and can truly handle a 63 node rack. We had problems where the 1734-AENT CPU bandwidth would be saturated, and IO on the end of the rack would not function, despite following AB design guidelines.

With PointIO be aware of analog output modules, these produce a lot of heat and if you sandwich them together they fail due to heat-transfer between modules. We've experiences this as well. There is a tech note to put CTM cards between analog outputs.

DeviceNet/ControlNet are quickly fading into oblivion. -- My interpretation

 

[kamenges]

Originally posted by Paully's5.0:

DeviceNet/ControlNet are quickly fading into oblivion. -- My interpretation

That may be a bit tricky with ControlNet and 1797 since I think ControlNet is the only current comm protocol supported for 1797 stuff. Have you heard if they plan to switch to Ethernet/IP for 1797? That would be cool.

The big reason I would like to see 1734 and Ethernet/IP in an Ex version is it would open the door to safety I/O in an Ex area.

Keith

 

[Paullys50]

That may be a bit tricky with ControlNet and 1797 since I think ControlNet is the only current comm protocol supported for 1797 stuff. Have you heard if they plan to switch to Ethernet/IP for 1797? That would be cool.

The big reason I would like to see 1734 and Ethernet/IP in an Ex version is it would open the door to safety I/O in an Ex area.

Keith

Not saying DeviceNet and ControlNet won't have their place, but Ethernet/IP is taking over wherever it can.

Didn't ask about Enthernet/IP for 1797 stuff.

 

[GTUnit]

I literally just got done with an 18 hour session diagnosing a 10+ yr old Flex I/O rack at a major distribution center of a large retailer everyone knows.

I was going to post a warning about Flex I/O to all but found this thread relevant so I'll put it here.

My conclusions are Stay Away from Allen Bradley Flex I/O like the plague.

There is an inherent design flaw in the backplane/bases which creates ALL KINDS of intermittent/ghost problems with Flex I/O.

The base to base connectors are unreliable and flaky.
AB tech support in was also very candid in saying that the connector design is notorious for easily bent pins in the field. I did not experience the bent pins issue but intermittent failures for no apparent reason.

I had communications drop out and modules stop working.
I replaced the entire backplane with parts from another piece of equipment.
Even then still not 100%.
After reseating connectors several times things started working.

There is a 20+ yr old PLC5 running next to the flex with not so much as a hiccup.

IMO AB Flex I/O is NOT a reliable system and it does not stand the test of time.

 

[Paullys50]

I literally just got done with an 18 hour session diagnosing a 10+ yr old Flex I/O rack at a major distribution center of a large retailer everyone knows.

I was going to post a warning about Flex I/O to all but found this thread relevant so I'll put it here.

My conclusions are Stay Away from Allen Bradley Flex I/O like the plague.

There is an inherent design flaw in the backplane/bases which creates ALL KINDS of intermittent/ghost problems with Flex I/O.

The base to base connectors are unreliable and flaky.
AB tech support in was also very candid in saying that the connector design is notorious for easily bent pins in the field. I did not experience the bent pins issue but intermittent failures for no apparent reason.

I had communications drop out and modules stop working.
I replaced the entire backplane with parts from another piece of equipment.
Even then still not 100%.
After reseating connectors several times things started working.

There is a 20+ yr old PLC5 running next to the flex with not so much as a hiccup.

IMO AB Flex I/O is NOT a reliable system and it does not stand the test of time.

A bit dramatic....10+ years of service and you spent 18 hours troubleshooting? While I will agree the connections from terminal base to terminal base can be tricky, same with inserting the modules but Flex has been a staple for AB for a long time. You can't compare it to PLC5 IO, where PLC5 remote IO isn't anything different as it's just chassis IO with an RIO adapter.

IMO AB Flex I/O is NOT a reliable system and it does not stand the test of time.

You haven't given substantial argument on this one...for future readers take this with a grain of salt.

 

[GTUnit]

A bit dramatic....10+ years of service and you spent 18 hours troubleshooting? While I will agree the connections from terminal base to terminal base can be tricky, same with inserting the modules but Flex has been a staple for AB for a long time. You can't compare it to PLC5 IO, where PLC5 remote IO isn't anything different as it's just chassis IO with an RIO adapter.



You haven't given substantial argument on this one...for future readers take this with a grain of salt.

Obviously you did not really read or comprehend what I wrote. You would be a bit dramatic too if you just dealt with hardware that drops out for no reason in different areas. No sign of a physical problem, no diagnostic information, no clear direction where to look. Used replacement hardware exibits similar issues.

The entire building was down on and off for 12+ hrs. Cost them over a hundred thousand dollars. Can be sure I'm doing to be dramatic.

What does "tricky" mean in an industrial grade product that's supports million dollar a day operations?
I translate "tricky" as unreliable.

There is also the issue on RevB comm adapters having a completely different dip switch setup as the RevE adapters. That was a nice surprise.

It's not the same as inserting modules. The modules don't have a problem. The BASES do. You don't seem to understand a simple comparison between hardware that has a lifetime reputation as being dependable (PLC5) and a newer product (Flex) that has known and recurring issues. (As explained by the manufacturer themselves)

That is all I am saying on the matter.