compactLogix RPI

I have a COmpactLogix 1769L16-BB1B with an added analog input and an analog output Point I/O

I have three applications using the same configuration of HMI, controller, and Point I/O... the only difference between the three is the speed of the application. The slowest one seems to operate OK... the other two operate at a similar speed to each other but considerably faster... they seem to be erratic... I have not been able to figure out what the heck is going on.

The digital Inputs seem to operate OK. The analog outputs only send a 4-20 mA signal to a VFD, and remain at the same amplitude unless the operator changes the speed ....but something seems to be going on with the analog inputs...

What effect does the RPI have on these point I/O analog inputs?


thank you

I think it's unlikely that a PLC configuration problem is the issue here. Could be wrong, but just doesn't sound like it to me, sounds more like a physical level problem.

What do you mean when you say the two faster systems are "erratic"?

What "seems to be going on with the analog inputs"? What symptoms are you seeing?

Is this a new system, or have they previously been working OK?

this stuff is new they take measurements of objects as they pass by on conveyors. they all have the same stuff in them, and the same programs.

One measures 30 objects a minute and the other two ...90 to 100. The 30 / minute works OK. The others reject good objects too often, and display incorrect results on the HMI

The logic uses about 6 one dimensional arrays which are compared and subtracted etc.. The arrays are filled with values from lasers connected to the analog inputs ...Point I/O .These arrays must be compared correctly and if I do it slow enough, they are. And everything is displayed correctly on the panelview.

There is only one program , no prioritized or periodic routines

At normal speeds the values on the HMI from simple subtraction are wrong. It should be updating them a row at a time down the screen. Sometimes it seems to skip a row or stay on one row and update the same value more than once... and it's rejecting objects it should not ... whether it is passing bad ones, I don't know ...I don't have much reason to think it's not.

I've been staring at this thing constantly for two weeks. I'm pretty bug eyed.
I know it's more likely my programming, but I'm starting to think the PLC is just screwing up some how... or just inadequate for the task

I don't understand how things like the RPI and the display refresh rate work but something is wrong, I think its speed related, and I don't know what it is

Does the amount of stuff displayed on the Panelview matter?
Does being online myself matter?
I assume the controller has to deal with communicating with this stuff... does that affect the execution of the logic?

thank you

What are the RPI rates at?
The update rate for the screen on the HMI? How many tags on the screen?
Are the values correct in the PLC, but incorrect on the screen?
What is the execution rate for the task / program ?

first of all thank you for taking an interest

What are the RPI rates at?

Click to expand...

For the input Module 3 mS

The update rate for the screen on the HMI?

Click to expand...

for the screen I'm most concerned with, the maximum tag update rate is .25

How many tags on the screen?

Click to expand...

For the same screen ...50 numeric displays

Are the values correct in the PLC, but incorrect on the screen?

Click to expand...

that's one of the $64,000 questions... I have not been able to ascertain that.

What is the execution rate for the task / program ?

Click to expand...

I don't know for sure.... I see for one that I don't believe I was looking at the most useful "property" I looked on the "Main Program" and the longest was just over a half millisecond the last one was about 432 ... I don't remember for sure

The Main task seems to tell you other things, but I am not online with it at the time

going deeper ...

this is just a GUESS based on what you've posted so far ...

what value do you have the filters set for on your analog input modules? ...

what about the RTS (Real Time Sample) rate settings? ...

just as an experiment, I'd personally be tempted to crank the filter setting up to its MAXimum allowable setting – and the RTS to its MINimum allowable setting - and see if the pain will go away ...

just as a quick tip –

the setting of the RPI (Requested Packet Interval) basically determines how OFTEN the module's signal gets sent to the processor ... (RPI analogy: how often does the paperboy deliver a newspaper to your doorstep?) ...

on the other hand ...

the setting of the RTS rate (Real Time Sample) determines how OFTEN the module recalculates a new/fresh value from the analog source ... (RTS analogy: how often does the newspaper company rewrite/print a fresh new edition of the paper?) ...

the basic idea of the analogy is that - just because you're getting a paper delivered each morning - doesn't necessarily mean that it's always a FRESH paper ... it could be just a stale copy of yesterday's news ... that's why the RTS setting is usually MORE important than the RPI setting when it comes to ANALOG signals - in a HIGH SPEED application ...

I covered these topics in gruesome detail several years ago ...

DO THIS ... read the full "newspaper" analogy in the following post (it's short) – and then if you want to go deeper, read the rest of the thread ... warning: the thread is long and detailed – but the ideas might indeed be causing the problems that you're having ...

http://www.plctalk.net/qanda/showthread.php?p=392002&postcount=4

once again, all of this is just a GUESS – based on what you've posted so far - but I have reasons for suspecting that this COULD be a factor in your problem ...

SUGGESTION:

if this doesn't help – and if all else fails – start posting SPECIFIC catalog numbers for the hardware that you're using ... I just GUESSED at one type of Point-I/O analog input module to post the screen shot – yours might certainly be different ...

if you get REALLY desperate, you might consider posting your program code (ACD file) ...

we're willing to help – but right now we're sort of just peeping through a keyhole at the problem ... we can always give much more DETAILED help – once we know the DETAILS of the problem ...

.

and here is a screen shot taken from the user's manual available at the following link ... again – I just GUESSED at a module number for an example – since you haven't posted that information yet ...

http://literature.rockwellautomation.com/idc/groups/literature/documents/sg/1734-sg001_-en-p.pdf

question: did you just leave the filters at the default setting of 60 Hz? ...

good luck with your project ...

.

Good grief... what a fabulous resource and what great guys!

1734-IE2C/C input

RTS 100 mS ...man that seems too long between newspaper printings to me! (your analogies are excellent !)

RPI = 3 mS

Seems to me the delivery boy would be delivering about 33 versions of the same newspaper

Yes the filter is 60Hz

1734-OE4C/C output


I will try what you suggested ... you just nail this stuff to the wall, don't you?
excellent.

thank you

that made a significant improvement ... I'm gonna monitor it for a while

thank you very much

Sorry for not jumping back in after my initial post, I've been away onsite for a week. Although I think Ron has helped you more than I ever could have anyway