Continuous vs Periodic Tasks in PLC

This has been entertaining.

I totally agree, lots of great input!!


The key question here is what is the project? For a lot
(I'm going on a limb and say the majority)
of projects it won't matter much

Almost everything matters here:)

This project:

Only 2 PLC's
... 1st Prototype Ships In ~6 Years
... over 108,000 hours of PLC work (54 man years)

PLC guys here, for the most part

... do no wiring, cabling, chassis building, packaging, drawings I/O testing etc...
... Most time is spent on the program
... Though we do end up gowned up once in a while for sure.


7 Very Senior Level Programmers, 2 Very qualified (9 total)

... All well rounded PLC/SCADA/Database/Network guys
... ~215 years of combined experience


It has pretty much any & everything I have ever seen on a project

... 5 volt PLC I/O cards TTL level
... Sending & reading 10 ms communication pulses on remote Point I/O with required accuracy of +/-0.5 ms
... Many PID loops with mili degree accuracy required

Many protocols
... ASCII 232 & 485
... Modbus Serial
... Modbus TCP (using code in PLC without adding special Modbus module)
... EthernetI/P
... DeviceNet
... EtherCat on Allen Bradley (trying to nix that)

~15 remote chassis (panels)
... ~40 total racks (AENTR cards)

~2,000 physical I/O

~100,000 virtual I/O
... Vacuum Transmitters, Oxygen Transmitters, Getter Ion Pumps, Neg Pumps
... Turbo Pumps, Chillers, Heaters, Environmental Controllers, Robots, Servos
... PQube Power Monitors, Weather Monitors, Barometric Pressure, Lasers
... Many others

Trending in Ignition @ 250 ms default for many tags
... & higher for a few special tags

Sil2 Safety mostly, possibly Sil3 for a few components
... 3rd party will read every line of safety code & test it endlessly.

Must be safety stamped for nearly every place in the world
... Taiwain, Japan, Korea, Europe, US

Not only do they need to pass countries codes
... Every customer has requirements that are additional
... 1 machine must fill all safety & customer requirements

Remember with the Rockwell guys - those that can, do.
Those that can't go into sales and sometimes tech support.
I realize that I'm painting with a broad brush and there are some very notable exceptions -
apologies to them.

The local rep I asked this question actually gave my first
... & only PLC class over 30 years ago on a brick SLC 500

... He gave me good rational advice on this
... and he never talks in absolutes
... Good guy!


That is like asking a saw manufacturer how to best build a house.

Your response to my comment "I get various opinions when calling tech help"
... Quite accurate :)


The Engineers that I have a hard time tolerating are the ones that say:
... "I called Tech and they told me it must be done this way"
... or worse
... "They said it can't be done..."

Front line techs on the help line are generally either beginners or perpetual intermediates.
... As soon as they get some experience they move up the ladder
... and you only get to talk to them again when issue is escalated.
 
Almost everything matters here:)

I meant for most projects out there. The size you mention is something that, in my opinion, could use some breaking into smaller PLCs to carry out specific tasks and report upwards.

This would potentially be slightly more expensive, but greatly improve the capability of the system as you wouldn't be dreading whether that additional rung you'll add will break the timing and cause all sorts of mischief. But this is just me.

Is this for a ship? You don't carry around 2000 IO's to be stamped in every place of the world.
 
I meant for most projects out there. The size you mention is something that,
in my opinion, could use some breaking into smaller PLCs to carry out specific tasks and report upwards.

This would potentially be slightly more expensive,
but greatly improve the capability of the system as you wouldn't be dreading whether that additional
rung you'll add will break the timing and cause all sorts of mischief. But this is just me.

This is my 3rd project for them
... As you could guess from the scope 1 or 10 PLC's could be called for & the price not be an issue
... Being intimately involved myself & 8 other programmers would prefer only 1 PLC
... It is highly structured
... Additional PLC's require coordination, produced & consumed standard & safety tags
... Asynchronous scan times of PLC comms cause issues


As you mentioned (paraphrased)
... "On a majority of projects it won't matter"

On this project 95% of the the project don't matter
... scan time, execution time, of 200 valves, pumps, etc not critical
... Only 5% needs attention to detail
... The fun part :)


All devices must have a unique identifier
... If there is a motor # 1
... There cannot be a Valve # 1 or Pump # 1
... Temperature probe # 1, E-Stop # 1
... anywhere else in the project

I have seen where there was
... Ball Valve 1, Solenoid Valve 1, Blocking Valve 1 in PLC # 1
... & PLC # 2 had the same
... If you told somebody to turn on Valve # 1 there were 6 possible answers



Is this for a ship? You don't carry around 2000 IO's to be stamped in every place of the world.

This project is 'Metrology'
... (not to be confused with 'Meteorology')
 
this has been entertaining. Personally, i know both periodic and continuous task have their place.
Periodic tasks are for determinism. Continuous tasks are for back ground things.
I think a more important question has fallen through the cracks. When is the i/o done?
A periodic task should have all the inputs it needs updated at the beginning of the task and its outputs done at the end. When are continuous task i/o updated? Sometimes there isn't much point in using input data that is 20 milliseconds old.

Also, back ground tasks are good for long computations that may take several scans to complete. A common example is sorting. A bit can be set or cleared to indicate when the sorting is done and the data is now sorted.

+1
 
In logix processors, I put PID loops and any totalizer calculations in periodic tasks but all logic in continous.
 
I totally agree, lots of great input!!




Almost everything matters here:)

This project:

Only 2 PLC's
... 1st Prototype Ships In ~6 Years
... over 108,000 hours of PLC work (54 man years)

PLC guys here, for the most part

... do no wiring, cabling, chassis building, packaging, drawings I/O testing etc...
... Most time is spent on the program
... Though we do end up gowned up once in a while for sure.


7 Very Senior Level Programmers, 2 Very qualified (9 total)

... All well rounded PLC/SCADA/Database/Network guys
... ~215 years of combined experience


It has pretty much any & everything I have ever seen on a project

... 5 volt PLC I/O cards TTL level
... Sending & reading 10 ms communication pulses on remote Point I/O with required accuracy of +/-0.5 ms
... Many PID loops with mili degree accuracy required

Many protocols
... ASCII 232 & 485
... Modbus Serial
... Modbus TCP (using code in PLC without adding special Modbus module)
... EthernetI/P
... DeviceNet
... EtherCat on Allen Bradley (trying to nix that)

~15 remote chassis (panels)
... ~40 total racks (AENTR cards)

~2,000 physical I/O

~100,000 virtual I/O
... Vacuum Transmitters, Oxygen Transmitters, Getter Ion Pumps, Neg Pumps
... Turbo Pumps, Chillers, Heaters, Environmental Controllers, Robots, Servos
... PQube Power Monitors, Weather Monitors, Barometric Pressure, Lasers
... Many others

Trending in Ignition @ .250 ms default for many tags
... & higher for a few special tags

Sil2 Safety mostly, possibly Sil3 for a few components
... 3rd party will read every line of safety code & test it endlessly.

Must be safety stamped for nearly every place in the world
... Taiwain, Japan, Korea, Europe, US

Not only do they need to pass countries codes
... Every customer has requirements that are additional
... 1 machine must fill all safety & customer requirements



The local rep I asked this question actually gave my first
... & only PLC class over 30 years ago on a brick SLC 500

... He gave me good rational advice on this
... and he never talks in absolutes
... Good guy!




Your response to my comment "I get various opinions when calling tech help"
... Quite accurate :)


The Engineers that I have a hard time tolerating are the ones that say:
... "I called Tech and they told me it must be done this way"
... or worse
... "They said it can't be done..."

Front line techs on the help line are generally either beginners or perpetual intermediates.
... As soon as they get some experience they move up the ladder
... and you only get to talk to them again when issue is escalated.

Without being too specific,, what in the heck is this application? I'm not entirely sure that I can comprehend being part of a project that large.
 
I'm a periodic man myself. Gotta have full control of the entire system.
IO tasks are run at 20ms (or whatever the lowest rpi is).
Control tasks are run at 20ms (or same as IO).
Time task = 100ms
Metrics task = 1000ms

For me, this means that the IO is updated for every control task
I can then see how much CPU is being used. If over 80%, then I increase the control task from 20 to 40ms

Works great for me and I can monitor everything.
 
Here is a link to an EtherCAT/EtherNet/IP Gateway solution & the rep's e-mail

https://www.anybus.com/technical-su...entation---x-gateway-classic?ordercode=AB7682

.... [email protected]

As I said I am typing to nix the EtherCAT on the Allen Bradley hardware
... But need to find compatible Ethernet/IP Flow & Pressure controllers

... It looks possible but not sure you get the EtherCAT speed that is so touted
... (though the speed is not generally needed for this part of the project)
... It also adds more hardware (points of failure) & special configuration

Curious, how are you doing that?

This intrigues me. Master or Slave?
I'm waiting for an AB EtherCAT solution.

EtherCAT is available on thousands of devices that do not have Ethernet/IP
... Mass Flow controllers, Pressure regulators, etc.
... This is why this is being brought up

The Semi Conductor Industry seems to be leaning toward Beckhoff PLC's these days
... and away from Allen Bradley
... Beckhoff looks to have much faster I/O speed & is way less expensive especially for Safety rated processors & I/O
... They also have more I/O options in some lines

Luckily our PLC group here has over 200 years of Rockwell experience
... and zero years of Beckhoff experience
.... So changing to Beckhoff is not likely here for a few more years

Need time to learn Beckhoff nuances on something small & not when not on a fast pace complex project.
... The PC's used look to be too proprietary.
... Need more time to investigate.
 

Similar Topics

Hello, Monitoring my system with the Studio5000 Task Monitor I see a yellow banner "Continuous Task is present, resulting in 100% CPU...
Replies
78
Views
35,037
I've inherited a L61 that was installed last year and the memory usage is well into the red. There is only one continuous task with a scan time of...
Replies
5
Views
3,762
Hey guys, I wanna to write a bit of code to return value to show if the task is in periodic or continuous mode? is there a tag in RSLogix 5000 to...
Replies
18
Views
28,581
Hi Guys, I'm new to controls and PLCs and this might have a trivial answer. I have a small project with a PLC and a inverter. The inverter...
Replies
25
Views
3,686
Can anyone explain to me why Rockwell thinks a Continuous task is better than a Periodic task? That is what it defaults to. It may cause your...
Replies
10
Views
2,905
Back
Top Bottom