Control Logix Guru's need your Insight

[dbh6]

@okie PC, ok now u understand the logic, as long as the TT bit is on the timer will essentially be latched on up until the 0.5 secs, make sense it been a long day lol, as for the structured text, now that you and ron mention it i did not know that you need to be for more additional language, something that i did not consider, specially when another programmer comes to trouble shoot and cant see the code with the structured text, so ill just take out that structured text routine, thanks on the heads up on that,

as for the pump pushbuttons, @okie pc, ill try it tomorrow with your method and let you know how it goes

once again thanks on the feedback guys

 

[OkiePC]

my point is that if you don't NEED to add those extra languages for some GOOD reason – then why saddle everybody else with having to deal with the expense – and the extra learning curve involved? ...

We just want them. I know Mike and I both can use FBD, so we'll just tell the boss to add it to the budget for this big project coming up.

We will upgrade his evrsi license to a concurrent Pro License if I can just get him to write the requisition when he gets back from vacation.

Then at some point, get another pro license. I would love to do some of this string and serial craap with ST, even though I am not familiar with the RA implementation.

I have used their FBD in it's early days and it was very good, just a few quirks with lines they had to iron out. Lots of our process control would be so much simpler on a couple of FBD sheets.

I wouldn't saddle an end user without some sort of forewarning or discussion as was done in our case, but perhaps this one routine is their remote portal since in our case, they intend to remote monitor and make adjustments for several weeks after installation. Or it's by God Perfect Code that shall never need inspection nor tampering.

 

[dbh6]

@Okie PC here is my conclusion thus far, when using your recommended button stuck syndrome logic where i used the timers .tt bit to trigger a push button to stay on for 0.5 seconds. The results are as follows, if i used a momentary push button on the HMI the issue was still present, when i used a maintained push button the issue got even worse as in the plc was missing the signal more often, even when i used a Set button on the HMI the plc missed out random times.

@ Ron i took out the logic for the fan animation as well just to try it out, but didn't really fix the issue. So im guessing their is something up with the communications through the 1756 enbt module,

I called Rockwell, FT view group and they said they have heard this issue before and that they recommend upgrading the firmware of the enbt module, mine is currently 4.008, ill try it out and let you guys know.

 

[OkiePC]

@Okie PC here is my conclusion thus far, when using your recommended button stuck syndrome logic where i used the timers .tt bit to trigger a push button to stay on for 0.5 seconds. The results are as follows, if i used a momentary push button on the HMI the issue was still present

You got comms problems.

My example is not a catch all. 1/2 second may be totally inappropriate in some cases.

The maintained button is worse because each time you hit it, the HMI will alternate the value it writes, so it's only going to work with my example logic every other time you touch it.

The set value command should work equally as well as a momentary button.

If it doesn't then, every single failure of the code to behave properly is an indication that a comms packet did not get through.

If this happens more than once a day you have a serious issue, and no amount of programming can fix a frayed ethernet cable or noise problem, or flaky switch somewhere.

Add a counter to your HMI button address, and put the counter on a branch above the OTU.

Go hit the button fifteen times...not too fast, this is only a PV+ after all, give it at least a second longer than the programmed hold time between presses. Then go look at your counter. It should match the number of presses.

The minimum hold time is critical. It must be longer than the sum of your scan time and the worst case communication delays. Typical values are from about 100 to 1000 milliseconds. 500ms should be a good number from which to test things.

 

[dbh6]

@Okie PC, just to let you know it is not a panel view plus terminal, the FT View SE is running on a PC.

So i did what you insisted, where i added the CTU branched above the OTU, i also made the hold time for the Pushbutton 500ms from 250ms. The PLC scan time is roughly 700 micro seconds so the 500ms is more is a good time, so when i tested it pressing the Push button 15 times with a good interval in between the times i was pressing it, it counter 12 and sometimes 11!! and sometimes 13 so it definitely was missing counts, so i really don't even know at this point

 

[dmargineau]

I was under the impression you were running an RSView SE application...
Your last post states FTView SE...
Which one is it?
It definitely seems to be a communications timing issue; your answer to my previous question inquiring about the networking topology was pretty "formal' to say the least...
Within a distributed HMI application (such as yours) it is critical that the networking topology allows for efficient data packets transfer.
Have you tried to run the application while the RS(FT)View running PC is patched within a low level subnet (without any LAN managed switching)?

 

[dbh6]

@dmargineau i apologize if i mislead you, I'am indeed running a FT view SE application version 6.10, when you compile to create the runtime application it creates a .cli file (client file), as for the networking topolgy, their is an unmanaged switch that i'am using to that switch the only components that are connected are the PC with the HMI application which is really my virtual machine, and also the control logix ENBT module is connected to the switch. So to clarify i'am running 2 virtual machines on one computer, one VM has all the logix 5000 components such as Rs logix 5000 software and rslinx classic while the other vm has the hmi application, and the HMI and plc are connected via rslinx enterprise. Their is no other components connected to the switch other than what i have just stated.

 

[dmargineau]

Hmmmm...FTV SE client running on a VM ?!...I have yet to attempt to accomplish this task...

 

[dbh6]

ok here is something that i realized, i used a set push button on the HMI, so i left the PLC with the responsibility of unlatching that set push button bit. So what i did was i have the hmi push button in series with a counter and also a branch under the counter that resets the push button, i clicked it 15 times and sometimes the results were 13 sometimes 14 so if the HMI is setting it their is no way! the plc won't get that signal unless their was a comms issue, i bet it has something to do with that RET instruction at the end of the routine, maybe depending on when in the scan the push button bit is set if it is half way through the scan the RET instruction might write a 0 to it, and on the next scan that value is a 0 so the plc misses it. So what i finally ended up doing is i put a simple code in the main routine disabled all the routines that i am calling and just have the hmi push button again in series with the counter and a branch that unlatches the HMI push button bit. The results it when i pressed the pushbutton on the hmi is that out of the 15 times i clicked it, it never!! missed it. So this leads me to believe their is something up when calling it from the main routine so what i will now try is make another routine and then call it from that routine instead if the main and see what happens

 

[dbh6]

Ok so I put the logic where I call the pump routine into another routine instead of the main routine and still the same issue, but I'am sure that it has something to do with how the parameters inside the routine are passed and returned. The reason why I'am saying this is because, as I have mentioned before I tested the push buttons without calling them into a routine and the counter never missed the amount of times I pressed the push button, so if anyone has a workaround to this issue your aid is much appreciated