Set/Reset hangs
- Thread starter Lucask
- Start date
I am using a 1515 CPU, and the cycle time is about 2ms.
Does the plc scan coils differently from the Move function?
As far as I know, no...
I should mention again, that it is working now, by just monitoring that (in this case) M100.4 is "False" and then "Move" 0 to "Test_Seq_2".
Before doing the above (without looking at M100.4 is "False"), when i had this problem. When testing this, it looked good, didn't think more about it.
But when in Auto mode and everything running, all of a sudden, this problem occured.
But the question here is why this is happening,
so you can take that into consideration when making the program.
Not talking about physical outputs, only the Internal memory.
Does the plc scan coils differently from the Move function?
As far as I know, no...
I should mention again, that it is working now, by just monitoring that (in this case) M100.4 is "False" and then "Move" 0 to "Test_Seq_2".
Before doing the above (without looking at M100.4 is "False"), when i had this problem. When testing this, it looked good, didn't think more about it.
But when in Auto mode and everything running, all of a sudden, this problem occured.
But the question here is why this is happening,
so you can take that into consideration when making the program.
Not talking about physical outputs, only the Internal memory.
drbitboy
Lifetime Supporting Member
In the same scan cycle?
And then the executive goes back to the beginning of [Network 5] and re-evaluates the entire rung/Network to false, so it does not execute the Reset instruction in the branch?
Is that what you are suggesting?
That would violate left-to-right, top-to-bottom, do you agree?
drbitboy
Lifetime Supporting Member
rQx
Lifetime Supporting Member
Change place of the move and coil in the parallell branch so that the coil executes first?
I have had problems with execution of parallell rungs in TIA portal before. You can also maybe have the outputs in series? Or make them execute in seperate rungs after each othee
I have had problems with execution of parallell rungs in TIA portal before. You can also maybe have the outputs in series? Or make them execute in seperate rungs after each othee
Last edited:
drbitboy
Lifetime Supporting Member
And you said you are seeing this on physical hardware and not on a simulator.
Ouch.
If I didn't think I recognized the programming environment from the PNGs, I would be asking what fly-by-night company makes that hardware. This is something I would expect to see from an early commit of a Github repo trying to make a RaspberryPI into a PLC.
There are warnings, disclaimers, and caveats about safety, injury, and death all over in the documentation, but they do this? If true, then that is unconscionable.
Is this common knowledge, i.e. don't use branches on this company's hardware?
Can you post a .ZIP with the entire project? I have a test setup and would be interested to see this. I started to do it by hand yesterday but TIA Portal crashed.
Last edited:
L D[AR2P#0.0]
Lifetime Supporting Member
Recode your original network 5 as an SCL network and test to see if the behaviour changes.
If I need to run 2 different sequences that needs to interact with each other , I would make a master sequence controlling the other 2, it gives me a better overview of what is going on.
I don't see anything in your network 2, that is preventing it from setting the done bit, when network 5 is trying to reset the bit
Now this is just pieces of a program, there might be mechanical things that would preventing this, but if it was possible to set both sequences to 0 in network 5 you would prevent network 2 firering again.
I think I have experienced something simuler to what rQx is describing,
but at the moment I was not sure if it was just me being low on sugar, but ever since then, in all my sequences the last thing I do in the network is setting the sequence step.
I don't see anything in your network 2, that is preventing it from setting the done bit, when network 5 is trying to reset the bit
Now this is just pieces of a program, there might be mechanical things that would preventing this, but if it was possible to set both sequences to 0 in network 5 you would prevent network 2 firering again.
I think I have experienced something simuler to what rQx is describing,
but at the moment I was not sure if it was just me being low on sugar, but ever since then, in all my sequences the last thing I do in the network is setting the sequence step.
drbitboy,
I want to know aswell if it something to look out for in the future.
Unfortunately I can't share the project.
If you are testing this, it probably is going to work.
When I did the testing this seemed to be just fine, but when running and executing other parts of the program, it started to happen.
Holmux,
Exactly, this is just part of the program.
This is not the actual program, (hence the name "Test_Seq_2") but this is just the steps to show my problem.
The only way these two steps can set/reset at the same time, is if i manually type 10 in one of them.
What I think is strange, is that even in Series with the move (before and after the Move block) the reset, doesn't even reset.
I was monitoring this, making sure that the "Set" was not TRUE. (It was not in that Step anymore)
But "Test_Seq_2" became 0.
Is the plc prioritizing something differently when scanning?
I want to know aswell if it something to look out for in the future.
Unfortunately I can't share the project.
If you are testing this, it probably is going to work.
When I did the testing this seemed to be just fine, but when running and executing other parts of the program, it started to happen.
Holmux,
Exactly, this is just part of the program.
This is not the actual program, (hence the name "Test_Seq_2") but this is just the steps to show my problem.
The only way these two steps can set/reset at the same time, is if i manually type 10 in one of them.
What I think is strange, is that even in Series with the move (before and after the Move block) the reset, doesn't even reset.
I was monitoring this, making sure that the "Set" was not TRUE. (It was not in that Step anymore)
But "Test_Seq_2" became 0.
Is the plc prioritizing something differently when scanning?
moggie
Lifetime Supporting Member
Can you add another MOVE onto the ENO of the existing MOVE in network 2 and MOVE 15 into a spare tag(INT).
Then do the same to the MOVE in network 5 and MOVE zero (0) into the same spare Tag(INT) and run the program.
Does your new tag change value from 15 to Zero ??
Then do the same to the MOVE in network 5 and MOVE zero (0) into the same spare Tag(INT) and run the program.
Does your new tag change value from 15 to Zero ??
JRW
Member
Set up a trace too
drbitboy
Lifetime Supporting Member
Exactly right, my version works as expected.
However, my Test_Seq_1 and Test_Seq_2 words are %MW110 and %MW112; yours have the prefix "Seqvenser" - does that mean they are stored in a different memory area? Is that other memory area contain static memory i.e. will those tags have the same value at the start of a scan cycle that they had at the end of the previous scan cycle?
L D[AR2P#0.0]
Lifetime Supporting Member
"Seqvenser" is the symbolic name of a global data block so on a scan by scan basis the tags in a global DB will behave the same as M flags.
Similar Topics
Hi,
I was testing my 1747-ASB for some I/O mapping last week. Today when I tried to power-on my chassis with 1747-ASB inside, the ASB just hang...
Hello,
I still do not have equipment in to test.
But the RTAC ladder does not have a timer reset coil.
I wonder if a reset coil on Timer.IN is...
Hi,
i am using DVP-14SS2 PLC, after program written to plc, when power is reset, plc doesn't run. always need to connect to pc for the run mode.
Hi need help why this “failure 5 emergency stop “ appears at every startup in the morning ?
Have to shut off main switch at least 10 times on...
I was loading a program onto an XE1e2 PLC and it got stuck on these two windows and won't progress. It won't let me connect from the PC to reload...