Set/Reset hangs

Lucask · May 9, 2020

Hi,

I have an odd issue.

I use the S/R instruction to enable another step to be executed in the program.
And when "Test_Seq_2" becomes 10, "Transferchain_7.3_Done" should go "False".

Sometimes this work, sometimes the "Set" keeps hanging.

The "Test_Seq_2" becomes "0", but the "Transferchain_7.3_Done" is still "True".

And "Test_Seq_1" is not in Step 10 when i am trying to reset the bit.

Could this be some scan-related issue?

And for good measure.
Same result as the above.

bkottaras · May 9, 2020

What about using a small time delay to initiate the Reset?
Say 50mSecs or so?
Even if it is a scan issue the delay will take care of it.
IF TestSequence=10 AND Not M101.2 -----> 50mSecs time delay
IF TestSequence=10 AND Not M101.2 AND Txxx=50mSecs THEN Reset M100.4

drbitboy · May 9, 2020

I would suspect Transferchain_7.3_Done is being changed on a rung not shown here.

As diagnostics:

1) add a set and reset of another bit, which is not used anywhere else, in parallel with the branches that S/R the Transferchain_7.3_Done bit.

2) If (1) doesn't tell you anything new, then two more separate unused bits and two counters:

one --( )-- in parallel with the Set branch that goes to 1 if the set branch is true,
one --( )-- in parallel with the Reset branch that goes to 1 if the reset branch is true,
Four new networks: put two separate counters on each new bit, one to detect a rising edge and one to detect a falling edge.

Lucask · May 9, 2020

Thank you for the replys.

drbitboy,

Have tried with the parallel coils, same result.

I tried the following:

IF "Test_Seq_2" =10 AND NOT "Conveyor_5.1_Outlet" AND NOT "Transferchain_7.3_Done" THEN
"Test_Seq_2:=0;
END_IF;

This way I was sure that "M100.4" was false.

But the question here is why, I guess.

I would have thought it scanned the entire network.

Jkoso · May 9, 2020

Any differenses if you put -(R)- on that MOVE ENO output?

Lucask · May 9, 2020

Actually.

I have tried doing what you suggest Jkoso, and also before the Move function.

Sometimes it works, other times it doesn't.

The only thing that seems to work, is making sure that the bit is actually "False" before Moving 0 into the step.

Lucask · May 9, 2020

And it is only in two networks it is written to.

drbitboy · May 9, 2020

are these networks running in physical hardware or in a simulator?

Also, what brand are we looking at (it seems like Siemens)?

Lucask · May 9, 2020

Physical.

The conveyors and other stuff did not behave as expected.

Then i started to dig in.

Lucask · May 9, 2020

When testing this on the simulator in Tia Portal, everything works as you expect it to.

bernie_carlton · May 9, 2020

In the first rung you test the value of Test_Seq_1 and in the second rung you test Test_Seq_2. Is it possible for both of theses be equal to 10 on the same scan?

Lucask · May 9, 2020

No,

"Test_Seq_1" becomes 20, after around 30 seconds (after executing other steps)
"Test_Seq_2" becomes 10.

Isn't it likely that it is as simple as scan related?

That is the only thing i can think of.

drbitboy · May 9, 2020

Lucask said:
No,

"Test_Seq_1" becomes 20, after around 30 seconds (after executing other steps)
"Test_Seq_2" becomes 10.

Isn't it likely that it is as simple as scan related?

That is the only thing i can think of.

what do you mean by "scan related?"

Lucask · May 9, 2020

What I mean is that 0 is Moved to "Test_Seq_2" before "M100.4" becomes "False"

Steve Bailey · May 9, 2020

Is there any chance that somewhere in the program you are writing to %M100 as a byte or word instead of turning on or off individual bits?

Set/Reset hangs

Lucask

Member

bkottaras

Member

drbitboy

Lifetime Supporting Member

Lucask

Member

Jkoso

Member

Lucask

Member

Lucask

Member

drbitboy

Lifetime Supporting Member

Lucask

Member

Lucask

Member

bernie_carlton

Lifetime Supporting Member + Moderator

Lucask

Member

drbitboy

Lifetime Supporting Member

Lucask

Member

Steve Bailey

Lifetime Supporting Member + Moderator

Similar Topics