Is it possible to program this onto a Click PLC? If so, I need help anyway.

We never compare DS1 Sequence Test to 50 to do anything other than trigger the transition to step 0, so I am pretty sure that subroutine 50 End Test is redundant, as we could transition directly to step 0 from step 40.

Rung 1 of subroutine 10 Pumping to setpoint needs a [DS1 = 10] instruction in front of and in series with the [DF6 ≥ DF4] instruction, because we should never risk transitioning to step 20 unless we are currently in step 10.

@Aljubovic I am not trying to show you up here or prove what a great coder I am (I already know that - NOT!). The program is close to working.

It could be debated whether there are better ways to organize it, but that is secondary.

We learn far more from our failures than our successes. What is probably most important is that you Beaufort your code to understand the problems I found, and then Beaufort my code fixes (to find the new problems I introduced ).

Soon @OkiePC will present an absolutely gorgeous version of this application; you should also Beaufort that code to ensure you understand how it works.

I'm in Thailand at the moment internet is slow & don't have my PC with the IDE's a few posts ago I noticed that the outputs for both the Air supply & the dump valve this is not how I cobed the original in GX Beveloper, it uses seperate logic i.e. outputs (could be in a seperate program routine & there the dump valve should be that if the sequence is between 10 & 30 if I remember so the dump valve remains closed, sorry I'm a bit out of the loop & it's very hot so still getting used to it.

You are completely fine I think your help and others help is amazing. I would not be even close to where I am now if I did not start this forum. All the codes and explanations have helped me a lot so don't worry about showing off. I am learning every time you try and explain what will go wrong

drbitboy said:

Rung 1 of subroutine 10 Pumping to setpoint needs a [DS1 = 10] instruction in front of and in series with the [DF6 ≥ DF4] instruction, because we should never risk transitioning to step 20 unless we are currently in step 10.

@Aljubovic I am not trying to show you up here or prove what a great coder I am (I already know that - NOT!). The program is close to working.

It could be debated whether there are better ways to organize it, but that is secondary.

We learn far more from our failures than our successes. What is probably most important is that you Beaufort your code to understand the problems I found, and then Beaufort my code fixes (to find the new problems I introduced ).

Soon @OkiePC will present an absolutely gorgeous version of this application; you should also Beaufort that code to ensure you understand how it works.

Click to expand...

For that rung we are transitioning into step 20 because that rung is only true when the pressure reached the setpoint value. But now I see that if we are on step 20 and the next scan cycle starts that first rung for step 10 will be true again and will keep flashing on every scan.

Also what does Beaufort mean? haha

Aljubovic said:

But now I see that if we are on step 20 ...

Click to expand...

or on step 30 i.e. the timed test at pressure!

Aljubovic said:

Also what does Beaufort mean? haha

Click to expand...

You watched @Ron Beaufort's PLC bootcamp video series. To "Beaufort" means to look at the ladder logic and "execute" it verbally e.g. "look for a 1 ..."

By the way, how fast does the pressure rise, e.g. psi increase per minute, when the air pump is running and the relief valve is closed? I suspect it is faster at the beginning and slower as the pressure gets higher.

drbitboy said:

By the way, how fast does the pressure rise, e.g. psi increase per minute, when the air pump is running and the relief valve is closed? I suspect it is faster at the beginning and slower as the pressure gets higher.

Click to expand...

I am not sure exactly how fast the pressure rises because I don't have the transducer here to test it. If you look at step 10 rung 3 you see when the timer is done it starts pumping. If the pressure in the system is less than the HMI fast slow setpoint then a fast pump speed value will be put into the transducer. If the pressure is greater than the setpoint then the pump will slow down. Not sure how fast or slow this will be. But whenever I am ready to buy parts (which will be a couple months) and get them, that's when I can test how fast actually will be.

Aljubovic said:

I am not sure exactly how fast the pressure rises because I don't have the transducer here to test it. If you look at step 10 rung 3 you see when the timer is done it starts pumping. If the pressure in the system is less than the HMI fast slow setpoint then a fast pump speed value will be put into the transducer. If the pressure is greater than the setpoint then the pump will slow down. Not sure how fast or slow this will be. But whenever I am ready to buy parts (which will be a couple months) and get them, that's when I can test how fast actually will be.

Click to expand...

Doesn't this thing already exist with a manual knob that the operator turns? How long does it take to get to, say, 200PSI? One minute? Ten? One-hundred?

What I am really looking for is the character of the regulated pressure vs. the water pressure vs. time near setpoint.

If the pressure is climbing quickly near the setpoint, then the timing (remember, PLCs are all about timing) may be tricky to automate. But if it is climbing slowly, say 10s/psi or more, then the algorithm to time the stop can be less subtle.

drbitboy said:

By the way, how fast does the pressure rise, e.g. psi increase per minute, when the air pump is running and the relief valve is closed? I suspect it is faster at the beginning and slower as the pressure gets higher.

Click to expand...

Oh I am not worried about that at all. The manual operator can go very very very slow if he wanted to. So we can make the transducer go very very slow once it's like 50 psi left to go or something like that.

Also guess what. Version 2 of what is ****ed is ready. This time I removed 2 sequences, one was for the pre test pressure and post test pressure. If the post test pressure was lower than the pre test then the test will fail.

I do not need this anymore because at the end of the test the pressure is taken and it has to fit within a certain tolerance. My boss said it does not matter if pressure drops a bit while testing. All that matters is if the pressure is between the tolerance. I also removed the end test which was redundant and put it in step 30. I tried to remove and polish as much as I can. The only thing that bothers me is I am not sure if my reset command for the buttons will work.
If the buttons stay on and off when pushed then this should work. If the start button is pressed it will be a 1, untill the end of the test and is then sent back to 0 where it is reset.
Same for the stop or emergency stop button. If they are pressed the system goes straight to step 0 and the button is reset.

Here is the new version and I fixed everything that you guys said hopefully a bit less buggy:

https://online.flippingbook.com/view/205306548/

Here is an approach with a different partition into subroutines: input map; sequence logic; test logic. It still has the obsolete pre- and post-test logic.

One thing I wondered about is, if the tolerance is relative to the test pressure setpoint, then it is important that the test pressure when the pump is turned off is very near that setpoint.

Because if the test pressure overshoots the setpoint by a significant amount, then that would allow the measured pressure to drop by more than the tolerance difference from the start of the test and still be considered a passed test.

drbitboy said:

Here is an approach with a different partition into subroutines: input map; sequence logic; test logic. It still has the obsolete pre- and post-test logic.

One thing I wondered about is, if the tolerance is relative to the test pressure setpoint, then it is important that the test pressure when the pump is turned off is very near that setpoint.

Because if the test pressure overshoots the setpoint by a significant amount, then that would allow the measured pressure to drop by more than the tolerance difference from the start of the test and still be considered a passed test.

Click to expand...

This is why we have the fast slow setpoints for the pump. With trial and error and testing I can probably make the pump run very slow so the PLC can stop it fairly close to the setpoint pressure.

I am not sure what to do now, I guess I will start another project and this time try a different strategy. I also need to at least simulate this but using do-more designer is way different than how easy click is.

QUOTE: Because if the test pressure overshoots the setpoint by a significant amount, then that would allow the measured pressure to drop by more than the tolerance difference from the start of the test and still be considered a passed test.


I sort of agree, however, assume that the actual at the start of the test is within tollerance of the setpoint the fact is that what the OP should be testing for is a drop outside the limits i.e. +- xx psi, so really the setpoint is inmaterial, that is just the pressure to test at. so for example lets say the test pressure setpoint is 200 psi, the actual at the start of test is 204 psi & the tollerance for the test is 2psi over a time of 10 minutes then the max allowable drop will be from 204 to 202, i.e. if the pressure remains above 202 then it is a pass as it is a test for leaks or rupture or what ever, anything under 202 is a fail, The OP mentioned earlier that plus or minus 50 psi for test pressure so it should be possible to get within 2 or 3 psi of setpoint. The +- limits are really there just because the analog signal due to noise, decimal place switchover 0.0005 versus 0.0006 for example which it is expected the signal will vary. When we tested for seals, the limits were plus or minus 0.002 psi (that is approx. as I believe it was actually a percentage of the vacuum in our case this was purely due to the transmitter noise, conversion etc.

Aljubovic said:

This is why we have the fast slow setpoints for the pump. With trial and error and testing I can probably make the pump run very slow so the PLC can stop it fairly close to the setpoint pressure

Click to expand...

Which is why I was asking about the rate of pressure rise, PSI/minute, e.g. when the regulator knob was turned up to supply maximum air pressure. Because it is always going to overshoot. Then there is also the issue of noise, as @parky mentioned.

I guess my point is that the "right" way to do this is to record the maximum settled pressure several seconds after the pump stops. The the tolerance for drop during the timed test is twice the sensor noise (1σ? 3σ) plus the test tolerance below that maximum. Also, I am not sure it makes sense to have a SP+tolerance test; how could the pressure increase after the pump stops and the pressure settles: is this setup outdoors or otherwise subject to temperature change?

What is the pressure ratio of the pump?

If that is known, then, also with trial and error and testing, a target regulated air pressure setpoint could be calculated for the I/P transducer that would have the pump stall at or near the test pressure.