Hello, i am trying to solve a mystery without even knowing if this that i am looking for is achievable.

First of all i have an input that is a paddlewheel flow sensor. 

What i am trying to build is the following :

When a certain amount of water has passed through the system i want to trigger an output. I am supposed to measure the m³  but not measure time depending on the flow rate. I mean i can easily measure time and every (y) hours trigger the output. The outcome would be almost the same because the flow rate will be the same every time, 0 or (x) . But i would like to know if this is possible. After some calculations i made i need to count 135 hours or 22.000.000 pulses but i dont know how to keep them in memory. 

I tried connecting a threshold trigger to the input and then use a mathematic instruction but i am stuck there. 

tasostop

Hello, i am trying to solve a mystery without even knowing if this that i am looking for is achievable.

First of all i have an input that is a paddlewheel flow sensor. 

The sensor can measure flow from

0.15 m/s (0.5 ft/s) producing a frequency output signal highly repeatable.

What i am trying to build is the following :

When a certain amount of water has passed through the system i want to trigger an output. I am supposed to measure the m³  but not measure time depending on the flow rate. I mean i can easily measure time and every (y) hours trigger the output. The outcome would be almost the same because the flow rate will be the same every time, 0 or (x) . But i would like to know if this is possible. After some calculations i made i need to count 135 hours or 22.000.000 pulses but i dont know how to keep them in memory. 

I tried connecting a threshold trigger to the input and then use a mathematic instruction but i am stuck there. 

The LOGO! is not a good tool for this job. The LOGO! uses Integer maths.

That means you may need to use multiple counters to track a big number. Let us say counter 1 tracks up to 10,000 and triggers counter 2. The counters can be set to retain its value.

Now for the hiccup: What is the speed of the incoming pulse? Counter 1 can handle up to 5kHz if connected directly to a 'special' input. Counter 2 cannot handle the same speed -- it will be at the speed of the scan cycle.

The LOGO! is not a good tool for this job. The LOGO! uses Integer maths.

That means you may need to use multiple counters to track a big number. Let us say counter 1 tracks up to 10,000 and triggers counter 2. The counters can be set to retain its value.

Now for the hiccup: What is the speed of the incoming pulse? Counter 1 can handle up to 5kHz if connected directly to a 'special' input. Counter 2 cannot handle the same speed -- it will be at the speed of the scan cycle.

  I thought the same about using counters that trigger other counters but i think that this is not a reliable solution. 

I have some technical data but the actual flow rate is decided from many variables and i think that i cannot test the system before it will be installed. I am just trying to find a way to start.

• Flow Rate Range: 0.15 to 8 m/s

- square wave

- frequency: 45 Hz per m/s nominal

So i get 45HZ per second for one litter of water that has passed through the particular pipe size. 

tasostop

I thought the same about using counters that trigger other counters but i think that this is not a reliable solution. 

I have some technical data but the actual flow rate is decided from many variables and i think that i cannot test the system before it will be installed. I am just trying to find a way to start.

• Flow Rate Range: 0.15 to 8 m/s

- square wave

- frequency: 45 Hz per m/s nominal

(13.7 Hz per ft/s nominal)

So i get 45HZ per second for one litter of water that has passed through the particular pipe size. 

Peak is 8m/s or 8 x 45 = 360 pulses / second. This translates to 2.78ms between pulses. 

Are you looking for an approximate volume or should it be just close to? Assuming 20ms scan, an inaccuracy of 7.2 pulses may be introduced each time Counter 2 is triggered. Over the 22,000,000 pulses that would occur 1,100 times which is approximately 8,000 pulses or 0.03% error.

It can be done, though not perfect. It is inconvenient to make a user input translate to the end count target. It is inconvenient to translate the target to the required count amounts.

I would have used a Rockwell Micro810 with screen if I was on a shoe string budget. But you may have a reason to want to use the LOGO! -- right?

I totally agree with you but this is not my decision. I figured out that i cannot be precise enough and that would be translated to loss of almost 500 litters of water every time i finish a cycle and i trigger the output. Unfortunately there is no other option for me, its the only solution i can use at this time. 

So in conclusion, i have  to trigger one counter for some number of pulses and then trigger another counter every time the 1st counter reaches a certain amount of counts.
I actually don't know any possible way to count and store the pulses. 

tasostop

I totally agree with you but this is not my decision. I figured out that i cannot be precise enough and that would be translated to loss of almost 500 litters of water every time i finish a cycle and i trigger the output. Unfortunately there is no other option for me, its the only solution i can use at this time. 

So in conclusion, i have  to trigger one counter for some number of pulses and then trigger another counter every time the 1st counter reaches a certain amount of counts.
I actually don't know any possible way to count and store the pulses. 

Why do you want to store pulses? Is that not what the counter is for. Perhaps you should show LOGO! program of what you do know. Here is a sample concept you can start with.

This model that you designed is great and would work but when it resets i wont  be able to know anything about stats. In order to calculate , i set the 1st counter to 1 m³ (20000)pulses and the other to the maximum amount of m³ that i will count and then triger the output. But it is required to have L/h and total m³ that have been through the system on the screen.I also need to trigger an alarm for a flow rate lower than a specific.

So i have to store the pulses somehow. The problem is that the counter has some limits and when they exceed i will have to reset it but then i lose the total m³ . I am also curious about how to show these stats in real time.

tasostop

This model that you designed is great and would work but when it resets i wont  be able to know anything about stats. In order to calculate , i set the 1st counter to 1 m³ (20000)pulses and the other to the maximum amount of m³ that i will count and then triger the output. But it is required to have L/h and total m³ that have been through the system on the screen.I also need to trigger an alarm for a flow rate lower than a specific.

So i have to store the pulses somehow. The problem is that the counter has some limits and when they exceed i will have to reset it but then i lose the total m³ . I am also curious about how to show these stats in real time.

Now you are morphing the requirement into flow rate measurements.
Specifically a moving average.
Recently it has been handled in an earlier topic: 
https://support.industry.siemens.com/forum/de/en/posts/299391 

I think it is at this point that I leave this discussion.
Good luck.