Jump to content

Calculating FPM (feet per minute) or just Feet on a press.


Recommended Posts

Calculating FPM (feet per minute) or just Feet on a press.

 

We have an old Comco Flexo Press from the early 90s, I think.  We are removing the old Totalizer and replace it with a SM70-J-R20 with attached program.  I am not sure if I am calculating the fottage correctly because it is different from the old totalizer.  Could someone point me in the right direction please.  And Thank you.

ComcoTest_V3.vlp

20221014_215302.jpg

Link to comment
Share on other sites

  • MVP 2023

Before anyone can help you on this, the thing that needs to be fully understood is the number of shaft rotations to output length.

In your program, what is E-stop?  Emergency or End?  Why the different formulas?...Trying different maths to see if it matches up to old? 

There are many things odd in the structure of your program that invite comment, but the distance relationship is a must know before all else.

cheers, Aus

Link to comment
Share on other sites

3 hours ago, Ausman said:

Before anyone can help you on this, the thing that needs to be fully understood is the number of shaft rotations to output length.

In your program, what is E-stop?  Emergency or End?  Why the different formulas?...Trying different maths to see if it matches up to old? 

There are many things odd in the structure of your program that invite comment, but the distance relationship is a must know before all else.

cheers, Aus

Yes.  The last three rungs and the first few rungs are the important ones.  The Stop/LED is not important at this time.  It took 2.5 inches for one revolution of the shaft.  The math it is using the same sensor for three different outputs (counters).  We are trying to see if the math it good.  What else do I need to do to capture a good foot?  I will check a few more things and be right back.  Thank you

 

Dragan

Link to comment
Share on other sites

At slow speed of the press 1 minute is about 507 rotations and according to the old counter it is about 106 feet.

5 rotations are about 1 foot.

At high speed for 1 minute is 1749 rotations @ 364 feet.

A * 2.5 / 12

The math is A = rotations of drive shaft.

2.5" is paper distance traveled in 1 rotation.

than divided by 12 for the footage.

Link to comment
Share on other sites

  • MVP 2023
6 hours ago, Gabriel Franco said:

2. Loss of precision due to formula calculations. Note that ML are integers, so decimals derived from calculations are truncated.

Yep.  Multiplying an integer by a floating point number is not a good idea.   

I would do this application in floating point (MF) land  Your program is simple at this point so it won't be hard to modify.  Look at the bottom of the Math functions for "Float".

When your math isn't working right the best way to troubleshoot it is to use individual function blocks.  You can't see inside a FORMULA block to look at the intermediate calculations.

I've done a little hacking on your program for the first calculation, using Visilogic 9.8.94.  You should be able to do the rest.

Joe T.

 

 

ComcoTest_V3 JT.vlp

Link to comment
Share on other sites

2 hours ago, Joe Tauser said:

Yep.  Multiplying an integer by a floating point number is not a good idea.   

I would do this application in floating point (MF) land  Your program is simple at this point so it won't be hard to modify.  Look at the bottom of the Math functions for "Float".

When your math isn't working right the best way to troubleshoot it is to use individual function blocks.  You can't see inside a FORMULA block to look at the intermediate calculations.

I've done a little hacking on your program for the first calculation, using Visilogic 9.8.94.  You should be able to do the rest.

Joe T.

 

 

ComcoTest_V3 JT.vlp 241.02 kB · 0 downloads

Thank you.  I am on it.  I will update.

Link to comment
Share on other sites

  • MVP 2023

I see you keep hours like I do.

You need to do your scaling in MF registers to accommodate the decimal point you need, but as you can see they revert to exponential notation when the numbers get big.  They only have 8 digits of precision.

You'll need to convert them to MLs for display purposes.  Have a look at the Float Convert functions in the Help, and post your code so I can hack a bit more.

Joe T.

Link to comment
Share on other sites

11 minutes ago, Joe Tauser said:

I see you keep hours like I do.

You need to do your scaling in MF registers to accommodate the decimal point you need, but as you can see they revert to exponential notation when the numbers get big.  They only have 8 digits of precision.

You'll need to convert them to MLs for display purposes.  Have a look at the Float Convert functions in the Help, and post your code so I can hack a bit more.

Joe T.

Okay thank you.  I was looking at after I sent the message.  Not to sound dumb but code = program?

Link to comment
Share on other sites

5 minutes ago, Joe Tauser said:

Yes

Ok I found it and will test it out shortly.  They are currently running, and the operator is a pain to work with.  You are awesome.  I work 12 hours and I am new at program, and this is my 5th press.  The other presses are working great.  This press is just different, and we really don't have a good place to put a sensor to measure the rotations.  We are on the same shaft as the old one just a few feet away.

Convert.jpg

ComcoTest_V5JT.vlp

Link to comment
Share on other sites

8 minutes ago, Joe Tauser said:

You'll see when you download it and look at the numbers.  A is the MF you enter into the block and B will be the numbers to the left of the decimal, which is what you're interested in.  C will be the decimal part.  You may or may not be interested in this number.

image.png.bd515b36280faf0090982552085a710e.png

 

Joe T.

 

Oh okay I see.  The image you see was not pulling up in the help file.   I have not yet tried it but does B and C auto populate?

Float.jpg

Link to comment
Share on other sites

  • MVP 2023
8 hours ago, draganjr said:

I have not yet tried it but does B and C auto populate?

As a general rule (is this ALWAYS the case? I'm not totally sure) the pieces on the left side of a Function Block are inputs to the function and the pieces on the right are the outputs of the function, so yes they auto-populate.

Link to comment
Share on other sites

21 hours ago, Flex727 said:

As a general rule (is this ALWAYS the case? I'm not totally sure) the pieces on the left side of a Function Block are inputs to the function and the pieces on the right are the outputs of the function, so yes they auto-populate.

Yes it did.  First time using it.  Working great so far.

Link to comment
Share on other sites

  • MVP 2023

I'm pleased it looks like things are working out ok and fully agree with everything already said.  But I also want to throw in something else.  To me your sensor needs to have it's shaft revolutions to sensor triggering checked to confirm it is a fixed proportion over all speeds.  This is easily done by looking at the sensor reading online whilst varying the shaft speed.   eg  100 of output = X   whilst 1000 of output =  X x 10  and so on, over all the speed range possible.

The reason is I wonder what metal your sensor target is made of (it looks to be aluminium) and whether the sensor is simply reading the steel screw that holds it in place on the shaft.  With higher speeds it may not be triggering the sensor correctly, or long enough for it to be read correctly, if it is only sensing the screw.  Not likely, but something to ensure is ok. 

Where the PLC controls the motor speed and also reads an output directly related to it, I always run this as a first check on this relationship by using a tiny bit of ladderwork for that machine that runs speed up and down in given steps and records results.

I would also possibly slightly increase the sensor gap from the target.  There's not much room there for any stray piece of waste etc floating around and getting caught.  On the other hand, the way it is now.....it has to get into the tiny gap in the first place, which contradicts what I'm saying!

Cheers, Aus

Link to comment
Share on other sites

10 minutes ago, Ausman said:

I'm pleased it looks like things are working out ok and fully agree with everything already said.  But I also want to throw in something else.  To me your sensor needs to have it's shaft revolutions to sensor triggering checked to confirm it is a fixed proportion over all speeds.  This is easily done by looking at the sensor reading online whilst varying the shaft speed.   eg  100 of output = X   whilst 1000 of output =  X x 10  and so on, over all the speed range possible.

The reason is I wonder what metal your sensor target is made of (it looks to be aluminium) and whether the sensor is simply reading the steel screw that holds it in place on the shaft.  With higher speeds it may not be triggering the sensor correctly, or long enough for it to be read correctly, if it is only sensing the screw.  Not likely, but something to ensure is ok. 

Where the PLC controls the motor speed and also reads an output directly related to it, I always run this as a first check on this relationship by using a tiny bit of ladderwork for that machine that runs speed up and down in given steps and records results.

I would also possibly slightly increase the sensor gap from the target.  There's not much room there for any stray piece of waste etc floating around and getting caught.  On the other hand, the way it is now.....it has to get into the tiny gap in the first place, which contradicts what I'm saying!

Cheers, Aus

Thank you for your feed back and I will look into it tomorrow night when I return.  I tried several sensors and it's true some where not working right.  I did move the current sensor over a little.  It's metal and I also thought of hiw wide I should make the flag for the sensor. 

Link to comment
Share on other sites

On 10/18/2022 at 4:46 PM, Ausman said:

I'm pleased it looks like things are working out ok and fully agree with everything already said.  But I also want to throw in something else.  To me your sensor needs to have it's shaft revolutions to sensor triggering checked to confirm it is a fixed proportion over all speeds.  This is easily done by looking at the sensor reading online whilst varying the shaft speed.   eg  100 of output = X   whilst 1000 of output =  X x 10  and so on, over all the speed range possible.

The reason is I wonder what metal your sensor target is made of (it looks to be aluminium) and whether the sensor is simply reading the steel screw that holds it in place on the shaft.  With higher speeds it may not be triggering the sensor correctly, or long enough for it to be read correctly, if it is only sensing the screw.  Not likely, but something to ensure is ok. 

Where the PLC controls the motor speed and also reads an output directly related to it, I always run this as a first check on this relationship by using a tiny bit of ladderwork for that machine that runs speed up and down in given steps and records results.

I would also possibly slightly increase the sensor gap from the target.  There's not much room there for any stray piece of waste etc floating around and getting caught.  On the other hand, the way it is now.....it has to get into the tiny gap in the first place, which contradicts what I'm saying!

Cheers, Aus

Good evening.  The sensor I am using is a Sick IME12-04BPSZC0S.  Switching freq is 2000 Hz.  I am using metal not aluminum, I did have aluminum and was not working out well.  I am assuming the higher the switching freq. the better.  The sensor does need to be card thick close to read at high speeds.  If not, it will not read at high speeds.  What is a good sensor to use in your opinion for high-speed counting.

I looked at other sensor we had in the shop, and this was the best one.   When we jog the press at low speeds the counters all count the same.  But again, at high speeds the new current counter falls back about 10 feet but gets greater over time.  I will give the actual number towards the end of my shift at work.

Thank you

Ragan

Link to comment
Share on other sites

58 minutes ago, Joe Tauser said:

The best sensor for high speed is an encoder.

You could put a toothed pulley on the main shaft and couple it to an encoder mounted next to it with a timing belt.  The PLC has a high-speed input on I0 designed for this.

Joe T. 

As Joe T. an encoder is the best way to measure speed/revolutions.

On 10/20/2022 at 2:13 AM, draganjr said:

a Sick IME12-04BPSZC0S.  Switching freq is 2000 Hz.

The switching frequency is not the problem. Is the way the sensor works.

those are inductive sensors the sensing depends on the material and distance this one was designed to be used mounted "flush", and even being within the range (3.4 mm) technically only the tangent point of the bar (a very small area will be in the range of the sensor). so, it is no surprising to me that it doesn't sense well.

I bet you that it will work better if its mounted parallel to the shaft and having some metal part attached or a longer screw pointing to the sensor

Although the accuracy won't be as good as an encoder, you will have a pretty good repeatability even at higher speeds

If you want to buy an encoder Sick has encoders that you can attach to the shaft, and they are like from very basic to very smart ones.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...