Flying Cut off saw

I have a customer that wants to put together a flying cut off saw that will be positioned hydraulically, whats the fastest anyone ohas seen a saw move? His line is moving at 450 FPM and thats way too fast by what I can see.

Will

Your gut instinct may be valid, but to know for sure you will need to do some calculations.

• how quickly can the carriage accelerate from 0 to 450 fpm and what distance does that represent?
• how long does it take for the saw to make its cut and what distance that represent?
• how quickly can the carriage decelerate from 450 fpm to 0 and what distance does that represent?
• add up those distances plus a bit for safety and you have total length for the flying cutoff
• how quickly can the carriage return to 'home' position ready for next cut and what material length does that represent @ 450 fpm?
• add up the total flying cutoff length plus the material length during return plus material length during accel to get the shortest cut length that you can handle

I have already looked at that, what I am after is actual applications that are at work right now. I know what isnt possible but I want to know what is possible.


Thanks,

Will

I just got through working on a flying cutoff saw.

The flying shear I saw went about 30 inches per second or 150 FPM. It looked pretty tame. 300 FPM or 60 inches per second would be another thing. I don't think 450 is practically possible. I am sure it can be done given a real desire to spend money.

The calculations Gerry suggests are a start, but the motion profile is the easy part. The hard part is to properly size the hydraulics. That gets tricky and it is easy to overlook something. You will not be able to control something like this with just a PLC.

In addition to Gerry's questions I would also ask
1. What is the customer willing to spend? Unless the customer has deep pockets don't bother.
2. What is the mass?
3. What is the cycle time.

So how good of a hydraulic designer are you? You must be fearless, foolish or just very good to consider such a project. On Ron's site I am asking about one of the main problems about the encoder reference that the shear must be geared to. This encoder reference must be nearly jitter free. This is a mechanical problem and you will have no control of this unless you do the mechanical design too. In Ron's hydraulic area I am asking about how long it takes for the pump to be at full stroke. It isn't well documented. A flying shear is a fast operation and the pumps respond very slowly. I found the one must add more accumulator capacity than what you might think. In my case one can't assume a 47 GPM pump is always producing 47 GPM.

There are just too many things that can go wrong.

Peter, I was waiting for you to answer this post, its good that you have. I am niether very foolish or very smart, but I can smell problems with customer expectations when I see them. I am at the very plelimanry stages with this project. Infact the fellow I am working with on the hydraulic side needs to get more information from the customer but I can tell you this much right now, there is no way we will get this thing working at 450fpm with the mass I was given and the cylinder lenght. Once again I think this is a case of someone dreaming up something before researching it.

Will

How can it be done??

willpower100 said:

I have already looked at that, what I am after is actual applications that are at work right now. I know what isnt possible but I want to know what is possible.

Thanks,

Will

Click to expand...

Will what is the product? How wide and thick? How flexible?

End cutting two by fours is probably a little different than plywood which is a little different than cutting coiled steel sheet.

Lessee 450 FPM
is 7.5 feet per second
is .75 feet (nine inches) in 0.1 second.
is .075 feet (0.9 inch) in 0.01 second
is 0.09 inches in 0.001 second

That saw has no time for dilly dallying without binding or bunching up the material if there is any width to the material.

What about a water jet cutter or other "non contact" cutter and if a diagonal cut is allowed?

Dan Bentler

This still can be educational. This can be a physics lesson.

willpower100 said:

I can tell you this much right now, there is no way we will get this thing working at 450fpm with the mass I was given and the cylinder length.

Click to expand...

Just for educational purposes, why don't you provide the mass and cut length? I don't see why you didn't provide the length and mass so people can try calculating the acceleration, force and power required. We can see if the application is really absurd. You can make a test or contest out of this. We did this with sinusoidal motion on Ron's site. If you don't want to do it here then do it on Ron's Fluid Power site. I can help with the answers.

willpower100 said:

Once again I think this is a case of someone dreaming up something before researching it.


Will

Click to expand...

I see this all the time. Too many customers have no clue about how difficult or expensive their wants are. Obviously this customer is clueless and is waiting for you to research this.

The calculations Gerry suggests are a start, but the motion profile is the easy part

Click to expand...

Allen Bradley has a 1771-QB Linear positioning module, which is normally used for apps like this one. According to the manual it has a scan of 2 ms. At 450 FPM or 90 Inches per sec you would be facing potentially a positioning error of 0.18".

Jiri, tell us......

How do you synchronize the cut off saw with the feed system or reference encoder using a QB module? Don't you think that is important? How do you follow a motion profile while accelerating without acceleration feed forwards?

The 1771-QB has a velocity or position based profiling and a transducer input. It is a closed loop operation. The transducer in this case would be a Temposonic or Balluff type linear position transducer. My understanding is that the module would not accept an incremental encoder.
Perhaps linear position transducer would be OK?
There is some kind of a limit on how large these transducers can be and the resolution will suffer at large length.
Best thing is to check with AB and Temposonic.

Not with a QB you don't

Jiri, aren't you aware there are other Delta and Rockwell controllers that have a very small chance of working in an application like this. The QB has none. I hope Willpower ignores this QB stuff and heeds the rest as a warning.

The Delta and RA 1756 class motion controllers support gearing, clutching or camming which the QB does not. The Delta and RA 1756 class controller support acceleration feed forwards which are essential to controlling large mass while accelerating and declerating. Both the Delta and RA 1756 class controllers have updates that are at least twice as fast as the QB. At 450 FPM you are traveling .180 inches per 2 milliseconds. That isn't very fine position resolution. Go to Ron's site and read about the mechanical problem thread.

The Delta and RA 1756 class controllers can mix an match inputs and outputs. This means there is a means of directly interfacing to a reference encoder. You can't use a MDT rod for a reference encoder. First because it is continous motion and second, the QB only uses PWM or gates MDT inputs which do not provide the resolution necessary for being a reference.

Finally, good luck trying to tune the system or even prove that that shear in cutting in the right place. Where is the trend or plotting feature?

Jiri, the QB is an old '80s design not intended for flying cut off saws.

Hey Peter,

Thanks for the explanation. I knew it was just a long shot.
Last time I was involved with hydraulic motion control was about 12 years ago then the 1771 still ruled.
I did mention the scan time as a drag.
I thought that since the carriage has to return home you might
not have to use the encoder??
Which 1756 module did you have in mind?

The 1771-QB equivelent / replacement in the 1756 family is the 1756-HYD02 2-axis Analog/LDT Module

The 1756-M02AE 2-axis Analog/Encoder Module is the analog motion card.

Dynapro built the QB with sawmill setworks in mind - as Peter said, about 20+ years ago. Nowhere near as demanding an application as this flying cutoff saw.

willpower100... long time, no see...

This kind of question, as you have posed it, always leaves the actual situation wide open to personal perception and interpretation by those that are trying to help.

A poorly defined question gets poorly considered answers.

This problem is solvable... however... we need some answers to some critical questions.

I wonder...
Is the material being cut being cut by a blade, or is it being cut by some sort of water-jet?

I've worked with both, flying-blade-cut and flying-water-jet-cut.

If being cut by a blade, does the blade assembly "grab & hold (hold while flying)" the material as it is being cut?

If being cut by a blade, is there some room (timing-room) for allowing the lateral motion of the blade-assembly to be a bit slower than the line speed?

Is there a discernable difference between items to be cut?

Time-wise, what is that difference? (leading-edge to leading-edge and cut-extend to cut cut retract)

If being cut by a water-jet, the head must move at a particular speed for the particular angle (the speed corresponds to the angle, SIN, COSINE sorta stuff) to make a square cut.

If cut by water-jet and the cutter is hydraulically driven then there MUST be a way to at least adjust the cross-rate. Otherwise, there must be some sort of position feed-back! (Feed-back control is by far the better way to go!)

Do you have any kind of position feed-back?

These are critical questions that need to be answered before a solution can be developed... even so, there may be other critical questions needing answers depending on the answers to these questions.