flying cutoff saw controller

[Daile]

Can anyone give me an example of a flying cutoff system running with a PLC as the main controller?

All saw controllers i have seen are either PC or microprocessor card type controllers. I put this down to PLC's not being fast enough to handle the position/speed control.

A number of people i have spoken to are of the opinion that PLC's aren't up to the job. My gut tells me this is not the case (with new PLC's being much faster) but id like to hear some feedback on what is the standard this day and age.

 

[Steve Bailey]

The PLC by itself probably is not fast enough to act as the motion controller for the servo drive. But many PLCs can support motion controller modules that are up to the task. I've designed one using that architecture.

 

[Leadfoot]

I have seen several cases where a PLC controls a servo that drives the cutter. The PLC needs line speed input and cutter position/speed to sync and cut.

 

[Daile]

Any idea what sort of specs the system had?

ie max line speed, sync time, carriage run length - all that stuff.

The system i am looking at has a max speed of 25m/min. the carriage has a mass of about 5000kg (it has two 250kW motors running two friction saw blades!) and a runway length of about 6.5m. The carriage is driven by a hydraulic ram with a moog valve and incremental encoders on the ram and on the material feed.

 

[Steve Bailey]

The system I'm familiar with needs to be able to match a line speed of 30 FPM maximum. Off the top of my head, I don't remember the mass of the moving mechanism, but it uses a Fanuc Beta6 motor. That's 6 Nt-meters continuous torque at 2000 RPM.

It cuts a 48 inch wide panel in about 17.5 inches of travel. It has about 6 inches of motion in which to accelerate up to match line speed before the saw blade starts to cut.

 

[Peter Nachtwey]

Just to get started.

Any idea what sort of specs the system had?

ie max line speed, sync time, carriage run length - all that stuff.

The system i am looking at has a max speed of 25m/min.

That isn't too fast.

the carriage has a mass of about 5000kg (it has two 250kW motors running two friction saw blades!)

That is a lot of mass.

and a runway length of about 6.5m.

Does that mean you have 6.5m to make a cut?

The carriage is driven by a hydraulic ram with a moog valve and incremental encoders on the ram and on the material feed.

Not enough info. I hate playing twenty questions.

This is an example of the kind of analysis done for willpower100's proposed flying shear. This document is just for the hydraulic part. It will give you an idea of the info required for just the hydraulic part.
ftp://ftp.deltacompsys.com/public/PDF/Mathcad%20-%20Willpower.pdf
This is a simulation done for real flying shear for one of our customers. Again this is just the hydraulic part.
ftp://ftp.deltacompsys.com/public/PDF/Mathcad%20-%20Hydsim-ps.pdf

The mechanical and hydraulic design are critical. The actual programming for a flying shear is this is really very simple with the right tools. If the mechanical and hydraulic parts are designed right the tuning is easy.
I have example programs that I provide to our customers.

 

[darrenj]

o.k. now i am interested!!

By flying cut off saw are we talking about a saw that moves along the y axis while cutting in the x axis?

Peter if thats a "simple" problem...i would hate to see a complex one!
I have never seen one of these units..any one have a pic or a vid of one?

Thanks
D

 

[Ken Roach]

Vic Croston at Globe Machine in Tacoma designed the first flying saw I'd ever seen, which was part of a continuous wooden I-beam assembly machine. He used an Electrocraft IQ2000 series motion controller back then (about 12 years ago). I think he uses ControlLogix and M02AE modules now.

Vic is still known in the engineered wood products business as the godfather of flying saws. The neatest ones I saw were opposite-cut dual saws for taking samples on the fly. The blade/motor assemblies would jump up and down on pneumatic cylinders instead of drawing all the way out of the way of the board.

Once upon a time, the biggest one he'd built until then was in the assembly shop, being tested and debugged. An assembly electrician walked underneath the 'caution' tape to install some conduit, and broke the beam of the trigger photoeye.

The saw jumped up, started the blade, cycled across the carriage, and was spinning down quietly by the time the severed conduit dropped to the floor.

The saw's nickname stuck, after that: 'Grendel'.

 

[Peter Nachtwey]

Peter if thats a "simple" problem...i would hate to see a complex one!
I have never seen one of these units..any one have a pic or a vid of one?

Flying shears are not complex. The mechanics must be designed right so the system is controllable. Otherwise it is just a matter of math as some tuning.

I don't have any pictures or graphs. I do have plenty of graphs that show the motion profile as a function of time. I have the programs too.

If you want a real challenging application then check out veneer lathes or curve sawing.

 

[Daile]

Wow, I haven't seen anything like that since uni Peter. In fact i probably haven't seen anything i did at uni since i left!

The way it stands at the moment is we have a functioning cutoff saw. The hydraulics and everything are existing. The existing control system is effective. A qbasic program however, is not the most readily accepted format and the hardware running the program is unreliable and apparently difficult to find spares for. The decision has been made to replace the controller. The saw is running in a plant that is controlled by S7 and fully networked via Profibus.

I could quite easily convert the qbasic program to ladder and dump it in a high speed PLC with a high speed counter and an analog output card...but will it be fast enough for the speed and position control functions? I am contemplating dusting off my old control theory books and having a go at simulating the control system in discrete time. This could take me some time though after ive cleared the cobwebs! I am just wondering if anyone has encountered this already.

Peter - i will have a good look at your examples and see if i can get my brain working again. Cheers.

 

[bwheat]

Peter has not said it yet but the Delta motion controllers will solve this application quite nicely. I did the electrical design for two flying cutoff saws in 1998 using a Delta motion controller with a profibus interface to a TI PLC. The PLC just sent commands and setpoints to the motion controller. The motion controller handled all of the "hard work" The system worked out well. After I left the company they built at least four more just like the first two.

This can be done with a CLX and a motion control card but if your existing system already has a profibus network you already have a way to send this controller all of the info it needs.

 

[Jimmie_Ohio]

Years ago I did an application for a flying cutoff. If my memory suits me, we used an Allen-Bradley PLC5 processor and interfaced to a Siemens (I think) drive.

The PLC gave the drive simple input commands like "run" and "enable". The drive returned outputs of "in synch" (with the line) and "cycle complete".

The PLC also controlled the simple cutting commands once the cutoff carriage was in synch with the line.

The majority of the complexity of the system was at the drive end. There are drive people who can fill this application quite well "off the shelf". It is not really a tough PLC application.

If I remember, our line ran at about 200-300 FPM. The carriage weighed about 1,000 lbs. and the length of the cutoff area was 60 feet.

The mechanics are the really important factors. Mis-sizing a drive is suicidal...

 

[kamenges]

If there is one thing I have come to realize over the past few years it's that pushing things around accurately with hydraulics takes some knowledge. I used to have the overly simplistic view that hydraulics 'just worked'. Well, if you are just pushing something slowly between two fixed stops that may be true. But if you need to account for any dynamics at all it's not quite so straightforward.
Daile, I strongly recommend you talk to the guys at Delta. They do alot of hydraulics and you can be sure they will get it right.

Keith

 

[Peter Nachtwey]

That is why I fanatical about the math and physics.

The mechanics are the really important factors. Mis-sizing a drive is suicidal...

A poorly designed system will impossible to control. The motion controller, even ours, is just a tool. It is the mechanical/hydraulic design that makes or breaks the system.

The majority of the complexity of the system was at the drive end. There are drive people who can fill this application quite well "off the shelf". It is not really a tough PLC application.

Drives don't normally have the ability to synchronize axes like this but there are so many flying shears or flying cutoff saws out there that any good motion control company should have many of these and have examples from which to start.

From the motion controller point of view the trick is to gear the shear carriage to the feed chain at a one 1/1 ratio. The gear ratio starts at 0/1 so the slave is not moving until triggered by the master reaching a certain position. At this point the slave starts ramping up the gear ratio as a function of distance the master has moved. During this time the master will move exactly twice as far as the slave because slave is ramping up. For example, if the master and slave are to be synchronized at position 0 and the slave need a distance of 4 to ramp up, the slave must start at -4 and must start ramping from up as the master reaches -8. As the master moves from -8 to 0, the slave will ramp up from -4 to 0. Afer 0 is reached there is a small distance allowed for settling and then the shear fires. After the shear goes down and up the slave must race back to -4 so it will be ready for the next cut. More advance applications will always cut at the same position because the machine is mechanically reinforced at this point. This means slave's start point will change as a functon of master speed. The faster the master speed, the longer the ramp distance for the slave because one wants to keep the ramp rate constant and the shear point constant. If the master speed doubles the slave start point may need to be set back to -16 ( why ) and the point at master causes the slave to ramp up changes to -32.

Think about it.