PDA

View Full Version : Cutting w/an Engraver



Phil Wolf
06-10-2008, 11:50 AM
I'm new to the world of lasers, so I thought that I might pose a question for all of you out there that have much more experience than I do.

I wish to fixture some foam shapes and cut them out. The problem that I'm having trouble with is that these shapes aren't flat, rather they are shaped similar to a "Pringle" potato chip. I've had discussion with several companies, some of which don't think that it will be an issue and other's that are telling me that it won't work.

I know that I'm somehow going to have to program the Z-axis, continuously alter the focal point, or extend the focal length (if that's the correct term) in order to allow the laser to cut along the contour without moving anything else.

I've attached some photo's for reference.

Any thoughts, ideas, input?

Phil

David Fairfield
06-10-2008, 12:20 PM
Hi Phil

I don't know of anything in the driver that allows the laser to change the focus on the fly.

However, the laser doesnt always need to be in precise focus to cut. Results will vary depending on the material.

Speaking of materials, looks like you have vac formed styrene. Styrene doesn't cut very well in a laser. You can do it, but it tends to melt and re-heal so you wont get crisp cuts as you do with wood. And if you're going in and out of focus, I expect the styrene problems will be greater.

You may want to consider a process where the material is cut flat first, then formed?

Dave

Phil Wolf
06-10-2008, 12:35 PM
Perhaps I should be more specific in the materials department. We are using a 4# polyethylene foam which is formed in a mold after being heated. aka, we put a foam sheet into an oven and let it warm up, then place it into a mold, and then squeeze it with a press. The foam then retains the shape of the mold once it has cooled.

David Fairfield
06-10-2008, 12:47 PM
ok gotcha. In my experience, foam sheet cuts very well. I've used it to make animal cracker shapes for an artist and it cuts crisply. Couldn't tell you the exact chemical name for the type I used, just your generic craft store colored foam sheets.

If you've gotten thumbs up from laser sales departments on cutting the 3d object you propose, I'd call back and ask if they'd be willing to run a trial sample for you. Best if you could be present while they do it. My sales guy let me test a bunch of stuff before I bought.

Dave

Phil Wolf
06-10-2008, 12:53 PM
I've sent out some samples for various companies to cut out for me, I'm just awaiting the return of them. I would love to test some out myself though. Do you know of anyone in WA that sells these, it seems that just about every company that I talk to is located in the southwest.

Gary Hair
06-10-2008, 1:15 PM
I would think a cnc router would be the best choice for this. No problem with X, Y and Z control.

Gary

David Fairfield
06-10-2008, 1:29 PM
Or how about using a die around the perimeter of the thermoform press? That way its formed and stamped out at the same time. It wouldn't have to be a hard material like steel since you're only cutting foam. High tech may be overkill for this application.

Darren Null
06-10-2008, 1:56 PM
Most of the lasers we have on here are fixed z-axis ones. Well sort of- the bed normally moves up and down to focus, but there's no provision for continuous adjustment on the z-axis except manually.

Robotty 3D machines do exist, but from what I gather about the cost of them, time on them would be quite costly.

You could -having cast your objects- cut them on a standard laser if you could hold them flat for the duration of the cutting and allow them to sproing back into shape afterwards. There are problems with this approach- 1) Successful re-sproinging of your shape; 2) You'd have to make your cast/s in EXACTLY the same place on the sheet every time (because you'd have to make a custom thingy to hold the sheet flat) and 3) There would be a very high chance of cut-and-sproinged bits getting caught in the head and messing up subsequent cuts...I'm assuming that this is going to be a volume job with multiple 'pringles' per sheet.

By far and away the easiest method of approaching the problem would be to either build the cutting device into the moulds themselves.

1)
TOP -v-----------------v-
BOTTOM -^-----------------^-
(Obviously they'd line up on the real thing. ASCII engineering drawings lack accuracy)

2)
TOP -|-----------------|-
BOTTOM -v-----------------v-
(The '|' indicates blades. Probably, with a bit of thought, they could be made retractable, so you could just push down on the top of the mould when the 'pringle' has set)

...or get a custom cutting tool made to slide down the outside of the mould and cut the 'pringle before opening the mould. I'm thinking a sort of ring-shaped razorblade mounted on a yoke big enough to fit over the top of the mould and a handle on top. You could always mount the tool into an array of them for doing whole sheets.

Both the custom cutting tool and 'build the cutting device into the mould' would require some initial expense, but it'd cut out long-term expense and you can do it in-house. Hope that's helped.

Doug Griffith
06-10-2008, 2:04 PM
I'd have a deep steel rule die made the shape of the oval form but without the inner core. I'd then slide it over the upper form and cut down through the substrate. It would be very easy and cheap to do.

Darren Null
06-10-2008, 2:20 PM
Yeah! That sounds better than ring-shaped razorblade thingy. Gonna have to bone up on engineering terms. Substrate. heh. 10 points.

Rodne Gold
06-10-2008, 2:20 PM
You could do it if you had a long focal length lens as the length of the cutting section of the beam will allow cutting all around the countour without having to fiddle with the X axis , an 8" lens would probably do as it will have a very large depth of field but will also have a large spot size.

Phil Wolf
06-10-2008, 3:18 PM
Wow, Thanks for all the input.

To address some things which were brought up...

CNC Mill: Not really the route we want to go since the foam isn't rigid at all and will vibrate and move around leaving a frilly edge. (We need a crisp clean edge)

Steel Rule Die: This is a path that we went down earlier, and why we are looking into cutting them out with lasers. I think the idea has merit, however, the powers that be above me in the company feel that there is a better way to do it and would like other avenue's pursued.

8" lens: This is intriguing to me, by "spot size" are you referring to the kerf? and if so, what is "large"? Also, where would I find one of these magical lenses that increase the focal length?


I also have one other question which I've been hesitant to ask as I'm sure it will spark all kinds of answers, which will probably not address my main concern of whether this is possible or not but I would like to know, and that is... What brand, kind (CO2, YAG, etc), and wattage would you recommend if I were to pursue this further?

Rodne Gold
06-10-2008, 3:39 PM
Well , due to the nature of foam and its propensity to melt around the cut , you might get a kerf as large as 1mm , but you can adjust for this.
You probably need a minimal amt of power for foam cutting , 25w at most
You would get much quicker results with a Galvo Co2 laser than an xY type machine . it steers the beam with galvo mounted mirrors rather than having a flying head.
I dont recon you would have to actually buy a enclosed type plotter laser engraver we use , a Galvo head laser unit will do , unless you want to use it for other stuff.

Doug Griffith
06-10-2008, 3:56 PM
If you're looking for different routes how about:

3-axis CNC with knife head ala vinyl cutter. Along with a matching contoured backer from Delrin or the like.

Modified 3 axis CNC with foam cutting wire ala wire EDM.

Matched steel die.

The first thing you should do is test flat pieces of your "substrate" in conventional laser cutters.

Sandra Force
06-10-2008, 3:58 PM
Why not cut it to the shape you want and then form it. Works on some of our projects with ABS.

Richard Rumancik
06-10-2008, 5:27 PM
The laser engravers most people on this forum are using are not ideal for this application. Even the higher-end and open-bed units don't have program control of the z-axis. These lasers generally run from CorelDraw which is a 2D drawing program. Usually robotic lasers are used to trim these kind of shapes.

The idea of using a long focal length lens might work, but it seems to me that you would be greatly out of focus in some areas. If you use this method you would probably set your focus point to the "average" height of the saddle, but that still leaves considerable distance to the upper and lower points of the part. It will have poor cutting quality at these locations.

Perhaps this could be improved by strapping the workpiece onto a rotary. You may have to modify or build a special rotary to make this work, but it would offer a few advantages. If I am visualizing this correctly it would reduce the magnitude of the focus error (up to 50%), and it would allow you to aim at the part surface in a more perpendicular manner for a better cut. I fully realize this part is not cylindrical but nor is it flat. If you built a rotary fixture mandrel with a "saddle" and some clamps to hold the uncut sheet (physical clamps on the outside and maybe magnets inside?) then maybe it would work well enough. You need enough "swing" clearance to rotate the sheet around 180 degrees or so.

You may need to leave a couple tabs on the part, or plan to quickly move the carriage out of the way when the scrap comes free.

To generate the required vector file, you may have to do a bit of "reverse engineering", trial and error iteration, etc. Not sure if you could your your CAD program to generate the file. Start with an approximate vector file and adjust from there. It is only needed to be done once.

You need to do enough experimentation to convince yourself that one method or another can be made to work successfully. (Otherwise, you will have to put your laser up for sale cheap. Contact me first . . . )

Doug Griffith
06-11-2008, 10:11 AM
One more option that may work:
A 3-axis CNC with a Synrad laser head (http://www.shopsabre.com/Laser%20Page.html). Most gantry style CNCs could be fitted with one. You could build a 3' x 3' system for under 30k with vacuum hold down.


This would allow control of the Z-axis.

Richard Rumancik
06-11-2008, 12:19 PM
Interesting machine, Doug. With the older Synrad tubes it was pretty difficult to mount them vertically like this on the z-axis as they were much longer. (Usually you had to mount horizontally like the Kern laser machines on the gantry and then use a mirror.) So on the ShopSabre, the laser beam is directed up, then bounced through the black tube with 2 mirrors? What is the white box on the front of the tube at the bottom end?

I still wonder how well the foam would cut with the beam hitting the part at a 45 degree angle at the base of the saddle.

Doug Griffith
06-11-2008, 12:44 PM
It would be great if the white box is for optical registration (but I doubt it).

I think the 45 degree angle cutting is required unless a 5-axis machine is used. This would be expensive and not efficient for large runs. Parts can't be ganged close together because of the swing of the 4 & 5 axis cutter head.

re: Why not cut it to the shape you want and then form it.
Probably due to registration issues. It may be a better idea to build a fancy forming tool that takes care of this.

Some notes:
1) Air assist appears to be missing but could be easilly added.
2) fume evacuation is also missing.
3) This type of setup would only work for cutting with no engraving.
4) Currently owned CNCs could "easilly" be retrofitted.
5) 3-axis programming has a learning curve. 5-axis is gnarly.

Cheers

Phil Wolf
06-12-2008, 12:29 PM
First off Doug, I love the fact that you used the word "gnarly" haven't heard that one in some time. Oh yeah and thanks for the link to the shop sabre site, I'll have to look into that more.

I think that a 5-axis machine is completely overkill. There's just no sense paying for capabilities that we won't ever need to use. That being said, what do you mean by a 45 deg cutting angle? In the case of the Shop Sabre, wouldn't the beam come straight out of the gantry so that it is perpendicular to the table?

I'm also not to keen on the rotary idea, not because I don't think it can be done, but because I would like to fixture more than one of these at a time. The end goal is to cut out roughly 400 of them per day. One at a time just won't cut it!

Thanks again for all your input,

Phil

Doug Griffith
06-13-2008, 10:38 AM
"Gnarly" kind of shows my age. I use "radical" too. And "sweet". Never "phat"

45 degrees is referring to the edge of the material as the laser hits it (ie. at the highest and lowest points).

It's seeming like a 3-axis machine with a Synrad is your best bet. I would invest in something like a MultiCam 1M and fit the laser head to it. This way you'd have a robust CNC machine with a fairly small footprint.

You may want to find out if ShopSabre's laser software supports the z-axis. I doubt it does. You may have to use routing software (ie. EnRoute) and activate the laser head with a relay triggered by spindle speed control.

Cheers

David Fairfield
06-13-2008, 11:30 AM
How about looking the other way, to a low tech solution? You could employ a minimum wage worker to trim waste material away with a knife, and smooth the edges with sandpaper. Its just foam and it beats shelling clams! :)

Dave

Scott Shepherd
06-13-2008, 11:50 AM
So the finished product looks like a Pringles potato chip?

Doug Griffith
06-13-2008, 12:15 PM
It looks like a pringles chip to me.


How about looking the other way, to a low tech solution? You could employ a minimum wage worker to trim waste material away with a knife, and smooth the edges with sandpaper. Its just foam and it beats shelling clams! :)

A large vacuum forming company I work with does almost everything this way. Sometimes there is no other solution. Having employees has its own set of problems and expense.

If a high tech solution is used, a human is needed to load and unload anyways.

Phil Wolf
06-13-2008, 1:14 PM
Actually, employing people to cut these out with scissors is what's being done now. And there is a reason we're trying to eliminate it. We have a small army of workers cutting these things out which takes a lot of time and skill to get them to all be the same size and shape. Employing several people to do this is kind of a waste of resources. I don't know what their wages are, but you really must think of the big picture. Wage + Benefits + Insurance (Workers comp on people that cut things out with scissors all day isn't cheap, especially when they all develop carpal tunel syndrome) is probably in the $15-$20/hr area. Multiply that by 6 workers at 8 hours per day, 5 days per week... an investment in a machine doesn't seem to expensive, especially if we can then utilize the employees in other areas that really need the extra help.

Not to mention that as you can imagine this isn't the most exciting job and the employees needed, need to be dedicated to producing repeatable parts in terms of shape and size. That just probably isn't going to happen with some random person that you hire from a temp service, especially if you pay them minimum wage. As Chris Rock said, if somebody is paying you minimum wage they're basically saying "if I could pay you less, I would"

Doug Griffith
06-13-2008, 1:37 PM
I've had to deal with trimming items on compound curves quite a few times which is why I find this thread so interesting.

Instead of scissors, you could build a dedicated 3-axis pantograph cutter similar to a Deckle. It would be a tricky endeaver though.

If you built a matched steel die and used a pneumatic single station punch press, one employee could bang them out in probably 5 seconds each (load,punch,unload). No carpal tunnel and they would be perfect. This approach would be under 20k.

David Fairfield
06-14-2008, 9:06 AM
Yeah I hear you about employees. But to continue my trend of looking for the simplest solution....

Instead of a three axis machine, how about a 2 axis router, with a bit long enough that it does not need to move up and down the vertical axis to maintain contact with the part edges?

I have a router bit on my tool board that will make a 3" deep groove I'm sure there are longer ones.

If the part is inserted into a jig, the router would then only need to trace the perimeter of the "potato chip."

Basically its your laser cutter idea, but the laser beam is replaced with a router bit solving the depth problem. You could go with a manual router, or a CNC might save some labor. As a side benefit, compared to lasers, those CNC routers are damn cheap! :)

Dave

Doug Griffith
06-14-2008, 10:17 AM
I don't know of any 2-axis CNCs out there but I could be wrong.

A long cutting bit would work and remove the need to program the z-axis but has issues. Long thin bits will tend to "warble" and can't traverse the part as fast since they flex and break easier.

A CNC would allow many parts to be done at a time while a manual jig would be one at a time. The jig itself (where a hand held router has a bearing that slides within a slot) needs some area around the part. This means the parts would have to be spaced farther apart and require more material and forming processes.

Using a router instead of a laser (or punch) introduces a secondary operation of cleaning up the part. Routers are messy.

Router spindles cost less than Synrad laser heads but still aren't cheap. Any decent router has a real spindle (3-phase) and not a hand held router. Hand held routers mounted in CNCs are very loud and a sloppy way to go.

Cheers

David Fairfield
06-14-2008, 11:56 AM
hm, yeah ragged edges would probably be a problem with foam. Wood cuts really nicely.

OK how about a water jet cutter? I know they are used to cut metal, but I've also seen them in an industrial bakery application, to cut cleanly through cakes and brownies. Apparently, focal length is less of an issue.

Dave

Ron Moorehead
06-14-2008, 12:28 PM
Hi Phil,

I saw in one of your post a WA is that for Washington state? Has anyone offered to let you try it on their laser in WA. If not I am in Olympia, lets talk.

Doug Griffith
06-14-2008, 1:49 PM
OK how about a water jet cutter? I know they are used to cut metal, but I've also seen them in an industrial bakery application, to cut cleanly through cakes and brownies. Apparently, focal length is less of an issue.

Talk about messy. It takes more than water to do the cutting and the fine grit would get embedded in the foam. Plus the expense of the equipment is out the door.

Cheers

David Fairfield
06-14-2008, 2:11 PM
I don't think the abrasive is necessary for all water jet cutting applications, at least not when cutting baked goods and factory assembled sandwhiches. I guess these foam potato chips have the consistency of sliced bologna, so the grit won't be a factor.

Dave

Jack Harper
06-14-2008, 11:19 PM
Water jet would be expensive and there is a change in kerf over distance. With that said a water jet or an abrasive water jet would cut the material with no problem. My system runs at 55KSI three times the speed of sound, that's about the speed of a bullet. No abrasive and almost no water has a chance to stay on the product. In fact toilet paper, known for its absorbency, is often cut to width with water jets with almost no moisture left in the material. With all this said, the system would probably only be worth it if it was part of a more complex automated production line with custom fixturing.

Doug Griffith
06-15-2008, 10:03 AM
Jack, I'm currently having parts cut with an Omax 55100 equiped with a Tilt-A-Jet. The machines are awesome and the 3/8" steel I require cuts like butter. Your machine is even bigger and must be a great asset to your business.

David Fairfield
06-15-2008, 10:11 AM
Anyone seen any home built water jet cutters? I guess a low power water cutter could be made based on a pressure washer or a hydraulic piston? Amazing the sorts of CNC machines some of these DIY enthusiasts have made from scrap. Just thinking out loud :)

Jack Harper
06-15-2008, 10:34 AM
Jack, I'm currently having parts cut with an Omax 55100 equiped with a Tilt-A-Jet. The machines are awesome and the 3/8" steel I require cuts like butter. Your machine is even bigger and must be a great asset to your business.

It is great asset. I can cut up to six inch thick steel with my tilt-a-jet head. It complemts my laser nicley, allowing me to create even more unique custom items. I needed the larger 80" x 160" for my laser bed that is 72" x 120". Having one in the shop keeps the cost down to be market competitive on my products. Of course it also seves as a business all to itself. I just finished a 4 mile long cutting job, cutting 4' diameter fiberglass gaskets for large oil pipelines. The oreder was shiiped to Saudi Arabia.

Doug Griffith
06-15-2008, 10:45 AM
Anyone seen any home built water jet cutters? I guess a low power water cutter could be made based on a pressure washer or a hydraulic piston? Amazing the sorts of CNC machines some of these DIY enthusiasts have made from scrap. Just thinking out loud :)

I think a home made water jet would be comparable to a home made laser using a laser pointer. I would like to see one myself though. Maybe a portable gantry system that goes back and forth across ugly driveway oil stains or for removing paint from stucco.

Scott Shepherd
06-15-2008, 10:57 AM
Problem is cost justification for some of those ideas. We're only talking 400 parts per day. That's a tiny volume when put in some of those applications. My guess is you could run 400 of them in 30 minutes, which would mean you'd have a machine sitting idle the rest of the day.

One thing that hasn't been brought up is the complex shape. You can't take a shape like that and just "cut it out from above" without some serious math. For example, a sphere wouldn't be cut round if you were cutting it from a fixed focal plane. It would be a parabola,ellipse, or something similar. Done the math to generate a parabola? I have for CNC machines and trust me, I should have studied more in school.

I know some solid modeling software has the capability to project that object through a flat focal plane and generate the shape for you, but I personally don't have that software.

With that quantity, I'd bet you can have a die made for less than $3000 and be done with it. You could probably also crank them out in seconds per part, leaving a lot of free time to do other things.

It's just too small of a project for a huge expense, in my opinion.

Doug Griffith
06-15-2008, 12:45 PM
One thing that hasn't been brought up is the complex shape. You can't take a shape like that and just "cut it out from above" without some serious math.

I'm guessing a 3-axis CNC created the forming tool in the first place so the perimeter of the part is already known. If modeled by hand, good luck.


With that quantity, I'd bet you can have a die made for less than $3000 and be done with it. You could probably also crank them out in seconds per part, leaving a lot of free time to do other things.

I agree. A matched steel die with a pneumatic press. The investment would be more than 3k though. Die + die set + press.

Scott Shepherd
06-15-2008, 1:56 PM
I'm guessing a 3-axis CNC created the forming tool in the first place so the perimeter of the part is already known. If modeled by hand, good luck.

Yes, it was done using 3-axis movement, so the shape is very different than one that would have to be replicated with a 2 axis movement. With a "fixed Z", it changes the shape you'll have to cut for sure.

I used to program CNC machines, full 3 axis (actually up to 8 axis in total, but that's a different story), is very different than the 2 axis and the shapes have to be different to get the final product the same. I used to do a lot of that type stuff and used a lot of formulas to create the tool path in 2-axis to simulate a 3-axis shape.

Much easier to do today than it was then, but still not easy unless you have some powerful software.

David Fairfield
06-15-2008, 2:49 PM
the material is foam, so the die and die press don't have to be industrial grade. Those scrap booking punches that make little valentine hearts might cut through this material.

assuming water jet and punch and die is too expensive, another couple of low tech ideas-

how about a hot wire? A wire formed to the overhead profile of the potato chip, electrically heated and pressed down over the chip while it's still in the former? Sort of like a punch and die, except the die is a hot wire

Or how about a cam that rotates the chip in the former, so the edges meet a spinning grindstone in fixed position? Saw a WW2 film clip of a Garand rifle factory, that's how they mass produced the wooden rifle stocks.

:)