Lithophane

[Donald]

Lately I've been machining a few lithophanes out of corian. I've found that it gives a much better finish than white acrylic. but the cost of corian dictate the side of the finished product since i purchase all of my corian off of ebay. I can get almost the same look and effect out of the white acrylic, but the acrylic tends to get to hot and melt back on its self.
I'm using an engraving tool with a .03 radius with a stepover of .01. and i have compressed air blowing on the tool and acrylic while machining. My question is: Has anyone ever used a cold air gun in any fashion? to keep material cool or even the tool? i was wondering if they are worth the price.

Again I'm using a engraving tool with a .03 radius, and the feed on the x and y are 3" per sec. and 2" per sec on the z. I guess after proofreading my post i'm really asking has anyone found a way to keep the white acrylic from melting and still attemping to make it profitable?
Any input would be appreciated.

[CNCadmin]

I would think that liquid would provide the best cooling that air.

[Donald]

I think your right that the liquid would disapate the heat better. But I'm worried about the liquid mess or the mist, I have a dust collector also and think that with the cold air guns ( http://vortec.com ) would help keep the tool cool and the air would also help blow the chips away from the tooling allowing the dust collector remove the debree. Has anyone used or made anything like these cold air guns?

[kong]

Cool air guns should work, but you need a massive compressor to run them. Something like 8.5 SCFM for the small models I think. What about a micro-mister like the Trico micro-drop? No mess with that one. I will be trying the lithophanes on acrylic in the next few weeks with the Trico unit, if you can wait that long, I will let you know how I get on.

[voddebosk]

Donald.

I have bin doing some lithopanes,and i use also white acrilic.
but you have to get the cast acrilic but not the extruded acrilic.
I reduce the spindel speed to about 16000 rpm and go as fast as 6000mmpm.

good luck

arjen

[Evodyne]

Originally Posted by Donald Again I'm using a engraving tool with a .03 radius, and the feed on the x and y are 3" per sec. and 2" per sec on the z. I guess after proofreading my post i'm really asking has anyone found a way to keep the white acrylic from melting and still attemping to make it profitable? Any input would be appreciated.

Donald,

Hi. I just collaberated with a fellow doing the same thing. He had to slow the travel down to a crawl to keep from melting and it took forever to run a job. Like you, he ran at a fixed rate on the X & Y. I wrote him a small piece of software that reads his g-code files and adds a feedrate to each line whose value is dependant on the Z depth. He is able to define a curve in the software that relates depth to speed. The output of the software is a separate, updated g-code file. I still have a bug or two to work out, but his feedback to me was that he can run a job with no melting, a better finish, and it's finishing about 30% faster. This may be something you would consider trying.

Lance
Evodyne Soft Systems

[Donald]

Hi Lance,
I'd be interested in knowing more about the software you wrote.
I have been able to crop my images into any shape that i wish, but it still will machine those invisible corners of the file. How can one just machine the cropped image and not those invisible corners?

[mcmmach]

Hi, basically most of the finished product is determined by the picture you start out with, your graphic editing program is the most important. Sometimes the picture from your graphic program will look bad but will cut very good, most of the time minor details don't matter at all. Watch out for redeye and glint from the flash, even a small amount can cause a big hole in someones eye.
Dean

[Evodyne]

Originally Posted by Donald Hi Lance, I'd be interested in knowing more about the software you wrote. I have been able to crop my images into any shape that i wish, but it still will machine those invisible corners of the file. How can one just machine the cropped image and not those invisible corners?

Hi Donald. O.K, I need to let you know I've never made one of these myself-just wrote the software to help another fellow out. So I'm short on practical advice as far as the actual machining. BUT if you have correctly cropped the photo the stuff you cut out is gone and just can't be part of your output. So I have to think you are missing something basic.

First off, are you cropping the photo in a separate "paint" program, then converting it to g-code? Or does your CNC/lithophane software allow you to open a photo, edit it, then run it?. (See I told you I'm pretty ignorant about this stuff). Well, at any rate, I'd try cropping/editing in a separate paint program and make sure you save the modified picture to a different file name, just so you KNOW it's a different version than what you started with. You can check the file sizes of the original and the cropped-if you've done it right the cropped will be smaller as there is less data.

Incidently, what software do you fellows use to go from picture to g-code? I'm curious.

The guy I worked with ended up with g-code that ran at one travel rate no matter what the depth of the cut. He was fine at shallow depths, but had melting issues on deep cuts. I wrote a program that looked at each line of his source gcode, read the Z value, and used a curve to to get the appropriate feedrate. I added the feedrate to each line of the g-code and saved it to a new file. The curve I made user editable, so you can vary the feedrate in different ways. It's very simple and only does that one thing, but he said it made a noticeable difference for him.

I wouldn't mind sending you a copy, but first I'd like you to send me one of your g-code files so i can try it here first-just in case there are differences in his generated code and yours. I'd hate to cause wasted material or worse. It's up to you. You can send a file or two to me at lance@evodyne.net.

Lance

[Donald]

Lance,
I'd be glad to send a file. most of my are very very long. Approx 500,000 pluse lines, but I may have a smaller file... I'll look through my stuff this weekend and send it via you email. I have a friend that uses Artcam and I've been attempting to use his software to make my code, but have used a program called Profiler and also the tiff converter that came with the router.

I just re-read the past sevral posts, ---- The only time I have a problem with melting is in acrylic and I can adjust my feedrate for that. If I can find a way to keep the acrylic cool than I can run faster, possible even to match what I do in Corian which is between 5 and 6 in per sec. It does slow down when it changes its Z height, and thats where I loose time.
You mentioned,

"I wrote a program that looked at each line of his source gcode, read the Z value, and used a curve to to get the appropriate feedrate."
I'm not sure what you mean when you use a curve to get the appropriate feed rate.

Again I'll look for that file this weekend and try to send it sometime saturday night or sunday.

Thanks again Lance,

Don

[Evodyne]

Donald,

You may send any size file-I've got Comcast and it is very fast and I've no email size limits.

Essentially it seemed that to give the most flexibility that the feedrate should vary by the Z depth, but perhaps not linearly. I gave the program an editor screen with a graph of feedrate vs. Z depth. The user can drage a little cursor around and change the shape of the curve. See the attached shots. Remember, the software the gentleman was using did not allow for anything but one speed. If yours already does, this may not be useful. By the way, the program processed a 300,000+ line file in like 10-15 seconds on my PC.

Lance