Anyone doing basic type programming

[mcmmach]

Hi, I am looking for someone that has some quickbasic or other type of basic programming experience to develop a program to modify basic text files.

The files can be very large, sometimes with 300,000 plus lines of code and 7+ megabytes in size.

The lines are always as follows, X00.000 Y00.000 Z00.000 (replace zeros with XYZ coordinate numbers)

I want to be able to add a feedrate to the line that changes depending on the depth of the Z cut.

The variables section would specify if Z is between 0 and 0.020 the feed rate = F150.0, if Z is between .021 and .040 feed rate = F120.0, etc with maybe 10 different feed rates that can be modified easily

Each line would change to X00.000 Y00.000 Z00.000 F000.0 from the above sample.

If anyone thinks they can do this please contact me so we can discuss the project and the terms.
Thanks, Dean
dean@mcmmachine.com

[Evodyne]

Dean,

Hi! I've been doing serious Windows programming for several years now and what you'd like done would be a snap. If you can email me some test files I'll send you back a test program and some performance benchmarks. You can play with it and, if it meets your needs, we can go from there as far as adding bells and whistles. I'm not interested in any money-I do it because it's fun.

If you send along the files tomorrow I'll have a test program back to you by the end of the week. I don't have any size restrictions on my downloads, but it would be best if you zipped all of the files together. Send 'em to me, Lance, at lance@evodyne.net. I look forward to hearing from you!

Lance
Evodyne Soft Systems

[JRoque]

Lance, way to go! Good to see folks helping out and not always looking for a kill.

Dean, very interesting idea. I suggested something similar to the makers of VisualMill. In my case, I thought the hardest part of any machine operation was to cut down into the material so having a way to control feedrate of Z would be great. I can cut horizontally at a much faster rate that I can vertically so having control over "plunge" rate would really speed up things.

JR

[Death Adder]

I'm trying to drum up source code to put together into a CNC CAD/CAM package of open source code. If you'd be interested this might be something that could go into the package. Check out the open source forum here to see what I'm up to.

If you aren't interested that's cool too.

Originally Posted by Evodyne Dean, Hi! I've been doing serious Windows programming for several years now and what you'd like done would be a snap. If you can email me some test files I'll send you back a test program and some performance benchmarks. You can play with it and, if it meets your needs, we can go from there as far as adding bells and whistles. I'm not interested in any money-I do it because it's fun. If you send along the files tomorrow I'll have a test program back to you by the end of the week. I don't have any size restrictions on my downloads, but it would be best if you zipped all of the files together. Send 'em to me, Lance, at lance@evodyne.net. I look forward to hearing from you! Lance Evodyne Soft Systems

[Scott_bob]

Originally Posted by mcmmach Hi, I am looking for someone that has some quickbasic or other type of basic programming experience to develop a program to modify basic text files.[/email]

Dean,

If this project can be done by searching and replacing in your editer, then you could do this yourself...

http://www.cnczone.com/forums/showthread.php?t=11136

It's great to see someone offering to do this for you, but if you want to tackel doing it yourself, this would be a good project to try...

Good luck,

[Evodyne]

Hi guys! Dean has been kind enough to forward a few files to me and I've already dug in. Death Adder-I haven't taken the time to dig through the open source forum before, but I am going to this evening. I'm curious to see where you are going. And Scott_bob, your right, these are .txt files and one could make the changes in an editor, but why? We could all take chisels and make the same things by hand as we do with our CNC machines, but the machines do it faster and more accurately. Tools, be they hardware or software, are there to make our lives easier and improve our productivity, right?

Lance

[MetLHead]

Lance,

I know you probably have your own programming style, but I thought I'd throw this out there. I uploaded a VB template for a program that I wrote for converting files:

http://www.cnczone.com/forums/showthread.php?t=11323

Don't know if it will be useful to you, but this is just the kind of thing I wrote it for.

Regards,
Scott

[mcmmach]

Originally Posted by JRoque Lance, way to go! Good to see folks helping out and not always looking for a kill. Dean, very interesting idea. I suggested something similar to the makers of VisualMill. In my case, I thought the hardest part of any machine operation was to cut down into the material so having a way to control feedrate of Z would be great. I can cut horizontally at a much faster rate that I can vertically so having control over "plunge" rate would really speed up things. JR

The main thing reason I need this is I have been cutting deep lithophanes and must slow down to @35 ipm or the plastic melts, on the upper layers I could do 150ipm+ but my current conversion software only allows 1 cutting speed so cutting a 8x10 is taking 4-7 hours.

[Death Adder]

Originally Posted by mcmmach The main thing reason I need this is I have been cutting deep lithophanes and must slow down to @35 ipm or the plastic melts, on the upper layers I could do 150ipm+ but my current conversion software only allows 1 cutting speed so cutting a 8x10 is taking 4-7 hours.

What if the program you wanted took a maximum feed rate and a minimum feedrate and interpolated between them over the Z range of your toolpath? That way the cutter would seem to gradually speed up and slow down as it cut and you'd only have to set 2 parameters. It should still be configurable enough.

I think the above would even make a neat addition to my collection of open source cad related programs. (whether I write it or someone else does.)

[Scott_bob]

It may have been easy to misunderstand my post...

What I am saying is that you could create an automation routine (using the software that my link outlines) yourself. I don't mean that you would manually do the conversion every time. That would be archaic (hammer & chisel). Software automation is not that difficult if you choose the right software. If you already know how to use visual basic or some other structured language then that may be your best route. But if you don't then this program I mentioned (in the link) is easy to use, cheap to purchase, and powerful in capability. It will not be as fast as the visual basic solution, but it will be far easier to create, and tweak afterward.

The learning curve on this macro program is far shorter than a software language. And you'd be surprised at what can be done with this little program. I once had to covert 3,000 Autocad files by opening, ansewering some various dialogues, save, close, open the next file in the directory. All this was done automatically, in a matter of hours with only an occational input from me.

Software automation is a key to accuracy and productivity, just like CNC has been to the machining industry.

[Evodyne]

Hi all.

Scott_bob, I feel I owe you an apology: reading back over your post and mine I sounded a little pissy. I definately will look into the macro link you posted-it might be just the ticket for a lot of us. I certainly didn't mean to come off as I did. Sorry.

Death Adder-you and I had the same thought! The only limitation with a max and a min is you'ld end up with a linear change in feed rate. Which might be O.K.

My thought was add two more points and define a Bezier curve. Along with the curve, the user can define how many breakpoints to have. Zero breakpoints means the curve is used directly to set the feedrate given a depth. With ten breakpoints the curve will be broken into ten feedrate regions, etc. For anything other than zero, the user could "tweak" the results in a tabular format.

Another point: consider I am at a shallow depth at a given feedrate and the next instuction puts me much deeper at the appropriate slower rate. It seems to me I'd want to slow to the lower feed rate before plunging, not once I reach that depth. This would work towards preventing the melting Dean mentions.

Conversely, the opposite seems to apply when going shallower from a deep cut: stay at the slower feed rate until already at the shallower depth, then speed up.

Does this make sense? A kind of feed forward. Of course this would only matter if one was making large steps with big changes in feedrate.

Anyhow, I'm having fun. Others may wish to take up the challenge as well and give Dean some choices.

Gotta go for now-I am at work and not working!

Lance

[Death Adder]

Originally Posted by Evodyne Hi all. Death Adder-you and I had the same thought! The only limitation with a max and a min is you'ld end up with a linear change in feed rate. Which might be O.K. My thought was add two more points and define a Bezier curve. Along with the curve, the user can define how many breakpoints to have. Zero breakpoints means the curve is used directly to set the feedrate given a depth. With ten breakpoints the curve will be broken into ten feedrate regions, etc. For anything other than zero, the user could "tweak" the results in a tabular format.

To each his own but I think that you are making it more complicated than it has to be. This is a big problem in most cad apps.

I think that feedrate regions would only complicate something that could be done more simply. I do like the bezier curve idea though. There might be applications where the feedrate regions idea would be necessary so maybe it's an option for those that need it.

Originally Posted by Evodyne Another point: consider I am at a shallow depth at a given feedrate and the next instuction puts me much deeper at the appropriate slower rate. It seems to me I'd want to slow to the lower feed rate before plunging, not once I reach that depth. This would work towards preventing the melting Dean mentions.

Yes, consider the following sequence:
N2X34.00Y23.00Z1.00
N4X40.34Y16.52Z-2.43

Now consider what it would look like if it had feedrate control:
N2X34.00Y23.00Z1.00F150
N4X40.34Y16.52Z-2.43F34

This would move the machine to <34, 23> at Z1.0 (probably above the block). The move would be at F150 because feeds take effect right away. Then it would travel to <40.34,16.52> ramping down to -2.43 at a feed of 34 (once again because feeds take effect right away)

Originally Posted by Evodyne Conversely, the opposite seems to apply when going shallower from a deep cut: stay at the slower feed rate until already at the shallower depth, then speed up.

Reversing my above example:

N2X40.34Y16.52Z-2.43F34
N4X34.00Y23.00Z1.00F150

This is bad because it will feed up from a low Z level to a high one at high speed. So obviously a lookahead would have to be employed so that either positive or negative ramps are slow speed as are deep cuts but shallower cuts are progressively faster.