Messages

1191

LazyTurn / Re: LazyTurn

« on: December 16, 2008, 01:21:11 PM »

Hi Guys:

  Heres a new version that takes into accoutn an overhang on the pullouts. Its also optomises some for speed considerations..

Art

1192

LazyTurn / Re: LazyTurn

« on: December 14, 2008, 01:46:28 PM »

Rich:

  Roughing as a step can leave larger chunks than onw wouel eant to ake in a finsh pass, so ID define "Fine-Rough" as a step that reduces the chucks to a size managable by a finish tool. ( typically the same tool as the finish pass woudl be generated with. Basically iits a safety pass to clean up the profile so that a proper finish pass can be applied without hurting the finish tool. Any tool can be used for any pass, but by allowing a fine-rough pass, youd know that the same tool now can do a proper finish pass without marring the surface.

  A proper surface finish requires we dont rub the stock,nor chatter from overstress on the tool, but come as close to the "finishdepth" of material removed on the entire pass, this will eliminate chatter and rub problems in the final finish as the finish tools stress is held as close as possible to a static condition during that final pass.

 Thats wht I planned for anyway.

  As to the fact that multiple rough passes may nbe necessary, that too is already int he code, but its completion requires the fine-rough stage to be a valid one as its inputs conditions will be used by the second or third rough pass. Its input conditions are calculated by the first rough pass. SO I need to progress a bit further before you begin to see secondary roughing. In all cases the tools will be unrestricted and any can be used on any stage.


  Im working now of some optimisations that Ive left for a long time, so Im hoping next version will be much faster than previosu version, and Ive also added auto-stitch that fixes some drawing where end's done meet to starts or overlaps exist. The program now tries to fix such things on the fly for even more file loads.


Art


Art
 

1193

LazyTurn / Re: LazyTurn

« on: December 14, 2008, 10:52:16 AM »

Willem:


 It should.. and Im still looking into a better algorithm to determine that.. Not as easy as it sounds to make that right on all profiles.. but Im getting close..



Art

1194

LazyTurn / Re: LazyTurn

« on: December 13, 2008, 08:35:34 AM »

Willem:

   No, when its done, you will just select the three operations, one by one, and when posting the program will take care of making the entire GCode file. At the moment all you have is roughing, so you can only do th efirst stage...

Art

1195

LazyTurn / Re: LazyTurn

« on: December 12, 2008, 05:05:02 PM »

Willem:
 
  Only as 1 tool for roughing, one for finish rough and one for finish..

Art

1196

LazyTurn / Re: LazyTurn

« on: December 11, 2008, 11:34:25 PM »

My thought on sectioning is that after a profile is generated ( the math here relies on generating the entire profile..) Id like to simply drag a slider that selcts a section, but that will be a posting section item, you select a section ( or the whole profile ) and then when it posts the code, it does so only for that section. It would then allow for a job to be cut in sections in any order you wish, graphiclly , without much hardhsip on the user... ( just a thought..)

Art

1197

LazyTurn / Re: LazyTurn

« on: December 10, 2008, 06:05:10 PM »

Hi:

  For all intents and purposes there is no real good reason, a radius of .0005 is effectively 0.0 in real world terms. It means an error tolerance of .001 ( inch or mm) in a
profile, so it has no real effect. Internally however, I just have to ensure low values are treated fairly for the sake of stability.
  Questions like this one go to stability in the end, and the theory of "If they CAN break it, they WILL" , which is somethign I have learned over the past
few years. It was my decision early on in LTurn to allow for this, and to profit from it by allowing it to be broken, shut off each avenue as I can, and then
I wont have to worry so much about support. In Mach3 I quite often use the axiom that "If it hurts to laugh,..dont laugh.." , but in LazyTurn I want people
to go ahead and break it if they like, if I can shut off that door, I will, and hopefully we'l have much less support questions from things that shouldnt be done, because internally
the program will correct whats wrong before you see it.

One of the reasons for example that more files now load, is that I change them as they come in , or just tell
you that they wont load. The program now "AutoTolerates" so that lines that dont join, automatically stitch
together as long as it doesnt create discontinuities in the math functions. This means the program will try a
 tolerance of .1 for example, and if it finds that this creates a conflict, it will jump back and set the toleance
to .01 and try again, then .001 etc.. until it either solves the problem, or rejects the file.

  By the time LazyTurn is complete, I will be very well armed to tackle some of these type of issues in LazyCam
which suffers greatly from tolerance issues. These sort of notifications therefore go toward educating me on the issues,
and even if they sound kinda crazy at times, they actuall do have their place in the scheme of things.

 :-)
  
Art

1198

LazyTurn / Re: LazyTurn

« on: December 10, 2008, 03:36:02 PM »

Thx guys..

  Its only because of your testing it is where it is. Ill work on making 0 the minimum tip radius ( sharp point in other words..)

Chip:  Not a bad idea at all, Ill keep that in mind when I look to inside turning.. be awhile though. Im being a bit picky on this one to make sure
all works as well as I can.. Your toolpaths are looking pretty good, and your right, with a very small pass depth, just cleanup woudl pretty much make the piece, but it'd be a bit nasty of a finish. Whats required I think is a rough to .1 then a .1 finish pass. But that too is likely to be awhile. When Im finished cleaning up what we have the next step of "rough/finish" is going to be a bugger mathmatically..

-- Ill check the other diamonds, I thought I caught them too, but maybe not.

Still not sure why the roughing screwup on that inside chase.. but Im sure its related to the undercut..


( I notice this thread seems to be getting popular.. must be allot more turners out there than ever before. )

Thx,
Art

1199

LazyTurn / Re: LazyTurn

« on: December 09, 2008, 11:44:04 PM »

Thanks Rich:

  Ill work on it in the morning.. looks better, just still a bit touched when doign undercuts. ( thats why somne programs dont allow undercuts..

 Not possibel to show in different colour.. 3d modeling is a tough art..

Art

1200

LazyTurn / Re: LazyTurn

« on: December 09, 2008, 07:10:12 PM »

Tim:

  No real restriction, it just has to have a shape, and the shape must be real. For example, an incribed circle of 1mm cannot have tip radii of 5mm, the radisu is larger than the size of the tool itself. Inscribed circle tells the program how large the tool is, the tp radius really shoudl be able to be set to zero, a sharp point on the tip, that will happen soon. An inscribed circle is , for those unaware, the largest circle that can be placed on the tool without extending off the edges. Tip radisu is the roundness of the pointy parts..

  SO the only real concern in the end is that I dont allow you to specify mathmatically impossible shapes. I hope to allow you to draw a shape in the end.. I wrote the collision detection to take into accoutn any shape at all, not matter how complex, though it will slow the algorithms that check the gouge. Slowness will be the main issue, though I havent as yet tried to optimise speed at all, Im more worried about accuracy and safety to be frank, Id rather 10 guilty men go free, than one innocent man go to prison, or Id rather have too much material remain, than too much taken away. Easy to fix one.. hard to fix the other ..

Art