Hello Guest it is March 18, 2024, 11:40:44 PM

Author Topic: LazyTurn  (Read 1359547 times)

0 Members and 1 Guest are viewing this topic.

Offline RICH

*
  • *
  •  7,427 7,427
    • View Profile
Re: LazyTurn
« Reply #830 on: February 06, 2009, 07:40:10 PM »
Hi ART,
I'll see what i can find.
RICH
Re: LazyTurn
« Reply #831 on: February 06, 2009, 07:53:16 PM »
Hi,
Here is the default.Can't seem to get, the so called errors as before..... 1 hour fooling around to Break It. Here's the default setting. 0.1 across the board. visual dia. Work's well.   Thanks,   Hank S.

Offline khalid

*
  •  213 213
    • View Profile
    • My Wood carving
Re: LazyTurn
« Reply #832 on: February 07, 2009, 09:49:56 AM »
Help me in setting up my Lathe..

In attached picture you see my machine setup..Its actually Milling machine with Indexer.. I will remove the Belt from the indexer gears and will install the Spindle motor to the protruded shaft... Now this will become the lathe..

As I generated the toolpath for one of the piece, I saw X and Z moves in G-code File.. Can some one give me an idea what will be the best arrangment for axis... My smaller axis is Y-axis and the larger one is X-axis..

Can you designate How I use it as a lathe? Following are some options:

1- My Z-moves (G-code One) as a machine Z-axis and the X-moves(G-code One) as the  machine X-axis
2- My Z-moves(Gcode One) as Machine Y-axis  and the X-moves as the Machine X-axis

Please some body give me an idea...:) It will be good for me  if somebody specify the G-code axes in the attached figure...

Thanks

My indexing work can be seen here.
http://www.cnczone.com/forums/showthread.php?t=39588&page=37

Offline ART

*
  • *
  •  1,702 1,702
  • Tough as soggy paper.
    • View Profile
Re: LazyTurn
« Reply #833 on: February 07, 2009, 10:15:24 AM »
Hi Guys:

 Lets me explain whats going on with the phot that was shown in that letter above.
The selection for visual diags was checked, and the # of passes was set to seom number less than the passes normally generated by the material swept volume calculator.

  As you can see in the drawing the stock has been chewed down each pass till the maxPassNumber was seen, calc'ing stoipped at that point. The photo, shown again here, shows a few things. The greenline is the amount of offset from the part used in the profile goungin checks, ( just a diagnostic for me.), the black line on top is the actual stock outline at that point, White dots indicate endpoints of the material stock outline to that point. The large red shape, is the swept tool volume of the last pass done. It is the shape of the tool , swept by the amount of that pass , and subtracted from the stock volume, it was successfull on that line as shown by the fact the tool shape is properly subtracted fromthe material.

  When you do a toolpath, if the end material is wrong, you will be able to find a particular pass where the shape of the tool is NOT subtracted properly, this can happen from a wrong insode/outside assumption at some point. If you do get a failed stock, try narrowing in on What pass it failed on. Try 100, if it fails, try 50, if that passes try 75, etc.. eventually youll find a point where the subtraction fails, that image is the interesting one to me.

  In the rough command dialog, the Max Depth number is not one your allowed to change, its simply a limitation, based on the tool, of how deep you can cut. Its read-only. A tool can only be used to cut to its inscribed circle depth, and the end max depth is the inscribed circle, but compensated by diamter or radius mode selection.

  This is actually a very good shot and shows success up to that point, If the user had selected a pass one number higher, then the photo woudl show the next line being subtracted..etc..

 Just thought Id explain the straneg images.. :)

Art



Offline RICH

*
  • *
  •  7,427 7,427
    • View Profile
Re: LazyTurn
« Reply #834 on: February 07, 2009, 10:45:04 AM »
Thanks for the explanation Art.
Hopefully i will  get to testing some later today.
Do you envision the diagnostic remaining in the program or is it just a temp thiing at this time?
RICH

Offline ART

*
  • *
  •  1,702 1,702
  • Tough as soggy paper.
    • View Profile
Re: LazyTurn
« Reply #835 on: February 07, 2009, 11:00:47 AM »
Hi Rich:

  I may leave it in, though I may hide it a bit in selection, its a handy way to see whats screwing up a swept volume calculation, and it may help in future with
diagnosing any strange problems, the SVC ( swept volume Calc , so I dont have to type so much in future :) ), is a very important subsystem and will be tied into allot
of the future calcs, its now about 98% working, so Ill need help in finding the last bugs in it to get it 100%, bu nothing is ever 100% so the diags will likely stay.
 Personal diagnostics that are only for me are already hidden and run only in debug compilations.

   SVC is known in the industry as being very hard to do , and its taken allot of research to get to the right point,which is why Ive been so quiet. I deleted dozens of versions over time, but this one is a keeper, if we can get rid of its pecadillos it will be a great way to find secondary roughing passes, and then, finally a proper finish pass.

  It seems to fail as the tool hits close to the center line, or passes it, but Im narrowing in on the failure. I believe its a failure to see inside vs outside properly in certain specific
line compbinations. Ive seen it fail so far mostly in areas where an intersection hits anothert intersection making it difficult to sense the proper winding of the tool. We'll see. :)

  All this has secondary application to real-time simulation as well perhaps as volumetrics being performed on a per pass basis, that woudl be very powerfull stuff if it can be applied properly.
Ive kept the fancy stuff out for now till it can be made bulletproof as I dont want to overly confuse issues involved in finding trouble.

Art

Offline ART

*
  • *
  •  1,702 1,702
  • Tough as soggy paper.
    • View Profile
Re: LazyTurn
« Reply #836 on: February 07, 2009, 12:24:30 PM »
Here's a newer version, shows volume of stock on tree, and volume removed per toolpath when the toolpath is truly sucessfull.

 Im narrowing on the failures and hope to find it soon.

Art

Offline RICH

*
  • *
  •  7,427 7,427
    • View Profile
Re: LazyTurn
« Reply #837 on: February 07, 2009, 01:42:20 PM »
Art,
Lets see if my understanding is correct. The attached shows three different senerios all with the same tool settings except for the numer of scans done.
_1_  -  RESULT OF 1 TO 5 SCANS
_2_  -  RESULT OF 6 SCANS
_3_  -  RESULT OF 9999 / SCANS

So how do i interpret?
 
RICH

Offline RICH

*
  • *
  •  7,427 7,427
    • View Profile
Re: LazyTurn
« Reply #838 on: February 07, 2009, 01:56:33 PM »
Ok, just tried the newer version. Same figures in last post #837 apply.
So now in the tree i see no volume was removed from the end volume.

Thus, until it shows that the tool volume was removed it still is a now go?
Bear with me, don't want to spin my wheels once i start going thru a whole bunch of files.

RICH

Offline ART

*
  • *
  •  1,702 1,702
  • Tough as soggy paper.
    • View Profile
Re: LazyTurn
« Reply #839 on: February 07, 2009, 02:24:22 PM »
Rich:

 1) I have to fix gouging tools, the dpeth is way too limited.. doesnt make sense using inscribed cicrle to limit that tools depth.

2) Your file is failing on load, no stock is being displayed, Ill look into that as well.

3) Your _1_ is working correct, the black box that IS the stock before begin revolved is shrinking properly pass by pass.

4) Your _2_ shows proper as well, as that line is correctly showing being subtracted from the black box to create a new stock shape.

5) this one is hosed, see how the balck box is now just a line shape on the left, soemwhere between pass 6 and the final pass , whatever it is, the balck box was destroyed by a filed subtraction in the SVC system.  Youd have to expeiment to find where that occured.

 Your main troubel is that no stock soild is being generated on any pass formthe solid, I suspect this is an OpenGL issue on your system, so Ill tighten the rules to see if I can find out why.

Art