GT Release 2.2Dev

[Ken_Shea]

Art,
That was it, flute length and I can't tell you if it was a change I made or something in the conversion calculation or lack of.

Good news is that should help you narrow down the culprit even if the culprit was me :)

Ken

[John S]

Art,
Thanks,
I'll put up with the lag if I can grow a longer tool.  ;D

[ArtF]

lol... I may not be able to handle that for you.. thats for rocket scientists...

Art

[ArtF]

Hi All:

Version 2.23 is now online with corrections to the tool import and export. ( Tools now export as text files you can edit, you can also import them. These are an IO type of file,
the program maintains its own database format, but allows you to export a tool library as backup, or import one. If duplicates are found in naming , then the names are updated
with a copy number. So its always safe to import any tool table.

  Taped and VBit tools are now profiling according to their class.. so be carefull of using tapers as the profiling will slowly move away from profile walls as the bit
moves deeper. This may sound wrong or seem wrong. My line of thought is that a vbit, is normally always under load on both tool flanks, but by shifting the path
to account for its taper on each pass , its load is shifted to a better position. It may be easier to handle the load on deeper profiles... but then I could be wrong,
it just seemed more elegant a solution which usually tells me its better.. :), I believe the load is actually identical per pass, but this should allow deeper total cuts.
 
  I include a photo to show what I mean , the picture shows 3 passes with a VBit.. the old way and the new one..

Anyway, things are getting more stable, the toolbox should be easier on you CPU load as I reduced the resolution of the tools..

Art

[Mooselake]

Not very important, but the current development link on the download page says 2.1xx.

Kirk

[ArtF]

thx.. fixed..

[John S]

Art,
Had a play and very nice, I like the way it's going.
Two questions:-

No matter how many times I watched the video, even used matchsticks under the eyelids to stop the blinking, how did you get from running the post to getting the code in notepad on screen ?

Second question, in the example you used a gear generated in GT will it be possible to bring a dxf in of a part not produced in GT but use GT's post to produce code ? Or is that for the CAD section later ?

PS Working on the Rocket Scientist but its an uphill struggle.  ;)

[Wod]

Art,
    I use Gearotic on a number of different machines, so I set my workbench Zero to the center of the grid. Is there an easy way to place a gear at 0,0 or do I just have to drag it around carefully and watch the numbers on the coordinates boxes?
Thanks,
Wiley

[ArtF]

Hi John:

To get the code in notebook, I paused the recording then opened the file. :) lol ( But I will add a "open file for display button shortly.)

You will be able to bring in anything to post code eventually, part of the CAD section is to make it bring in complete DXF's.
GT can already bring in DXF's, if one looks youll notice all the spokes and indicators are dxf's. In fact a user could for a long time design
their own spokes and indicators if they follow simple rules that I used when making my own.  If a spoke or indicator is added to their
respective folder, they will automagically appear in GT.. BUT, dxf's come in many flavours and GT will only load the ones Im using from vectric which
are a small subset of dxf's.
  My plans for GT's development have sharpened this summer considerably, I wondered for some time if the world needed yet another CAD
program, I originally just wanted gears for myself and others. After long discussions with my dog Buddy, in our many walk and talks....Bud and I have
stumbled across an idea for CAD that is both unique and interesting from a programming perspective that no-one has yet done well (or at all) to my mind.
Ill go into it when I start to release tests of it ( just in case its in actual fact impossible for one at my level of coding..)

  I cant get into CAD though until I fulfill an obligation to redo the 4th axis interface which to my mind is currently dismal..so its the next job. Then I must add
Bevels to both 3d and 4th axis. After analysing the bevel situation, Im of the mind that bevels less than 45 degrees should be 3d posted, and more
than 45 degrees shoudl be done on a 4th axis ( or 3d but better on 4th.). SO my aim now is get 3d working well, then 4th axis, then bevels, then ..finally
a CAD module. Bevels then will be able to be made on a standard CNC with no 4th axis if required.

  Needless to say, this will all take time. Youll see 4th axis this season, and probably bevels as well. I suspect highly CAD will be the result of my playing
in the 2015 offseason just as this years WorkBench was actually written last spring and summer, though its maturing fast now that its released. Bugs of course
will be removed as we go.

  Sorry for the verbosity.. Im long winded today, but figured an expanation of where Im headed was in order.... :)

Art

[ArtF]

Wiley:

  Ahh..so YOUR the one.. lol

I knew that eventually someone would use other than bottom left. SO I was waiting for questions that relate to such usage.
You've made an excellent point, so I will make it so any single gear will be sent to a position of  0,0 if the system is set to a centered
origin. For multiple gears though, I will probably then have 2nd, 3rd..etc..enter the table as they currently do..

Once I do this, try it out and give me feedback on how that works for you.

Art

[ArtF]

Wiley:

  Done for next version. Ive implemented it as follows.

IF your origin is set to center of CNC table..when gears are sent to the bench it will check to see if only one object was sent.
If that is true AND the bench was previously empty, then the one object now on the bench will be placed at a center of 0,0,0.
Otherwise objects will be placed using the pseudo intelligence the program uses by default...

Let me know how it works out for you after next update.. probably later in the week..

Art

[ArtF]

John:

  As of next version a checkbox is available that opens the posted file in whatever program you have set for opening such
programs..

Art

[Wod]

Art,
    Sorry to be 'THE ONE', but I do appreciate your quick response.
Wiley

[ArtF]

Hi All

  Figured Id post a teaser photo of the new 4th axis module, hopefully not long from release..

Art

[ArtF]

Hi Guys:

  Thought Id mention.. In looking at the photo in the previous post, it can be hard for many to envision what the toolpaths mean, what they are showing..

  This is because when displaying a 4th axis path, you have to see it in 4 dimensions, not 3 as in Workbench which is easy to envision. Each dot in the toolpath
is the position of the bottom tip of the tool , so one thing you can be sure of when viewing not only the photo in the previous post, but the 4th axis in the old module
is that if the gear can be seen to have no toolpath dots in it, then the gear itself ( screaming to be released from the blank ) is not chewed or gouged by the tool
in the running of such a path. This, of course assumes proper centering and proper A axis rotation direction. Since the positioning of the 4th axis in relation to the X and Y axis are all important parts of the job, I can never be sure  (nor can you) that the code does as youd think when you run it unless you can clearly visualize what its doing.

  In the next release you will be able to simulate the code so you can see what to expect. This will run much like a movie with the A axis ( or whatever letter you use) rotating,
and the tool moving in X, Y and Z as it would when cutting. You will be able to stop or slow down the tool , rotate the screen to look at the tool vs the gear to satisfy yourself that no gouging is occurring.

  Since every tooth is the same you should only have to inspect the simulation of one tooth. However, since you can , if you wish, use different tools for rooting and tangential and other steps, the system CANl do all rooting before doing tangential ( for example), but only if different tools are used, if the same tool is used then rooting will be followed by tangential on each tooth one by one. ( To make the job as fast as it can be..).

  We'll work together as you ( and I ) get used to the new way of posting to see if we can make it even more efficient, but Im hoping this simulation capability will help
you visualize the process and more easily point out any troubles we may face. So.. we'll see how it all works out in a week or so. :)

Thx
Art