Plane question

[So not a Guru]

We do a lot of 5-axis work, and we always program with all of our planes at the same origin. I usually create a 3d point at my 1st origin, then after creating any other planes, using any method  (geometry, solid face, create from etc), I'll re-position the planes origin using my original point.

But the "Origin (in view coordinates)" are not all the same, even if my original 3d point was created at Mastercams world origin.

 

Why is that?

 

Zeke

[Colin Gilchrist]

Because the coordinates for the origin of each plane are relative to the way that plane is oriented. Notice the phrase "in view coordinates"? That means that the XYZ values are relative to the Tool Plane itself, not the WCS.

 

The planes features has always worked this way. The "easy" fix is to do exactly what you are doing: associate the origin to a piece of geometry. That way if you need to shift the origins of all your planes, you can just transform the point, and all the origins move with it...

[crazy^millman]

Well I have had some discussion with some people at CNC Software and have a hard time with the way it is and has been accepted as. You work from a point that point is that point no matter which way you turn it or spin in. 0,0,0 is still 0,0,0 if I am kicking an axis over in A,B,C. T-C Plane is not correct IMHO and needs to be fixed. Your base point is your base point and if you change the angle that base value should stay the same for WCS and T-C Planes period. You change your WCS to a new point in Space that point is now your WCS point. All values for T-C Planes should now reference that point for their X-Y-Z values not the world origin WCS. Math is math and in this case the math doesn't work and yes it may work and been the way it is, but I for one say it is wrong and have brought it up to be looked at. I don't remember it being like this in previous versions. I do have a Dual Boot XP 32 bit/Win 7 64bit system that has V9 on it I can check it out on.

[sweeper]

One more reason most other CAM systems use a tool axis and not a tool plane.

[So not a Guru]

Well I have had some discussion with some people at CNC Software and have a hard time with the way it is and has been accepted as. You work from a point that point is that point no matter which way you turn it or spin in. 0,0,0 is still 0,0,0 if I am kicking an axis over in A,B,C. TC Plane is not correct IMHO and needs to be fixed. Your base point is your base point and if you change the angle that base value should stay the same for WCS and T-C Planes period. You change your WCS to a new point in Space that point is now your WCS point. All Vales for T-C Planes should now reference that point for their X-Y-Z values not the world origin WCS. Math is math and in this case the math doesn't work and yes it may work and been the way it is, but I for one say it is wrong and have brought it up to be looked at. I don't remember it being like this in previous versions. I do have a Dual Boot XP 32 bit/Win 7 64bit system that has V9 on it I can check it out on.

Let me know what you find out, I'm curious.

 

Zeke

[crazy^millman]

Well I base my conclusion off of learning math almost 30 years. When the base point changed you were now working in a different coordinate system. All your coordinate calculations came from some reference and if you are doing your formulas for answers using a base coordinate of 0,0,0 then everything will solve according to that point as a base for the solutions. Now we get to some point where to got broken in Mastercam and it got passed off as we need to accept it as it is what it is. Sorry If I do the math for any calculation using a base point of 0,0,0 and then want to go off at any angle from that point the base point for the calculation will always stay 0,0,0 and the end result will change. A^2 + B^2 = C^2 is still the same today as it was when the formula was first created. The box greys out and shows the values relative to the WCS you are using, but then when you go review it in the WCS manager they are different which further supports it is broken. It gets the job done an if you don't pay attention to what is really going on you miss it, but like Zeke is seeing after paying some attention things are not adding up. I would think anyone from a mathematically background would freak out seeing it why it as surprised me others have not raised this question before and just said it is what it is when if you do the math it makes no sense it is what it is.

 

Again I am the crazy person that presented this question a few months ago. I will bring it back up and see what the official response is.

[Newbeeee™]

^^^you don't sound crazy to me Ron

[Watcher]

Ron, crazy? No...

[So not a Guru]

Thanks Ron, I was beginning to wonder if I was not understanding something correctly. I don't think either of us is nuts...well, at least not on this point.

 

Zeke

[crazy^millman]

Yeah the 10 foot pole rule seems to have kicked in. :laughing: :turned: :scooter:

[jeff]

Well I have had some discussion with some people at CNC Software and have a hard time with the way it is and has been accepted as. You work from a point that point is that point no matter which way you turn it or spin in. 0,0,0 is still 0,0,0 if I am kicking an axis over in A,B,C. T-C Plane is not correct IMHO and needs to be fixed. Your base point is your base point and if you change the angle that base value should stay the same for WCS and T-C Planes period. You change your WCS to a new point in Space that point is now your WCS point. All values for T-C Planes should now reference that point for their X-Y-Z values not the world origin WCS. Math is math and in this case the math doesn't work and yes it may work and been the way it is, but I for one say it is wrong and have brought it up to be looked at. I don't remember it being like this in previous versions. I do have a Dual Boot XP 32 bit/Win 7 64bit system that has V9 on it I can check it out on.

Being a newbie when it comes to planes, my brain just exploded reading that.

[JoshC]

Maybe I am not seeing it the same way, but who cares what the values read in planes. lets say I have a 4 sided block and I am doing 3+2 style work using planes. You can set the planes to any location you want, so if you want all planes to go to the same spot just pick the point. Are you guys disliking that you see values in the planes page for XYZ locations which reference from a WCS TOP?  If so, just ignore them! ​

 

The only advantage i am thinking you are looking for is to be able to just type xyz 0 and have the plane automatically locate to the same spot as the previous plane, but how could it do that if your new planes XYZ are completely out of wack or completely different than the original. 

 

some of the 5 axis machines can position the part it virtually any orientation, So plane 1 may have X going positive direction and plane 2 could potentially have the plane going the opposite way, so X+2 would become X-2 and so the math would not work correctly if you used positive values for both.

[JoshC]

Maybe I am not seeing it the same way, but who cares what the values read in planes. lets say I have a 4 sided block and I am doing 3+2 style work using planes. You can set the planes to any location you want, so if you want all planes to go to the same spot just pick the point. Are you guys disliking that you see values in the planes page for XYZ locations which reference from a WCS TOP?  If so, just ignore them! ​

 

The only advantage i am thinking you are looking for is to be able to just type xyz 0 and have the plane automatically locate to the same spot as the previous plane, but how could it do that if your new planes XYZ are completely out of wack or completely different than the original. 

 

some of the 5 axis machines can position the part it virtually any orientation, So plane 1 may have X going positive direction and plane 2 could potentially have the plane going the opposite way, so X+2 would become X-2 and so the math would not work correctly if you used positive values for both.

I am not sure if this makes any sense as this is fairly complex to type out, but i created a simple image below showing what i am referring to. Both plane 1 and 2 are set to the same origin values, but you will notice they end up in different locations.

image free hosting

[mayday]

I really dont use planes in 5 axis. the software is smart enough to position on it's own. Just stay away from 5 axis curve unless you have custom tools. Very old school tool path. Much better selection in advanced 5 axis tool paths that produce a very clean  path

[jlw™]

So no 3+2 for you then?

[crazy^millman]

Maybe I am not seeing it the same way, but who cares what the values read in planes. lets say I have a 4 sided block and I am doing 3+2 style work using planes. You can set the planes to any location you want, so if you want all planes to go to the same spot just pick the point. Are you guys disliking that you see values in the planes page for XYZ locations which reference from a WCS TOP?  If so, just ignore them! ​

 

The only advantage i am thinking you are looking for is to be able to just type xyz 0 and have the plane automatically locate to the same spot as the previous plane, but how could it do that if your new planes XYZ are completely out of wack or completely different than the original. 

 

some of the 5 axis machines can position the part it virtually any orientation, So plane 1 may have X going positive direction and plane 2 could potentially have the plane going the opposite way, so X+2 would become X-2 and so the math would not work correctly if you used positive values for both.

 

Josh kick one of the planes over to a 30 or 45 degree angle and the 0,0,0 will change. Get up into the 50, 60 or even 100 planes programs and the get back to me how nice it is to have the planes all have the same number so you know this group relates back to this WCS or that WCS. That was the way I remember it being. I move the part to 0,0,0 and made my planes for 3+2 they all respected the same 0,0,0. Now they don't and again why does the planes in the operation show that value from the WCS, but not in the operation manager?

[crazy^millman]

I really dont use planes in 5 axis. the software is smart enough to position on it's own. Just stay away from 5 axis curve unless you have custom tools. Very old school tool path. Much better selection in advanced 5 axis tool paths that produce a very clean  path

 

Use the retracts in the advance 5 Axis toolpaths. They don't respect the planes, but if they did then you would not have to go to 3 places and use the line to control the Axis direction. Also pay attention to the backplot as you will get the wrong position from the tools if you are not using the correct planes for the 5 Axis toolpaths. I will sometimes get a random position move that I have to use toolpath editor to remove if I don't use a plane close or exact to the plane direction I am working in.

[crazy^millman]

I am not sure if this makes any sense as this is fairly complex to type out, but i created a simple image below showing what i am referring to. Both plane 1 and 2 are set to the same origin values, but you will notice they end up in different locations.

image free hosting

 

That is to simple of an example kicks those planes over and then review the results in operations that use those and see the difference in the plane manager for that plane.

[crazy^millman]

So no 3+2 for you then?

 

All the time I use 3+2 with the opti-core and rest material from stock models all the time.

[mayday]

Use the retracts in the advance 5 Axis toolpaths. They don't respect the planes, but if they did then you would not have to go to 3 places and use the line to control the Axis direction. Also pay attention to the backplot as you will get the wrong position from the tools if you are not using the correct planes for the 5 Axis toolpaths. I will sometimes get a random position move that I have to use toolpath editor to remove if I don't use a plane close or exact to the plane direction I am working in.

user difined direction does, just draw a line and select it. it grabs that vector for retracts. 3+2 or full 5

[crazy^millman]

user difined direction does, just draw a line and select it. it grabs that vector for retracts. 3+2 or full 5

 

What I was saying, but what does that retract number relate to? Back to the WCS not to the line you are using or the plane. Now use line for your Title Angle Control and limits then Conical limit with user defined and you have 3 planes where you need to pick the same line. If these uses planes you could use a plane and use the defined plane for all 3 and not have to do all the extra work for picking the same line 3 times when using these 3 options together. It would also flow with how things have been. Not trying to stay old school, but looking at consistent process across the board. Click on planes in the New 5 Axis toolpaths in a large file and click on the planes in old school 5 Axis toolpath. I am seeing upwards of 30 seconds waiting for the NCI to update. Not consistent see the same thing with the old school surface toolpaths and the new surface toolpaths.

[jlw™]

All the time I use 3+2 with the opti-core and rest material from stock models all the time.

That was actually at mayday, how does one not use planes in 5 axis?

[crazy^millman]

That was actually at mayday, how does one not use planes in 5 axis?

 

Most times I can get away on basic parts with only using one plane, but when you start getting into the couple 100's of operations then yes I agree be real hard to not use more than one plane. Notice I said plane and not WCS. I think that s where some people get a misunderstanding. The WCS is your 0,0,0 for your operations and if that number is changing in Planes Manager or planes used the same WCS, but as different angles then how do they become different 0,0,0 for working in standard Mathematical process? I taught the subject back in the 90's when I was an apprentice and the person they hired to teach this subject was lost. Funny being the apprentice teaching half of the subjects. I am teaching this to our girls right now as we home school and Euler math supports what I am saying here as well.

 

Funny how it seems to stay pretty stable for a while then it will start to just freak out. I think that is where Zeke got to why it was driving him crazy. Keep it pretty simple again never see it, but get into some really involved projects and some weird things start popping up.

 

[mayday]

I cant speak for all machine types, But on an Integrex and a trunion style machine, I let the geometry tell the system how to position. 90% of my programming is simultaneous 5 axis milling using parallel to multiple curves and morph between 2 curves most of the time. I only use curve 5 axis for custom tool support like form tools or grinding wheels.

I can use X,Y,Z or vector lines for retract planes.

notice the tool paths all use top,top,top although I'm in 360 degree space

 

[Rstewart]

^^^^^^

15/64" ball-nose