[Pat Marine]

Hi folks,

I am currently milling, drilling and tapping about 60 side ports in a 6.5" OD 316 Stainless Tube. The holes have a 1" bore and then a series of stepped registers, for mating flanged port inserts. I am chamfering all the flat edges but since there are so many of the ports, I would love to be able to chamfer the rounded outer edge where the outer register diameter meets the tube OD. It seems to me that the 3D arc operation should be able to do this. I imagine, I would need to program in the x/y axis sweep of the register OD (radius 1.5") with its centre on the port bore axis and then also (depending on which quadrant I was programming) the y z or x z axis sweep respectively ( radius 3.25") with its centre on the tube axis centreline. I also worked out an xyz coordinate half way along the sweep.

I tried this with no success. I have worked on fagor machines in the past where you were able to separately input profiles in the xy yz and xz independently so was hoping that the 3D arc operation was similar.

Any advice would be greatly appreciated.

I know that there was a 3D mode option available for the Ultimax III which might have helped in this situation (dont have that section of the manual) but the local (UK Ireland) Hurco rep says that it is no longer available for our machine.

[Pat Marine]

hi guys,

have been searching furiously for a solution to the below stated problem, since I haven't got a response, and am out in the cold here... looks like its the needle file so if I don't figure it out in the next day or so...
I thought that the 3d arc milling or even the helical milling might help me but I think the problem would be that if the tube section is viewed in the xz axis the tool path would be a concave curve (not a straight line as in a true helix form) and a convex curve formed by the tube od in the yz axis. am I over complicating this or would a simple helical arc work?
excuse my ignorance but I am relatively new to this and have none else to ask...

[Pat Marine]

hi guys,

have been searching furiously for a solution to the below stated problem, since I haven't got a response, and am out in the cold here... looks like its the needle file so if I don't figure it out in the next day or so...
I thought that the 3d arc milling or even the helical milling might help me but I think the problem would be that if the tube section is viewed in the xz axis the tool path would be a concave curve (not a straight line as in a true helix form) and a convex curve formed by the tube od in the yz axis. am I over complicating this or would a simple helical arc work?
excuse my ignorance but I am relatively new to this and have none else to ask...

[67Cuda]

I'd like to help, but I've not come across needing to do the task you are asking about.

[Pat Marine]

thanks for responding 67 cuda,

I suppose, to put it more succinctly, without my previous ramblings, can the standard 3D arc operation cut in three planes at the same time... the manual says that an arc must cut in at least 2 axes simultaneously, (which tantalisingly suggests it might be able to do three at the same time?).

[67Cuda]

Hi Pat,

Here is some information that my help you.




3-D Arc
To create a three-dimensional arc segment, press the 3D Arc (F5) softkey
on the Mill Contour New screen. The following screen appears:
Figure 4-65. Mill Contour 3D Arc Screen
Important
Do not confuse this Mill Contour feature with the 3-D Milling
option. The 3D Arc feature is always included in the Ultimax
software; however, the 3-D Milling option must be purchased
separately. If 3-D Milling is available on the machine, refer to the “3-
D Part Programming Option” chapter in this manual for more
information.

The following fields define a 3-D Arc:
Fields Definitions
X, Y, and Z End Define the end point for the arc segment.
X, Y, and Z Center Define the center point for the arc segment.
X, Y, and Z Point Define the plane of the arc when it is 180° or
greater. If the arc is less than 180°, these values
are not needed.
Feed Rate from the previous segment. This value can be
kept or changed.
This diagram shows the relationships of the coordinates to each other:
End Point
X, Y, Z Point
X
Y
Z
Center Point
Start Point
Figure 4-66. 180° 3-D Arc

To calculate the centerline of cutter movement, remember these points:
• Depending on the axis being worked, X can be interchanged
with Y or Z.
• The X and Z points must be calculated for an X-Z arc.
• The Y and Z points must be calculated for a Y-Z arc. (All
arcs travel through at least two axes.)
• To calculate the centerline in the X axis for a ball-nosed end
mill, use this formula:
X
R
R
a X X ) X
i
= ( s − c + c
Here are the elements of this formula:
Xa = Actual centerline dimension of cutter in X axis
X s = Arc reference starting point in the X axis
Xc = X center point
Ri = Radius of arc minus ½ cutter diameter
R = Radius of arc
Hints and Tricks
Follow these guidelines when programming 3-D arcs:
• Never program cut right or cut left cutter compensation into
a 3-D arc block.
• Always program the centerline of cutter movement.
• Lower the tool calibration point by manually changing the
Tool Zero reference point to one-half the cutter diameter.
For example, when using a ¼" ball-nosed end mill with zero
calibration of 2.2500" (ball tip touched to work surface), the
new zero calibration is 2.3750" after the manual change of
reference. Remember this change when entering Z Up and
Z Start dimensions (which must include the value of this
manual change to the reference point).

[Rich Carlstedt]

I have never run a Hawk, but it seems to me that what you want is a Full 3 axis machine.
My KM3 Hurco with Ultimax 11 is called a 2 1/2 axis machine, even though I have the 3D package and can do full helix's.
I believe 3 D Helix is standard with any Ultimax, and the 3 D package only gives repeatable moves of an arc within a block, instead of doing multiple offsets which you could do with your present basic package !
What that means is I can get two variable axis's and one linear axis.
A helix program does threads wonderfully, as they have a constant pitch. You cannot do a variabl pitch helix/thread unless you take many segments into play
I cannot get (in a single segment) three axis's that varies in non-linear movement ( Any curved [variable feed ?] movement is non-linear)
In math terms, you are asking for all three movements to be Sine waves

I have done intersections of round objects but it took many blocks, each with a calculated Z start/end to get what you want.

RICH

[Rich Carlstedt]

Pat
The only way I know to chamfer or bevel the edges on your pipe
is as follows, and it only uses a YX arc.
Say your pipe is on the X axis and the hole is X0, Y0

Get a ball endmill of about 1.5 " , ( or custom radius) and start it at say Y3 and the Z depth would be just enough to chamfer the lower part of the hole.
As the cutter moves from +Y3 to Y-3 (example) when it encounters the hole it will chamfer that edge. Now have it do a YZ arc coming up (top of pipe) and then back down to the same Z depth as before as it travels to -Y .
This will give you a chamfer, it just won't be a 45 degree one, and will be based on the ball cutter radius

Rich

[Pat Marine]

Thanks 4 the responses guys.

Rich I may try your method next time this comes up. I basically came to the same conclusion, that I needed a non linear arc in the z, on a full 3 axis machine. It looked surprisingly close on the screen when I tried to use a simple helix, but it didn't touch the edge all the way around the circumference. oh well I had to resort to the needle file. thanks again for the advice

[Pat Marine]

Hi Rich,

Turns out the client required an additional tube, and I have really grown to despise the needle file recently so I tried again to figure this out. The tube is 1950 mm long this time, with an additional 12 ports. I still had the set up in tact, so I decided to try to play around with the chamfering again.

Your recent suggestion of using a ball nose on a single plane got me thinking, but it occurred to me that a 45degree chamfering tool would be easier to describe mathematically, than a ball nose cutter. I played around in AutoCAD for a bit. I know the xy coordinates of any point on the edge, since they are described by the diameter of the outer register, and I know the z axis coordinates since they are defined by the outer tube diameter. By setting a z depth on a 20.5mm chamfering tool, I was able to work out the changing tool diameter for any point along the register outer diameter. When I plotted a number of points along the part arc, it turns out the tool centre transcribes a definable arc also. Its centre point is well away from anything logical (for me at least) but it works out, and the arc looks something like an involute spline with respect to the register diameter. In practice I needed to play around with the exact curve a bit, and split it up into a few spline curves instead, but the method in general seems to work.

I have attached a picture of the finished product ( one of many ). Thanks again for putting me on the right track.

[Pat Marine]

Not sure why It didnt attach last time. Lets see if it works this time