Machsupport Forum
Mach Discussion => General Mach Discussion => Topic started by: nso on January 06, 2013, 02:44:44 PM
-
hi every one
first of all ,sorry for my bad english
at this video ,cnc wood turning machine working with 2 cut blade ...but how can do that?Each tool individually moves. The question is: How did this with Mach 3?
My first thought is that I need to add an axis like Y or A or C But the problem is creating gcode..i dont have any best idea
and that is my problem.
so if any one can help me or say comment make me happy.
http://www.youtube.com/watch?v=dT3Kxw3sQ6Q
-
Simple answer is I think you are not going to be able to do it unless you can produce suitable code and that will not be easy I am afraid, also the toolpath view will not be correct in Mach so you wont be able to use that, reason is your toolpath view is X and Z and you will have an additional axis but it will not show.
Hood
-
Just to add a bit to show the complexity of the code that would be needed. Mach can not control two axis to move independently of each other at the same time so you would need to have the moves broken down into very small moves and be able to calculate where each was at any given point and where it should be going at any given point.
I would say forget about it in Mach.
Hood
-
hi dear hood
like always ...You're the first person to give the answer ,thanks.
-
may be can do it at cam software like dolphin, OneCNC or Visualturn etc.....just i need think more... ::)
-
I think you would need a specialised CAM programme for that.
Hood
-
I think the controller would need a dual planner. I cannot see how a CAM could pull it off with a single planner.
Just a thought, (;-) TP
-
I believe Mach could drive the machine with two separate cross-slides, driven by two separate motors, in sync with the Z axis.
The Mach profile can be set-up as the traditional Z, X, plus a B axis.
Providing the code is the “trick.” ……….. (And, not worth doing, IMO.)
Consider you program an MDI move for the tool to travel from one corner of a “virtual” rectangular box to the opposite corner.
If the start point is the top of box, right/rear corner, and the end point is bottom of the box, left/front corner, the X, Y, and Z axis motors move simultaneously from start to finish. The three axis’ dimensions of the “virtual” rectangular box may all be different dimensions, but the motion is a straight line (within the tolerances of the machine) from start to end.
Mach integrates the steps; all we provide is; G1 X5 Y6 X-1 at the start and X-3 Y-2 Z-4 for the end point values on the next line of the program. We just made an X, Y, Z move with three axis’ moving in sync.
Nothing new so far; this is the “magic” of a CNC machine we use without a second thought.
In the Lathe problem, the two cross slides are riding on the Z axis; X and B must both be in sync with the Z, even though they may be moving at different rates, different directions, or number of steps with each other.
In other words, each incremental step size chosen for the Z, will be similar to the description of a
“Virtual rectangle box” motion from one corner to the other; but it will be Z, X, B moving in sync.
The “bad” part of this code programming is that the Z increment (distance travel) is the “baseline” for the programming.
The Z axis must move some given distance, to produce a corresponding X and B move. In real life, sometimes the Z is stationary, while X is making a cut, like a facing motion, or moving very slowly in tiny steps in relation to the X, as when cutting the OD of a sphere.
For each increment of the horizontal Z path, each corresponding displacement value of the “X”, and the displacement value of the “B” must be programmed on the same line.
One possible method would to use CamBam. Output the dxf for the Z,X profile, and divide the Z(X axis in CamBam) line into however many increments needed. You can then save the CSV file of X (the Z axis) and Y (the X axis) coordinates. CamBam can store the point pairs in a separate file.
Do the same for the Z,B profile, and save this CSV file.
(Note: CamBam may be able to combine the two profiles, but you would have to ask on the forum.)
The two profiles can now be combined using MS Excel.
Paste the first Z,X CSV file into into Excel. You will have two columns. Now paste the second Z,B CSV file into the same spreadsheet, in the next two columns. Make sure to keep the common Z “Zero” point aligned vertically.
You will now have an Excel spread sheet with (4) columns,
Z, X, Z, B.
Delete the duplicate Z column after making sure the Z “Zero” columns match exactly vertically before you delete; add in columns for the G-code letters Z, X, and B in the appropriate position. Save the spreadsheet as a .txt file, and you have a G-code text file that Mach should be able to load and run.
Some CAD programs may be able to do this without the bother of using Excel, but I am not into CAD lathe programming and do not have knowledge of such details.
--------
Unless I am missing something, I see NO viable reason to build a lathe to do this kind of operation.
Two passes with different tools will do the same thing, and utilize the functions such as cutting radii, linear slopes, vertical slopes, and do it well.
This goofy dual slide set-up removes the smooth motions of a standard CNC lathe; you cannot do a vertical slope; you cannot thread, or plunge a slot. Also, such a set-up more than likely would be a jerky motion, as the feed rate “vectors” (three axis) would be jumping all over the place, as all the vectors are tied to the Z axis traveling some given distance. Some moves simply could not be done realistically, like cutting a vertical face. And, it certainly is not easily programmed.
I also do not believe such a machine makes a big cycle time improvement, as the motion cannot be as smooth as a standard Z, X lathe profile.
The machine in the picture probably has a very high "Gee Whiz" factor, to impress onlookers.
-
BR549 and mrprecise44 thanks for your comment .maybe i can do it with your advice.. :)
-
I have seen these things in action (Production Wood turner) What you saw in the video is at idle speed (;-). It would outrun a straight up 2 axis lathe about 20 to 1 in production. That rotary ring is a cutter as well to cut the square stock round as it goes IF needed.
The ones I saw WERE dual planner controllers where the #2 cutter followed the master as a seperate entity. For these programming was very simple as you just program each pass seperately and the machine applied an offset from #1 to #2 and ran the code.
Just a thought(;-)TP
-
Hi BR549:
I just watched the video, and got a better picture of what goes on.
Never saw such a machine myself.
The leading ring (steady rest) is a major contributor to it's productive capacity. It appears the leading tool pre-cuts the OD, for the trailing tool, which does the major in/out excursions in the X axis. When the trailing tool is making the big in/out profiles, the leading tool basically holds steady at the current OD.
In the beginning of the video, the leading tool prepares the profile like a standard lathe, while the trailing tool just sits there backed off. The trailing tool doesn't remove the bulk of stock; that is done by the leading tool.
When the cutters move towards the tailstock, they do not go further than keeping the steady rest ring on the square. The two tools trail the ring by a 4 or 5 inches I am guessing, so that is the max reversal distance for the carriage.
Each tool seems to be moving in a regular 2-axis lathe profile manner.
I must say, it is a slick operation to see, and I can see how it gets such productive capacity.
Perhaps with a little creative programming, Mach could drive a lathe like this. Just linking all the profile segments together some simple?? way. Just the thing for a new Wizard.