PDA

View Full Version : CNC turning on a gantry machine



mike hide
04-24-2006, 03:40 AM
I was wondering if anyone has mounted a slow rotating spindle on the main machining bed so that one could essentially do on a gantry machine what otherwise would be done on a CNC lathe ?

I am thinking mainly in terms of a woodworking machine.

cbass
04-24-2006, 06:53 AM
I think its called a 4th axis (or a rotary table). Look around on this site and you'll see many people doing it.

Carlo

keyne
04-24-2006, 12:03 PM
search "indexer"

joecnc2006
04-24-2006, 12:19 PM
Sample....

http://www.cnczone.com/gallery/showphoto.php/photo/453/cat/500/ppuser/223

mike hide
04-24-2006, 01:07 PM
Thanks joe ...This is similar to what I had in mind except based on ignorance I thought that the forth fifth and sixth axes were angular move
ment of the spindle iin three planes .

What I had in mind was a similar set up only in the long axis ["X" axis] where the machine would first do the basic turning along the length of the workpiece as would be done normally on a lathe ,varying diameters molding shapes etc.

when that aspect was finished then the machine would come back and detail the turning along it's length. For instance lets say in the initial " turning" a half round form was made on the diameter ,then the machine in the second operation might come back and detail the half round to make it into an "egg and dart" molding

Geof
04-24-2006, 03:36 PM
What is 'egg and dart'?

I have in mind doing what you are describing (I think). Setting up a rotary axis parallel to the X axis then carving the periphery of long cylinders; pillars when you stand them upright. This project is in its early stages and all I have so far is the gantry machine.

mike hide
04-24-2006, 04:01 PM
I have included a couple of shots of "egg and dart" molding ....regards mjh

Geof
04-24-2006, 05:37 PM
Yes that would look neat around the top collar of a 2' diameter 10' high pillar.

ger21
04-24-2006, 07:37 PM
What I had in mind was a similar set up only in the long axis ["X" axis] where the machine would first do the basic turning along the length of the workpiece as would be done normally on a lathe ,varying diameters molding shapes etc.

when that aspect was finished then the machine would come back and detail the turning along it's length. For instance lets say in the initial " turning" a half round form was made on the diameter ,then the machine in the second operation might come back and detail the half round to make it into an "egg and dart" molding

No need to do this in two seperate steps. All can be done in one step, although you may need to do a roughing pass to get it clost to shape, then change tools to do the finishing pass(es). May depend on what CAM software you use, though.

mike hide
04-25-2006, 12:27 AM
Yes I agree an initial "roughing out "operation would be required otherwise the depth of cuts could be too much, and could cause the operation to bog down .

Not being familiar with CNC operations If I may some basic slightly off subject Questions .

If you are "skimming" a flat surface and say the cutter [gantry] is traversing up and down the X axis, when the cutter gets to the end of it's travel does it jog over and cut the next tract in the opposite direction or does it disengage traverse to the start point jog over and start cutting the new tract in the same direction as the preceeding one .

Same situation except there is a web across the workpiece ,does progress in the x axis virtually stop while the cutter climbs over the web and then continue on the other side . seems to me there would have to be some anticipatory [if there is such a word] action on the part of the software to prevent the cutter balooning over the web.On the other hand is the software sufficiently sophisisticatedto recognize such a situation and perform the operation in the y axis instead or is that up to the operator to optimize the cutting path?

mike hide
04-25-2006, 06:39 PM
I think this concept , as far as wood working goes is a viable solution to a sepatate CNC lathe . Just make a turning spindle superimposed on the gantry machine bed the 4th axis . This would make a separate lathe unnecessary at least as far as woodworking goes.

Digressing , spindle turning on a wood lathe is totally different to turning on an normal engine lathe . Thin material sections on a wood turning tend to part company at the most inopartune times due to the nature of the material spindleturning involves a slicing action . Turning wood on a metal lathe and using normal metalworking techniqies simple does not work very well.

However what we are talking about is using the spindle with endmills or router bits,in my opinion that should work.

Tell me why not you are not going to hurt my feelings

Geof
04-25-2006, 06:57 PM
Are you meaning you would have this spindle just driven continuously but with the cutting action performed by a router bit rather than a chisel as would be done on a wood turning lathe? If that sentence is understandable.

I mentioned above that I am working toward something like this and it is not an entirely original idea. Quite a long while ago I met guy who had a very successful business making large wood columns for expensive houses. He had a big modified metal lathe; big as in 20 feet long with a swing of 30 inches. To turn the outside of his columns he mounted them between the chuck and tailstock and had a large router on the toolpost. The spindle drive was modified so it could turn very slowly and similarly the feed drive was also modified so it would feed at up to 1-1/2 inches per revolution of the spindle. He had a way of attaching a template along the back of the machine and had a system for disconnecting the screw on the cross slide so he could have the slide follow the template and turn tapers or barrel shapes. He also had an indexing system so he could lock the spindle and just move the carriage and with different cutters in the router could cut flutes the length of the column and these could also follow a template.

It sounds a bit like you are considering doing something that is along these lines but starting with a gantry router instead of a lathe. With full CNC of course you would be able to do much more detail.

phillby
04-25-2006, 08:22 PM
Some one was building a gantry out of steel to mount over a wood lathe in the last six months or so but I haven't seen anything on it for a while.

I am also interested in this Idea.

mike hide
04-25-2006, 08:44 PM
yes I envision the basic spindle turning slowly and being shaped by the router bits in the gantry spindle, the software determining what cutter etc feedrate etc . After theat the software goes into a different mode and also the Lathe spindle .

Perhaps a series of cutters would then be used by the gantry spindle to operate on the "turned" workpiece in a similar fashion as the normal milling mode only the "turned piece being lpositioned radially by the "lathe" and in the X Y and Z axis by the forth axis software which also would control the depth of cut etc for the particular tooling .

If this could work then it would certainly make the forth axis earn it's living . this is the case of trying to teach an old dog new tricks so what is the consensus am I barking up the wrong tree?

Geof
04-25-2006, 10:08 PM
I don't think you are barking up the wrong tree because I am planning on doing something along the line of what you are saying. But as one old dog to another I think you should find a few CNC shops to visit or a technical school that has CNC courses. Your background is in woodworking not CNC machining and for what you are thinking about you really need to have a good idea of what a CNC machine equipped with a fourth axis can do and how it is controlled. A lot of what you are, I think, viewing as separate operations get rolled into one big and fairly complex operation on a CNC machine.

mike hide
04-25-2006, 10:54 PM
Well kind of been there done that kind of thing . did a 5 year indentured apprenticship in aerospace in the UK more years ago than I like to recall. the first two years having to learn the rudiments of operation of most machines on the shop floor .CNC was in its infancy ,the computers were things called slide rules and computers and modern day electronics were yet to come .

So it has been a steep learning curve at tiimes even to learning how to type [still pretty rudimentary in my case].

The concepts I propose are basically out of necessity financially and space wise I will do well to end up with one machine that is why I came up with idea of incorporating the "lathe" into the gantry machine .It saves money and space and uses the same software . I am resonably confident I can build such a machine and am trying to get a handle on the electrons bit and evaluate mach 2 software to see if it can run it.

One thing I have noticed is many on the group talk about woodwork and metalwork in the same breath. As far as I am concerned they are worlds apart unless you are working something like lignum viti which is as hard as some metals . spindle Turning of a wood lathe is more of a slicing action and the tool angle varies all over the place ,I do not believe metaltooling would work well on wood although a router cutter designed for woodworking would work quite nicely.

well you mentioned that the egg and dart would look good at the top of a column ,you might look at the order of columns .The Romans had the same idea a couple of thousand years ago, bat us to it I am afraid

JRoque
04-25-2006, 11:01 PM
This is a good idea. I'm planning to add a 4th axis to my gantry machine for this purpose. The lathe part of this, however, would be better if driven by a timing motor rather than "slowly moving" as suggested. This would give you all the control you need to do serious carving on round objects - for example, engraving.

JR

ger21
04-25-2006, 11:11 PM
yes I envision the basic spindle turning slowly and being shaped by the router bits in the gantry spindle, the software determining what cutter etc feedrate etc . After theat the software goes into a different mode and also the Lathe spindle .

Perhaps a series of cutters would then be used by the gantry spindle to operate on the "turned" workpiece in a similar fashion as the normal milling mode only the "turned piece being lpositioned radially by the "lathe" and in the X Y and Z axis by the forth axis software which also would control the depth of cut etc for the particular tooling .

If this could work then it would certainly make the forth axis earn it's living . this is the case of trying to teach an old dog new tricks so what is the consensus am I barking up the wrong tree?

You don't have to "turn" the piece before doing the milling operations. As was already mentioned, It's all done as one operation. The exact process would depend on the software you use to create the gcode. Usually roughing passes would be performed, either radially around the part as it turns, or lengthwise while it's held in a fixed position, then rotated a fixed amount. You can usually specify a tolerance which the roughing passes will be taken to from the finished part. Then you can change tools, and do the finishing passes.

ger21
04-25-2006, 11:18 PM
If you are "skimming" a flat surface and say the cutter [gantry] is traversing up and down the X axis, when the cutter gets to the end of it's travel does it jog over and cut the next tract in the opposite direction or does it disengage traverse to the start point jog over and start cutting the new tract in the same direction as the preceeding one .


Depends on what software you use to create your g-code with, but most should give you the option to do it how you want. Although some software will give you these options for the roughing passes, but the finishing passes may be back and forth only. Entirely dependant on the CAM package.




Same situation except there is a web across the workpiece ,does progress in the x axis virtually stop while the cutter climbs over the web and then continue on the other side . seems to me there would have to be some anticipatory [if there is such a word] action on the part of the software to prevent the cutter balooning over the web.On the other hand is the software sufficiently sophisisticatedto recognize such a situation and perform the operation in the y axis instead or is that up to the operator to optimize the cutting path?

Again, totally dependant on the software being used. Usually, the less money you spend, the less options you'll have. As the price of the CAM package goes up, you'll usually have more types of machining stategies to use.

Geof
04-25-2006, 11:26 PM
A nice feature on some of the commercial machines with fourth axes is what is called circular mapping and it is intended for what JR mentions; carving or engraving on round objects. If you have a something that would be carved in the X Y plane, for instance a picture you have processed through V-carve or something like that, you can map it onto the surface of a cylinder. Part of my concept is to take pictures similar in nature to the prints done by the artist MC Escher and convert them into engravings around columns and pillars. A feature of these paintings is that the left side leads you into the right side so when they are wrapped around a column there is no start or finish. The picture would appear complete no matter from which direction the column was approached. Because the circular mapping is a machine control feature it means that off the shelf software such as Vcarve that is oriented towards 2-1/2D engraving can be used to create a (sort of) 3D image.

mike hide
04-25-2006, 11:41 PM
Hi JR The reason I suggested slow rotation is I anticipated roughing out the turning first before doing the details .I anticipate in my case using a pretty acute angled Vee cutter to do the details which I think would not be suited for general roughing out. I anticated the 4th axis motor would slowly rotate the spindle turning to optimize the router cutter working . After that the motor would coordinate the angular position of the "turning"with the other axes to perform the incised detail cuts.

What I would like to do is produce something like that shown in the photos in cluded or at least do the basic donkeywork.

ger21
04-25-2006, 11:43 PM
The roughing is done the same way as the finishing, just with a larger tool. You specify how much material to leave for the finishing pass to remove, and the roughing passes will take the piece down to size.

mike hide
04-26-2006, 12:11 AM
I was just thinking rather than changing tooling to suit the particular task at hand wheather it would be feasible to mount say three spindles all preloaded with particular cutters in the Z axis and just program the offsets into the computer program.when that particular spindle is in use ......probably over the top dont you think

mike hide
04-26-2006, 12:16 AM
Good thought could be useful Will check it out thanks mjh

mike hide
04-27-2006, 02:33 AM
A nice feature on some of the commercial machines with fourth axes is what is called circular mapping and it is intended for what JR mentions; carving or engraving on round objects. If you have a something that would be carved in the X Y plane, for instance a picture you have processed through V-carve or something like that, you can map it onto the surface of a cylinder. Part of my concept is to take pictures similar in nature to the prints done by the artist MC Escher and convert them into engravings around columns and pillars. A feature of these paintings is that the left side leads you into the right side so when they are wrapped around a column there is no start or finish. The picture would appear complete no matter from which direction the column was approached. Because the circular mapping is a machine control feature it means that off the shelf software such as Vcarve that is oriented towards 2-1/2D engraving can be used to create a (sort of) 3D image.

Been thinking and doing a little reading on your last entry, in particular the vcarve tutorial. So correct me if I have it wrong .....which I am frequently but loath to admit.

The v carve exercise shows the operation on a flat surface and I assume if the "lathe" arrangement we have been talking about is the 4th axis It would be required to rotate the "turning" so that the cutter was always presented with a surface normal to it's axis thereby fooling it into thinking it was a flat surface .

I have photoshop and played with it enough to know that images etc can be mapped onto irregular surfaces in addition photoshop can output EPS files .

reading about vcarve they do mention angling the cutter in some situations .This bothers me a little because does that mean that the spindle has to be able to be angles relative to the workpiece ,if it dose are we then talking about a 5 axis machine ?

Another nagging could be problem is looking at the pictures and the turned section only,although everythis round in the y z axes there is considerable slope in some sections in the z x axis . perhaps it would be better to do a stacked arrangement and machine each major section on an individual basis .Glued up no one would know the difference and I do not see any lessoning of qualityof the assy ....mjh

Geof
04-27-2006, 11:03 AM
...The v carve exercise shows the operation on a flat surface and I assume if the "lathe" arrangement we have been talking about is the 4th axis It would be required to rotate the "turning" so that the cutter was always presented with a surface normal to it's axis thereby fooling it into thinking it was a flat surface .........
.......reading about vcarve they do mention angling the cutter in some situations .This bothers me a little because does that mean that the spindle has to be able to be angles relative to the workpiece ,if it dose are we then talking about a 5 axis machine ?

Another nagging could be problem is looking at the pictures and the turned section only,although everythis round in the y z axes there is considerable slope in some sections in the z x axis . perhaps it would be better to do a stacked arrangement and machine each major section on an individual basis .Glued up no one would know the difference and I do not see any lessoning of qualityof the assy ....mjh

Let me respond to your last point referring to the pictures you posted showing ornate carvings. I think replicating these on a CNC could be very difficult; this is way more complex and very different to what I am thinking of.

Refering to my V-carve idea around a cylinder I think you have interpreted my explanation the way I meant it. You simply take an image from a flat plane and wrap it around a cylinder with the appropriate diameter so that your circumference equals the Y distance of the flat plane. The cutter is always working on a surface normal to its axis, the G-code produced by the Vcarve software is completely normal it is just the CNC machine controller internally converting linear Y moves into angular A moves and as a I said this is a feature built in to some machines.

Now regarding your comment about angling the cutter. Yes, to be able to angle the cutter relative to both the X and Y axes you would need a 5 axis machine. However, using cylindrical mapping it is possible to "angle" the cutter relative to the "Y" axis. The normal surface your cutter is working that you mention above is tangent to the cylinder at that point and you can think of this tangent as a "Y" axis. But the spindle axis is only normal to the "Y" axis when the spindle centerline is exactly on the cylinder centerline. If the spindle centerline is moved away from the cylinder centerline then the tangent at the point of contact is at an angle; from the point of view of the surface of the cylinder at that point the cutter is at an angle. The amount of angle depends on the diameter of the cylinder and the offset between the spindle center and cylinder which is done using the Y axis. Therefore at least in one plane my concept can accomodate angling the cutter.

There is an additional aspect in my concept that could also be altered to possibly change the appearance of the carving mapped onto the cylinder and this is related to how Vcarve uses the Z axis. Engravings from Vcarve are not really 3D. I think they are best described as 2D with variable depth. In Vcarve the Z coordinate comes from how they treat the brightness of the image; darker is deeper or something like that. Whatever it is called the engraving depth does vary and when mapped onto a cylinder it would be possible to choose the top of the engraving for the diameter used for the cylindrical mapping or the deepest Z cut could be used or something in between. These different choices could make a difference in the texture of the engraved surface.

mike hide
04-27-2006, 03:09 PM
In response to your reply, I agree there are differences in our aspirations but the basic paths to get there are similar. From my aspect even if I end up withe the basis forms cut ,which I am reasonably confident can be done on this arrangement, and even with a shallow engraving on the turning I would be happy . The rest of the carved detail I can do having been doing this type of hand carving for quite a few years now. Right now I would like to design a machine that could do this and perhaps more .The 4th axis would have to be driven in a similar way to the circular mapping mode of the modern CNC machines rhat you mentioned . I know the hardware for the 4th axis is available at a resonable price [hobbycnc et al] but now the question remains could software such as Mach two or three be modified to do so or would it take customized software to do it ?

In your second paragraph you mention angling the cutter by offsetting it from the centerline of the cylinder. In the case of a column if I understand what you are saying the detail would look different according to the direction you approach the column or are you sugesting offsets both sides of the centerline .

You mentioned that Vcarve is essentially a 2D program with various depths of cut , true, but if you come to think of it thats what all carving is , just a play on light, highlights and shadows . Many times the shadows [the deep cuts] are accentuated with contrasting paint or dark stains . I have seem highly decorative pieces with none of the preceeding and they are practically life less. Again a highly carved detail molding used as a crown molding looks "muddy" when viewed from twelve feet away in these cases the carved detail needs to be more robust and on a larger scale to show to best advantage.

Geof
04-27-2006, 04:22 PM
Yes we are on more or less parallel tracks but I think yours is the more challenging. What I am planning on doing is using off-the-shelf processes in new and maybe unconventional combinations. To me it seems you want to do much more creative stuff; treating the machine just as a tool to help you release the creativity. One thing about your track is that when you are successful you are unique and more or less impossible to copy. Because my creativity, such as it is, is restricted to put existing processes together differently I can be copied by anyone.

Regarding angling the cutter yes either side of the centerline.

I like your comment; "the carved detail needs to be more robust and on a larger scale to show to best advantage." This is partly why I bought the machine I did. I expect that the type of engraving I want to do will not stand up to really close scrutiny. It is more or less impossible to get the type of finish that is possible by hand without either using multiple cutters, very small steps between tool paths and/or doing quite a bit of hand finishing. All this takes time. However if the engraved items are large, intended for viewing from a distance of 10 feet or more and located so that really close viewing is not easily possible all the artistic imperfections inherent in machine engraving are fuzzed out. Fewer cutter changes and larger steps will reduce the machining time and hand finishing can probably be omitted.

carlnpa
04-27-2006, 06:34 PM
Is there any low end software to create 4th axis code like sheetcam for 2.5D or meshcam for 3D?
I think the mechanics may be alot easier than the code.
Thanks
Carl

ger21
04-27-2006, 07:02 PM
In response to your reply, I agree there are differences in our aspirations but the basic paths to get there are similar. From my aspect even if I end up withe the basis forms cut ,which I am reasonably confident can be done on this arrangement, and even with a shallow engraving on the turning I would be happy . The rest of the carved detail I can do having been doing this type of hand carving for quite a few years now. Right now I would like to design a machine that could do this and perhaps more .The 4th axis would have to be driven in a similar way to the circular mapping mode of the modern CNC machines rhat you mentioned . I know the hardware for the 4th axis is available at a resonable price [hobbycnc et al] but now the question remains could software such as Mach two or three be modified to do so or would it take customized software to do it ?


Mike, all you need is a 3 axis router with rotatry axis to hold and turn the workpiece. I think your making this far more complicated than it needs to be. There's no need to try to modify mach3, it will work just fine. The problem for you may be creating the 3D model of your carving, and then getting a CAM program to output g-code your happy with.

You should download and play with the demo of Deskproto, it should be able to do what you're looking for. www.deskproto.com It may give you a better understanding of the process.

mike hide
04-28-2006, 11:32 AM
Mike, all you need is a 3 axis router with rotatry axis to hold and turn the workpiece. I think your making this far more complicated than it needs to be. There's no need to try to modify mach3, it will work just fine. The problem for you may be creating the 3D model of your carving, and then getting a CAM program to output g-code your happy with.

You should download and play with the demo of Deskproto, it should be able to do what you're looking for. www.deskproto.com It may give you a better understanding of the process.

I probably am making it too complex basically due to my own ignorance of what is available and what it"s capabilities are . I do appreciate the guidance and will take a close look at the suggested program. thanks mjh

mike hide
04-30-2006, 01:27 AM
I probably am making it too complex basically due to my own ignorance of what is available and what it"s capabilities are . I do appreciate the guidance and will take a close look at the suggested program. thanks mjh

Ger, if you don't mind > I have an early version of turbocad which I am somewhat familiar, however it is not up to the task of getting from drawing to a cnc machined item .

Do I need a program such as "Rhino" or the like to produce STL files for "Deskproto" and the like need to drive a program like mach 2 or 3 ?

presumably the path is rhino or similar program > code making program>mach 2 or3 or simialr cam program

If this is correct would you mind giving me an idea what progams will do his and what I need as a minimum. thanks mjh

mdreitzusa
04-30-2006, 04:19 AM
mike hide
here is a pic of one of whiterivers machines that looks something like what you are talking about.
http://www.cnczone.com/forums/attachment.php?attachmentid=12713

ger21
04-30-2006, 08:12 AM
Ger, if you don't mind > I have an early version of turbocad which I am somewhat familiar, however it is not up to the task of getting from drawing to a cnc machined item .

Do I need a program such as "Rhino" or the like to produce STL files for "Deskproto" and the like need to drive a program like mach 2 or 3 ?

presumably the path is rhino or similar program > code making program>mach 2 or3 or simialr cam program

If this is correct would you mind giving me an idea what progams will do his and what I need as a minimum. thanks mjh

Basically any CAD program that can export a 3D model as an .stl will work ( or even a .dxf, depending on the CAM program.)

Yes, Rhino would work, but there are cheaper options available. Newer versions of TurboCAD **should** work.

Yes, the path you mentione is correct.

CAD > CAM > Mach3

Rhino = CAD
Deskproto = CAM

The important thing when choosing a CAD program, is to make sure it can exprt files in a format that your CAM program can read. But, there are inexpensive file converters available that can help to overcome this.
http://www.micromouse.ca/

Deskproto is one of the cheapest CAM programs that support 4th axis work. You might also want to take a look at MeshCAM, www.meshcam.com , Version 2, which is not released yet, will have 4th axis support. It may be a while before it's released, although there is a beta available I think.
http://www.meshcam.com/

mike hide
04-30-2006, 10:56 AM
mike hide
here is a pic of one of whiterivers machines that looks something like what you are talking about.
http://www.cnczone.com/forums/attachment.php?attachmentid=12713

Thanks fr the shot. Yes the principle is exactly what I had in mind . It wpould appear the machine shown is milling some kind of soft material . I see a more robust design for milling wood as the tool loads will probably be much higher in particular with deeper or wider cuts .I appreciate your example for a while I had thought I would have to reinvent the wheel ,but then there is really very little that is new ,except all these programs that is ....

mdreitzusa
05-01-2006, 04:45 AM
heres a link to whiterivers site.
http://www.whiterivermfg.com/
have a look around he does a lot of wood work as well as foam.

Art Ransom
05-01-2006, 09:02 AM
I am in the process of building a column lathe, 4 axis, it will have a 24" by 11'6" capacity. Using Mach3 and haven't finilaved on CAD software. I will be posting details on mt site, www.turningaround.org in CNC section. I have been using a Legacy Ornimental Mill but got tired of its limintations. I figure the mill will cost 4K with all hardware and software. Live on south edge of Dallas,TX. Adress and phone # on my site. I started a Legacy group and have all the Legacy links on the web and about 75 tips from the group. The mill should be able to do more than the Legacy, better, faster and cheaper. Several from the group have expressed interested in me building one for them but also with router ability. For all you turners the front page of site should grab you!