[Evodyne]

Here is a link to a company with another take on a six-axis milling solution. There are two (avi) videos in the downloads section. Just food for thought...

Evodyne

[whateg01]

Interesting.

Dave

[Mcgyver]

the vid is impressive, thanks for the link. i think the people that were able to interpret a 3d complex design and translate it into 6 axis of movement must have been from another planet.

[Bloy2004]

It sure would be nice to see inside those axis mechanisms!!
What size motors?... What kind of bearings?....How are they geared?......and many many more questions are raised

[Evodyne]

It sure would be nice to see inside those axis mechanisms!!
What size motors?... What kind of bearings?....How are they geared?......and many many more questions are raised

Yeah, that's the shame of it: they raise more questions than they answer. Must be some pretty neat packaging. I'm guessing maybe harmonic drives tied to high speed servos. The harmonic units have no backlash and tremendous gear reduction-or tremendous torque multiplication. Thus a tiny motor could still whip a fair amount of weight around.

But a 10:1 reduction belt drive feeding a second 10:1 reduction would get you into nearly the same territory much cheaper. Or perhaps a very "tight" pinion and spur gear assembly.

Where did I pack all of those pulleys and belts....

Evodyne

[Splint]

An interesting machine but I can't see any real advantage over a 5 axis machine.

Splint

[Joe McLean]

An interesting machine but I can't see any real advantage over a 5 axis machine.

Splint

The 6 axis CNC Machining System has been developed to rapidly manufacture functional metal components, directly from a component’s computer model, three dimensionally. It actually replicates the computer model, simplifying the manufacturing process, using robotic three dimensional component presentations. This allows manufacturing operations to be undertaken, without having to reset position. It provides a total system solution to automatically machine complex shaped parts directly from a computer model, in metal, in 'one hit'.

The machine has six degrees of freedom (3 linear + 3 rotary) allowing complete freedom of movement in Cartesian space. It has been developed specifically for manipulating components three dimensionally against a process tool. This is achieved using the continuous instantaneous centre, created at the intersection of its rotary axes, as a 3D “real world” datum to emulate the “virtual world” Cartesian datum in the CAD. It can in fact machine a "golf ball" on a "tee" since it provides access not only to the top four but also to the lower four quadrants of a sphere.

This takes it beyond 5 axis machines into a CNC machine controlled by a post-processor that has a primitive intelligence built around the mathematics’ of the manipulator's movement algorithm. Superimposed rules of behaviour written over this algorithm actually turn it a robotic system. This simplifies programming and removes the requirement for a skilled machinist, enabling operation by personnel having only knowledge of CAD.
Productivity Benefits:-

v Eliminates fixtures and their cost.
v Eliminates “setting-up time”.
v Eliminates “down-time” between operations, reducing time to manufacture, manufacturing costs, Increasing productivity.
v The system is ideal for large or small batch component production and being robotic, is re-programmable for production of other components.
v It can also be used as a 3D digitiser and may be fitted with alternative process tools

I hope this helps with your understanding

Joe

[Joe McLean]

It sure would be nice to see inside those axis mechanisms!!
What size motors?... What kind of bearings?....How are they geared?......and many many more questions are raised

For answers to these question rather than chat please see spec freely available on web site

[madmickiii]

Def very interesting, take a look at hexapod type machines, maybe some of the principles work with these cyba thingies lol.


Mike

[greybeard]

".....robotic three dimensional component presentations...."

er, yes.

".....This is achieved using the continuous instantaneous centre...."

er, sorry ??

"..a CNC machine controlled by a post-processor that has a primitive intelligence built around the mathematics’ of the manipulator's movement algorithm...."

That certainly takes it beyond a 5-axis machine, and probably past me as well.

"I hope this helps with your understanding"

Sorry, Joe, but it doesn't for me.

However, I still admire the ingenuity, and the machine's capabilities.

[jneutron]

"I hope this helps with your understanding"

"Sorry, Joe, but it doesn't for me.

Yah, I agree. I'm gonna hafta read that a few more times.

My 7 axis machine and my code seems less complicated than that..

Cheers, jn

ps...well, 11 axis actually, but three are real time force controllers and one is weld power.

[underthetire]

Kinda reminds me of
[ame=http://www.youtube.com/watch?v=-Toh9EFNQHk&feature=fvwrel]MORI SEIKI NMV5000 DCG 5[/ame]

Guess holding it on a "t" would be ok for tiny stuff. Neat idea anyway.

[Joe McLean]

Ok Greybeard,

Please forgive me I'm also a grey haired Engineer passing on my years of self taucht CNC development free of charge for those who wish to learn. Those who don't please disregard.
It is my experience that those who don't could possibly become dinosaurs or Ludites
Should you wish to learn more of Continuous Instantaneous centres see http://www.roble.info/robotics/mobil...obots-1se3.htm its good for first principles in Mechanics - Linkages

Do your best old man, theres a whole new world out there for you in "one Hit" 3D machining. You can get there quicker.

[greybeard]

"Should you wish to learn more of Continuous Instantaneous centres see http://www.roble.info/robotics/mobi...Robots-1se3.htm its good for first principles in Mechanics - Linkages"

Yes, I would like to learn more. Perhaps you could fix the link, as it appears to have a slight malfunction.

The grey-haired non-engineer

[Joe McLean]

Sorry
try this http://www.roble.info/robotics/mobil...bots-1se3.html

When you're stretching the limits, pal, everthing is in the dark.
Pick up a brush and get on with it

[TomB]

Sorry
try this http://www.roble.info/robotics/mobil...bots-1se3.html
....

The link still does not work for me. I get to a roble.info page that is for sale and is currently advertising Lowes garden stuff. Further I would like to read the paper and I am a retired mathmatician/engineer so I should be qualified to understand it.

I have seen several videos about these type machnes and I really did follow Lance's explination of moving coordiante systems, but I still wonder if these machines are really 6-axis uints. To my eye they have 5 control movements, x, y, and pitch, roll and yaw (airplane terms but ok placeholders). However the pitch roll and yaw are applied in series so the modeling transforms have some extra sumation terms. I just wonder if those extra sumation terms are being used to fake a 6th degree of freedom and if so does that limit how every tool bit cutting surface can be made tangent to the desired cut. Just a curosity queston.

[jneutron]

I just wonder if those extra sumation terms are being used to fake a 6th degree of freedom and if so does that limit how every tool bit cutting surface can be made tangent to the desired cut. Just a curosity queston.

Odd, your use of roll, pitch, and yaw..I do the same.

I can see the need to orient the tool cutting surface with respect to either the desired surface, or the excess material surface. It does tend to complicate things when you also have to consider the cut vector.. I find the translation a headache at times.

On occasion I need to create pseudo-axis' to point a tool in a direction the primary axis' do not support. Most of the time, matrix math works just fine, but it can easily overreach your ability to visualize..

That video was really cool. I wonder if that roll around was really necessary or just showin off to impress..(it worked..)

Cheers, jn

[greybeard]

I had a very good English teacher at school, and enjoyed mathematics, but if I heard someone talking in the style that that article is written, I should avoid making eye contact, and quietly walk away.

I went back and forth through several pages, but I'm none the wiser for trying. :tired:

[Evodyne]

I had a very good English teacher at school, and enjoyed mathematics, but if I heard someone talking in the style that that article is written, I should avoid making eye contact, and quietly walk away.

I went back and forth through several pages, but I'm none the wiser for trying. :tired:

John,

Hi again! I too read the article. And I have to agree with you-pretty stiff (sorry Joe). I read volumes of technical publications and published papers-in my opinion some are written to be understood and some are written to show everyone how bright the author is. Here's Lance's simple explanation based on what I think I know (which may easily be incorrect):

Say your driving your car and you turn. You make a perfect turn with a constant radius. In your mind the you visualize the car turning about a point-it's center of rotation. A single, non-changing point. Now I follow you and due to my less polished driving skills, I make the same turn in an elliptical path. You watch me and determine that there is no single center of rotation, but several-one for each instant of time depending on my steering angle and car position at any given point in time. Thus there are several instantaneous centers that change position as I travel. But, you reason, my motion is continuous, so you decide that there is but one center of rotation for all instants that moves continuously in sync with my movements and is defined for any and every point of my rotation. Clear as mud?

Basically it's a point (possibly moving) about which all motion is rotational only.

Look at the picture I drew-basically it's a side view of Joe's machine. In your head mentally rotate the two arms to some position. No matter where you put them the three axis intersect at a fixed point, marked "center". Relative to that point all motion of the work piece is strictly rotational. Note too that "center" is fixed at a real, fixed, permanant point in space-Joe's datum.

Now note how the XYZ cartesian coodinate system that describes the part rotates with it. That's a problem for us: we quickly lose our bearings.

But note that if I shrink you (sorry) and glue you to "center" you are always looking down the object's X axis when you look at the part. From your perspective it doesn't move.

The final piece of the puzzle is that the cutting tool is always perpendicular or "normal" to the piece-the part that is currently being milled is always pointing upward.

So...if I was a post processor I'd look at the next point to be machined. I'd figure the rotations of the arms and chuck needed to make that point aim upward. Knowing these angles I could figure a 3D line from "center" to the point-to-be-milled on the workpiece. And I could do this in machine (not workpiece) coordinates. Finally I could move the spindle to put the cutting tool onto the part at those machine coordinates.

Or something somewhat kinda like that! Take care.

Lance

[turmite]

Lance......huh? You guys may not realize, it but it is humbling to read, and sometimes actually get to listen, to some of you talk about this stuff. This makes me wish I had taken all the math I could have taken when i was in school. Instead I avoided it like the plague. :tired:

Lance I am curious if you have tested the Mach software? Would Mach be able to handle something like this machine withou the G-Rex that is coming up?

Mike