Why not a Gantry Setup??

[Hellbringer]

Hi,
I am working on designing a new mill.... I was thinking about making a gantry setup..... What are the disadvantages to doing a gantry setup vs (what i am going to call) conventional setup.....

BTW i would be cutting aluminum and some steel....

Mike

[RobWright]

A gantry set up will be significantly less rigid than the conventional column setup, and this will affect the tolerances of the product. Also with a gantry the motors have to move the very heavy spindle around, this means more expensive, powerful motors must be used and also more wear on the linear axis parts such as ballscrews and slides. In most cases for these benchtop cnc machines the product being cut out it much lighter that the spindle setup so it is logical that the stock is being moved not the spindle. Also the counter argument being that in conventional setups the spindle still has to travel up against gravity, however this can be overcome with a counterweight where as on counter weight can be used to offset spindle weight in a gantry setup.

I hope this is helpful, but im an no expert, im still learning.

[awerby]

That can be a lot more rigid than a gantry, since the beam carrying the Y and Z axes doesn't have to move along X. Instead, you move the table underneath it in the X directions. The main disadvantage is that it takes a bit more floor space, but it saves having to make a giant casting, which is the basis for most conventional mill designs.

By the way, you can use a constant-force spring (think of a tape measure) instead of counterweights in a gantry (or bridge) design to help handle the weight of your spindle.

Andrew Werby
www.computersculpture.com

[Zach_G]

Pound for pound and cubic inch of workspace, a gantry will be SIGNIFICANTLY more ridgid than a conventional machine. The difference comes from the large moment arm attaching the spindle to the column of a conventional mill which is effectively eleminated by a gantry setup. But smaller conventional bench mills can be overbuilt to narrow the performance while not increasing the weight beyond what most hobbyists would consider a burden or excessive expense. For comparison sake lets consider the pros and cons of a gantry vs conventional mill assuming equivalent specs of ridgidity, work envelope and speed. Here's what I would expect to see in a gantry:

• Lighter weight
• Smaller machine footprint
• Extra complexity (more parts to manufacture and align, possibly need second slaved drive to prevent "racking" on one axis)
• Decreased accesibility of work envelope (everything must fit between gantry columns)

I myself decided to go with the gantry for a metal working machine and am currently waiting on some money, space and shop access to fall into place before I go forward with the build. Tried to mitigate the cons as much as possible with this design but certainly isn't the only way to do a gantry.

[awerby]

Originally Posted by Zach_G Pound for pound and cubic inch of workspace, a gantry will be SIGNIFICANTLY more ridgid than a conventional machine. The difference comes from the large moment arm attaching the spindle to the column of a conventional mill which is effectively eleminated by a gantry setup. [Sure, but I think you're missing the point I was making. Yes, a thousand-pound gantry will be more rigid than a thousand-pound "C" column, but you don't have to push that C-column (or the equivalent bridge) around, or make it change direction and go the other way. With a gantry, every pound you add counts against you as inertia. You can make a C or a bridge as big and heavy as you want with no penalty, but this isn't the case for a gantry. To work, they have to be as light (but rigid) as possible.] But smaller conventional bench mills can be overbuilt to narrow the performance while not increasing the weight beyond what most hobbyists would consider a burden or excessive expense. For comparison sake lets consider the pros and cons of a gantry vs conventional mill assuming equivalent specs of ridgidity, work envelope and speed. Here's what I would expect to see in a gantry: Lighter weight Smaller machine footprint Extra complexity (more parts to manufacture and align, possibly need second slaved drive to prevent "racking" on one axis) Decreased accesibility of work envelope (everything must fit between gantry columns) I myself decided to go with the gantry for a metal working machine and am currently waiting on some money, space and shop access to fall into place before I go forward with the build. Tried to mitigate the cons as much as possible with this design but certainly isn't the only way to do a gantry.

[You're correct there. I think that design (as pictured) suffers from the same problem as most gantries; it's overweight and under-rigid. Hanging everything on those big vertical moving members gives you moments of leverage that work against you. You're constructing a big rectangle that wants to be a parallelogram (or a series of them, depending on its mood). A design that eliminates those in favor of raised rails on the sides seems like a better bet to me - see the Mechmate design, which works that way.]

Andrew Werby
www.computersculpture.com

[Hellbringer]

a little more info on where i am comming from and my idea.....

the mill that i am drafting right now is somewhat like Zach G's with a little different ways of moving the axis... the gantry does not move (this way i can weld or overly bolt the gantry to the base to help with flex) the spindle on the gantry moves left and right for the x axis and the whole gantry moved up and down (using to steppers to offset the weight) and then i have the table move front and back for the y axis... the way i look at it i am moving more differnt parts then the same part over and over again trying to help with ridgidy. (i will try and get a pic up so you guys see what i am talking about....)

why i like the gantry setup over a column:
with the column setup the table is supported in a small portion of the of the complete table (size of the saddle) making it so if you get a good amount of weight on the end of the table and extend it to its extent in anyone way there is a possiabliity of the table flexing up and down do to the weight being so far away from the support (the saddle)

with a gantry (atleast my setup) the rails run the whole lenght of the machine (bigger then the table) so no matter where the table is on the machine it has the same support on the rails....

also with the gantry setup you can have more cutting area in a smaller package.... (sounds weird right).... my machine is 30" wide (across the y axis) and i have 23.75" of cutting travel on the y... on a column mill (if you have a 8in saddle) your table would have to be 31.75 in wide... well you say 1.75" thats not a big different but a 8in saddle trying to hold a 31.75in table you are going to have a flex issue if not you machine tipping over.... to make it worth while you are going to need a 12-14in saddle to try and hold the weight of the table (if not even bigger) meaning your machine is now you have a 35-37" machine with the possibility of table flex...

just my thoughs...... what do you guys think

Mike

[slashmaster]

Originally Posted by Hellbringer also with the gantry setup you can have more cutting area in a smaller package.... (sounds weird right).... my machine is 30" wide (across the y axis) and i have 23.75" of cutting travel on the y... on a column mill (if you have a 8in saddle) your table would have to be 31.75 in wide... well you say 1.75" thats not a big different but a 8in saddle trying to hold a 31.75in table you are going to have a flex issue if not you machine tipping over.... to make it worth while you are going to need a 12-14in saddle to try and hold the weight of the table (if not even bigger) meaning your machine is now you have a 35-37" machine with the possibility of table flex... just my thoughs...... what do you guys think Mike

You can have a few inches more than that 23.75" of cutting travel if you have the table partially come out of the saddle but of course it becomes less rigid when you do that. Sometimes I wonder how well it would work out if you had casters on springs pushing on the floor on each end of the table, so you could have a table a mile long that wouldn't flex so much and wouldn't tip.

[awerby]

[QUOTE=Hellbringer;635826]a little more info on where i am comming from and my idea.....

the mill that i am drafting right now is somewhat like Zach G's with a little different ways of moving the axis... the gantry does not move (this way i can weld or overly bolt the gantry to the base to help with flex) the spindle on the gantry moves left and right for the x axis and the whole gantry moved up and down (using to steppers to offset the weight) and then i have the table move front and back for the y axis... the way i look at it i am moving more differnt parts then the same part over and over again trying to help with ridgidy. (i will try and get a pic up so you guys see what i am talking about....)

[You're losing me when you're talking about the whole gantry moving up and down. I can't see why you'd want to do that; you lose rigidity, and you've got a lot more weight to move than if you simply moved the Z-axis back and forth and up and down, like most people do. What are you gaining this way? Maybe the picture will make more sense than what you're saying here...]

Andrew Werby
www.computersculpture.com

[Zach_G]

Hey Hellbringer, what you have in mind is definitely the most ridgid setup possible. A fixed gantry or bridge mill with a supported table is certainly more stout than a moving gantry. The only reason I went with a moving gantry is to eliminate the extra support structure and to reduce the machine footprint, with a sacrifice for some ridgidity and an increased mass to accelerate, but I think my nema 34 steppers will be up to the task Correct me if I'm wrong but the attached picture is the axis setup you're going for.

Depending on the dimensional configuration of your machine, I definitely agree that moving the entire Y axis along the Z is much better than having a Z column move along the Y (unless there's very limited Z travel) simply because having an extended Z column from the bridge will create a large moment. Might as well just do a conventional column mill at that point.



Shun the nonbeliever, (awerby) for the metal cutting gantry cometh! That is assuming I ever get around to building the thing I might try and do a FEA comparison to see just how bendy it really is...

[Hellbringer]

yes that is what i am working on i guess it is called a bridge mill (learn something new everyday)....

thanks for the insight
Mike

[CalG]

A gantry passes over stuff.

Call 'em what you like, and build 'em strong.

both designs suffer two obstructions to the work.



CalG

[Andre' B]

Many of the best high performance mills are bridges.
One advantage is the mass being moved by the X and Y servos is closer to the same then any other configuration. That helps to make nice round circles at speed.
And if the Z is just a ram then it can move very fast which is nice for high speed hard milling where you tend to get lots of very small cuts were the Z is moving up to a safe level when repositioning for the next pass.