[louieatienza]

The longest section that will be machined will be ~6'. This is doable if you open up the side doors on our CNC and do incremental moves. Since there will be holes, you can use the position of the last hole to set a new coordinate system each time. Than it is rinse and repeat. Machining components is not my concern. I have enough experience, with 1.5 years of tool and die school, as well as as 2-3 years experience with all CNCs and CAM. My only concern is rigidity of the machine. I am starting to think though if I have a 175 lb slab of aluminum going 6' across on the Y, it should be stable enough to handle machining high density foam.

That being said, I'll be posting my whole design. There is a lot I don't want to type out, and I'll be posting pictures once I finish my current revamp.

As far as the router head is concerned, we will be using Bosch MR23EVS (as that is what is current CNC). It has an 18k spindle with a vaccum and will work just fine.

FYI: The design in the video is very similar to mine, with various tweaks.

I would be surprised if you didn't have at least .002" TIR in the router itself... not to mention the likely inaccurate taper. I wouldn't be the only person here in the forum who's seen that in Bosch routers of late. Just saying, in the quest of "perfection"...

[OneWound]

I would be surprised if you didn't have at least .002" TIR in the router itself... not to mention the likely inaccurate taper. I wouldn't be the only person here in the forum who's seen that in Bosch routers of late. Just saying, in the quest of "perfection"...

Interesting. Where does this .002 come from? The taper or?

[dmalicky]

I understand that reasonably accuracy is probably fine at .030, but I'm going to holder better tolerances. Otherwise there is no reason to build this CNC (what I am designing is replacing another CNC).

1. How many people are on the project, and what is their experience level with machining/fabrication and design/build/develop projects? Some examples of their past design/build projects would be great to know. There are two people on this project. Me and the person handling servos, controllers etc. The other person is a CE major. I am designing the CNC. There is already an existing frame that this will mount to, therefore covering the 10' requirement. A lot will be answered once I finish my revamp of the current design. I am also the person responsible for the assembly specs. That being said, everything that has been designed is able to be machined by a Haas VF2. It may be a headache on some pieces, but it is doable. To note, this will be reviewd by lots of engineering students. As well as this forum. And maybe a couple professors on campus.
2. How long has the team already worked on this project? I've worked on it...20-30 hours of actual work time? I don't keep count.
3. How many total hours/week can the key team members dedicate to it in the future? As much as possible
4. How many weeks does the team have to finish it? Probably a month
Hopefully I'll be posting the final design at the end of this week.

FYI: Making all machined components out of steel is not out of the question, but I'd rather avoid it for now.

Thank you, that's very helpful info. To be honest, I'm concerned about the match between project scope (this is a very ambitious project) and project resources (1 person and 1 month to do all mechanical work, including redesign iterations and parts procurement, and I assume you need to get passing grades with a full courseload). This isn't meant as a criticism of you -- actually, I think it's awesome that you have the shop training and experience that you do -- you will be a great mechanical engineer. Rather, I'm trying to help find a path to project success, which I think means de-scoping the project a lot.

I realize I don't know you or much about your capabilities. Maybe you are super productive, experienced, and can dedicate tons of hours a week to this. And maybe the scope is less than I realize, if you can reuse most parts from the old machine. 3 clarification questions:
1. It's great you want to work on it "as much as possible" -- what does that mean in hours per week -- 10, 20, 30...?
2. What is the most complex design/build project that you've done up to now?
3. What year are you, in particular, have you taken Mechanics of Materials and Machine Design yet?

On the gantry cross-section, in Mechanics of Materials there is the Moment of Inertia, I, and the Polar Moment of Inertia, J.
The gantry is under combined bending and torsion. To resist that, it needs a high I and J.
For it to accelerate quickly and achieve fast run times, it needs to be reasonably light weight.
The best solution for both of those is a hollow square or rectangular tube, not a solid bar. Solid is good for pure tension or short-column compression; for any other loading, tubes rule.
See these posts for numbers and theory:
http://www.cnczone.com/forums/diy-cn...ml#post1413406
http://www.cnczone.com/forums/diy-cn...ml#post1419700

To machine the 6' section, I'm unclear how you will 'rinse and repeat' to achieve a flat rail mounting surface. Translating the X and Y is fairly easy, like you described. Translating and achieving a coincident Z surface is a lot harder (maybe you have a plan for this already). The back of the gantry crossmember (tube or bar) is not precision machined, so when it slides down, the surface-just-machined may no longer be parallel to the table. Will you indicate the surface and shim the new mounting to 0.001" resolution? Or indicate and then write code to follow parallel to the old surface?

[OneWound]

Thank you, that's very helpful info. To be honest, I'm concerned about the match between project scope (this is a very ambitious project) and project resources (1 person and 1 month to do all mechanical work, including redesign iterations and parts procurement, and I assume you need to get passing grades with a full courseload). This isn't meant as a criticism of you -- actually, I think it's awesome that you have the shop training and experience that you do -- you will be a great mechanical engineer. Rather, I'm trying to help find a path to project success, which I think means de-scoping the project a lot.

I realize I don't know you or much about your capabilities. Maybe you are super productive, experienced, and can dedicate tons of hours a week to this. And maybe the scope is less than I realize, if you can reuse most parts from the old machine. 3 clarification questions:
1. It's great you want to work on it "as much as possible" -- what does that mean in hours per week -- 10, 20, 30...?
2. What is the most complex design/build project that you've done up to now?
3. What year are you, in particular, have you taken Mechanics of Materials and Machine Design yet?

On the gantry cross-section, in Mechanics of Materials there is the Moment of Inertia, I, and the Polar Moment of Inertia, J.
The gantry is under combined bending and torsion. To resist that, it needs a high I and J.
For it to accelerate quickly and achieve fast run times, it needs to be reasonably light weight.
The best solution for both of those is a hollow square or rectangular tube, not a solid bar. Solid is good for pure tension or short-column compression; for any other loading, tubes rule.
See these posts for numbers and theory:
http://www.cnczone.com/forums/diy-cn...ml#post1413406
http://www.cnczone.com/forums/diy-cn...ml#post1419700

To machine the 6' section, I'm unclear how you will 'rinse and repeat' to achieve a flat rail mounting surface. Translating the X and Y is fairly easy, like you described. Translating and achieving a coincident Z surface is a lot harder (maybe you have a plan for this already). The back of the gantry crossmember (tube or bar) is not precision machined, so when it slides down, the surface-just-machined may no longer be parallel to the table. Will you indicate the surface and shim the new mounting to 0.001" resolution? Or indicate and then write code to follow parallel to the old surface?

Yes I have taken mechanics of materials, no to machine design.

For the gantry section, I do agree that tubes are superior. The problem lies in tolerances however. From the tube manufactures that I would use, the twist and tolerance stack up would be too great for me to machine the tube nice and flat. That's why I am using aluminum bar, it doesn't have the same twist tolerances as aluminum bar for example. I understand I am sacrificing weight, and if it becomes a problem I can always machine material away.

In regards to machining it, measuring Z offset for each section will be critical (and done with a Reinshaw probe). Once one section is machined flat and you move it, the flat section can be used to level the 6' piece. While this will suck, a piece can also be made to support the ends of the bar stock while it is held in the vice.

I don't know how much time I'll devote to this. There is no set deadline for completion of this. However, I am close to finishing this design as I've already done two iterations of it.

I don't know the most complex design I've worked on. I do understand it's a big project, but that's half the fun.

[DonKes]

Two iterations, YOu are just getting started.

Bank to fundamentals.

If you keep looking for perfection, you will never get to the finish line.

What are the tolerances of the end product you are producing?

Don

[OneWound]

Two iterations, YOu are just getting started.

Bank to fundamentals.

If you keep looking for perfection, you will never get to the finish line.

What are the tolerances of the end product you are producing?

Don

The tolerances depend on the part. This machine will produce carbon fiber molds for experimental cars that use solar power. Ie, some parts will have more critical tolerances than others and I can't predict them currently

[dmalicky]

It's important to know the critical tolerances that the users will need. Else the customer requirements cannot be defined, and the machine will likely fail to do its job. The engineer's first job is to define the customer, then the customers' requirements, then the specs to meet the customer requirements, then develop the design. Of course there massive background research at every stage, and lots of iteration among those tasks.

That's great you have looked at the tolerances. It's true the stated tolerances for tubes are worse. But:
- Actual tubes are almost always better than the spec. A500 steel is quite straight and flat, in practice, as it goes through cold finishing. Alum, not so much, as it is hot out of the extruder, but still plenty straight to make work.
- ~All commercial routers and most of the heavy duty DIY routers use tubes for their gantry (usually steel, but alum can work). It is a solveable problem, not difficult. This is an example of why background research is so important.
- A standard solution is to attach (epoxy, bolt, and/or rivet) sacrificial strips for leveling purposes. Commercial and some DIY routers weld on the strips, but then the gantry needs to be stress relieved.

Machining a solid aluminum bar is not without its own problems. T6 has substantial internal stresses, may warp after machining. It will definitely warp if you machine enough to lighten it, and unless you understand structural mechanics very well, lightening it will turn it into a noodle for stiffness. Almost no one uses solid thick bar stock for a structural member like this -- engineers can find much better solutions.

That leveling process is clever, and I agree it will be a pain. As you may know, if you apply the any prestress to the bar during fixturing, there will be spring back, and the surfaces won't be parallel once released. If it's a solid bar, its self weight will be an issue, since solids have a poor ratio of inertia/weight. Standard practice is to find a job shop with a big enough machine to do it in 1 setup. It's not expensive, and your project has the budget.

As DonKes said, 2 iterations is the beginning, especially for an engineer new to a field. A sound design would probably go through at least 10 major iterations/explorations. But only after months of background research to develop expertise and learn standard solutions. I would spend at least 100-200 hours on the design before ordering parts, and I've done this before.

From your posts, my impression is you are skilled at machining, but a beginner at major design/build projects. If so, there's nothing at all wrong with this stage -- we all start there. It's admirable that you want to undertake an ambitious project like this. It is indeed a lot of fun. But the project needs to be a success -- and so far, I don't sense this project is headed that way. The substantial project budget would go to waste on an unsuccessful machine. My best advice is to redefine the project scope so it is a better match to your stage. Maybe find a reasonable way to improve the existing machine -- what are its weak points? Maybe just a new Z axis. Or a better drive system for another axis. Or a much smaller scale and scope CNC machine. You'll learn a lot from something like that, and the result is much more likely to be a high quality contribution. And then you're better prepared for the next, bigger project.

[OneWound]

It's important to know the critical tolerances that the users will need. Else the customer requirements cannot be defined, and the machine will likely fail to do its job. The engineer's first job is to define the customer, then the customers' requirements, then the specs to meet the customer requirements, then develop the design. Of course there massive background research at every stage, and lots of iteration among those tasks.

That's great you have looked at the tolerances. It's true the stated tolerances for tubes are worse. But:
- Actual tubes are almost always better than the spec. A500 steel is quite straight and flat, in practice, as it goes through cold finishing. Alum, not so much, as it is hot out of the extruder, but still plenty straight to make work.
- ~All commercial routers and most of the heavy duty DIY routers use tubes for their gantry (usually steel, but alum can work). It is a solveable problem, not difficult. This is an example of why background research is so important.
- A standard solution is to attach (epoxy, bolt, and/or rivet) sacrificial strips for leveling purposes. Commercial and some DIY routers weld on the strips, but then the gantry needs to be stress relieved.

Machining a solid aluminum bar is not without its own problems. T6 has substantial internal stresses, may warp after machining. It will definitely warp if you machine enough to lighten it, and unless you understand structural mechanics very well, lightening it will turn it into a noodle for stiffness. Almost no one uses solid thick bar stock for a structural member like this -- engineers can find much better solutions.

That leveling process is clever, and I agree it will be a pain. As you may know, if you apply the any prestress to the bar during fixturing, there will be spring back, and the surfaces won't be parallel once released. If it's a solid bar, its self weight will be an issue, since solids have a poor ratio of inertia/weight. Standard practice is to find a job shop with a big enough machine to do it in 1 setup. It's not expensive, and your project has the budget.

As DonKes said, 2 iterations is the beginning, especially for an engineer new to a field. A sound design would probably go through at least 10 major iterations/explorations. But only after months of background research to develop expertise and learn standard solutions. I would spend at least 100-200 hours on the design before ordering parts, and I've done this before.

From your posts, my impression is you are skilled at machining, but a beginner at major design/build projects. If so, there's nothing at all wrong with this stage -- we all start there. It's admirable that you want to undertake an ambitious project like this. It is indeed a lot of fun. But the project needs to be a success -- and so far, I don't sense this project is headed that way. The substantial project budget would go to waste on an unsuccessful machine. My best advice is to redefine the project scope so it is a better match to your stage. Maybe find a reasonable way to improve the existing machine -- what are its weak points? Maybe just a new Z axis. Or a better drive system for another axis. Or a much smaller scale and scope CNC machine. You'll learn a lot from something like that, and the result is much more likely to be a high quality contribution. And then you're better prepared for the next, bigger project.

If a customer requirement must be developed, I'll state it here as +/- .010. Since I am apart of the customer as well of the developer, I do hope this is adequate.

In regards to fixing the CNC, that is a no-go. It's a machine that was built in 1995 and needs a new Z axis assembly, new bearings (that aren't made anymore), bad servos, no soft stops, and the list goes on.

While I understand I'm in over my head, I'm here to complete a task. The task being design and build a CNC gantry so we, we being a student organization, can machine 4x8 foam molds starting in the fall of 2016 (I was just told this information last night). I feel that, with enough questions and y'alls input, it should be doable.

I doubt that I will machine the aluminum thin enough for it to wrap. The only thing I would be doing would be taking ~.05 off each side, and drilling mounting holes (I'll post pictures tonight). In regards to prestress, there won't be any prestress (i.e. From the bar hanging unsupported from one end) because it'll be supported when it's being machined. That being said, this probably won't be the final solution. I'm starting to wonder if a custom (machined, of course) I-beam that is thicker on the top to allow for the mounting of linear rails. It'll be in 4 sections, mated together with connecting plates that use bolts and dowel pins for alignment and than welded. This would be much easier to machine, as well as handle.

[OneWound]

Pictures, as promised. Anything circled in black is areas where dowel pins will be used for alignment. Another note, the X axis plate (labeled in picture) will have mounting plates installed to connect it to the Y axis and it will also have holes for dowel pins. This has not been modeled, however.

Also attached is the CNC I will be replacing. The only thing being kept will be the frame, as it is the only usable part..Attachment 315358Attachment 315360Attachment 315362Attachment 315364Attachment 315366

[louieatienza]

I'm not sure I'd rely on just one linear rail on your gantry. I probably wouldn't rely on the ballscrew absorbing any loads perpendicular to the gantry axis either. Also if you need that amount of Z travel, I think you'd be better served having the rails on the moving carriage and the bearing blocks on the saddle. But this only in conjunction with the other suggestions I made.

[OneWound]

I'm not sure I'd rely on just one linear rail on your gantry. I probably wouldn't rely on the ballscrew absorbing any loads perpendicular to the gantry axis either. Also if you need that amount of Z travel, I think you'd be better served having the rails on the moving carriage and the bearing blocks on the saddle. But this only in conjunction with the other suggestions I made.

What is the benefit to having the rails on the moving carriage? Wouldn't that cause more weight to move?

Second question, I'm assuming that you are implying you want the ballscrew between the two linears rails so it sees less force?

[louieatienza]

What is the benefit to having the rails on the moving carriage? Wouldn't that cause more weight to move?

Second question, I'm assuming that you are implying you want the ballscrew between the two linears rails so it sees less force?

I have a Z carriage weighing about 20 lbs with an older Bosch 1617EVS installed. I use a 1/2"-8, 8 start screw (1TPI or 1.0" lead) and a 380in-oz stepper with 48V. I have absolutely no problems rapiding it at over 200ipm.

On your drawings, I only see 1 rail on the gantry. And only 1 block? And if you're going to have the saddle extend down so far, then why bother with gantry legs? As to the ballscrew, you do not want any side loads on it, or it can cause binding/premature wear. The profile bearing blocks are actually designed to take up some error in height difference from rail to rail, so I would predict between that and the large unsupported lever of a Z plate, you'd have way more deflection than the accuracy you plan to achieve.

With your supposed budget, I'd have two bearing blocks per rail, and two rails, on the gantry, and same on the Z carriage.

[OneWound]

I have a Z carriage weighing about 20 lbs with an older Bosch 1617EVS installed. I use a 1/2"-8, 8 start screw (1TPI or 1.0" lead) and a 380in-oz stepper with 48V. I have absolutely no problems rapiding it at over 200ipm.

On your drawings, I only see 1 rail on the gantry. And only 1 block? And if you're going to have the saddle extend down so far, then why bother with gantry legs? As to the ballscrew, you do not want any side loads on it, or it can cause binding/premature wear. The profile bearing blocks are actually designed to take up some error in height difference from rail to rail, so I would predict between that and the large unsupported lever of a Z plate, you'd have way more deflection than the accuracy you plan to achieve.

With your supposed budget, I'd have two bearing blocks per rail, and two rails, on the gantry, and same on the Z carriage.

Actually, I'm updating my design relatively close to what you are suggesting. I still need to do some calculations to determine some aspects of it, as well as mull over design decisions.

[dmalicky]

Thanks for the pictures. Yes, the old design is not really upgradable, except the table is salvageable. I can see your tool and die background in your new design.

OK, I understand you have an assignment. This is now reasonable since you have until Fall. Can you work over the summer on it (that will be needed)? It's also good that you're open to a different gantry cross-section. There is a tendency for new designers to "lock-on" to their early designs, resistant to change. This is especially true once the design is put into 3D CAD, since that took some work and it looks "real". It's hard to give up that work, because it feels like we're moving backwards -- but the opposite is true. This tendency is a major handicap, since it prevents exploration of much better concepts -- concepts that can actually save a lot of work, reduce cost, and meet or exceed the customer requirements. 2 quotes that have served design engineers well:

-- "If you want to have good ideas you must have many ideas. Most of them will be wrong, and what you have to learn is which ones to throw away." -- Linus Pauling

-- "Nothing is more dangerous than an idea, when it's the only one we have. -- Emile Chartier

One key to good design is being un-attached to our prior work. Rather Buddhist. This helps us be open to the best ideas. Your project is still in the early stages. The best strategy at this point is to expand your options -- then you can pick the best ones.

I suggest researching more about CNC routers -- theory and benchmarking. What designs have commercial routers settled on? And well-executed DIY? The hardest thing about designing in a new field is becoming an expert. The expert can hone in on great designs quickly. The beginner can't tell the difference. The faster the designer gains expertise, the better the design path.

Even 0.050" off the side can cause warp, if you are shooting for very high straightness -- remember the highest residual stresses are on the surface. Unless it is evenly supported all along its length, there will be prestress.

On I-beams, look up the polar moment of inertia (J) of one, and compare to a square or rectangular tube of similar dimensions and weight/foot. You may be amazed in the power of the closed section. Then imagine what else you could do with those 4 plates, as one option.

Think about how you will level the rail surfaces of the existing frame. Also consider bracing the long green supporting beam in the middle (bolted, not welded) with triangulation. 10' is a long length for a precision unsupported beam of that size.

I wouldn't count on the dowel pins for perpendicularity of the gantry to the X axis -- the gantry is too long to be precisely squared by pins in a small pattern. Perpendicularity is usually achieved by the sync of the 2 X-drive systems (almost always rack and pinion for an axis greater than ~5').

As Louie mentioned, the gantry will need 2 rails, with 2 blocks each. And the ballscrew is only for axial force/position. For best stiffness at the cutter, the ballscrew is ideally placed as low as possible -- just above the lower of the 2 gantry rails.

[louieatienza]

Actually, I'm updating my design relatively close to what you are suggesting. I still need to do some calculations to determine some aspects of it, as well as mull over design decisions.

If I were you, and you have your mind set on using an aluminum billet, you might want to scrap the piece you have now and just get a drop-off piece of Mic-6 cast aluminum tooling plate. Far easier to machine, and already Blanchard ground to within .005" flatness from 3/4" to 4". All you have to do is machine the edges and ends, and drill and tap your holes. Done.

[OneWound]

Thanks for the pictures. Yes, the old design is not really upgradable, except the table is salvageable. I can see your tool and die background in your new design.

OK, I understand you have an assignment. This is now reasonable since you have until Fall. Can you work over the summer on it (that will be needed)? It's also good that you're open to a different gantry cross-section. There is a tendency for new designers to "lock-on" to their early designs, resistant to change. This is especially true once the design is put into 3D CAD, since that took some work and it looks "real". It's hard to give up that work, because it feels like we're moving backwards -- but the opposite is true. This tendency is a major handicap, since it prevents exploration of much better concepts -- concepts that can actually save a lot of work, reduce cost, and meet or exceed the customer requirements. 2 quotes that have served design engineers well:

-- "If you want to have good ideas you must have many ideas. Most of them will be wrong, and what you have to learn is which ones to throw away." -- Linus Pauling

-- "Nothing is more dangerous than an idea, when it's the only one we have. -- Emile Chartier

One key to good design is being un-attached to our prior work. Rather Buddhist. This helps us be open to the best ideas. Your project is still in the early stages. The best strategy at this point is to expand your options -- then you can pick the best ones.

I suggest researching more about CNC routers -- theory and benchmarking. What designs have commercial routers settled on? And well-executed DIY? The hardest thing about designing in a new field is becoming an expert. The expert can hone in on great designs quickly. The beginner can't tell the difference. The faster the designer gains expertise, the better the design path.

Even 0.050" off the side can cause warp, if you are shooting for very high straightness -- remember the highest residual stresses are on the surface. Unless it is evenly supported all along its length, there will be prestress.

On I-beams, look up the polar moment of inertia (J) of one, and compare to a square or rectangular tube of similar dimensions and weight/foot. You may be amazed in the power of the closed section. Then imagine what else you could do with those 4 plates, as one option.

Think about how you will level the rail surfaces of the existing frame. Also consider bracing the long green supporting beam in the middle (bolted, not welded) with triangulation. 10' is a long length for a precision unsupported beam of that size.

I wouldn't count on the dowel pins for perpendicularity of the gantry to the X axis -- the gantry is too long to be precisely squared by pins in a small pattern. Perpendicularity is usually achieved by the sync of the 2 X-drive systems (almost always rack and pinion for an axis greater than ~5').

As Louie mentioned, the gantry will need 2 rails, with 2 blocks each. And the ballscrew is only for axial force/position. For best stiffness at the cutter, the ballscrew is ideally placed as low as possible -- just above the lower of the 2 gantry rails.

What are the benefits of a rack and pinion system over a ballscrew? The only thing I can would be potentially longer service life. As noted before, I may go away from the long aluminum section and instead go to dual linear shaft (yet to be determined, I still need to do calculations.) If I do have two linear shafts that span the 5' gap (with a ballscrew in between them, of course), would it be suggested to still have a gantry piece to span the 5' for improved rigidity?

In regards to leveling the x rail, I'll be leveling it to the bed (there is an iron frame underneath the foam/wooden bed). I'll get it as accurate as I can, but I don't trust that bed is 100% flat. Once one X linear rail is set, however, the other linear rail will be parallel to the first linear rail (dial indicator/Faro arm will be used).

[louieatienza]

What are the benefits of a rack and pinion system over a ballscrew? The only thing I can would be potentially longer service life. As noted before, I may go away from the long aluminum section and instead go to dual linear shaft (yet to be determined, I still need to do calculations.) If I do have two linear shafts that span the 5' gap (with a ballscrew in between them, of course), would it be suggested to still have a gantry piece to span the 5' for improved rigidity?

In regards to leveling the x rail, I'll be leveling it to the bed (there is an iron frame underneath the foam/wooden bed). I'll get it as accurate as I can, but I don't trust that bed is 100% flat. Once one X linear rail is set, however, the other linear rail will be parallel to the first linear rail (dial indicator/Faro arm will be used).

The main benefit of rack and pinion is performance to price ratio. You can try pricing a precision ground ballscrew for your 6' gantry; it would blow a significant portion of your budget, and you'd need probably at least a 36mm screw to prevent sag. This screw will be VERY heavy, and you'll need a powerful servo to overcome the inertia of the screw. I would have concerns using a rolled ballscrew of that length and diameter, especially at higher speeds. I'd then consider a high lead ballscrew, and use a smaller servo with planetary gear reduction, as you probably couldn't spin the rolled ballscrew as fast as a precision ground one. I would NOT recommend unsupported linear shafting.

Like I mentioned, a helical rack/pinion set would give you the speed you desire and need, with very good accuracy (I'd say less than .005"). I would implement the spring pinion and servo planetary reducer from Atlanta Drive, though this would require some precision in setup. They do make a helical rack that bolts on with your profile linear rail, making alignment of rack to rail easier. There is at least one member here (spoiledbrat) in the CNC router project log subforum that utilized their system for a 4 x 8 gantry CNC.

Many high end commercial portal style machining centers use rack and pinion, although they employ two pinions and two motors, master and slave, that electronically preload against each other or work in tandem depending on the move.. Check out the bad boy of the bunch:
https://www.youtube.com/watch?v=DtlaBssb0gE
https://www.youtube.com/watch?v=nNP0NnVzqno
That's maybe .001"

[louieatienza]

Just to give you a scope on accuracy. Here's F Zimmermann's low cost model. Even here, positional accuracy is around .002" for the 8', and .003" for the 12". And I'm pretty sure this is way beyond the scope of your design.

FZ30 / FZ35

[OneWound]

Just to give you a scope on accuracy. Here's F Zimmermann's low cost model. Even here, positional accuracy is around .002" for the 8', and .003" for the 12". And I'm pretty sure this is way beyond the scope of your design.

FZ30 / FZ35

I can't remember the link, but I saw a CNC with a positional accuracy of .001 and .005 of repeatability. This was on a 4'x8' model.

PS sorry for the late reply, work was busy this weekend

[OneWound]

The main benefit of rack and pinion is performance to price ratio. You can try pricing a precision ground ballscrew for your 6' gantry; it would blow a significant portion of your budget, and you'd need probably at least a 36mm screw to prevent sag. This screw will be VERY heavy, and you'll need a powerful servo to overcome the inertia of the screw. I would have concerns using a rolled ballscrew of that length and diameter, especially at higher speeds. I'd then consider a high lead ballscrew, and use a smaller servo with planetary gear reduction, as you probably couldn't spin the rolled ballscrew as fast as a precision ground one. I would NOT recommend unsupported linear shafting.

Like I mentioned, a helical rack/pinion set would give you the speed you desire and need, with very good accuracy (I'd say less than .005"). I would implement the spring pinion and servo planetary reducer from Atlanta Drive, though this would require some precision in setup. They do make a helical rack that bolts on with your profile linear rail, making alignment of rack to rail easier. There is at least one member here (spoiledbrat) in the CNC router project log subforum that utilized their system for a 4 x 8 gantry CNC.

Many high end commercial portal style machining centers use rack and pinion, although they employ two pinions and two motors, master and slave, that electronically preload against each other or work in tandem depending on the move.. Check out the bad boy of the bunch:
https://www.youtube.com/watch?v=DtlaBssb0gE
https://www.youtube.com/watch?v=nNP0NnVzqno
That's maybe .001"

In regards to sag over a 10' span..I was plan on using a 25mm lead screw, you think that would sag under its own weight?

Also, would there be concern for sagging over a 2" 5' 1566 linear shaft (under its own weight)?

To everyone else, I hope to have a CAD model sometime this weekend...I've been busy the last few days :tired: