[Sea Shadow]

Hello everyone,

I've been a long time lurker on the forums and am finally getting ready to build my own machine, my budget is currently ~$6k but I'm willing to allocate more as necessary. My goal is mostly wood routing with the ability to do some aluminum and brass.

I am still just doing rough sketches of my design, but I'm leaning towards the idea of a fairly large router with a rail length of ~5.5'x10'x1.5' to achieve a cut area of 5'x9'x1'. That way I have the option to add a tool rack and ATC spindle later on. I am leaning towards the use of HiWin HGW25 series rails with a pair of blocks per rail, and also having fixed blocks/moving rails for the Z axis.

The machine will have dual screw for the X axis and single screw for the Y axis (Gantry), along with single screw for the Z axis. The tablet itself will be made from a mixture of 3060, and 3030 extrusion.

I was looking at the prices/weights of different extrusions and I found myself stuck with a decision to make for my gantry. Do I want to use a single piece of 3060, or should I go for multiple pieces of 2040 and 2020 series extrusion? The span is 1700mm

Current options for my gantry:

• A single span of 3060.
• A pair of 2040 extrusions bolted together with plates along the joint making a square beam.
• A pair of 2040 extrusions and a 2020 extrusion bolted to form a "C" shape with the leadscrew travelling in the channel between the two.

In any case I'm looking at mounting the linear rails to the top and bottom of the gantry so that I can make the front to back depth as thin as possible. If this is a bad idea and I should mount both rails to the front of the gantry, please say so.

How does this rough outline sound so far?

[blau_schuh]

Have you downloaded the 8020 deflection tool yet? You may want to play around with it...

I think your 3060 single rail is going to deflect quite a bit at that length, may not matter with what you're doing. Sounds like you may be doing a dual screw though, so in that case it wouldn't matter since you can just add a support in the center.

I have roughly a 72" run and with a decent sized load in the center (75lb point load) I in theory will see something like 0.0007" of deflection.

[Sea Shadow]

Thank you for your response, I have played with the 8020 deflection tool but was unable to figure out how to handle combined beam load.

I realize now that my first explanation may not be that clear. The table itself is going to be a pair of 3060 beams on edge down the length of the bed, with 3030 support pillars and cross beams to make up the rest of the table.

My dilemma has to do with the gantry itself, I'm trying to figure out what would offer the greatest strength for a roughly comparable price. The span is ~1700mm

[blau_schuh]

ahh i see now...

I assumed a worst case point load with each 3060 length taking the whole load of the gantry and it came out under a thou and i was happy. Sounds like your gantry width is similar to my frame length but you're load at least from gravity is going to be way less since it's just a spindle you're whipping around and not a whole gantry. And you're linear rail load is evenly distributed. Not sure what the side load (perpendicular to the 6" side) would be when doing heavy cutting though.

I think a single 3060 would be fine if all 6 holes on each side are tapped and fastened to some heavy plate.... I currently have my 52" wide gantry fastened to two pieces of 40 series (roughly 1530 equivalent) since that's what came with the linear actuator I'm using. I may switch at some point to 3030... not sure a single 3060 is an option due to linear actuator hole spacing.

I think the rails on top and bottom are OK too...

Have you thought about a single 3060 span and marrying a steel or aluminum stiffener plate (or c channel) on the back side?

3060 ... $$$$ starts to add up quick. I've got $1500 (easily) into extrusions so far.. probably $900 into just the 3060. ouch... $6k is a good budget .

[pippin88]

Consider a big rhs steel beam.
Google "moment of Inertia shs" and compare it to the 3060.

Both steel and aluminum extrusion require making a flat surface to mount rails for best results.

15mm or 20mm rails will be more than strong enough, so look into the cost of them. The advantages of bigger rail include bigger mounting holes. Depending on application the larger height might be good (allow ballscrew mounting alongside easier) or bad (overhang of Z axis)

[wizard]

You are looking at a seriously large machine and frankly I don't think 80/20 has anything suitable for it. As pipped above mentioned you need to look into a steel beam and it will likely end up being a rather large beam. Ultimately the size of the beam will be impacted by your expectations but any serious aluminum or brass work will require a far stiffer machine that one suitable for wood. You also need to look closer at the risers holding that beam up in the air. All of this will have to sit upon a non trivial frame to support the work properly. I suspect you will over run your budget with such a large machine.

You might want to consider shrinking the machine to a few inches larger than the sheet goods you expect to work on. This to control costs. If needed you can implement a carousel or other solution to do tool changing.

[ger21]

I am still just doing rough sketches of my design, but I'm leaning towards the idea of a fairly large router with a rail length of ~5.5'x10'x1.5' to achieve a cut area of 5'x9'x1'. That way I have the option to add a tool rack and ATC spindle later on.

A rail length of 66" is most likely not nearly long enough to have a 60" cut length.
What you should be doing is deciding exactly how far you want the tool to travel, and work out the rail length based on your requirements.

I'll second the big steel beam. Easily the stiffest for the money.

[Sea Shadow]

... Have you thought about a single 3060 span and marrying a steel or aluminum stiffener plate (or c channel) on the back side?...

Consider a big rhs steel beam.
Google "moment of Inertia shs" and compare it to the 3060.

Both steel and aluminum extrusion require making a flat surface to mount rails for best results.

I had not considered either option, both are definitely worth considering. A married stiffener could also double as a support for a cable carrier. And a steel beam might just be a better option over all, though how would one create a flat surface to mount the rails?
EDIT Ok so I found this thread that was talking about it. So it looks like my 2 options are to pay for machining, and pour a self leveling epoxy. I can already tell that I am going to need to do quite a bit more research into this.

Ultimately the size of the beam will be impacted by your expectations but any serious aluminum or brass work will require a far stiffer machine that one suitable for wood. You also need to look closer at the risers holding that beam up in the air. All of this will have to sit upon a non trivial frame to support the work properly. I suspect you will over run your budget with such a large machine.

I will primarily be using this machine for wood. I would be willing to take it pretty slow with lots of shallow passes for any aluminum or brass that I would do. As for the table and supporting frame, I already planned to overbuild it, that way I can upgrade the machine later should I so desire. Though realistically, what kind of center point loading should I be using for my gantry calculations, and what is acceptable displacement for wood and for metal? I had based my preliminary ideas off of a center point load of 150 lbs and that led to deflection of .0012 in and .004 in, but that was w/out accounting for the linear rails and the compressive/tensile strength they would contribute. Also not including any reinforcing plate. I really like blau_schuh's idea and am leaning towards marrying a steel C channel to the back of the beam.

A rail length of 66" is most likely not nearly long enough to have a 60" cut length.
What you should be doing is deciding exactly how far you want the tool to travel, and work out the rail length based on your requirements...

Well the 60" travel isn't a hard and fast requirement, more to the point I'm just working with what I have. I have a full set of rails that I received in trade for some work.
I have a pair of 3 meter long rails, and a pair of 1.7 meter long rails. Both are HiWin HGW 25 series along with wide blocks (standard length wide body). I also got a pair of THK 15mm rails that are 18" long in a separate trade (standard blocks and bodies, can't remember the series off the top of my head). The HiWin rails are new and the THK rails are used but have no noticeable play. So really my 6k budget doesn't need to cover the linear rails, just everything else (minus a host computer).

[wizard]

I had not considered either option, both are definitely worth considering. A married stiffener could also double as a support for a cable carrier. And a steel beam might just be a better option over all, though how would one create a flat surface to mount the rails?
EDIT Ok so I found this thread that was talking about it. So it looks like my 2 options are to pay for machining, and pour a self leveling epoxy. I can already tell that I am going to need to do quite a bit more research into this.

There are several ways to flat surfaces for mounting rails. You also need to consider what you accuracy expectations are. Finally there is a lot of variability in the quality or straightness of steel tubing, some is pretty good and may be good enough for you needs.

I will primarily be using this machine for wood. I would be willing to take it pretty slow with lots of shallow passes for any aluminum or brass that I would do. As for the table and supporting frame, I already planned to overbuild it, that way I can upgrade the machine later should I so desire.

The problem here is again expectations but with those long spans you may have significant problems machining metals well. For example some series of brasses and bronzes are very difficult to machine. There can be a big difference between machining thin aluminum sheet and some of the alloys out there.

Though realistically, what kind of center point loading should I be using for my gantry calculations, and what is acceptable displacement for wood and for metal? I had based my preliminary ideas off of a center point load of 150 lbs and that led to deflection of .0012 in and .004 in, but that was w/out accounting for the linear rails and the compressive/tensile strength they would contribute. Also not including any reinforcing plate. I really like blau_schuh's idea and am leaning towards marrying a steel C channel to the back of the beam.

I'm no expert in stress calculations but those numbers don't seem realistic at all. This especially if you are expecting 1.5 foot of Z axis travel. That sort of stick out on the Z will cause significant twisting of whatever beam you use. Here in lies a problem there are so many smaller machines out there that people have a good idea what is needed to produce a good machine. Here you are talikg significantly larger than most DiY machines so honestly it will pay to do some engineering.

Well the 60" travel isn't a hard and fast requirement, more to the point I'm just working with what I have. I have a full set of rails that I received in trade for some work.

One thing I missed before is your Z travel which is significant this will require a longer X axis to provide for solid and stable uprights to support that high gantry. Your X carriage/saddle might need as much as 1.5 feet between the linear bearings (assuming traveling gantry here). This impacts the overall length of your machine.

I have a pair of 3 meter long rails, and a pair of 1.7 meter long rails. Both are HiWin HGW 25 series along with wide blocks (standard length wide body). I also got a pair of THK 15mm rails that are 18" long in a separate trade (standard blocks and bodies, can't remember the series off the top of my head). The HiWin rails are new and the THK rails are used but have no noticeable play. So really my 6k budget doesn't need to cover the linear rails, just everything else (minus a host computer).

This makes me think you might have reasonable access to machining services. That just makes thinking about steel as a reasonable consideration. This machine isn't small by any means and the structure will impact cost. The "free" slides will reduce your money outlay significantly allocation more money to the machines structure. I'd still try to save some bucks by considering steel over aluminum. Beyond that once you get the engineering done for the long beam and the long Z, I susoect you will find that you need a rather large beam

[pippin88]

There was a great thread on gantry construction a while ago, with lots of analysis by dmalicky http://www.cnczone.com/forums/diy-cn...questions.html

Your 18" of Z travel - how much of that is to allow workpiece height? E.g. the build I'm doing has 8" of Z travel, but only 4" below the gantry. This is to allow a 4" workpiece and a 4" tool to clear it.

[Sea Shadow]

Thank you everyone for the advice and the suggestions. It is clear that I am going to have some serious planning ahead of me.

Also thank you Pippin for the link. I have spent the past several days reading it and some of the other associated links. To answer your question directly, the rails are 18" long but I'm only going for 12" of travel. And I really only want 6-10 inches for the work piece. I will have the cutter on an adjustable plate so that I can move it for extremely tall jobs, but it will predominantly be at the lower height.

I think my best bet is going to be to take some time to model my ideas and then present them for further review. A lot has changed since I first started this thread, and I appreciate all the help that I have gotten. I am going to have to completely overhaul my current plans based on what I have learned.