Fixed Gantry Design

[CatalystGilles]

Hi all,

Stumbled across this whole world a few days ago and have been putting a lot of thought into the idea of building my own router. I come from an engineering and machining background so I have a lot going for me there, but routers and the diy part with respect to cnc stuff is very new. My current idea that I have been building upon is a fixed gantry router with about 24" X, 18" Y, and 6" of Z. I would like it to be as rigid as possible without breaking the bank. Below is a render of my initial design in solidworks. The base would be out of 2" x 2" square tubing and with the gantry being 2" x 4" for the supports and 4" x 6" for the bridge. The table would be either 3/4" or 1" aluminum setup like a fixture plate. The linear guides shown are just ones I found a part file of grabcad, so they may not actually be that style when built. Overall my design is very bare bones at the moment and doesn't have stepper placement, mounting holes of everything, etc.



Couple questions.

With regards to the tubing for the frame. Do you guys know of any builds that used this to a high degree of success? Is it possible to keep the tolerances required for everything to run smoothly?

Also, are there any past threads on choosing a diameter for ball screws based on the length/weight requirements?

Similar Threads:

• Build Thread- Hybrid Design - Fixed Gantry Router for High Speed Aluminum Milling - Metal Nibbler 5
• Fixed gantry vs moving gantry
• Fixed Gantry Design Question
• Newbie- Small Fixed Gantry Design
• New Machine Build- 8020 Fixed Gantry Design

[ger21]

On the back side of the gantry, run the diagonl brace to the top of the gantry beam, instead of below it.

[wizard]

Hi all,

Stumbled across this whole world a few days ago and have been putting a lot of thought into the idea of building my own router. I come from an engineering and machining background so I have a lot going for me there, but routers and the diy part with respect to cnc stuff is very new. My current idea that I have been building upon is a fixed gantry router with about 24" X, 18" Y, and 6" of Z.

This is a very nice size for a fixed gantry machine.

I would like it to be as rigid as possible without breaking the bank.

Ultimately what you are willing to pay is up to you. I would lean towards making a rigid frame first and cutting corners on things you can easily upgrade latter. For example implement an old router for the spindle or implement acme lead screws over ball screws. A good machine tool frame often outlasts the stuff that is bolted to it and will almost certainly outlast the electronics.

Below is a render of my initial design in solidworks. The base would be out of 2" x 2" square tubing and with the gantry being 2" x 4" for the supports and 4" x 6" for the bridge.

I would tend to go slightly larger that 2" square tubing and probably add additional support for the tubing carrying the X axis linear slides. Generally you want tubing thick enough to hold the screws that mount the linear slide rails or you buy tapped linear slide rails and use through bolts.

The table would be either 3/4" or 1" aluminum setup like a fixture plate. The linear guides shown are just ones I found a part file of grabcad, so they may not actually be that style when built. Overall my design is very bare bones at the moment and doesn't have stepper placement, mounting holes of everything, etc.

You might want to make provisions for a thicker table.

Couple questions.

With regards to the tubing for the frame. Do you guys know of any builds that used this to a high degree of success? Is it possible to keep the tolerances required for everything to run smoothly?

Square steel tubing is use in all sorts of machines and special machines for automation. There are several examples of very nice machines in this forum.

As for tolerances it is really up to you and what you want. If you want to do a good job on a welded steel frame you need to spend a bit of money. First the frame needs to be stressed relieved and then machined on a mill to flatten and square everything up. Another option is hand scraping everything in, you can also do things like epoxy leveling.

In the automation industry welded steel is the common approach to machine building. Here we almost always weld pads to the steel tubing to provide for linear rail mounting and assembly mounts. At some point in a machine design you are pretty much forced to use bolted together sub assemblies. In the case of your design you need to consider how you would machine the surfaces for the X Axis rails with the uprights welded in place. This largely depends upon what machinery you have access to locally. If you have the machinery to do it then welded in place uprights are the way to go. Those uprights should have in place mounting pads for the gantry beam which get machined at the same time. Doing it all in one setup keeps surfaces square and parallel as required. Having your frame machined simplifies a lot of things when it comes to final fit up of the machine. The gantry beam should have mounting feet that match up with the pads machined on the uprights.

You can of course use other techniques that take the expense of going to a machine shop out of the equation but this will result in a lot more work on your part. There are a couple of realities that you need to be aware of. For one square tubing isn't always square, at least not to the precision a machinist might expect. Think parallelogram. Second the surfaces of square tubbing are not always flat, they can be concave (most often) or sometimes convex. This will raise hell with profile rails if you try to bolt them directly to the tubing. Thus one of the goals with machining or leveling the mounting surfaces is to get the rails to set in the same plane.

By the way don't assume that a set of components all nicely finished from the machine shop will just bolt together with out any and fitting or tweaking. It really comes down to what you expect from the machine.

Also, are there any past threads on choosing a diameter for ball screws based on the length/weight requirements?

There are a lot of threads that touch upon this but be aware your goals or expectations are a factor. Also the types of drives you use impacts lead screw selection. In the end you need to trade off a lot of factors to get the results you want. I strongly recommend getting the engineering data available from the various ball screw makers so you have a handle on the various limits of operation.

You might also note that I talk about mounting pads and feet for the gantry beam. I bring this up because I doubt that you can get the beam square enough by welding it in place. Better to have the flexibility of a bolted connection there to allow adjustments to squareness of the Y & Z axis.

In a nut shell a good part of machine design is answering the question: how will I build this thing?

[CatalystGilles]

On the back side of the gantry, run the diagonal brace to the top of the gantry beam, instead of below it.

Good catch. This also made me notice I can make the Y axis plate wider to spread out the linear guide carts as they can go past the bracing.

First, thank you for the very in depth reply. It is exactly what I was looking for when I came here.

This is a very nice size for a fixed gantry machine.

Ultimately what you are willing to pay is up to you. I would lean towards making a rigid frame first and cutting corners on things you can easily upgrade latter. For example implement an old router for the spindle or implement acme lead screws over ball screws. A good machine tool frame often outlasts the stuff that is bolted to it and will almost certainly outlast the electronics.

This is something I have thought about, but I think I would rather save up longer to get everything I really want at once.

I would tend to go slightly larger that 2" square tubing and probably add additional support for the tubing carrying the X axis linear slides. Generally you want tubing thick enough to hold the screws that mount the linear slide rails or you buy tapped linear slide rails and use through bolts.

Currently my thought was 1/4" wall for the 2" square tubing and probably 3/8" for 4"x6" on the bridge.

You might want to make provisions for a thicker table.

What do you have in mind? 2"? I need to talk to a colleague of mine, because he may be able to find a scrap piece I could have in the range of 24"x36".

Square steel tubing is use in all sorts of machines and special machines for automation. There are several examples of very nice machines in this forum.

As for tolerances it is really up to you and what you want. If you want to do a good job on a welded steel frame you need to spend a bit of money. First the frame needs to be stressed relieved and then machined on a mill to flatten and square everything up. Another option is hand scraping everything in, you can also do things like epoxy leveling.

In the automation industry welded steel is the common approach to machine building. Here we almost always weld pads to the steel tubing to provide for linear rail mounting and assembly mounts. At some point in a machine design you are pretty much forced to use bolted together sub assemblies. In the case of your design you need to consider how you would machine the surfaces for the X Axis rails with the uprights welded in place. This largely depends upon what machinery you have access to locally. If you have the machinery to do it then welded in place uprights are the way to go. Those uprights should have in place mounting pads for the gantry beam which get machined at the same time. Doing it all in one setup keeps surfaces square and parallel as required. Having your frame machined simplifies a lot of things when it comes to final fit up of the machine. The gantry beam should have mounting feet that match up with the pads machined on the uprights.

You can of course use other techniques that take the expense of going to a machine shop out of the equation but this will result in a lot more work on your part. There are a couple of realities that you need to be aware of. For one square tubing isn't always square, at least not to the precision a machinist might expect. Think parallelogram. Second the surfaces of square tubbing are not always flat, they can be concave (most often) or sometimes convex. This will raise hell with profile rails if you try to bolt them directly to the tubing. Thus one of the goals with machining or leveling the mounting surfaces is to get the rails to set in the same plane.

By the way don't assume that a set of components all nicely finished from the machine shop will just bolt together with out any and fitting or tweaking. It really comes down to what you expect from the machine.

There are a lot of threads that touch upon this but be aware your goals or expectations are a factor. Also the types of drives you use impacts lead screw selection. In the end you need to trade off a lot of factors to get the results you want. I strongly recommend getting the engineering data available from the various ball screw makers so you have a handle on the various limits of operation.

You might also note that I talk about mounting pads and feet for the gantry beam. I bring this up because I doubt that you can get the beam square enough by welding it in place. Better to have the flexibility of a bolted connection there to allow adjustments to squareness of the Y & Z axis.

My basic plan for manufacturing the frame would be weld up the base, stress relieve if possible (Same contact for the table might be able to do it), and machine the top and bottom so that they are flat and parallel. I should be able to reach everything with the bridge being removable/level-able which is a fantastic idea. Today I have been looking at methods of making it possible to level and tram everything.

In a nut shell a good part of machine design is answering the question: how will I build this thing?

I think my biggest issue is gonna be stopping at a reasonable point and not making it overbuilt and out of my budget. There is a possibility I may be working at a different company in about 9 months, which has everything and more to make a project like this far easier/achievable. If that is the case I will probably upgrade it to a 24"x36" table with the relative travels.

[handlewanker]

Hi.....your linear rails on the Y axis are not all that well supported and this is where most of the cutter forces impinge on.........fixing them at the ends only will allow them and the frame to twist when you mount it on an uneven surface.

I would suggest welding the Y axis linear rail mounts on top of multiple cross tubes welded to the side rails.

In the present design the cross tubes can be welded to the underside of the linear rail mounts and the side rails.

BTW.....you should start by designing the machine from the Z axis motor mounting and work outwards from there........a 24,000 RPM water cooled spindle is not practical for milling and a mi;lling spindle needs to have a large motor belt drive to get the torque.

I assume by the build strength you want to cut harder material than wood or plastic.

The concept sketch is too basic to really judge the construction on.
Ian.

[CNCfun&games]

Your objective in building a fixed gantry design is a very smart decision.

Your asking for input is wise because there is a lot of brainpower available here on the Zone.

The first CNC machine I built was a Bridge Mill & I have built several since then.
That first machine ran for 15 years error free. I started an overhaul 3 years ago because the DOS computer died.

I'm offering a few suggestions because I made too many mistakes on that first build!

Keep your design simple for economy sake. Select stock steel slab sides and a steel tubular cross beam at least 24 " long.
Use round linear rails, .75 or 1.00 in.

You should be able to build the main structure with only a Drill Press.

Keep us posted on your result......I'll also be watching.

W. Smith, Mission Viejo, CA.

[A_Camera]

I think that a fixed gantry design is the best way to go in every DIY CNC. It is not as sexy looking as a moving gantry, but it is easier to make and will be more robust and rigid than a moving gantry can be. The only disadvantage a fixed gantry has is that it takes up a little more space, compared with a moving gantry type with the same work area, but it is much easier to build and easier to square and keep the accuracy over time.

The things I'd do differently:

1. I would turn the table 90 degrees so that the direction of moving is along the shorter side. This makes the machine shorter but wider, and in my opinion better looking. I don't think it has any practical advantage, you can do it any way you prefer but my preference is this, and I think it is a more logical orientation as well if the Y axis movements in positive direction are less than the X axis movements.
2. I would increase the Z linear guide separation quite a bit, about twice as much as it is in your design, but at least as much as the fixed part of the Z plate is, including the missing motor holder. I think that would give a better rigidity.
3. The beam must be moved backwards, otherwise you will lose precious work area. The center of the spindle motor shaft and depending on the fixture and the spindle motor, that can be over 2".
4. Increase the support for the table slides. You can add cross tubes, but better if they are supported from below, side to side across the bottom. Especially under the spindle and along the X beam, the forces will be very large in every direction.
5. Move the diagonal beam support up to the top, especially the two facing forward, but even the two at the back work better if moved up. That is where you need most support and to reduce flexing you must move the supports as high as possible, but at least to the center of the linear guides.
6. The most important part of a fixed gantry design is along and under the beam and of course the beam itself. Basically that is the place which must be REALLY rigid, as rigid as possible, since that is where most of the forces are concentrated, so you must spend time on designing that part very carefully.

[handlewanker]

Hi,....the advantage of having the table across the frame instead of passing under the gantry means the gantry can be more beefy, virtually a solid wall for the Z to slide up and down on.

In this mode it does mean the Z axis slides and spindle are sticking out from the face of the X axis more, but the support it gets from not having to hang down in the air off the X axis slides as in a conventional CNC router is far better.

A column mill has this configuration and the stick out for the head and spindle is quite well supported all the way from top to bottom by the column.

Attached is a pic of a commercial CNC .....router?....mill?......that more or less has this layout, even if the table does pass under the X axis slightly.
Ian.

[wizard]

catalystgilles;

As I alluded to above and many people have also focused on with further comments, your layout with 2" square tubing for the X axis isn't going to cut it. Even if your goal is to machine wood I'd be reluctant to say go for it.

The problem is your 2" squares tubes have to spare a relatively long distance to get 24" of machining area. They will tend to flex and vibrate over that distance. Larger tubing will help but you need to consider a frame design that supports what ever tube size you do use. That means bracing between the rails to tie everything together.

Due to this need to improve the strength of your basic frame I'd strongly suggest going to a larger cross section on your square tubing. Heavy wall 3" would be a minimal. You could weld up a frame out of 2" tubing that would be pretty strong if you put a lot of bracing and reinforcement into it but in the end it becomes a waste of effort and time. Plus you will have more distortion and built in stress, in a heavily welded up frame. The larger tubing doesn't completely dismiss the need for bracing to tie everything together either. However the larger tubing will minimize what you need to throw into the frame to make it stiff.

By the way I'm assuming here that you want to do more than wood with this machine. Your expectations play a big role in what is suitable. The other factor is what you have available, if you have pile of 2" square tubing laying around it may very well pay to use it with more effort placed on stiffening the frame up.

Another thing to consider with your frame, weight can be friend or foe! If you need to move the machine around a lot or even transport it off site a heavy frame is a problem. However weight in a CNC machine is generally a very good thing. It can held dampen vibration but more importantly the mass can help anchor the machine as it drives the axis back and forth. You need to decide which way to trade off here but a heavy frame is generally a good thing. If you expect to machine aluminum or other softer metals a heavy stiff frame can be a requirement for passable cut quality.

[CatalystGilles]

Thank you to everyone for the replys. Definitely some needed thoughts that I am going to work on implementing. The X axis mounting would definitely be reinforced/redone to help support it more. Ideally I would like to find a blanchard ground table to mount everything to, but that is obviously very dependent on finding one for a reasonable price. My current plan is to continue working on the design over the next few weeks and probably post some updates on here when it is more refined.

[handlewanker]

Hi.....you mention the work envelope of 600 X 450 ....(24"x 18")..... which means the base plate you need will be twice this length.....a characteristic of fixed gantry routers.

You'd better start saving for a Balnchard ground steel plate at 1200mm X500mm depending on the configuration of the table and at least 20mm thick if distortion from flexing is to be avoided......that will weigh a ton.

The problem is you need to have a mill big enough to handle a steel plate that size.....even a pedestal drill will be hard pressed to handle that.

All of a sudden the steel tube all welded frame starts to make sense.

If you choose to have the table long length across the frame you can make the frame and gantry all in one piece welded from tube etc......but if you choose to have the table long length pass under the gantry you would need to have the gantry as a bolt on structure due to the need to be able to mill and drill the Y axis linear rails seats after welding.
Ian.

[wiremonkey]

Smart, making the rear support go to top of gantry!

If I might add, your bed, the long'ish 2" x 2" spans, will be a weak point. Maybe a tried and true truss, only upside down, on the bottom of those spans? Wouldn't cost too much, just more tube, and some patient welding.

I like the design.

[wiremonkey]

Ah, I see you already addressed the long Y bed issue.

[wiremonkey]

Has anyone tried repurposing a cast iron table saw top for a table? Hmmmm. I think the idea has merit. Anyone?

[wiremonkey]

Take a look at Wade O's design on youtube, it's brilliant and I think you'll see the brilliance of his X axis stepper placement, inside the tube!

[wizard]

Has anyone tried repurposing a cast iron table saw top for a table? Hmmmm. I think the idea has merit. Anyone?

I actually salvaged an old table saw table from some worn out tooling of my fathers. I haven't found a good use for it yet. The problem is those tables as surprisingly thin. At least on the smaller saws they are. A commercial quality table saw might be a different matter, the facilities people at work have a rather massive cast iron table saw but that is something you move with a large forklift. It is something like: #4 Saw Standard, this guy. That is a solid machine but then you have to pay for it and probably have more than a few wood workers chastising you for the sin of converting it to a router. WireMonkey's suggestion to look into a Wade'o design would likely be cheaper than a Northfield conversion. If yo search hard enough used machine base frames are out there to save yo a few dollars, it just takes a bit of searching and maybe the willingness to strip a lot of junk off the top.

[Goemon]

You didn't say what you plan to use it for, what materials you want to cut and what size of parts you intend to make? This is key for some of your questions and vital for the design stage.

I ended up buying a number of componants twice for my fixed gantry build because I didn't understand the relavence of the specs to my intended purpose. E.g. If you want to cut steel with a 300lb milling head then you need a stronger frame with larger motors and larger rails and ball screws with a lower lead than if you want to lightly engrave aluminum with a 10lb high speed spindle.


Also, you didn't say what your budget was. You will get drastically different recommendations if you have $10,000 to spend on parts vs someone trying to get it up and running for less than $1,000.