Design question: mounting of supported rails on t-slot extrusion

[Buoyen]

Hello.
I have a question for you all. Made a drawing of a router with 60mm extrusions and SBR20 rails. The rails are 30mm between the mounting holes and 45mm total width. So it means they can be mounted with t-nuts in the 60mm extrusion. Then I startet wondering: would 45mm extrusion be better and just bore and tap in the extrusion?

Please see the pictures. And thanks for any input on this.

Similar Threads:

• Need Help!- 3 axis cnc router based on supported rails conception design and parts
• Where to post linear supported rails to sell.( question for ger21)
• Newbie- Extrusion, Hiwin rails, and mic6 cnc router...need help with design
• How to ensure supported rails are even?
• Newbie- Supported Rails Design: Y & Z considerations

[wizard]

A couple of things:

1. Unless we are talking a very small machine 45mm t-slot is very small
2. Tapping the extrusion isn't the norm. Depending upon how things line up you may have a hell of a time doing a good job.

For a little more highly biased info:
1. I really don't like mounting linear components directly to most T-Slot extrusions. In my mind you will be better off bolting to thick solid paced of aluminum bolted to the t-Slot. You will then have a flat surface to drill and tap to your hearts content. The more you expect precision wise the worst the t-Slotted extrusions are.
2. I don't understand your pictures.If you really want to use T-slot the part functioning as your gantry beam need to be rigidly bolted to the uprights. There needs to be the use of gusset or other reinforcements
3. A gantry beam is unsupported between uprights. You need to consider this when sizing beam size. Beam cross sectional area will vary based on that distance to span

[A_Camera]

Hello.
I have a question for you all. Made a drawing of a router with 60mm extrusions and SBR20 rails. The rails are 30mm between the mounting holes and 45mm total width. So it means they can be mounted with t-nuts in the 60mm extrusion. Then I startet wondering: would 45mm extrusion be better and just bore and tap in the extrusion?

Please see the pictures. And thanks for any input on this.

Use 60mm extrusions and don't tap 45mm extrusions. If you can, you could add an iron plate between the rails and the extrusion for increased rigidity, but otherwise you are fine, unless your machine is huge. I have a similar solution and milling aluminum with fairly high high speed without problems, but my machine is quite small, the table is only about 450 x 300mm, mine is fixed beam, moving table design.

[Buoyen]

Thanks to both of you.
The gantry beam extrusion is 510mm long. There is one 30x60 extrusion standing behind the gantry beam to help stiffen the "upright" on each side and I also want to add some angles between them and the gantry beam extrusions.

Is it only because of rigidity you would have a plate between or would the plate be more flat than extrusions are? I don't have equipment to do surface milling of the extrusions..

Here the x-axis rails are 500mm and the y-axis rails are 900mm. I thought about turn that around, but it would lead to a very long gantry beam - if you understand. Bad idea?

[wizard]

Thanks to both of you.
The gantry beam extrusion is 510mm long. There is one 30x60 extrusion standing behind the gantry beam to help stiffen the "upright" on each side and I also want to add some angles between them and the gantry beam extrusions.

Didn't see that the first time through.

Is it only because of rigidity you would have a plate between or would the plate be more flat than extrusions are? I don't have equipment to do surface milling of the extrusions..

The problem with extrusions is getting the T-NUTs to seat properly and in doing so keep your linear rail straight and in the right plane. You have already complicated this by having the linear rails on the top and bottom of the gantry beam rails. If you bolt an aluminum plate to the face of the gantry rials you immediately have established one plane. You can then bolt your rails to the plate concentrating on getting them parallel to the table. This would give you more clearance under the gantry beam also.

The trouble you have is that when you are mounting you linear rails on two different extrusions, you end up with many different directions in which you can have alignment issues. I just see this as the hard way to put together a gantry beam with extruded aluminum. Simply going to a bigger extrusion (or a steel beam) is also a possibility which can give you one flat surface to mount your two linear rails on.

Here the x-axis rails are 500mm and the y-axis rails are 900mm. I thought about turn that around, but it would lead to a very long gantry beam - if you understand. Bad idea?

You can orient things anyway you wan't, after all it is your machine. Most people prefer the long axis on the "X" but others have good arguments for the long axis being on the Y. Usually those good arguments involve fitting the machine into a limited space, or easy access for a common operation.

[Buoyen]

Great input.
Now I've read some more and discovered the FEA analysis that was done in a another thread. Trying to source an extruded 150x150 or 200x200 aluminium square tube with a cross inside for reinforcement and the mount the rails on the front face of taht. It would make it a lot easier to put together for sure.
Talked to a supplier that maybe could cast a mold for it. Don't know the cost yet though.

Something like this, 150x150 with 10mm thickness, 5mm thickness cross inside.

[wizard]

Great input.
Now I've read some more and discovered the FEA analysis that was done in a another thread. Trying to source an extruded 150x150 or 200x200 aluminium square tube with a cross inside for reinforcement and the mount the rails on the front face of taht. It would make it a lot easier to put together for sure.
Talked to a supplier that maybe could cast a mold for it. Don't know the cost yet though.

Aluminum is very expensive relative to steel. The cost of aluminum is why I generally recommend steel to builders.

Something like this, 150x150 with 10mm thickness, 5mm thickness cross inside.

You can have sheet metal cut and bent to fit the tube and rivet or bolt that reinforcement in place. You can get similar results by orienting internal gussets with in the tube, say every 6-10 inches. Or you can god with a really heavy wall tube but weight then becomes a problem.

I really don't see casting a tube as being cost effective, at least it wouldn't be in this country. You don't want the project to sink under the cost of the gantry tube.

[Buoyen]

Completely true.

So today my head started working... Found this at work. Has to be machined flat (or epoxy for levelling - the area under the SBR rails is quite big), but if I can get a used one of these for free that would be enough aluminium for the whole machine... The dimension is 160x107mm. The reinforcements are not diagonally, but I can add bolts also if that would stiffen the beam further?

[A_Camera]

Thanks to both of you.
The gantry beam extrusion is 510mm long. There is one 30x60 extrusion standing behind the gantry beam to help stiffen the "upright" on each side and I also want to add some angles between them and the gantry beam extrusions.

Is it only because of rigidity you would have a plate between or would the plate be more flat than extrusions are? I don't have equipment to do surface milling of the extrusions..

Here the x-axis rails are 500mm and the y-axis rails are 900mm. I thought about turn that around, but it would lead to a very long gantry beam - if you understand. Bad idea?

I would turn those Y axis linear guides 90 degrees so that the gantry is resting on it in the right orientation, the way it supposed to be. I don't think it is a good idea to mount the way your design shows. My machine is mostly build out of 90x45 and 60x30, plus some extra for making it stiff and rigid. It has it's own feet, so it isn't really a bench to design, but it's not shaking when I run a violent g-code with high speed and a lot of direction changes. Your Z seems to be very high, so you have to consider that as well. The spindle motor should, in my opinion be installed on an plate. My original Z design looked similar to yours but after installation I realized it was wrong, extended with an extra plate first but decided to make a completely new plate in the end. That's better than a sandwiched extra plate. Anyway, I don't know what kind of Z clearance you planned for. I have about 140mm and that's more than enough for me, I can put a heavy vise on the table as you can seen in the video, and still have enough clearance.

[Buoyen]

Clearance down to spoilboard is 18,5cm in the picture. I might lower it as well to make the machine stiffer. But I don't really know what it'll be used for yet so want to have some room.

[wizard]

Completely true.

So today my head started working... Found this at work. Has to be machined flat (or epoxy for levelling - the area under the SBR rails is quite big), but if I can get a used one of these for free that would be enough aluminium for the whole machine... The dimension is 160x107mm. The reinforcements are not diagonally, but I can add bolts also if that would stiffen the beam further?

Free always wins out in my opinion.

I'm not sure what you mean by adding bolts. They wouldn't stiffen the beam that much. For an extrusion like that simple internal gussets would likely be all that you need. You could probably get away with the gussets in just the two outside channels.

By the way there is another approach to stiffening things up in a beam, you could fill it with epoxy concrete. That however makes for one heavy beam. However for a machine base it might not be too bad especially if you only fill one channel. You do need to look out for sag induced by the heavy weight of the fill if you go this route.

[Buoyen]

Thanks again. Really appreciate the input on this!

Epoxy is a good idea - in combination with foam to keep the weight down I could make a diagonal of epoxy inside for additional strength.
There's also a 230x160mm version of this extrusion, but we don't have som many of those.

I have another question. Per my drawing above: On the longest axis (I called it Y) the rail and bearings are placed just beneath the table. I guess the optimal placement of this would be at the same hight as the cutter (vertical distance zero - zero twisting torque)? And if they are placed slightly below as I have here, should the vertical distance to the drive nut be equal? Or less? Does is mather if the nut is below or above the linear bearings?