Choosing between alternative X axis bearings designs

[eci22]

Before choosing the bearing position you need to fix the gantry.

Separate tubes are weak.

You should have one large tube.

Your pros and cons are correct. There is no "best" for this design choice. They each have pros and cons. It partly depends on the other design compromises you make, for instance how much stick out the spindle will have due to other decisions.

If you choose good rigid gantry uprights, for example large tubes and not thin plates, you can get back the lost space under the gantry due to rails underneath.

For me, the hassle of trying to align rails in alternative B and the increased complexity and weight of the carriage make it not worthwhile. I prefer alternative A.

Thanks pippin88, I think am leading to the same conclusion as you. My diagram wasn't the best, the rails will be on the tube going across only (along the x axis), the diagram shows that it also goes across the small section of box that is going in the Y direction but this won't be the case since this space will be used to house a motor and leadscrew bearing support, so the rails only sit on one section of tube.

If you are going to weld up the gantry from steel box section then you can use the epoxy levelling method to get a flat surface to fit the rails to in option A, but you should make it more solid. My own gantry is made from two pieces of 100x50 box section welded together to give a 200mmm high face for the rails to fit on. As always with a DIY build, your best option depends very much on what tools or other workshop facilities you have access to. Option A makes the Z-axis assembly much simpler to build as well.

By "more solid" do you mean by adding bracing? This is something I will be doing for sure. Yes they will be welded, though I am considering welding the highest x-axis guide box to the top of the frame and bolting the one closest to the ground to make fine adjustment in reference to the welded one, haven't thought this completely through yet. Do you have pictures, or a build log ? it would be great to see your machine Thanks for your input

[JAZZCNC]

ADJUSTMENT ADJUSTMENT ADJUSTMENT... I've said this many many times in posts but I'll say it again because cannot be stressed enough.

With this fixed gantry design, you are going to need plenty of adjustment. If you weld up the gantry completely and especially if welded to the base frame you won't be able to adjust the planes the rails sit on in relation to the other planes.
Any extra time spent thinking about how you can build in adjustment will pay back big time when setting up the machine and the accuracy you will get from it.

Regards the 50mm with 3mm box section then like I said in other thread, it's spaghetti-like properties mean you really will need to brace it good. Not so much for strength but for rigidity and resonance. The thin wall means it's resonant which causes vibrations at the tool, yes you can fill it with sand or concrete-like some have done, but this won't help a poorly designed and braced gantry.
This stage is critical to get right because it's the foundation everything else references from, Wobbley foundation = Shit machine simple as that.!

There's 3 key areas to building a great machine, get any one of them wrong and the machine will suffer.
#1 Base frame
#2 Strong Z-axis
#3 Good electronics.

[Kitwn]

By "more solid" do you mean by adding bracing? This is something I will be doing for sure. Yes they will be welded, though I am considering welding the highest x-axis guide box to the top of the frame and bolting the one closest to the ground to make fine adjustment in reference to the welded one, haven't thought this completely through yet. Do you have pictures, or a build log ? it would be great to see your machine Thanks for your input

You do better following Jazzcnc's advice rather than mine on making the structure more solid. He's had years of experience and I've only built the one machine. However, you might consider restricting the 50x50x3 stuff you've already bought to just the base frame and start again with the gantry design.

I have been thinking about putting out some details on the forum as I'm in the middle of finishing the Z-axis at long last. Not so much a build-log on how to build a good machine but more about the compromises I've had to make to fit the available budget and footprint and being very remote from suppliers and professional workshops. It works, but I wouldn't advise anyone to copy it.

If I were starting from scratch I'd brew a nice big pot of tea, open a packet of Hobnobs (or the Australian equivalent) and read everything Jazzcnc and Boyan Silyavski (if you'd like to see some really heavy engineering) have written and a number of other build-logs before buying the first nut and bolt.

Kit