Welcome! Log In Create A New Profile

Advanced

Why is know now welding their printer frames CoreXY would benefit?

Posted by Crunch69 
Why is know now welding their printer frames CoreXY would benefit?
July 20, 2020 12:00AM
Why is know one welding their frames? A core xy would be ridged and square. Not only will the strength and squareness be a bonus you would require a less parts. By welding you would not need so many corner brackets bolts t nuts and ,plate bracketsspinning smiley sticking its tongue out

Edited 2 time(s). Last edit at 07/20/2020 12:02AM by Crunch69.
Re: Why is know now welding their printer frames CoreXY would benefit?
July 20, 2020 01:45AM
it doesnt happen often but there is at least one being built that way ATM,
[reprap.org]
Only the way to go if you have the tools & skills, i'd just buy a hifi stand & work with that.
Re: Why is know now welding their printer frames CoreXY would benefit?
July 20, 2020 02:38AM
Because more people have screwdrivers than welders.

Also if you don't know what your doing you will warp the frame...
Re: Why is know now welding their printer frames CoreXY would benefit?
July 20, 2020 04:57AM
As linked to in Mechabits' reply, I'm (slowly, LOL!!) building a CoreXY with a welded frame. What you say is true, a welded frame will be very rigid, and will be less susceptable to losing it's "square-ness".

What many don't realise is that the process of welding introduces a lot of movement in the metal that you're welding, so it can be increasingly difficult to keep the frame square and true while you are welding it up.

Even if you have all the material cut exactly to size (+-0.01mm) and all cut perfectly square, when you make any welds, even just little "tack welds", as the weld cools it will shrink and thus "pull" whatever joint you're welding out of square. An experienced welder (which I am not) may be able to compensate for this by deliberately miss-aligning the joint to begin with and using the welding-shrink effect to "pull" the joint square, or as I did, you can create a weld-sequence that will hopefully counteract this joint-shrinkage by welding opposite corners of a frame in a sequence (so perhaps the 1st tack-weld pulls a joint out of square, so a tack-weld on the opposite side of that joint will help to pull the joint back into square).

To give you an idea of how much movement a simple Tig-tackweld can give to a 90degree joint, a single small tack (about 3-4mm) placed on the outside edge of the 90degree corner of 2 lengths of 500mm long 25mm box-steel will likely "pull" the joint out of square so that the ends are 5-6mm out-of-square (depending on how big the tack is). Of course you can clamp the frame down rigidly and tack all the joints which will keep everything square, but when you release the clamps, unless you've managed to make every one of those tacks exactly the same size, there will still be some movement as the tensions created by each tack will differ. Plus it is only possible to clamp down part of a cube-frame, so there will still be difficulties when it comes to welding the 2 "halves" of the frame.

What I did was to just clamp up each "half" of the whole frame, tack-weld up that half while checking the square-ness and compensating with tack-size, then tack welding both halves of the whole frame, getting it as square as possible. Then it was time to fully weld the joints, this has even more potential for "pulling" joints out-of-square, so again a weld sequence was used, welding only the outside seams, and welding diagonal-opposite seams in turn. All the while checking for square on all the joints after each weld, and again after the welds had fully cooled. It was by doing all this checking that it became clear that the direction that you weld in also has an effect on the movement of the joint as it cools, ie if you are welding a joint in the X/Y plane, the direction that you weld it in will also cause movement in the Z plane.

Then I welded out the side-seams of each joint, again checking for where the frame was being "pulled" out-of-square when each weld cooled/shrunk, and then finally the inner seams were welded out (again working 1st on the seams that would "pull" in a desired direction to keep/re-make the frame square). Even with all of that, when I'd finished the frame was out of square by about 1mm, which required some percussive attention (a few judicious taps with a dead-blow hammer, LOL!!)

So that's the bad news, it's not as simple as it first appeared to me. But (maybe?) the good news is that the movement gets less as the frame gets progressively stiffer as the tack/welds are made. But conversely, the stiffer the frame gets, the harder it is to get it to move into square if it has got out of alignment during the welding.

But all-in-all, I think the effort is worth it. You can end up with a very stiff frame for a 3d printer. But that in itself can add challenges, the frame is then fixed, so any additions you want to make, or any mounting holes you need to have, you either have to have drilled/tapped before the frame is welded up, be able to re-mount the whole frame in a mill/pillar-drill to drill afterwards, or you have to drill/tap by hand. Where-as if you have a non-welded frame, it's a lot easier to make adjustments or drill/tap holes.

I guess there are a lot of factors, both positive and negative, to making a welded frame. I decided to try it out as I had the equipment needed, and the time to work on it, but I've not finished yet, so I don't actually know (although I have faith, LOL!!) that it will work great. I have some model-engineering experience, but this is the 1st 3d printer I have built, so I am learning a lot, and no doubt if I was to do it again there's many things I would do different (making a complete 3d model of it would probably be at the top of that list, then I could've drilled/tapped many of the mounting holes much more accurately)

Hope this rambling helps someone who's contemplating a welded frame

Tim

(edited for spelling and to add detail of welding the side&inner welds)

Edited 1 time(s). Last edit at 07/20/2020 12:55PM by TE78.
Re: Why is know now welding their printer frames CoreXY would benefit?
November 11, 2020 04:21PM
If I were going to weld a printer frame, I would plan on CNC machining the hardware mating surfaces and mounting holes. Then use linear bearings for the motion systems.

It wouldn't be easy or cheap, but it would be VERY nice.
Re: Why is know now welding their printer frames CoreXY would benefit?
November 11, 2020 04:55PM
A rigid frame is a good thing, and even if it isn't square it's OK. The frame doesn't have to be square, just the linear motion guides that will be attached to it. It's easier to mount the guides square to each other if the frame is square, but if it isn't you adjust your mounting techniques to ensure the guides are square.

2020 t-slot that is commonly used is very flexible and all the corner brackets in the world won't change that. A lot of designs grossly overuse corner brackets to try to compensate for the flexibility of too-small t-slot, and to compensate for not having the ends of the t-slot cut squarely. If you can get the ends cut squarely or milled, you can bolt the pieces directly to each other and know that the result will be square and rigid. If you take all the money you would have spent on dozens of corner brackets and nuts and bolts and put it into larger cross section t-slot with square cuts you'll end up with a more rigid frame.

Thin walled aluminum tube is more rigid that t-slot, and cheaper. If you're planning to use 2020 t-slot, it would probably be better to buy tubing instead. You can always drill holes in it to mount parts if needed.


Son of MegaMax 3D printer: [www.instructables.com]
Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]
Re: Why is know now welding their printer frames CoreXY would benefit?
November 12, 2020 03:45AM
Another interesting question along those lines, why aren't more people casting their frames with vibration damping epoxy+granite composites, in 3D printed molds?
Re: Why is know now welding their printer frames CoreXY would benefit?
November 12, 2020 04:55AM
Quote
golfwolf
Another interesting question along those lines, why aren't more people casting their frames with vibration damping epoxy+granite composites, in 3D printed molds?

Because it is overkill ! A good choice for REAL CNC though.

But a slab of concrete is a good base to build a rigid AL frame on it.

Note a flexible frame is a blessing for the majority as it allows for a poorly designed, mounted, aligned, guiding system ! smiling smiley


"A comical prototype doesn't mean a dumb idea is possible" (Thunderf00t)
Sorry, only registered users may post in this forum.

Click here to login