Hi all,
after several weeks of waiting, I finally got the T8-2 lead screw for the Z-drive that I wanted. After mounting it, the screw tends to stick to the walls in the tunnel where it is supposed to move freely. I loosened and fastened the long M8 screw that holds the gantry array together, loosened and fastened the screws that keep the brass nut to the gantry, but am not yet happy with the outcome.
If I fasten the long M8 screw too much, the lead screw sticks. Loosening it improves the situation, but the lead screw still does not go straight to the top bearing respectively the motor axis.
N.B., I did not yet lubricate the lead screw, not wanting to cover an undesirable assembling flaw. Outside the MPCNC, the nut runs totally freely and smooth over the whole thread length.
Is this situation here uncommon, or standard?
TIA for any good advice.
BTW, I am not yet sure if I like the new website design. Getting any virtual points for nothing reminds me of some bangbad environment… But that’s another topic. jftr.
There should be a lot of clearance around the leadscrew. Are the spacers in backwards perhaps? Each spacer should have a larger hole for the lead screw and a smaller hole for the screw that holds the assembly together.
Also if the T8 nut is fastened down firmly and crooked then it can bind. It is recommended that the T8 nut be held loosely. But it doesn’t sound like that’s your problem.
Jamie,
thanks for your reply. The spacers are placed correctly. My impression is that the base of the nut does not sit really flat, that is coplanar with the machine bed, in the gantry. I think I’ll experiment a little there tomorrow. Midnight here now in Germany, time to log off. Have a nice sunday!
Not sure how that top piece is supposed to work, but you really don’t want the cable chain attached to the top of the z like that. It will cause issues down the road.
Okay, went and looked at the thingiverse page.
One, it looks like your z is waaay too tall. That’s a lot of stuff sticking out above the bottom black piece. Two, there should be a bearing in that black piece, and three it looks like your bearing in the top black piece isn’t centered. Your top down shot shows this.
The “black piece” is there just to keep the tube array up. Its regular place is just below the coupler/timing belt gear. When moved completely down, there are just a few centimeters of tube left. Of course, there wiill be a bearing in the lower black part. Right now it isn’t to show the misalignment I have here. The bearing in the top part is in the right place, it is the lead spindle that is off axis.
I’m always unsure about mechanical problems, but I’ve seen it said that the screws in the leadscrew nut are really just to keep it from turning. I think loosening the bolts and using the bearings to keep it aligned is a good choice. Without lube, any pressure will quickly make it bind. So I would also be careful to judge it without the lube. I know it moves smoothly when loose, but there isn’t 2kg of weight on when you do that.
Sort of like tuning a violin or even better a piano - I also had that idea - what use does it have now to adjust everything and then see it all flexed again under the load of the router / spindle.
So., I’ll rig the router first now.
Previously, I had attempted already to achieve the optimal squaring of the gantry axes by loosening / tensioning the bearing-screws of the gantry. Maybe that had some undesired effect on the geometry of the rest of the system.
Time to have them all servo driven, each screw, and having some software taking care of that job…
So, back to the workbench.
btw, I have done lots of things with the first version of the machine, with the same chain mounting on it. I never saw any issues that I would have related to the cabling. The plastic of the chain does not add much to the weight of the cables.
Re the centering of the top bearing again: The viewing angle is a bit odd in the from-the-top view, that’s why I added the view from below which shows it much better.
And: correcting the top part above the nut (of course less than “snugged down”) is less important here than the invisible part in the gantry array. Ideally, the leadscrew should run up and down without any deviation from the ideal axis through the nut’s thread.
It looks like there is an asymmetrical gap between your first and second pictures.
If the top of the xyz part is not flat where the leadscrew nut sits then the nut will want to cock sideways a bit and the leadscrew will not be straight when it’s unsupported. With the coupling on the top end of the leadscrew, the leadscrew will be held vertical but the nut will want to bind if it is tightened down, or even the weight of the z axis could push it enough to make it stick. Usually lube is enough but you could also try shimming or sanding the area on the xyz part where the nut sits.
Jamie, congrats, you’ve got a very sharp and trained eye or two! Just came up from the workshop after taking the pictures below. I used a piece of thin (0.2mm) shimming feeler gauge tape to lift the lower side of the nut, et voilá, everything looks fine and moves smoothly.
After taking off the nut, I saw that its bed is a bit asymmetrical, thinner on the hanging side. Maybe due to the printing / slicing / ion storms, whatever. It is difficult to picture that, here are some attempts.
I think just sand it. A shim would work but has more chance of getting loose and jamming everything up. Sanding (or an Xacto blade or something) is a clean, low-risk finished product assuming you can reach all the areas you need to reach.
Thanks. The best way to sand off fdm-made parts is to wet-sand them under flowing water. That guarantees the plastic does not melt and smear. Not sure yet if I am ready to unmount the part for a few micrograms of PETG. At least, it’s the way. And the shim sheet is stainless steel, glueing that to brass ensures a good bond. But I don’t rush it. Push it too hard and it will tilt over. Thanks again. Ah, sanding off some of the brass…?
No.
Or? The higher the tech, the lower the solution…
I was thinking coarse sandpaper with your hand. Coarse meaning 220 or lower. Sand a bit, test fit, then sand more, test fit, until you’re done. You won’t accidentally generate enough heat to melt the plastic.
Success.
The seat of the nut is a pretty narrow niche. So, accessing it with sandpaper was not really the way to go.
The next step was to use a medium coarse file. Better, but did not reach the concave recesses of the nut’s resting plane.
What do I have a collection of sharp chisels for? Not only for woodwork, scraping the PETG off was a charme - very fine layers came off, a narrow and a wide tool led to a perfecly planed and trimmed surface, and the leadscrew aims directly at the top bearing’s opening now. Smooth rotation, goal reached.
Thanks to all who contributed!
Yes, wood chisels are the way to go to manually “edit” a 3D print. It works much better than any other method.
In your case though, I would have gone with the classic “heat and press” method. much faster and easier to do.
Problem solved anyway, so it’s time to carve something to celebrate!
Thanks for your message. Yes, I had considered that, but the stepwise approach - we are talking here about a deviation of <2°, looked more relaxed and comfortable to me. It also left the chance to augment the surface with epoxy resin, if everything goes wrong.
In other situations, I prefer the very-soft solder method too.
Waiting for the right timing belts now, the ones that I had were 20mm too long (loop). That’s why they are called timing belts, they time your progress…
But that’s another topic. jftr.
