Odd circle test results (Page 4 of 24)

Essentially the idea is to fit a piece into the right side of the carriage which replaces the bottom Teflon block. It will be like a jigsaw puzzle piece, an exact negative of the contours. I was trying a difficult solution just to see what would happen

so why not just upgrade carriages with Lawsy set and linear bearings? also I have heard some in other topics talk of how the Tin (not steel) frames warp in places from all the torque of so many different motors and shafts mounted in it. even more with some mods. SD needs more than 16 gauge tin for frames heck thats what many cheap home furnace cabinets use and listen how they rattle with just 1 squirrel cage blower running. even 14 gauge is too thin for rigid support against twisting etc. maybe 10 ga or thicker that can be welded of cast aluminum etc even angle iron. just because a 110lbs weakling can stand on it means nothing. honeycomb cardboard paper 1 inch high like used in some packing will support over 300lbs but I wont use it as any type of frame to construct a precision machine on. lol

Analyzing Lawsy's carriage set, I now think I understand what Jago and Wardjr are trying to accomplish.  And it makes sense.  When printing my circles or arcs, it is very consistent that the problem is noticed when the extruder tries to change direction after a length of travel. 

Theory: If the carriage system works using linear bearings of Lawsy's Carriage set, instead of the white ride bar (shown in attachment), it proves that we are not dealing with backlash due to timing belt tension issues, but instead a design flaw in the carriage system.  This design flaw would imply that a degree of movement is not being accounted for.  One side of the carriage is being lifted during direction changes.  This could definitely cause the head to stop linear movement in that direction for a short period while the carriage is in the motion of lifting or falling back onto the rod.

Is this what you're testing now?

To the comment of frame flexing, holy cow this would be a scary deal if that were the case.  I can't find my T-square, otherwise I'd test this too.

Jaques,

Do you understand what I'm trying to accomplish? Unfortunately, my research load has ramped up on me today (I found several bugs in the code for my experiment that could drastically effect my results, so I need to fix this before participants come in). As a result, it would be several days before I get a chance to mess with this.

You mentioned you have years of 3D experience. Make a model that mimics the arcs in the right side carriage. Remove the teflon block on the bottom and top, slide your piece into place, and shim it with cardboard/whatever if the fit is too lose. Ensure the piece cannot move left and right. Then, re-install the top teflon block, leaving the bottom block out. In a moment I'll edit this post and show you a crude drawing wardjr sent me that should make the idea clear to you. The shaded part is the part you will have to design.

Take off the right side of your printer and remove the teflon blocks. It's the only way you will truly understand the issue. You are more motivated to solve this with the stock carriage than I am also, so I feel it's right to pass this down to you. I would do it if I had the time, but I don't want to keep you folks waiting for days. You also have more 3D experience so whipping this up should be no trouble. Use your imagination.

The goal is to prevent side to side movement of the carriage. Your piece should BARELY touch the rod on top--you don't want to make the issue WORSE by causing additional binding. You can't damage anything, so feel free to try whatever.

Make sure the part you design has a hole for the screw to attach it to the carriage in the same manner as the original Teflon block. If you don't do this, you'll just waste your time, because the part will flex back and forth and defeat the whole purpose. I learned this last night.

I hope this helps. I'm not bailing, but my research is obviously more important and I don't want to hold you up on solving this if this is the solution. If you get it to work, I'd be super thankful.

I've attached a photo. Look at your carriage, and the photo. Then let me know if you have additional questions.

Here is a crude drawing of my own with annotations. If you have a lot of 3d experience you should be able to whip this up in no time at all. Remember, barely touch the rod. You want it to slide smoothly. If you have lithium grease apply some to the rod and slide your part back and forth. Do not lose the Teflon blocks because your part will not hold up long.

The black part is your piece. Below that is the stock carriage. I hope this makes sense.

That's the reason I don't have time. Getting the part to fit requires a lot of iteration, because measuring it with calipers is difficult and not very accurate because you don't have room to measure it with the calipers totally straight. If I could get the right side off without entirely destroying it, this would be trivial. But good luck with that. I'm fairly certain the rods are glued in and you'll break the carriage entirely trying to get them off.

You don't need engineering experience.  Just experiment. Learn as you go. Do you want circles with the stock carriage or not?

You're gonna have to suffer a bit for them is all I can say. Expect pain. But you'll be learning.

nowhere to get filament locally? i guess i'm lucky i live in a big city... i can get makerbot abs any time i need it, but i've never done it. it's expensive as hell.

Well, order some up. You're gonna need it to keep printing anyway. Get it from Octave, not Solidoodle, if they will ship to you. $30/kg.

We all do, but do you have any evidence at all that this will work? Because I think we all have wasted enough time to now try to design something with several iterations without knowing if that's even it.
And I'm not sure what is it you're suggesting to make. Replcement for the white things? Something that goes with them? Those drawing are a bit confusing, sorry.
I thought you wanted to replace your y motor first? have you abandoned that idea or did I miss something?
And you seem to have discussed some stuff with wardjr, what was it? We don't seem to catch up. I think by the white things only having two "points" of contact you mean very small surface than needed by a "point"?
Sorry, but a lot of what you said doesn't seem to be clear to some of us. Before deciding to pass on this task to someone else, maybe you should explain exactly what it was you were trying to do. Or if you want just do it when you have the time, we'll wait. We've gotten used to waiting.