Ultimate z axis accuracy?

I need to print another to see. I think Marlin had an problem with missing steps because of a software issue that generated varying step pulse widths, which seems to have been fixed recently. I am thinking those small displacements in XY are due to that. The small prints that I have done since the fix look better.

So far it has cost me about $1000. so I don't know how "budget" that is in today's market. It is about 3-4 yo.

Edited 1 time(s). Last edit at 01/08/2020 10:42AM by ruggb.

... "industrial" starts at around $5000 ... but the definition is pretty fuzzy

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

But how much do I add for my labor and design modifications?
I did build it myself from internet plans I modified.
$1000. is just parts

Edited 1 time(s). Last edit at 01/08/2020 01:42PM by ruggb.

Not sure if this is a hijack or a valid entry in this thread...let me know.
I have an FT5-2 printer with a 2-screw, 4 linear guide-rod vertical drive Z setup. It prints 0.2mm layers more than meeting the quality I expect. I now have a second used FT5-2 that needs the mods I did on first one, but I want to go to a belted Z drive on this one, per Rehorst approach.
I need to research the details, but have a question about the general approach, please.
Everything I've found s far uses 2 belts/guides. For lateral stability without using additional guides of any sort, would using 3 belts/guides work to give a stable bed? See rough diagram with belts at 3 locations and driven by one stepper..... any advice much appreciated.

It looks like it should work just fine.

It's like using 3 screws with two guide rails. Put the guide rails near two of the belts/screws and let the cantilevered end of the bed be supported by the third screw/belt.

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

Quote
ruggb
Based on the performance of my CoreXY printer, I keep thinking y'all are overthinking this Z drive.
I have 3 screws driven by 1 motor/belt/pulleys ....................................................

Same here. Everyone laughed at me when I elected to use 8mm x 1 single start lead screws. But with a lead of 1mm, it means that using a 1.8 degree motor I have 200 full steps per mm which means that the resolution is 5 microns per full step. Typical positional accuracy of a stepper motor is +_ 5% non accumulative. So as far as the motor is concerned one can say that the positional accuracy in Z is +- 0.25 microns. And with the very shallow helix angle that a fine pitch lead screw has, there is no chance that the weight of the bed will cause the screws to rotate when power is disconnected, so the bed will never drop.

Edited 1 time(s). Last edit at 02/21/2020 09:36AM by deckingman.

---

[somei3deas.wordpress.com]
[www.youtube.com]

Quote
the_digital_dentist
It looks like it should work just fine.

It's like using 3 screws with two guide rails. Put the guide rails near two of the belts/screws and let the cantilevered end of the bed be supported by the third screw/belt.

Are you describing option "A"? Would option "B" be any better? I was planning on using only the three belts moving linear bearings on rails, and using no additional guides at all.

I think option A is the better of the two. It's best to keep the screws/belts close to the guides.

If the screw/belt is far from the guide, the lifted object acts as a long lever to apply torque to the bearing block.

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

I have one just like that.
I have a nozzle mounted piezo probe/endstop that gives me a std dev of 0.002mm. I had to custom make the mount for it, but I love it.

If your looking for high accuracy Z drive, as has been used for the last 40+ years in CNC machines, are ball screws, if you spend the money, you can ones with an accuracy of 1um, and a closed loop stepper motor. you can go either expensive proprietary or cheap open source like a mechaduino