Thanks for the update. Sometimes, one might think that your projects are flawless, but as many improvements are discovered, new issues arise, therefore learnings that we enjoy when are shared. 🖖🏻
It's around 500$ to build.... That's what I told my wife anyways... 🤪. One a more serious note you can build one for 2-3k USD depending on size and options selected.
@@HevORT Hahaha that's more what I would expect. Very impressive. Clearly for a select audience. I'd love to have something like that, but also I don't have enough time for my hobbies as is. ;)
Did you consider to use metal 3D printed parts instead of the plastic printed ones? Im wondering the limitations of those parts as well as draw backs, like machining required afterwards or surface problems
Have you considered using sensor to map the z ball screw wobble vs z height for each screw and then compensate for that in software rather than build these isolator attachments?
No I did not, I am a mechanical engineer. How would you do that? My guess is that the total oscillation will be the sum of the three screws per their respective current clocking. The nut tilt high variation would need to be addressed some how too. The programming effort sounds a lot more intense than the solution I cam up with. But once again, I am a mechanical engineer.
I know you're trying to release them as a 3d printed solution anyone can do but... ironically based on my recent fb comment, have you considered having them 3d printed from metal and potentially machining the bearing surfaces. This would eliminate the creep issue and some flex but not really fix the lever arm issue
Aren't 16mm ballscrews completely overkill for a 3d printer? Especially with the additional linear rails? You have an extremely rigid setup and then you decouple it anyways?
The ballscrews provide the Z reference and play no role into XY stability of the bed. The rail is taking care of the XY stability. In this design decoupling happens on the ball screw only. Why ballscrews? When using Z hop a typical Benchy will generate 7 times the standard Z travel distance that it would normally take to print the part. This is only a small benchy. larger and more complex parts can generate huge amount of additional Z movements. These small repetitive movements are ideal to rub the lubricant away. After more than three years of using the same ball screws on thousands of ZHop prints, I am still getting 0.0014 mm of backlash over 20 samples (measured with Beacon). Why 16mm? The screws that we buy are cheap, they are manufactured using a rolling process. the material is crushed to shape. The impact of that pressed threading process will affect material all the way to the core on smaller diameter screws. This is why the smaller diameter of these cheap ball screws are difficult to obtain with decent radial runout value. 16mm screws also provides a lower thread angle for the same pitch, increasing the resistance to being back driven by the bed's weight.
@@HevORT I think turning on "avoid crossing perimeters" in the slicer is more effective for reducing nozzle contact with printed area than grinding up the Z axis with constant Z hops.
Imo the auto-bed leveling pivot points and the rails are significantly reducing the bed's rigidity... I think that's why the rest of the printer it is so over-built in an attempt to compensate...
Just to validate your statement, give me an example of what you think is a good Z-Tilt(self-leveling) enabled system? And please, dont mention printers where the leveling is achieved by racking the printer.
@@geometerfpv2804 I made a jig to cut the silicone pads to around 4-5 millimeter thin so they squish very little and get stiff quick. It's rigid enough for my purposes.
The 3-point separately driven Z axis is unnecessary. - Superfluous mechanical elements (hinges, magnets, linear rails) - Redundant electronics (+2pcs stepper motors, +2pcs motor controllers, more expensive motherboard due to + controllers) - Unnecessary procedure (multiple probing, synchronization after each power up) All this can be saved by connecting the 3 axes with a ribbed belt and setting them precisely. Less money less time.... You already have it... I wrote this for others
I believe the 3 point system will be adopted more as non-planer 3d printing becomes mainstream, these projects lay the groundwork for future innovation
![[TH] 2024 PMSL SEA W2D2 | Summer | โหดจัด ซัดไม่เลี้ยง](http://i.ytimg.com/vi/TJa9d7tpSP4/mqdefault.jpg)
Thanks for the update. Sometimes, one might think that your projects are flawless, but as many improvements are discovered, new issues arise, therefore learnings that we enjoy when are shared. 🖖🏻
Improvement or over-engineering... Thin line.
when nextruder follow up video :3
I would love to see someone do a sun/orbiter geared extruder without the concave idler and extruder gear.
Would be very interested in that video too
Sheesh that has to be a pricey beast.
It's around 500$ to build.... That's what I told my wife anyways... 🤪.
One a more serious note you can build one for 2-3k USD depending on size and options selected.
@@HevORT Hahaha that's more what I would expect. Very impressive. Clearly for a select audience. I'd love to have something like that, but also I don't have enough time for my hobbies as is. ;)
Did you consider to use metal 3D printed parts instead of the plastic printed ones? Im wondering the limitations of those parts as well as draw backs, like machining required afterwards or surface problems
Have you considered using sensor to map the z ball screw wobble vs z height for each screw and then compensate for that in software rather than build these isolator attachments?
No I did not, I am a mechanical engineer. How would you do that? My guess is that the total oscillation will be the sum of the three screws per their respective current clocking. The nut tilt high variation would need to be addressed some how too. The programming effort sounds a lot more intense than the solution I cam up with. But once again, I am a mechanical engineer.
I know you're trying to release them as a 3d printed solution anyone can do but... ironically based on my recent fb comment, have you considered having them 3d printed from metal and potentially machining the bearing surfaces. This would eliminate the creep issue and some flex but not really fix the lever arm issue
Yes, I am considering this option ;)
Yes, either that, or «just» make them on a cnc as a flat aluminum piece, anyway thank you for showing us
Aren't 16mm ballscrews completely overkill for a 3d printer? Especially with the additional linear rails? You have an extremely rigid setup and then you decouple it anyways?
Welcome to these community 3D printer projects 😆 we all seem to like overkill
The purpose is defeated by putting them on horizontal linear rails with ball joints to get a gimmicky auto bed leveling feature...
The ballscrews provide the Z reference and play no role into XY stability of the bed. The rail is taking care of the XY stability. In this design decoupling happens on the ball screw only.
Why ballscrews?
When using Z hop a typical Benchy will generate 7 times the standard Z travel distance that it would normally take to print the part. This is only a small benchy. larger and more complex parts can generate huge amount of additional Z movements.
These small repetitive movements are ideal to rub the lubricant away.
After more than three years of using the same ball screws on thousands of ZHop prints, I am still getting 0.0014 mm of backlash over 20 samples (measured with Beacon).
Why 16mm?
The screws that we buy are cheap, they are manufactured using a rolling process. the material is crushed to shape. The impact of that pressed threading process will affect material all the way to the core on smaller diameter screws. This is why the smaller diameter of these cheap ball screws are difficult to obtain with decent radial runout value.
16mm screws also provides a lower thread angle for the same pitch, increasing the resistance to being back driven by the bed's weight.
@@HevORT I think turning on "avoid crossing perimeters" in the slicer is more effective for reducing nozzle contact with printed area than grinding up the Z axis with constant Z hops.
@@hd-be7di On slower printers, perhaps.
16mm ballscrews in a 3d printer?
Can you make the dad rock louder please? Still hearing some voice
Sure thing son!
Imo the auto-bed leveling pivot points and the rails are significantly reducing the bed's rigidity... I think that's why the rest of the printer it is so over-built in an attempt to compensate...
Just to validate your statement, give me an example of what you think is a good Z-Tilt(self-leveling) enabled system? And please, dont mention printers where the leveling is achieved by racking the printer.
@@HevORT None. Manual bed leveling with groove locked knobs. Probe for tramming and easy Z offset.
@@hd-be7diyou think having the bed floating on springs or squishy silicone is rigid? 🤔
@@geometerfpv2804 I made a jig to cut the silicone pads to around 4-5 millimeter thin so they squish very little and get stiff quick. It's rigid enough for my purposes.
First?
yes! and because of that you had the chance to watch this video in SD as youtube is painfully still crunching the HD version! lol
@@HevORT haha. It was in portrait mode on my phone so I didn't need much more lel.
@@HevORT haha :D
The 3-point separately driven Z axis is unnecessary.
- Superfluous mechanical elements (hinges, magnets, linear rails)
- Redundant electronics (+2pcs stepper motors, +2pcs motor controllers, more expensive motherboard due to + controllers)
- Unnecessary procedure (multiple probing, synchronization after each power up)
All this can be saved by connecting the 3 axes with a ribbed belt and setting them precisely. Less money less time....
You already have it... I wrote this for others
It all depends on what you are trying to do and the size of the printer you are targeting.
I believe the 3 point system will be adopted more as non-planer 3d printing becomes mainstream, these projects lay the groundwork for future innovation
Yeah but then you need manual leveling
This is a great printer, but it is over-engineered.
Yup! :)