Welcome! Log In Create A New Profile

Advanced

Two rods on one (Delrin-) ball-effector

Posted by o_lampe 
Two rods on one (Delrin-) ball-effector
March 21, 2018 04:44AM
As a spinoff from the discussion about improving Mag-rods, I (re-) invented the concept of using one Delrin ball for two rods.
The rods are held in place by springs or elastic cord between carrier and effector.

The Delrin balls are so light compared against magnet and steel ballstud, they can be much bigger in diameter and still save effector weight.
I made a small napkin sketch to find worst-case angles, when the effector is at the far side of a tower.
IMHO two 8mm OD rods could share a 16mm ball at +/-22° angles without interfering.

I've seen this concept in the past, but don't remember where.
Any pointers, opinions?
Re: Two rods on one (Delrin-) ball-effector
March 21, 2018 07:16AM
Re: Two rods on one (Delrin-) ball-effector
March 21, 2018 03:51PM
Should be looked at indeed.
One ball shared by two arm sets (of two arms) like in this SpiderBot, OK but one ball shared by the two arms of one set is not OK.

Edited 3 time(s). Last edit at 03/21/2018 03:59PM by MKSA.


"A comical prototype doesn't mean a dumb idea is possible" (Thunderf00t)
Re: Two rods on one (Delrin-) ball-effector
March 22, 2018 04:33AM
Thanks Nitram, that's what I meant. But they still use steelballs/magnets.

I've also seen plastic balls in Nylon or PTFE in 5/8" size.
For convenience we could still use magnets to attach the rods. Just add one magnet/ironplate combo in the string between carrier and effector. No spring or elastic rope. Three magnets instead of twelve and only three plastic balls on the effector will make a huge weight difference.

The magnet/ironplate would have no air gap, so the magnet could be smaller too, without loosing holding force.

Edited 2 time(s). Last edit at 03/22/2018 04:36AM by o_lampe.
Re: Two rods on one (Delrin-) ball-effector
March 23, 2018 04:43AM
Anyone in EU interested in a group buy? The Delrin ball seller in UK has a pretty weird price table, like 5 balls=10€ / 100 balls= 27€
Re: Two rods on one (Delrin-) ball-effector
March 23, 2018 02:10PM
I did buy Delrin balls in the past at amazon.com from the U.S. - try a search for "delrin precision bearing balls".
A source in Germany could be Kugel-Winnie.
Re: Two rods on one (Delrin-) ball-effector
March 24, 2018 05:24AM
thumbs up Danke Heinrich smiling smiley
I just ordered enough POM balls for three printers.
If it doesn't work out, I have to print a marble machine winking smiley
Re: Two rods on one (Delrin-) ball-effector
March 30, 2018 10:10AM
I made an effector for the 16mm Delrin balls. It fit's the 55mm rod distance from the Smart Effector, but it looks so small.


I hope, I can fit all the fans and z-probes without fouling the tensioner-string.
Attachments:
open | download - 16mm_effector.JPG (90.6 KB)
Re: Two rods on one (Delrin-) ball-effector
April 02, 2018 03:11AM
I made some rods from aluminum tube. The only stuff I could get before easter.
The sleeves have dishes in the size of the balls and they fit snuggly on the effector. The rubber rope is pretty loose, but during some test runs, I couldn't manage to provoke a fail. I raised jerk to 1200mm/min and acceleration to 6000mm/s² and upped the speed to 120mm/s. It had to print some nasty honeycomb-infill, the whole printer was shaking, but the rods stayed in place. To make things worse, I added the top heavy Cycloid extruder: no fail. See short video here.



I'm pretty pleased with it. It is exceptional light weight compared to magnetic rods ( and the fans will not stop or brake from magnetic pollution )
PS: I waited until April 2nd to publish it, so expect me to be dead serious...

Edited 1 time(s). Last edit at 04/02/2018 03:16AM by o_lampe.
Re: Two rods on one (Delrin-) ball-effector
April 02, 2018 08:14AM
I will be interested in seeing how the accuracy is in printing. Looks very smooth movement in the video.
Re: Two rods on one (Delrin-) ball-effector
April 02, 2018 01:08PM
Impressive, but I'm assuming that the rods are being held in by compression force from the retainer springs/tubes? What happens if the rubber wears out?
Re: Two rods on one (Delrin-) ball-effector
April 03, 2018 03:14AM
Quote

What happens if the rubber wears out?
I've bought the rubber rope from a shop specialized in boat-sailing equipment. This rope is UV resistant and made for heavy duty stuff. I expect it to live longer than the rest of the printer, but it's also cheap and easy to replace.

I also made an adapter for the Duet3D Smart Effector. I want to test the nozzle contact probe behaviour. Maybe even use extreme parameters to see, when the rods will loose contact. First I have to modify the carriers to work with rubber rope, so stay tuned for more video. ( It looks pretty dramatic, when a rod slips off grinning smiley )


Re: Two rods on one (Delrin-) ball-effector
April 03, 2018 03:19AM
Quote
Billhc83
I will be interested in seeing how the accuracy is in printing. Looks very smooth movement in the video.

Me too smiling smiley
I expected to see the rods dance around on the balls, but they are so smooth and have a very low friction coefficient ( almost like PTFE )
Re: Two rods on one (Delrin-) ball-effector
April 03, 2018 07:26AM
It seems, the bowden tube has a much bigger influence than expected. The black effector with the heatsink ontop falls apart ( literally ) whenever the bowden tube pushes or pulls on the heatsink. I have to design a new effector, probably with a Piezo probe inccluded.
Re: Two rods on one (Delrin-) ball-effector
April 03, 2018 03:51PM
I like the first incarnation. I think the distance from the balls to the nozzle in the second incarnation will mean any tilt in the effector will be greatly magnified.
Re: Two rods on one (Delrin-) ball-effector
April 04, 2018 01:52AM
The Smart-Effector adapter is just a quick and dirty design to test nozzle contact probing. I had to put it on long legs to clear the connectors ontop of the PCB.
I already had a former version installed and it worked well. (Deviation ~ 0.01 - 0.008)
But I didn't like the twisted mounts for the ballstuds.


Edited 1 time(s). Last edit at 04/04/2018 02:36AM by o_lampe.
Re: Two rods on one (Delrin-) ball-effector
April 21, 2018 06:11AM
I made a new adapter which holds the hotend below the alu-effector. Now the Bowden tube has no lever to twist the effector off the rods.
I'm loosing ~30mm of print height, but at least it works.
I think of adding a small magnet in the top rod-sleeves, to keep the rod in place when they accidently slip off the Delrin ball.


Re: Two rods on one (Delrin-) ball-effector
April 21, 2018 03:53PM
Quote
o_lampe
I made a new adapter which holds the hotend below the alu-effector. Now the Bowden tube has no lever to twist the effector off the rods.

Wonder whether a flying extruder would be good with this?
Re: Two rods on one (Delrin-) ball-effector
April 22, 2018 02:11AM
Quote
frankvdh
Quote
o_lampe
I made a new adapter which holds the hotend below the alu-effector. Now the Bowden tube has no lever to twist the effector off the rods.

Wonder whether a flying extruder would be good with this?

A Zesty Nimble would be perfect for this.
Re: Two rods on one (Delrin-) ball-effector
April 22, 2018 02:21AM
A flying ( or in my case dangling ) extruder will definitely help to decrease the forces of the Bowden tube, while the cable of any remote drive might be too stiff.
Anyway, I won't spend money on a remote drive unit, just to find out. But donations for testing are always welcome winking smiley
Re: Two rods on one (Delrin-) ball-effector
April 25, 2018 09:24AM
Here's a bit of homebrew rubberband theory:
The goal is to keep the rubber tension and thereby rod-ball friction to a minimum without loosing contact.
I'm currently using a thick, strong rubber. Same length as the rods.
I could achive the same tension with a thin, short rubber and I have the gut feeling, it would be the safer option because the spring rate graph of the thinner rubber is steeper.
When the nozzle bumps into something, the steeper spring rate will increase tension much more, keeping the rods and balls in contact.

Am I on the right path?
Re: Two rods on one (Delrin-) ball-effector
April 25, 2018 10:15AM
Probably not. How deep is the spherical depression at the end of a rod? Lets say it is 1 mm. Will your new rubber stiffen enough to exert a force of the same magnitude as the old rubber when it is elongated only by about 0.5 mm? Probably not.

Edit: It would need to be much more stiffer and its length would need to be precisely set.

Edited 1 time(s). Last edit at 04/25/2018 10:17AM by hercek.
Re: Two rods on one (Delrin-) ball-effector
April 26, 2018 02:55AM
I envisioned a thin rubber, almost being stretched to the limit in normal operation. 0.5mm stretch would increase force alot more, I thought.
Re: Two rods on one (Delrin-) ball-effector
April 26, 2018 03:42AM
Check it out with a hanging scale. Tie one end of the rubber to the scale. Pull the other end to stretch the rubber. Read the force on the scale. Stretch the rubber by 0.5 mm more. Reed the force again. Compare whether there is a significant difference.

In my opinion, this approach is too sensitive to the stretch length. It will be tedious to set it up just right. The problem is that you want a significant difference in the pulling force when it is stretched to length L compared to the length of (L+0.5 mm). Moreover even if you set it up right then it may rattle by a bit less than the 0.5 mm (during which the force increases significantly).
Re: Two rods on one (Delrin-) ball-effector
April 27, 2018 02:45AM
This weekend I'll run some tests with different stretch-scenarios. What would be a good part to print, to show any errors?
A single wall cylinder? Vase mode or interupted by horizontal holes?
Re: Two rods on one (Delrin-) ball-effector
April 27, 2018 03:45AM
I would try a simple star with sharp corners(*) so that you have there quick direction changes. Only shell, spiral vase mode.

(*) E.g. a straight extrusion of an isotoxal star. Go with a red filament and a star inscribed in regular pentagon for an added fun factor. Just kidding. Anything with sharp corners and straight in the vertical direction.
Re: Two rods on one (Delrin-) ball-effector
April 27, 2018 11:30AM
What would be the best way to machine the rods to fit the 16mm ball? All I can think is a 16mm ball mill.
Re: Two rods on one (Delrin-) ball-effector
April 28, 2018 04:01AM
I made the rod sleeves from ABS and used Acetone to smooth the surface, while pressing it on a Delrin ball. Delrin is acetone imune.
The top end rod sleeves are running on Haydns steel ball studs. I've built and described a smoothing method here.
But I later used a 90° hand deburr tool to match the length of the rods.


Edited 1 time(s). Last edit at 04/28/2018 04:07AM by o_lampe.
Sorry, only registered users may post in this forum.

Click here to login