Xnaron Printable Magnetic u-joint option for Rostock Max

[Flateric]

I only sent that as an example pic so that you would clearly see what I had used. Home depot does sell them in the exact same diameter as the magnet and arrow shaft.

The deeper part of the cup faces the ball of course.

And yes, you have it right however I have moved away from using the heatshrink tube and moved to using the kapton tape covered in plumbers goop because it has better adhesion and staying power. The shrink tubing seems to relax after a bit of use and it's proximity to the hotend.

The kapton tape wrapped many time stightly around the whole works serves to provide strength and hold everything in place. The plumbers goop servers to protect it from unwrapping and allows you to use light lubricant on the cup and balls without loss to the assembly strength by weakening of the glue.

[disneytoy]

Hi Gang!

I've been really fascinated by the magnetic arms since I first saw the BarryBot. I've researched for a year and am building my first Delta. In fact I think most of my electronics and motors get delivered Monday!!!

I have a request, if someone can help me out. At the moment I do not have access to a 3D printer. I was wondering if someone here who has printed several iterations of the end effector, carriage adapters, etc, might like to donate a set to the cause. Mainly just the plastic printed parts. These can even be less than perfect versions. I just want to get up and running, then I can print new parts if needed. It's the chicken or the egg scenario. I need some printed parts to be able to print, but can't print until I have the parts.

I think I'm at about 90% there on everything. I really think the magnetic route is the way to go. I'm excited to start getting printing and learn first hand. I have so many ideas I'd like to try designing and printing.

PM me if you may bew able to help me out.

Thanks!

Maxi
San Diego

[Eaglezsoar]

Has anyone tried to braze the ball to the screw? Does anyone know of a reason why this would not work?

[edward]

My fear would be too much heat. But, I can count the number of times I've brazed things on one hand. Maybe with a jeweler's torch?

When I welded mine I was careful about putting in too much heat and distorting the screw or changing the surface finish of the ball. Even then I had to polish the balls to remove a bit of roughness that appeared.

I would be curious to see a brazed result, though.

[Generic Default]

Does the way you attach the ball to the screw really matter that much? I just put some loctite glue in the head of the cap screw and stuck the bearing ball on for each one, and none of mine have ever broken. I didn't even clean the balls with acetone like it said to. I've glued 18 balls into screws that way with no problems so far.

[edward]

Well, not really, as long as they are all seated in the same position.

I think all of the discussion about alternate (read: no glue) methods of joining the ball and screw has arising due to the fact that a non insignificant number of people are having problems with detachment. Whether it be cleaning, glue choice, or whatever, some of them have asked for other options.

[DavidF]

Does the way you attach the ball to the screw really matter that much? I just put some loctite glue in the head of the cap screw and stuck the bearing ball on for each one, and none of mine have ever broken. I didn't even clean the balls with acetone like it said to. I've glued 18 balls into screws that way with no problems so far.

I think its important to make sure the ball is concentric with the stud and the center of the ball to the base of the screw head the same length other wise it may cause some issues with rod length. I'm not against jb weld or super glue as a DIY solution, but a marketable product needs to be welded.

[ddseeker]

Couldn't Neodymium Sphere/Ball Magnets be used instead of a Ball Bearing and Disc Magnet with printed Boom End?

N45 Disc magnet: 13,200 gauss

N42 Sphere/Ball: 13,200 gauss

http://catalog.monroeengineering.com/vi ... order=desc

Or would the lack of support from the spherical pocket in the Boom End make the arms less stable?

(was just thinking ball to ball would remove friction of ball to printed part)

[barnett]

I was thinking ball magnets like this could be used along with printed arm ends w/cylinder magnets in order to provide a stronger connection with less weight. Magnet to magnet vs magnet to steel.

I'm no engineer, but it seems like I still want the receiver to be a cup shape for the ball or else it would be wobbly.

[barry99705]

Couldn't Neodymium Sphere/Ball Magnets be used instead of a Ball Bearing and Disc Magnet with printed Boom End?

N45 Disc magnet: 13,200 gauss

N42 Sphere/Ball: 13,200 gauss

http://catalog.monroeengineering.com/vi ... order=desc

Or would the lack of support from the spherical pocket in the Boom End make the arms less stable?

(was just thinking ball to ball would remove friction of ball to printed part)

No, don't use spherical magnets. The chrome coating over the pressed powder insides isn't that robust.
[img]https://lh5.googleusercontent.com/-MySP ... 182906.jpg[/img]

The attraction is also stupidly strong because of the small contact point. The damaged point is from the two magnets I have hitting each other. There's no way to stop this from happening, you can't physically pull them straight apart, I have a bodybuilder friend that can't do it. You have to roll the poles apart before you can get them apart.


I also found that the paper seal you find in the top of a liquid medicine jar won't keep ferrofluid in when you hold these magnets to the lid. That crap is a pain to clean up! It will also stain white ceiling paint....

[disneytoy]

Didn't BarryBot use sphere 1.2" magnets and he ballend milled some bolts to receive them on the arms/end effectors. That is what I was going to do, then I see everyone doing the opposite. Steel balls and cylndrical magnets.

[geneb]

That's because it's a lot easier for Joe Builder to glue a ball into the partial pocket formed by a socket head caps screw than it is for them to use a ball nose mill to precisely machine a pocket in the end of a SHCS. This is especially true if you don't have the facilities to properly use said ball nose mill.

g.

[disneytoy]

Swivel Ball Socket Head Spring Clip Clamp Mount for Camera Flash

Not that this part would work, or it would be cost effective, But they're most likely is a ready made solution for us that has other applications:

[img]http://www.thebigstorage.com/images/ebay/hk/E63068.jpg[/img]

Also:

FEET FOR SWIVEL SCREWS

http://www.carrlane.com/catalog/index.c ... 2853524459

[img]http://www.carrlane.com/SiteData/Featur ... -FSS_p.jpg[/img]

FEATURES: Type D swivel feet are ideal for clamping in a slot or an inside corner. Available in a choice of finishes:
Steel, zinc plated yellow chromate
Steel, black oxide finish
Made in USA.

PATENTED DESIGN: Feet for Swivel Screws are an important part of our patented left-hand-thread swivel-foot system. The ball end of the body screws into the foot for easy installation and removal. Once the ball threads pass beyond the foot threads, the foot can rotate and pivot freely (20º total angle). Foot threads are left-hand to positively prevent accidental backout during use. Delrin® feet are optional. Note: Delrin® feet are not threaded – they snap on and off.

SIZES: Available in five sizes to fit swivel-screw bodies from 1/4-20 to 5/8-11 (M6 to M16 in metric) — see Dimensions page for part numbers.

MATERIAL (STEEL): 12L14 steel, zinc plated yellow chromate (black oxide finish optional). DFARS and RoHS compliant.

[lordbinky]

I was thinking ball magnets like this could be used along with printed arm ends w/cylinder magnets in order to provide a stronger connection with less weight. Magnet to magnet vs magnet to steel.

I'm no engineer, but it seems like I still want the receiver to be a cup shape for the ball or else it would be wobbly.

I thought the worst part of using a ball magnetic with the cylinder magnet is that the attraction would change with the angle of the arm (besides the potential to be repulsive if you got one backwards) giving your inconsistent force/friction on the joint.

[barnett]

I thought the worst part of using a ball magnetic with the cylinder magnet is that the attraction would change with the angle of the arm (besides the potential to be repulsive if you got one backwards) giving your inconsistent force/friction on the joint.

Oh I get it now, the poles would move relative to each other. Yeah that seems bad.

[ddseeker]

Any opinions on this three ball design...

http://www.qualup.com/spiderbot.eu/prod ... etail.html

[Nylocke]

I modified this to work with Rostocks and Rostock Mini's, http://www.thingiverse.com/thing:153055 if you, Xnaron, want anything added to the thing, I would be happy to oblige, 'Im quite tired from school and the stress of designing the working carriage in CAD tonight, so there may be some mistakes or the like....

[lordbinky]

The three ball design of the spider is neat in appearance. I'm not sold on it being an functional improvement for a couple reasons though.

I though 3 large bearings weighed more than 6 smaller ones, so it increases the weight of the platform.

The small and long magnets don't create as much magnetic force, lowering the maximum platform weight.

I'm not sure that the magnets don't affect each other while that close on the same object. It isn't good if their position or grip strength changes with movement.

It looks neat, but I don't see a functional trade off for the restrictions that design imposes. I could see a high jerk setting ripping the arms free in that design long before it would with the present one in use. *shrug* I'd be glad to be wrong though.

edit: And why call it a spider bot?! Spiders have EIGHT legs damnit!

[enggmaug]

I'm just about to jump in.

I however live in France, and thus tried to get parts from european countries, with dimensions using metric system.

I ordered 12x10mm ball bearings on ebay from a french reseller, and a pack of 15 cylindrical magnets 10mmx10mm grade N48 (I could not find N52) from a german shop.
I'll buy a 8mm carbon fiber tube on saturday from a Kite shop in Paris.

For the rest, I'll redesign the printed parts using metric system, and input my own ideas on how to get the job done.

I'll let you guys know how it goes.

I did not finish reading all 30 pages in this thread, but I want to thank ev1, and Xnaron above all, for what seems to be an amazingly great upgrade.
Without the Forum and you guys to animate it forward, the RostockMax would not be such a beautiful machine. Hey, I'm proud to be part of that community !

[enggmaug]

Hey, I got back home and found the BBs in my mailbox.
I Soldered them to bolts, printed some test parts for my adaptation.

I don't have the magnets yet, but I tried my set up on one - very strong - Hard drive magnet.

It seems that there are a lot more friction than with the stock arms U joints.

What do you guys reckon, is it normal ?

[Eaglezsoar]

Hey, I got back home and found the BBs in my mailbox.
I Soldered them to bolts, printed some test parts for my adaptation.

I don't have the magnets yet, but I tried my set up on one - very strong - Hard drive magnet.

It seems that there are a lot more friction than with the stock arms U joints.

What do you guys reckon, is it normal ?

It is normal to feel as though there is more friction and the joints are sometimes smoothed with a heated ball, not a hot ball but just warm enough to smooth the socket the
ball fits into, I personally feel that the balls should be inserted in the sockets and rotated around with a drill for a SHORT time to smooth out the cup. What you are feeling is not
all friction, a lot of it is the magnetic attraction which takes a while to get used to how it should feel. I am not a fan of the heated ball method because I know I would get the ball too
hot and ruin a good cup end. Do a little light smoothing, hook everything up and let it do some printing for a while, they should break in after a while. You definitely want the magnetic
attraction to be strong and all arms exactly the same length. I'm not sure about the strength if you used regular solder to connect the balls to screws. Regular tin/lead solder is not
very strong. Brazing with brass or bronze may work, JB Weld Epoxy works, drilling a threaded stud works, spot welding works but if you used regular solder be sure to check them for fracturing/stress cracks. As indicated lead is a very weak metal is really not designed for things that get a lot of movement. Just my opinions for what they are worth too you. Happy Printing!

[Generic Default]

I've been using magnetic arms since last July. I designed my own effector plate and joint cups; the joint cups use a cylindrical fit instead of a spherical fit because it has less surface area to mess up. I printed mine with trimmer line nylon because it's cheap and has a lower coefficient of friction than ABS or PLA. I also put PTFE grease in the cups, but I don't know if it's made a difference or not.

I just glued my chrome balls into the cap screws with loctite glue, and I've never had a ball pop off. It's strong enough that I can screw the cap head screws into the un-threaded nylon effector plate by twisting the chrome balls (all of the torque is on the loctite glue).


I'm hoping to get a few CNC machines this summer so I can make kits for 3d printers with this kind of stuff. I'll definitely make Teflon (PTFE) ball joint cups with a lathe for a precise, low friction arm system.

Nylon has a coefficient of friction of around 0.15 to 0.2
PTFE has a coefficient of friction of around 0.04 to 0.1
ABS varies because it's a mix of stuff, but I've read it's between 0.35 and 0.5.

Check out the cross sectional view of my cup design. I use 0.42 inch (10.5mm) brass rod arms for more rigidity, so my cups are slightly bigger than xnaron's.

[Eaglezsoar]

Generic Default, have you put these on Thingaverse and if not would you be willing to share the stls for the cups, effector and cheapskate mounts. I would be willing to pay for your
version, just send me a PM.

[Generic Default]

I'll try to get the files up by Saturday, but I'd rather just post them as attachments in this thread instead of putting them on thingiverse (which means Makerbot owns my design?)

For now here is a picture of how it all looks. The magnets are 3/8th by 3/8th inch (9mm or 10mm should also work). The 7/16th (0.4375 inch or about 11mm) brass tubes were about $3.00 each, I got them at an Ace hardware store.

Magnets - $20 ish
Brass Tubes - $20 ish
Printed parts - $1.50 ish

If I were to offer a $170 pre-assembled complete kit similar to what's shown in the image above but with a triple extruder (that doesn't exist...yet) would anyone be interested? This would likely be sometime around June, but no guarantees.

[Eaglezsoar]

If I say I would be interested does that mean you won't post the files?
I would be interested but I would have to see what's in my hobby budget at that time.
I would still appreciate the upload and I understand why you don't want makerbot getting them.