Is anyone interested in inexpensive parts for zero backlash magnetic ball and socket joints?

[Haydn Huntley]

I have been experimenting with them on my printers, and it was almost the same price to have 1000 of these made, as it was to have 50, so I have a lot left over to share with the Delta printer/Maker community!  :-)

Is $1 each reasonable enough?

These are made out of chrome plated steel (for maximum magnetic strength).

The screw at the base is the usual M3x10.

The ball at the top is 3/8", which is very close to 10mm.

You can use a 10mm socket to easily tighten them without marring the balls.

Besides the carriage, I have tested and working designs for the effector and rod ends, which I'd be happy to share on a non-commercial basis.

The carbon fiber rods came from TriDPrinting, and I also found an inexpensive source for extremely strong N50 magnets.

If you need some of the parts printed, I can even help with that.

Air mail and handling in the USA for *any* quantity is a flat $4.

I'm one of those weird people who enjoys logistics, and should be able to ship them out the next day.

[Stan Williams]

I am interested in 24 of the steel balls with screws.

[Dejay]

Hmm, well since you said you enjoy logistics, would you ship to Europe and for how much? ;)

[Brian Halkett]

Haydn,

I'm interested in 36 of them plus the rod end, effector and carriage files.

Let me know your paypal info and I'll get the $ off to you.

Brian

[Doug LaRue]

I would like a set of 12 balls and 12 magnets along with the effector and rod end files.

I will not use them in a commercial design and may talk to you later about that since I'm attempting an educational project although there are fees involved but not for the printer.

Doug

[Malcolm Williams]

I'd like a dozen ends and the magnets if you have them handy. Let me know what I need to do to make this happen.

[Haydn Huntley]

@Dejay,  No problem with shipping to Europe.  It looks like shipping for one set is $9.45, and for two or more sets $12.75.

[chri...@live.ca]

I am interested in 18, how much would shipping to Calgary Canada be?

[Adam P.]

I'm interested in 36 of these. Please add me to your list,

[Shai]

Interesting! What's the reason for the flat hex on the bottom of the ball? To make it easier to weld it to the ball?

Are your magnets powerful enough so that if the hot end bumps into a layer of plastic that curled up, it wouldn't detach?

[Malcolm Williams]

Correct me if I'm wrong, but I believe it is simply for tightening without harming the ball.

On Sat, Jun 28, 2014 at 4:18 PM, Shai <shai...@gmail.com> wrote:

Interesting! What's the reason for the flat hex on the bottom of the ball? To make it easier to weld it to the ball?

Are your magnets powerful enough so that if the hot end bumps into a layer of plastic that curled up, it wouldn't detach?

--
You received this message because you are subscribed to a topic in the Google Groups "Delta robot 3D printers" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/deltabot/485cfVFrFFU/unsubscribe.
To unsubscribe from this group and all its topics, send an email to deltabot+u...@googlegroups.com.
To post to this group, send email to delt...@googlegroups.com.
Visit this group at http://groups.google.com/group/deltabot.
For more options, visit https://groups.google.com/d/optout.

[Haydn Huntley]

Hi Shai,

These are not welded, they're machined.  The flat hex on the bottom of the ball serves two purposes:

1)  It makes for a standard platform, so that the ball will always be the exact same distance above it.

2)  It makes it easy to grip to tighten.  It isn't a good idea to grip the ball itself with pliers, because one might scratch it.  You can easily tighten these with a 10mm socket and your fingers.

In my designs, I have nut traps to hold 3mm nyloc nuts, so I use the 10mm socket to turn the ball studs until they seem reasonably tight.

Because there is the nyloc nut, it won't come loose over time.

I've tested these magnets on my modified Kossel Clear printing at 90mm/sec for over a month and it never popped loose during a print.  Non-printing moves were at 200mm/sec.

I also tested some smaller 6mm ones, and they also never came loose during a print, but they sometimes popped off when homing the mechanism!

By the way, my carriage design works beautifully with your carriage wheels!

[Haydn Huntley]

@Malcom, You're right!

[James Armstrong]

Sorry, reposting without duplicating images.

I will take 16 (12 plus spares). Let me know how your handling payments / shipment (paypal?).

- James

[Josh D]

I would like 12 I am located on the east coast US.

[Brad Hopper]

I think these look beautiful and the hex foot is extremely handy for sturdy level and tightening.

If you make more though I have a suggestion. It would be great to have even deeper "neck" under the ball with additional clearance for rotation. It looks like the rotation might be sewhat constrained but the foot area.

Printed carriages/effectors are easy to make sloped to maximize the useable ball area but mine and others I'm sure are flat with simple drilled holes. Mine is an aluminum water cooling block so it would be hard to machine the slope, and the carriages I have are carbon fiber, again easy if flat hard to do sloped.

[Gareth Chen]

I'll take 18, and magnets if you can supply them. Also the source files for your modified carriages, as I'm trying to use Johann's new printed recirculating sliders.

On Saturday, June 28, 2014 3:56:24 AM UTC-4, Haydn Huntley wrote:

[Haydn Huntley]

@Brad,  Thank you, and next time I have a bunch of these made, I'll have them manufacture them as you describe.

My first printer is a Kossel Clear, so all of the parts were cut out of acrylic, and like yours, couldn't easily be modified.

For that printer, a jeweler friend of mine TIG welded 3/8" ball bearings to stainless steel M3x8 SHCS screws.

They have a smaller area of contact (the head of the M3x8), but they allow the joint to swivel more, and they have worked quite well in my KC.

They were made individually, so they are slightly more expensive at $20 for a set of 12, and I only have a few sets.

If you're interested, I can post a picture.

I've never seen carbon fiber carriages!  Can you post a picture or two?  Sounds like you've built an amazing system!

:-)

--Haydn

[JohnD]

Hey Hayden - if you have 12 left, I'll take a set!

[Jean-François Talbot]

If there are left, I'll take 2 sets, which makes it 24.

Jeff

[Shai]

Haydn, what size magnets do you use? I'm assuming they're probably rather large?

We're working on a custom designed countersunk magnet for very strong magnetic bonding to the steel ball with limited friction.

[Ian Lee, Sr.]

I'll take 24 of the ball & screws.

I like your effector and glass mounts also.  If you want to send them or post them somewhere that would be great.  I've been working on this same problem but with JB welded balls directly to the effector/carriage.  Being able to remove the balls and move them to the next iteration will be great.

[Ian Lee, Sr.]

@Haydn - if you want to see Brad's printer, just search the forum for "Hotbot".

[Brad Hopper]

Shai, I think the strongest (and cheaper to make probably) magnet for arms will be a potted magnet. I.e a simple magnetic cylinder in a steel cup housing which touches the back of the magnet, and extends "just the right amount" past the front of the magnet to allow a ball to sit in the cup and be held as close as possible without touching the magnet. The steel polepiece increases the concentration of the field a whole lot, and steel costs less than custom ND for sure.

[Haydn Huntley]

Hi Shai,

I'm using cylindrical N50 magnets, which are 10mm in diameter and 15mm high.

I like to have a small air gap between the ball and the magnet (approximately 0.1mm), so that there is no friction between them.

Also, I like to have a nice, deep cup to hold the ball.

Brad's idea of using potted magnets sounds like an excellent idea.

I'm surprised that your carriages are laser cut acrylic and not printed.  By printing them and the effector, one can angle the balls and sockets to maximize stability, plus curves can be nice looking.

[David S]

Please put me down for 24. If you have the files available for the effector and rod ends, I'd really like to use them. 

On Saturday, June 28, 2014 2:56:24 AM UTC-5, Haydn Huntley wrote:

I have been experimenting with them on my printers, and it was almost the same price to have 1000 of these made, as it was to have 50, so I have a lot left over to share with the Delta printer/Maker community!  :-)

Is $1 each reasonable enough?

These are made out of chrome plated steel (for maximum magnetic strength).

The screw at the base is the usual M3x10.

The ball at the top is 3/8", which is very close to 10mm.

You can use a 10mm socket to easily tighten them without marring the balls.

Besides the carriage, I have tested and working designs for the effector and rod ends, which I'd be happy to share on a non-commercial basis.

The carbon fiber rods came from TriDPrinting, and I also found an inexpensive source for extremely strong N50 magnets.

If you need some of the parts printed, I can even help with that.

Air mail and handling in the USA for *any* quantity is a flat $4.

I'm one of those weird people who enjoys logistics, and should be able to ship them out the next day.

[C-line]

Hi Haydn-

I'd like 12 balls + 12 magnets please. Also, I am already on the list for a set of the motor vibration dampers, so please combine shipping if it makes sense to do so.

Regards /James

[Raymond Ma]

I'd also like 24 if they are still available. Thanks a lot!

[Billy Zelsnack]

I'm surprised that your carriages are laser cut acrylic and not printed.  By printing them and the effector, one can angle the balls and sockets to maximize stability, plus curves can be nice looking.

How does having the balls at an angle increase stability?

[Haydn Huntley]

The ball and socket joint offers a certain amount of freedom, but it is always less than a 90° angle, because the ball has to be attached to something at its base.

In a Delta printer, the effector is always well below the carriages, so the two most extreme positions the joint could need to accommodate are when the effector is straight below one carriage, and when the effector is all the way on the other size of one carriage.

By angling the ball on the carriage 30° to 45° downwards, it allows that range of motion.

Angling the balls on the effector is similarly helpful.

I wish I had a few photos or sketches I could use to illustrate this -- a picture is sometimes worth 1000 words!

[Billy Zelsnack]

On Sunday, June 29, 2014 2:36:39 AM UTC-4, Haydn Huntley wrote:

The ball and socket joint offers a certain amount of freedom, but it is always less than a 90° angle, because the ball has to be attached to something at its base.

In a Delta printer, the effector is always well below the carriages, so the two most extreme positions the joint could need to accommodate are when the effector is straight below one carriage, and when the effector is all the way on the other size of one carriage.

By angling the ball on the carriage 30° to 45° downwards, it allows that range of motion.

Angling the balls on the effector is similarly helpful.

I wish I had a few photos or sketches I could use to illustrate this -- a picture is sometimes worth 1000 words!

I understand how the angles can increase potential range of motion. What I don't understand is how it can increase stability. Do you mean in the event that the joint exceeds its range of motion?

[Haydn Huntley]

@Billy, I'm sorry I didn't answer the question you actually asked!  :-)

Yes, when the joint is near the edge of its range of motion, it is also in a position where something unexpected (like a bit of malformed plastic sticking up) can possibly pop it out.

By having the balls angled properly, the balls will be near the center of their range of motion most of the time, and this makes the joints more stable.

I think I should have originally written:  by printing the carriages and effector, one can angle the balls and sockets to maximize the range of motion, and perhaps improve the chance that the ball will stay in the socket, even if something unexpected happens.

In the parts I designed, I decided to copy other people's (Xnaron's: http://www.thingiverse.com/thing:87839), which seemed to have a fairly deep socket for the ball to be held in.  This costs a little potential range of motion, but I really didn't want the ball to ever pop out by accident, especially during a print!

I better get my designs posted before I go to bed...

[Haydn Huntley]

Here are links to the OpenSCAD files for the effector and carriage:

https://www.youmagine.com/designs/kossel-plus-magnetic-effector

https://www.youmagine.com/designs/kossel-plus-magnetic-carriage--2

I got an error message from YouMagine when I tried to post my rod end design a few minutes ago.

It is late, so I'll just post these now, and attempt to re-post the rod ends again tomorrow.

[Martin Bogomolni]

I would like 12 please.   

-Martin

[Doug LaRue]

The effector downloads correctly but the carriage download button fails with XML issues. But getting individual files works when you go to the other tab.

Now that I have linear rails on my almost Mini Kossel, it looks like some modes to get a carrage mount but it's helpful to have a starting point. Thanks for the files and I'll keep my eys open for the magnet holding rod ends. Thanks again.

Doug

[Haydn Huntley]

@Dough, I'll see if re-uploading the design files for the magnetic carriages fixes the error messages on YouMagine's site.  If not, then I'll just make them available on my own site.

I just finished posting the rod ends:  https://www.youmagine.com/designs/magnetic-ball-joint-rod-ends

[brent sagnotti]

12 please, great work Haydn!

[Sage]

Nice!  Even though I've all but given up on magnetic arms.. the dream is too strong.
I'll take 24 balls please.
Thanks Haydn.
-Sage

[Haydn Huntley]

@Sage, In my experience there are three easy places to screw up, which will keep this from working well:

1)  Make sure that the arms are exactly the same length.  Build and use a jig.  Insert a 3/8" (or 10mm) ball bearing in the socket to make this easier.  When gluing the arms, you can use super glue for the first end, but I recommend using something which sets slower for the second end.  I used tiny round pieces of tape as shims to get the arms all the same length, and of course start out by figuring out which one is the longest, otherwise you'll have to file it down.

2)  Carefully print and/or clean out the sockets, so that there is nothing interfering with them being spherical.  I use a 3/8" ball end mill and rotate it with my fingers.  I used to use a half-round file, and that actually worked pretty well.

3)  Use strong enough magnets.  :-)

I'll have more magnets in a few weeks, and I could make a set of rod ends to go with them, if this will help you to succeed.

Building rods takes a lot more time and attention, so I cannot do that inexpensively.

If there is sufficient interest, I could setup my workshop so that I can build complete sets with a standard length.  The rods I'm currently building are 296.9mm from the center of one ball to the center of the other, and I use 288mm lengths of carbon fiber from TriDPrinting.  If I did a run of ~10 sets of these, I could produce the rods with magnets (but without the ball studs, because you'd already have them or would buy them separately) for something like $10 per rod ($60 for a set of 6), plus postage.  If this is interesting to folks, contact me...

[Haydn Huntley]

@Doug, Sorry I misspelled your name before!  I re-uploaded the files for the magnetic carriages and have verified that they all download okay now.  Thank you for letting me know about this problem!

The three are:

https://www.youmagine.com/designs/kossel-plus-magnetic-effector

https://www.youmagine.com/designs/kossel-plus-magnetic-carriage--2

https://www.youmagine.com/designs/magnetic-ball-joint-rod-ends

Enjoy!

[Haydn Huntley]

At this point *all* of the orders paid for so far are packaged, sealed, stamped and sitting in my mailbox, waiting for the mailman on Monday morning.  :-)

[nickel...@gmail.com]

Haydn,

I would like a set of 12, if there are any left.

Thank you! 

[James Wei]

I'd like to get 24 if there are any more avalible.

[Haydn Huntley]

The weekend's orders were actually mailed at the post office on Monday morning -- it was getting to be too much to fit in my mailbox.  :-)

Monday's orders are more modest, and waiting for the postman to pickup this morning.

Still more available...

[George Hahn]

I'd like to go for 12, thanks!

[Doug Fletcher]

Im interested in 36.  I se some also asked for magnets so if you are alos able to supply those... 36 of them as well!

Doug Fletcher

[Haydn Huntley]

I'll be traveling for July 4th weekend, so I'll be unable to respond to requests from Thursday July 3rd through Sunday July 6th.

I should be able to catch up on Monday July 7th.

[sho]

Hi Haydn

I'm interested in 36 please. When you are ready ;)

@Doug for the magnets if Haydn don't have them, you can look here http://www.supermagnete.de/eng/ (E.U)

[Adam P.]

Thanks for the delivery, Hayden. Very fast service. Looking forward to the magnetic revolution haha.

[Adam P.]

Haydn, have you shared your effector, rod and carriage designs?

[Haydn Huntley]

@Adam P., Yes!  But for the sake of convenience, here are the links again:

https://www.youmagine.com/designs/magnetic-ball-joint-rod-ends

https://www.youmagine.com/designs/kossel-plus-magnetic-carriage--2

https://www.youmagine.com/designs/kossel-plus-magnetic-effector

[Adam P.]

Hayden what's your source for the magnets?

[Adam P.]

Thanks, Hayden. I have that stuff on order now.

[James Le]

Very intriguing. Just how fast can you really push these things? Any reason why you don't go above 300 mm/s travel? 

[Ian Lee, Sr.]

What jig do you use to assemble the rods?  Do you assemble from plastic tip to plastic tip or with a ball in the socket?  What is considered the rod length?  Magnet end to magnet end or the center to center of the balls?

[laurent muchacho]

Would you post to the UK? 

Anyone else from the UK interested in doing a group buy Haydn

L 

On Saturday, June 28, 2014 8:56:24 AM UTC+1, Haydn Huntley wrote:

I have been experimenting with them on my printers, and it was almost the same price to have 1000 of these made, as it was to have 50, so I have a lot left over to share with the Delta printer/Maker community!  :-)

Is $1 each reasonable enough?

These are made out of chrome plated steel (for maximum magnetic strength).

The screw at the base is the usual M3x10.

The ball at the top is 3/8", which is very close to 10mm.

You can use a 10mm socket to easily tighten them without marring the balls.

Besides the carriage, I have tested and working designs for the effector and rod ends, which I'd be happy to share on a non-commercial basis.

The carbon fiber rods came from TriDPrinting, and I also found an inexpensive source for extremely strong N50 magnets.

If you need some of the parts printed, I can even help with that.

Air mail and handling in the USA for *any* quantity is a flat $4.

I'm one of those weird people who enjoys logistics, and should be able to ship them out the next day.

[Haydn Huntley]

@Ian, I don't have a picture of my jig handy, but it isn't anything fancy.  Just two blocks clamped to a piece of aluminum L-channel on my work bench.

I assemble them with 3/8" ball bearings in the sockets, because that is the metric of interest -- ball to ball distance.

The rod length is the distance from the center of one ball to the center of the other.

So, if you want a distance of 300mm, and your balls are 9.7mm in diameter, then you want to make your jig 309.7mm wide.

[Haydn Huntley]

@James, I'm currently using an Arduino Mega to run my printer, with 1/32 stepping, running Répetier, and I wasn't sure it would be able to step that fast.

Also, for the things I print, it can only rarely accelerate to full speed, so it doesn't make much of a difference in the overall printing times.

Lastly, I assumed that the harder it accelerates, the more wear and tear there would be on the carriage wheels.

My next printer will use a 32-bit microprocessor, and I'll push it a littler harder.  :-)

[Gareth Chen]

So I've been looking at your designs, and I'm curious if the rod end causes the ball to actually rub up against the magnet. If so, I'm concerned about long-term wear on the magnet finish, as the nickel plating isn't particularly durable. Have you looked into adding delrin or teflon washers as a bearing surface?

[Brendan Halliday]

Hi Haydn,

I'd be interested in 24 of these, how much to Brisbane, Australia? :)

On Tue, Jul 8, 2014 at 11:21 AM, Gareth Chen <gareth...@gmail.com> wrote:

So I've been looking at your designs, and I'm curious if the rod end causes the ball to actually rub up against the magnet. If so, I'm concerned about long-term wear on the magnet finish, as the nickel plating isn't particularly durable. Have you looked into adding delrin or teflon washers as a bearing surface?

--
You received this message because you are subscribed to the Google Groups "Delta robot 3D printers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to deltabot+u...@googlegroups.com.
To post to this group, send email to delt...@googlegroups.com.
Visit this group at http://groups.google.com/group/deltabot.
For more options, visit https://groups.google.com/d/optout.

[Haydn Huntley]

@Gareth, Actually I've carefully tested my rod end designs and there is a small gap between the ball and magnet.

It is possible that with a different printer, they would contact, but I haven't heard any complaints so far.

You could try printing a pair of the outside pieces, insert the magnet and ball, and then use an ohm meter to verify they aren't touching (that there is no conductive path between them).

[Martin Bogomolni]

Got 'em!   time to print some arm ends up.

-Martin

--

[Haydn Huntley]

@Gareth, If you do need more distance between your magnets and ball studs, you can increase the value for magnetOffset from 3.4mm to something larger in the OpenSCAD file.

[Adam P.]

Hayden, 

I see that your carriage is meant for roller wheels and for 1515 extrusion. Do you have any recommendations for parts for just the carriage? I am planning on building a 2020 MK with printed sliders and separate magnetic carriages. 

Thanks, 

Adam 

[Lloyd Joseph]

Haydn,I would be interested in 24 if you would post to Melbourne Australia? Paypal would be my easiest payment method. (Speed of freight is not important as these are the very first parts for my delta printer!)