Xnaron Printable Magnetic u-joint option for Rostock Max

[thenewguy]

I checked quickly before I left for work and the rod was around .372- .374. The balls and magnets were about the same size as well. I used a bit but it didn't do much it was just shy of letting the rod set in place. I guess that's going to be my option in the end.

I figured it was a setting on kiss. That's why I'm asking. I tired increasing and decreasing abs in the extruder and that didn't help too much either.

[Broose]

I figured it was a setting on kiss. That's why I'm asking. I tired increasing and decreasing abs in the extruder and that didn't help too much either.

I assumed from your previous post that you tried scaling it up in Kisslicer to fit the rods and then the balls didn't fit correctly in the sockets (too loose). If so, I don't know of any other way to increase the size of the hole with a kiss setting that doesn't also increasing the size of the socket, hence the suggestion of the drilling option.

[barnett]

Either the arms will fit and the balls will have to much play in the socket or vise versa.

I had this same problem. My 8 yr old saw me filing away at it and suggested the solution: make your own. The one I used (stl & scad) is attached. If something is not fitting for you, just tweak the parameters in the scad.

arm_end.stl

(310.93 KiB) Downloaded 503 times

arm_end.scad

(733 Bytes) Downloaded 489 times

[Eaglezsoar]

Xnaron: I have the geared motor you used on your Bowden extruder and will be printing the printable parts soon.
Could you give me the steps per mm you use with your Bowden extruder so that I may be in the ballpark when calibrating?
Thanks in advance!

[thenewguy]

Xnaron: I have the geared motor you used on your Bowden extruder and will be printing the printable parts soon.
Could you give me the steps per mm you use with your Bowden extruder so that I may be in the ballpark when calibrating?
Thanks in advance!

I am in the same boat. Have the motor on the way and the gear in the link. I'm looking for a write up though on any other parts used like what size screws ect.

Also I found a trick with the sockets for the balls. If you have a little play in them I take an extra ball heat it up over the stove. Once it gets pretty hot I place it in the socket for a sec or two until I see a very small amount of abs melt at the edge of the ball and quickly place under cold water. Just melting out any imprefections. Balls are now very tight in socket with zero play. Hopefully it don't throw off any measurements.

[doctorgonzo]

Also I found a trick with the sockets for the balls. If you have a little play in them I take an extra ball heat it up over the stove. Once it gets pretty hot I place it in the socket for a sec or two until I see a very small amount of abs melt at the edge of the ball and quickly place under cold water. Just melting out any imprefections. Balls are now very tight in socket with zero play. Hopefully it don't throw off any measurements.

You can also use a drop of acetone for this process if you're using ABS...it may be more repeatable. Just a thought.

[gestalt73]

Hey guys, a few quick questions to Xnaron and those who have already done the mod:

1. How many magnets do we need if we're following Xnaron's instructions and link? Is it one magnet per ball end? (12)

2. It's mentioned that the ball actually rests against the plastic, and not the magnet. Is there any concern about repeated use over time wearing down the plastics of the caps and changing the geometry of the arms?

3. Would it help / harm to put a dab of lithium grease in the ball magnet interface to decrease wear and friction?

[Broose]

Hey guys, a few quick questions to Xnaron and those who have already done the mod:

1. How many magnets do we need if we're following Xnaron's instructions and link? Is it one magnet per ball end? (12)

12 is correct

2. It's mentioned that the ball actually rests against the plastic, and not the magnet. Is there any concern about repeated use over time wearing down the plastics of the caps and changing the geometry of the arms?

The loads are very low. Mine have run for a few months and it kind of burnishes the surface (its shiny at points) but its not really wearing

3. Would it help / harm to put a dab of lithium grease in the ball magnet interface to decrease wear and friction?

I don't do that because i find I need to clean tiny metal chips off periodically from the magnet or ball. If you can assure you won't have that, then I don't see that it would hurt.

[Flateric]

In my continuing quest to quiet and kill all extra vibration in my machine I have another tip.

First number and label your arms for the location and orientation if you remove them from your machine. This will save you tons of recalibration time.

Second, if you are using hollow arms, such as CF or aluminium, fill them about 1/4 up with very fine sand. The sand does wonders for absorption of unwanted vibrations. Machine will be quieter and print smoother, trust me.

[thenewguy]

In my continuing quest to quiet and kill all extra vibration in my machine I have another tip.

First number and label your arms for the location and orientation if you remove them from your machine. This will save you tons of recalibration time.

Second, if you are using hollow arms, such as CF or aluminium, fill them about 1/4 up with very fine sand. The sand does wonders for absorption of unwanted vibrations. Machine will be quieter and print smoother, trust me.

Very interesting! I have many years of car audio experiences maybe I can help : )

The best way to help vibration is basically adding weight or pressure against the vibrations. Elimination of the source is really the most ideal way. But wavelengths from speakers was a different story!

I know you started with the motors like me. : )

I still have a lot of noise since mine is in a wooden enclosure acting like a speaker box. It's pretty intense! But so much better! i also did the extruder posted in the first page. Big difference in print quality and noise!

Thanks again for your work xnaron!

The wooden enclosure just don't look cool it does work! I only print abs so far and it's nice. Stopped my abs from curling right away! If I leave the window open I get it just on that side so that's proof.

Back to your idea. Try filling your cf arms with great stuff "the foam". It adds some pressure even though its cf! But adds weight to the whole surface area. Not to much weight as sand. It could work better i don't know? But you want things to be light as possible from what I keep reading. Since you sparked that idea. How about filling the center holes of the x,y,z beams with sand or foam? ; )
Also the underside of of the bed under the wood. Use sheets of tar with aluminum backing. Just adding weight to kill vibration!

I may have a few things to do when I get home. : )

[itilguru]

Either the arms will fit and the balls will have to much play in the socket or vise versa.

I had this same problem. My 8 yr old saw me filing away at it and suggested the solution: make your own. The one I used (stl & scad) is attached. If something is not fitting for you, just tweak the parameters in the scad.

arm_end.stl

arm_end.scad

A thousand thanks for these files. I was finally able to print a set of caps and finished this mod!

A few ideas for people having mixed results:
Make sure every end has reasonably close to the same pulling force. One weak connection will inevitably pop loose at the worst time. Mine are very strong now, but started kind of weak, but since I had them all the same I thought that was normal.

You can use acetone to polish the cup ends using a spare bearing. I made a few extra with screw mounts, and they're incredibly useful. One you've removed the bumps, rinse in water.
Someone here suggested heating up a bearing to remove the bumps. I tried it, but it caused me to get a bearing stuck inside a cup and nearly ruined that arm. But it did show me that all the arms were too weak, so that led me to better connections later.

Instead, I used a careful end cup melting process followed by using a cold bearing on a screw to polish the end. Put the magnet inside first! Worked great. Do this before you glue the ends on the arms though. I didn't, and had to remake one arm.

My careful process was lighting the ABS for three seconds. Not recommended indoors. Heat gun would have been better.

The arms are awesome, and I'm printing more upgrades now. I'll use modular connections and make a set of custom use effectors - air powered syringes, other extruder designs, dual extruders, etc.

[PartDaddy]

I think someone mentioned this or something similar somewhere in the forum. I'm contributing this concept of how I would tackle the magnetic arms because I really liked Xnaron's work.

The ball stud is screwed into both ends of the arm rod. This allows a jam nut to secure the center to center distance of the ball studs very precisely. The rod doesn't have to be carbon fiber. Aluminum tube or even solid aluminum rod may be a better choice.

The neodymium magnets are on the platform and carriage. The magnet is adjustable in distance from the cup that receives the ball stud. There's a variety of ways you could achieve adjustable force of attraction between the ball stud and neomagnet. The attraction force adjustment is necessary to set holding force and friction between the ball stud and cup. The carriage (not shown) also has a similar magnet assembly for adjusting the magnet for controlling force/friction between the ball stud and cup. There doesn't have to be an elaborate adjustable magnet assembly. Like Xnaron, you can test, print, test, print and find a desired force and just draw printable parts to hold the magnet.

Material choice can make a big difference. Acetal is a good choice for making the cup. And a chrome ball for the stud. The drawing files are not very good and not scaled properly to be useful. I just wanted a drawing to show everyone and add to the discussion and improvements of, Magnetic Arm style Parallel Kinetic Mechanisms.

If I get the time, I will make a printable prototype version of this too and share the files on our git. Great work Xnaron!!

NewMag Delta Robot Platform Assembly

[Eaglezsoar]

PartDaddy, your idea is a great one. It's just so difficult for the average user to create such parts.
The Xnaron project has been posted for a long time now and I am surprised that someone has not
come up with a complete kit to offer for sale. I think a lot of it has to do with the time factor, no one
has enough time to create such kits and still "play" with the RMax.
The magnetic arms offer accuracy and the ability to add and remove effectors with ease, IMHO they
are the best of the best and represent the best mod you can do to a delta printer.

[lordbinky]

The ONLY thing I find as a slight issue with Partdaddy's take on the magnetic arms is the magnets are on the platform, the carriage carrying magnets is fine. One of the benefits of the magnetic arm setup is that the platform is easily swappable, and the magnets being a part of the platform either increase the cost per platform to have magnets on multiple platforms, or increased swap time by removeing the magnet/cup assembly and swapping that with the platform.

To restore that advantage in this design without going back to the original arrangement, I would like to see what people would think about methods to make swapping the magnet cup assembly faster to attach/detach from multiple platforms while staying as secure as a fixed setup.

[Eaglezsoar]

The ONLY thing I find as a slight issue with Partdaddy's take on the magnetic arms is the magnets are on the platform, the carriage carrying magnets is fine. One of the benefits of the magnetic arm setup is that the platform is easily swappable, and the magnets being a part of the platform either increase the cost per platform to have magnets on multiple platforms, or increased swap time by removeing the magnet/cup assembly and swapping that with the platform.

To restore that advantage in this design without going back to the original arrangement, I would like to see what people would think about methods to make swapping the magnet cup assembly faster to attach/detach from multiple platforms while staying as secure as a fixed setup.

In the PartDaddy design it appears that the magnet is not permanent to the platform, in other words you can remove the magnets and place them on a different platform
without too much time or difficulty. I don't think you would have to buy new magnets for each platform that you have.

[lordbinky]

AH! I see it now. The cup doesn't have to be affixed to the platform stud if it is metal. Thank you eagle. In case anyone else has that same issue I made a cheat sheet, blue is metal, green is the magnet, purple is the low friction material.

[Flateric]

The disadvantage of having the magnets mounted rather than on the arm and the added hardware this also adds I discovered early on in my initial unsuccessful tests before Xnaron was successful with the use of more powerful magnets.

The magnets need to be of the cylindrical type to really be effective and keep the strongest magnetic attraction in an even consistent manner through the entire movement. I had initially tried ball shaped magnets that were very strong but had a extra strong sweet spot that introduced subtle but noticeable irregularities in movement in the final print. Waves is the best way to describe it.

I also had tried hemisphere magnets in steel cup to focus the strength of the magnet in one direction. This made the above problem even worse.

Being that the magnet is shaped like a cylinder it seem logical and easiest to attach the arm and magnet which also allowed for the least amount of hardware to make this connection happen and keep the cone of magnetic force aligned to the ball it was rotating on best aligned at all times. Also the cylindrical shape of the magnet allows a larger magnet and the additional strength of a larger (longer) magnet to be used without affecting the height from carriage to ball (when magnet is mounted on carriage or effector) to be increased. Which also would lower build height and increase the effectors weight and the costs if you had multiple swappable effectors. Since small chrome balls will always be cheaper than strong cylindrical magnets (especially when they a longer in length than the diameter of the ball which is kept small for weight considerations)

Even longer and stronger magnets may be used while keeping the stock arm length of 269mm by simply shortening the arm length to accommodate (leaving the firmware untouched, something that was more important than nowadays with version 0.90 of repetitive soon too be released which allows adjustment within the host and no exit and reef lashing)

And finally one of the biggest goals on top of all of the above concerns from the very start was to make the project as accessible and as cheaply constructed as possible for anyone without any special machine tools or skills anywhere. My current and favorite arms are currently just carbon crossbow bolts cut to length, with cylinder magnets at the ends and then small nylon cups that Rona and Home Depot sell as snap caps in their hardware nut and bolt aisle for 0.12 cents each. This assembly is all bound together with Kapton tape and then given a good but thin coating of silicon (plumbers goop is the brand I use)

The nylon cups are chosen because they are the easiest, fastest, and most consistent way of creating a none slip cups shaped surface that retains the balls with almost no friction. But the biggest feature is the consistent size they provide. Makes all of my arms exactly the same with no variation in length or friction to mess things up.

Just thought I would add more to the discussion and some of the thinking and testing that happened before the resulting successful magnetic arms.

[Flateric]

Almost forgot, the nylon cups also create the exact same spacing and force at each arm end due to them being a premade product.

All of this in the end contributes to a lighter weight end result with fewer components to go wrong.

[Eaglezsoar]

The disadvantage of having the magnets mounted rather than on the arm and the added hardware this also adds I discovered early on in my initial unsuccessful tests before Xnaron was successful with the use of more powerful magnets.

The magnets need to be of the cylindrical type to really be effective and keep the strongest magnetic attraction in an even consistent manner through the entire movement. I had initially tried ball shaped magnets that were very strong but had a extra strong sweet spot that introduced subtle but noticeable irregularities in movement in the final print. Waves is the best way to describe it.

I also had tried hemisphere magnets in steel cup to focus the strength of the magnet in one direction. This made the above problem even worse.

Being that the magnet is shaped like a cylinder it seem logical and easiest to attach the arm and magnet which also allowed for the least amount of hardware to make this connection happen and keep the cone of magnetic force aligned to the ball it was rotating on best aligned at all times. Also the cylindrical shape of the magnet allows a larger magnet and the additional strength of a larger (longer) magnet to be used without affecting the height from carriage to ball (when magnet is mounted on carriage or effector) to be increased. Which also would lower build height and increase the effectors weight and the costs if you had multiple swappable effectors. Since small chrome balls will always be cheaper than strong cylindrical magnets (especially when they a longer in length than the diameter of the ball which is kept small for weight considerations)

Even longer and stronger magnets may be used while keeping the stock arm length of 269mm by simply shortening the arm length to accommodate (leaving the firmware untouched, something that was more important than nowadays with version 0.90 of repetitive soon too be released which allows adjustment within the host and no exit and reef lashing)

And finally one of the biggest goals on top of all of the above concerns from the very start was to make the project as accessible and as cheaply constructed as possible for anyone without any special machine tools or skills anywhere. My current and favorite arms are currently just carbon crossbow bolts cut to length, with cylinder magnets at the ends and then small nylon cups that Rona and Home Depot sell as snap caps in their hardware nut and bolt aisle for 0.12 cents each. This assembly is all bound together with Kapton tape and then given a good but thin coating of silicon (plumbers goop is the brand I use)

The nylon cups are chosen because they are the easiest, fastest, and most consistent way of creating a none slip cups shaped surface that retains the balls with almost no friction. But the biggest feature is the consistent size they provide. Makes all of my arms exactly the same with no variation in length or friction to mess things up.

Just thought I would add more to the discussion and some of the thinking and testing that happened before the resulting successful magnetic arms.

These snap caps are not available in the States and Home Depot Canada will not ship to the States. Would you be willing to sell these at say $1 a piece if we pay the shipping
to the States? I personally need 30 of them.

[Flateric]

I got your pm and I know they are available there as well, I'll get you the actual name of them when I go to homedepot today. They are used for a cover that goes over the head of bolts when you want to then snap a plastic colored cover over it. The nylon portion is the cover over the cap bolt then the colored cover snaps over this nylon cap. So that it hides the bolt or also servers to prevent the bolt from catching on clothing or scratching skin as you rub against it. Such as on boats and around pools, that sort of application.

[Flateric]

Here we go these guys....

http://www.leevalley.com/en/hardware/pa ... 41316&ap=1

And these are the covers homedepot sells for them, so they must sell the cups they snap onto.

http://t.homedepot.com/Tools-Hardware-H ... chNav=true

[DavidF]

Im still waiting on my magnets to show up to do some experimenting. Welding the balls to threaded studs has beeen proving difficukt to maintain concentricity. I may try making up a different jig for resistance welding them in the next couple of weeks but something else hit me in the mean time. Why not just have some magnets made to suit the purpose? I have my rostock assembled and just need to finish off the electornics and calibrate it, then ill gat back to playing with some mag rods.....

[Eaglezsoar]

Here we go these guys....

http://www.leevalley.com/en/hardware/pa ... 41316&ap=1

And these are the covers homedepot sells for them, so they must sell the cups they snap onto.

http://t.homedepot.com/Tools-Hardware-H ... chNav=true

After viewing the cups that are sold by Leevalley, I think I am finally starting to understand. Please correct me if I am wrong.
You started with the arrow shaft and placed the magnet on it, the cup goes on the top of the magnet centered over the hole in the cup.
You then placed a piece of adhesive heat shrink about an inch and 1/2 long over the cup, magnet and arrow shaft and applied heat to heat
shrink the assembly in place. Do I finally have it right?
The LeeValley diameter on the cups is 11/16" which is a little larger than the 3/8" magnet, would that matter?

[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.

[Eaglezsoar]

Flateric, thanks for helping me understand.
I also figured out why I can't find the cap bases in the States.
Here is what you found in Canada: http://www.homedepot.ca/product/6-8-sna ... tsk/956431
The USA HomeDepot do not carry anything from H.Paulin, the round screw covers that are carried have
small protrusions that fit the hole in a phillips screw. I talked to the manager at the local HomeDepot and
he stated that the snap cap bases are not sold anywhere in the States which means that we are SOL.
Thanks for all the info but I am going with the printed version. I have found if I pass a 3/8" stone on a
Dremel down the printed version, it makes everything fit as it should.
Thanks again Flateric and I won't be bothering you about this subject anymore.
Bet ya I can find more subjects to pester you about!