Welcome to the SeeMeCNC Community

Tinyhead wrote:

Nylocke wrote:Looks like filament starvation, is that PLA? You should probably use ABS for the joints.

It was filament starvation. Sitting there, the PLA would ooze from the nozzle, but as soon as I put pressure on it at the EZstruder... it would just jam up.

I've almost got my setup ready to go. I didn't have any 3/8 x 3/8" cylinder magnets, so I ended up using 3 x 3/8" short magnets and stacked them together. It ends up being .300" thick altogether, but the attraction feels pretty strong.

My rods are going to be aluminum, but I'm a bit worried about the weight that will be throwing around. I haven't found anyone local that carries 3/8 carbon tube and the ones online are a bit out of my price range. I've seen people use the archery arrow shafts, but I haven't found any that are 3/8. They all look to be around 5/16".

20140403_221724.jpg

20140403_221750.jpg

20140403_221803.jpg

I spec really strong n52 grade magnets. I haven't tried lesser magnets and it is possible youmay run into issues. If it is all you have you can try it. I used hollow aluminum tubes on mine. Checkout local hobby shops for them. Arrow shafts won't be the correct diameter.

I used arrow shafts on my set and what I did was make a plug/adapter/sheath to increase the diameter of the ends. As long as it's all centered right I see no difference. If it isn't all centered right, it's another way to drive yourself mad at a later time. To make the part I just threw two cylinders into OpenSCAD and compensated with any printing differences in the parameters to get it just right.

Ok. Got some aluminum tube and made all my rods 278.5mm on the nose from ball to ball. I ended up putting 4 of the magnets together instead of 3 and it gave me a bit more holding power. Then ends are now glued into place to the rods. Now... the only things stopping me from making the install is the fact that it was so difficult to get my stock set up right in the first place. One of the u-joint axles on the cheapskate is warped and it needed some VERY fine tuning to get specific screw tightness correct across the axle to allow the u-joints to move freely. I don't want to put in the new arms and find the quality has gone down, or I messed something up and suddenly I've got to try and get back what I had. I'm really very impressed with the print quality I'm getting right now with the stock set up and I'm just wondering now... do I fix it if it isn't broken?

Tinyhead wrote:Ok. Got some aluminum tube and made all my rods 278.5mm on the nose from ball to ball. I ended up putting 4 of the magnets together instead of 3 and it gave me a bit more holding power. Then ends are now glued into place to the rods. Now... the only things stopping me from making the install is the fact that it was so difficult to get my stock set up right in the first place. One of the u-joint axles on the cheapskate is warped and it needed some VERY fine tuning to get specific screw tightness correct across the axle to allow the u-joints to move freely. I don't want to put in the new arms and find the quality has gone down, or I messed something up and suddenly I've got to try and get back what I had. I'm really very impressed with the print quality I'm getting right now with the stock set up and I'm just wondering now... do I fix it if it isn't broken?

If you don't use the new magnetic arms you will always have a lingering doubt wondering if they would be an improvement or not. I would have to make the change to see what the results are or I would be constantly wondering.

On a side note, I just received my First150 order from Misumi USA which had my set of 3/8" cylinder magnets (full size) and tossed it into a spare rod end just to see how much stronger it would be and was pleasantly surprised to find that the 4 thinner magnets together held onto the ball tighter than the full 3/8" cylinder magnet.

Also, no kidding about not letting them snap together. It pulled one right out of my fingers and clapped together, breaking off a chunk from one magnet. Amazing how powerful these little things are.

Tinyhead wrote:On a side note, I just received my First150 order from Misumi USA which had my set of 3/8" cylinder magnets (full size) and tossed it into a spare rod end just to see how much stronger it would be and was pleasantly surprised to find that the 4 thinner magnets together held onto the ball tighter than the full 3/8" cylinder magnet.

Also, no kidding about not letting them snap together. It pulled one right out of my fingers and clapped together, breaking off a chunk from one magnet. Amazing how powerful these little things are.

They will also take a chunk out of your skin if you get some under there when they decide to snap together. An experience I had to learn the hard way.

I found some el cheap crossbow bolts at sportsman's that were to spec, I never ended up using them for the MAX that was planned to get the mod, but I did use the joints (without the bolts) for my Rostock Mini with an additional mod shaft coupler for the default carbon arms for the Rostock Mini.

Well I took the plunge.

Removed my stock arms and assemblies... and I'm kinda glad I did. A few of the arms were very stiff. I'm surprised I wasn't seeing errors in my prints. One of the u-joints was almost seized in place, and yes, they were all lubed. Either way, it needed some TLC. I only have about 10 days of print time on the machine, so that's some pretty quick wear. With all the other people with joints seizing up though, it shouldn't be much of a surprise.
Installed the new effector... man that's nice and easy. Just snap into place and BAM! Done. I made a second effector for the E3D, but haven't assembled it yet. I'd like to make another effector with a dial indicator for easier future calibration too. Being so easy, it's hard not to!

There was quite a difference in my delta_radius with the new arms and with the upgrade, I also lost about 18mm of Z height. I have to go back and read, but I'm not sure if that's typical, but I'm guessing it must be. Changed my effector spec in the configuration.h as specified before the calibration and now I'm printing.
I'm running off a 6hr print that I had completed earlier today with everything stock so I'll be able to see how it adds up. Exact same file with settings so it should be a direct comparison. I'll throw up pictures tomorrow after work. I'm hoping to see a quality difference, but I'm a bit scared there is going to be a scaling issue. I tried taking some callipers to it with the hot end whipping around and at first glance it seems like it's printing too big. 7.3mm instead of just 7mm kind of thing. I guess I'll be able to check for sure in the a.m. though.

Well here are the pictures of my part result:
There are a few horizontal 'artifacts' on both of the prints, which would make me think slicer/gcode error, but they're not always in the same place. Very close though. Must have been something to do with printing speed.

I was incorrect in the assumption that parts were going to be too big. They're within .02mm of my original part (and about 0.1mm or less out from actual part design).

Flat walls are very slightly smoother with the mag arms than the stock arms. It's hardly noticeable. I guess it goes to show how well the redesigned stock arms really are. If there is one appreciable benefit to the upgrade, it's ease of assembly/disassembly and changing out hot ends, little to no maintenance and you don't have to worry about any u-joints jamming up.

In the end, I'm still quite glad I did the upgrade (especially after finding those failing u-joints). It's been on my list since the beginning and now I can tick it off.

EDIT: After about 10hrs of printing... I lost my first ball. Effector fell down into the print. Luckily I was there and it was only about 20 minutes into the print. I can grab the galls with pliers to tighten the nuts and I thought that would be a good test. They held nice and strong. Toss a little bit of heat on them though, and they pop right off. I'm definitely going to tack weld a few sets of balls.

Hello all,

This is my first post to the forum, just finished assembling my first 3d printer (Max v2). I have already started on making my own set of magnetic arms, thank you xnaron for the stl files for the parts.

I have been thinking of another way of attaching the balls to the cap screws other than the JB weld. Thought of everything from tack welding with the MIG, brazing, friction welding. I ended up soldering them using a specialty kit from bernzomatic that I had kicking around in the shop. I sanded the top of the cap screw to remove the coating and expose the bare metal. I then cut a small bit of the solder off and placed it in the Allan wrench recess. Then I coated the top of the cap screw with the supplied flux, adding a little blob in the center to hold the little clipping of solder in. The ball was sanded where it would contact the cap screw and flux was applied to that area. I then pushed the cap screw against the ball in a aluminum faced vice to prevent damage. I took my MAAP gas torch to the assembly and heated until the solder flowed out around the mating point of the ball and screw. Heat was removed and allowed to cool ( leaving in the aluminum faced vice cooled faster than if removed, even after 5 min the first assembly I made that I removed right away was to hot to handle. But if left in the vice, cool to the touch in a min, the joys of heat sinks.

I tested one of the assembly's by clamping the screw in the vice and hitting with hammer, took 3 wacks to break the bond.
So with these done off to make the rest of the parts

That was a great idea and has to be a lot stronger than just the epoxy. I would have thought the solder would have to be hand fed around the ball while heating but you have proved that the solder would flow out from the center
to the outside of the ball. Brilliant! Thinking outside the box proved that there are many ways to do things. Thanks for this post!

RichardMac wrote:Hello all,

This is my first post to the forum, just finished assembling my first 3d printer (Max v2). I have already started on making my own set of magnetic arms, thank you xnaron for the stl files for the parts.

I have been thinking of another way of attaching the balls to the cap screws other than the JB weld. Thought of everything from tack welding with the MIG, brazing, friction welding. I ended up soldering them using a specialty kit from bernzomatic that I had kicking around in the shop. I sanded the top of the cap screw to remove the coating and expose the bare metal. I then cut a small bit of the solder off and placed it in the Allan wrench recess. Then I coated the top of the cap screw with the supplied flux, adding a little blob in the center to hold the little clipping of solder in. The ball was sanded where it would contact the cap screw and flux was applied to that area. I then pushed the cap screw against the ball in a aluminum faced vice to prevent damage. I took my MAAP gas torch to the assembly and heated until the solder flowed out around the mating point of the ball and screw. Heat was removed and allowed to cool ( leaving in the aluminum faced vice cooled faster than if removed, even after 5 min the first assembly I made that I removed right away was to hot to handle. But if left in the vice, cool to the touch in a min, the joys of heat sinks.

I tested one of the assembly's by clamping the screw in the vice and hitting with hammer, took 3 wacks to break the bond.
So with these done off to make the rest of the parts

Thanks for posting this! I have another set to make up soon and will try this method. One thing you need to be careful of is that you don't scratch the ball on the surface the PLA cups will traverse. This will cause increased friction and wear. I see you used the aluminum to protect it in the vice...that was good. I will see if I can find some of that benzomatic flux. What kind of solder did you use or did it come with the flux?

Brendin

The solder will naturally flow to the mating edges due to capaliray action, same as sweating copper pipe.

The solder comes with the flux, the solder is the magic. Flux is flux for the most part, but what makes up the solder is the important part. I think this one has a higher content of silver than other solders.

Double check each one after they cools. I was threading on in to the effector and the ball popped off the cap screw. Must have not gotten it up to full temp. You want to heat it up just until the solder flows into the mating surfaces. But don't heat up to the point that the chrome starts to debond from the steel


Richard

I'm just wondering, do the chrome steel balls lose any magnetism from being heated so much? Pardon my ignorance on metallurgy and magnetism!

Generic Default wrote:I'm just wondering, do the chrome steel balls lose any magnetism from being heated so much? Pardon my ignorance on metallurgy and magnetism!

The chrome balls are not magnetized so the answer to your question is no.

Hi guys,
what do you thing of this, almost free, solution?

I'm considering the following magnets too:
http://www.ebay.com.au/itm/10-Strong-Ro ... AQ:AU:1123

I can't tell what exactly is going on in that picture, but it looks like all metal on metal contact.

The threaded rod might work great for rigidity, but it probably weighs more than other types of arms. I used brass tubes on mine.

I don't know how the magnets will work with steel around them. Steel isn't very permeable for magnetism. Also, steel rubbing on steel might have more friction and more wear than you want.

Here is my effector plate with the soldered balls to the cap screws. Have done a few prints and all looks good so far.

RichardMac wrote:Here is my effector plate with the soldered balls to the cap screws. Have done a few prints and all looks good so far.

Very nice effector. Are you interested in making and selling these?

RichardMac wrote:Here is my effector plate with the soldered balls to the cap screws. Have done a few prints and all looks good so far.

Looks good. Can you post some details on how you have that fan setup?

The effector is a one off ( well actually 3 off but I broke my M3 tap threading the holes on the second one, waiting on a new tap).
I laid out the design on a sleet of 1/8" aluminum and cut by hand with a hack saw, then drill press for the holes, and made the bends using a bench vise and clamp.
The piece is a little rough on the finish, but gets the job done. if I had a quicker way of making the plates, I would consider it and figure the total cost of supply's . But at this time I would not be looking to sell the plates.

The fan set up is the same fans that came in the Kit for the max V2. same printed shrouds that were the first two items to be printed ( PEEK and layer ).
the aluminum plate basically replaces the stock effector, I used Xnaron's effector SLT design to make a flat design that would be bent. I added a small tab where the layer fan shroud would attach, same as the stock item.

Here is a video of stud welding the SHCS to the ball bearing for the ZeGo delta bot. Looks like a good solution if someone wanted to produce a bunch of them
http://vimeo.com/93456248

Here's a cool ball bearing u-joint design that doesn't use magnets, but uses string to provide the tension.

https://www.youtube.com/watch?v=RM1ytIy ... e=youtu.be

I do like the easy disassembly with the magnetic joints, though.

Broose wrote:Here is a video of stud welding the SHCS to the ball bearing for the ZeGo delta bot. Looks like a good solution if someone wanted to produce a bunch of them
http://vimeo.com/93456248

This is the way to do it here, I can't believe that I didn't think of this since I have used a similar technic for bonding 2 bearing together to create "hurricane balls"

https://www.youtube.com/watch?v=cvq8laPb498

Flateric wrote:

Broose wrote:Here is a video of stud welding the SHCS to the ball bearing for the ZeGo delta bot. Looks like a good solution if someone wanted to produce a bunch of them
http://vimeo.com/93456248

This is the way to do it here, I can't believe that I didn't think of this since I have used a similar technic for bonding 2 bearing together to create "hurricane balls"

https://www.youtube.com/watch?v=cvq8laPb498

I wonder what Harbor Freight welder would have the power to do something like that?