Idea for magnetic joint

Hey guys,

I'm currently designing a delta printer with stuff that is scrap/surplus at work... I'm lucky, I work as an engineer in an Automotive parts manufacturers.

Anyway, whilst looking for a ball joint solution I have taken an interest in magnetic solutions.. They are quick release, low to zero backlash and, well, cool.

I really like the versions where a 10mm ball ended cutter creates a socket from a hex head bolt, I can replicate these easily in the work-shop. These got me searching on ebay and I came across hundreds of neodymium magnets that have a countersunk hole in them, presumably at a 45. I also Google'd '10mm ball stud' and put two and two together..

Perhaps we could use steel ball studs and insert them into one of these chamfered magnets?? I don't propose that the surface of either the ball stud, or more pertinently the magnet, represent a good bearing surface, but with a Teflon interface or a silicone based grease would create a very low friction, and reasonably good joint. Perhaps a ball end mill could be used to finish the surface of the magnet?

What do you guys think?

[www.ameritoolmfg.com]

[www.first4magnets.com]

Jai

Edited 1 time(s). Last edit at 11/03/2014 06:30PM by jaistanley.

Quote
jaistanley
Perhaps a ball end mill could be used to finish the surface of the magnet?
Jai

"20. Can I cut, drill, or machine neodymium magnets?
The Neodymium Iron Boron material is very hard and brittle, so machining is difficult at best. The hardness of the material is RC46 on the Rockwell "C" scale, which is harder than commercially available drills and tooling, so these tools will heat up and become damaged if used on NdFeB material. Diamond tooling, EDM (Electrostatic Discharge Machines), and abrasives are the preferred methods for shaping neodymium magnet material. Machining of neodymium magnets should only be done by experienced machinists familiar with the risk and safety issues involved. The heat generated during machining can demagnetize the magnet and could cause it to catch fire posing a safety risk. The dry powder produced while machining is also very flammable and great care must be taken to avoid combustion of this material."

[www.kjmagnetics.com]

Also once the plating is damaged, the iron will corrode.

My first ever Cherry Pi design used counter sunk magnets and steel balls. The holding force was never enough. I never tried countersunk with magnetic balls as by that time I had started to use the machined bolts and 10mm ball magnets. They work ok, up to a point. The limiting factor is the speeds you can reach, around 80mm/s reliably in my case. Some people have reported premature wear on the magnetic balls. Good lubrication is key. My latest iteration of effector attachment uses printed sockets on the effector and carriages with 8mm drilled balls on the rods. All held together with a length of spectra line and a spring. It works great and it gives you all the advantages of magnetic attachment with none of the drawbacks. I reliably print at 100mm/s speeds now. The limiting factor now is the extruder not being able to keep up. I'm working on that now.

Andy

Edited 1 time(s). Last edit at 11/04/2014 02:31AM by AndyCart.

Aaah, I was concerned about machining the magnets.

Guess I'll go for decent ball joints, or a magnetic joint based on spherical neodymium magnets.

Thanks guys!

Jai

I went with 1/2 inch (12.7mm) spherical magnets and a ball end mill to cup some allen-drive M5 SHCS's for my joints. Stuck a bit of waxed paper under each screw for lubrication, and they have been fine through three spools of filament so far. The only grief that I had was caused by the CA glue not holding the spheres tightly, which I believe that I have solved. Never had a rod pop off pushing 120mm/second, except for when the CA let go and released the magnet/rod from the effector plate or carriage.

With that said, I still think that the spring solution with the rods in compression is probably the better solution if you can master getting the rods in place under spring tension.....