Planning new Cheapskates...

[critical_limit]

Hi,

here is my conversion of some Skates I´ve seen anywhere in the internet.

Made my own ones for the Rostock. Balls are Delrin 8mm, all other parts are printed.
At the Moment it´s only a idea. Will it worth a try?

Anyone tested these type of skates already?
They will be louder than the originals, but when I´m finished planning (and testing), they will be without any play to the boom because they are preloaded to the boom a little bit. Have no Idea how long they will last with printed ABS, but if it works fine we can discuss other materials.

So, any comments are welcome....

[Polygonhell]

I printed Johann's version and tried them on 15mm extrusions, short version is they are easily the best printed bearings I've tried. The downside is they are not as smooth as real bearings or even the existing cheapskates.
The issue is you can feel the transition between the balls, my best guess is the seam creates an uneven point.

[Mac The Knife]

Also, the recirculating ball systems I've seen use two different size balls,,,, the smaller diameter balls are more of a spacer, and since they only contact the load bearing balls, the system doesn't get bound up.

[critical_limit]

Ok. Thanx for the replies.

After I finished my magnetic-ball-end mod, I will give it a try. Printing with 0.08mm Layerhight and .25 Nozzle should be ok. Found 8mm high precision POM-Balls. And lubricated with PTFE spray they should run very smooth.

Not as smooth as "real" bearings, but we will see. First I have to finish the magnet end ball mod.

But I´m tired of adjusting the normal skates. Every 2-5 prints I have to re-adjust, because they came loose and have to much play.

Which high-precision system you guys using? Havnt found any mod here for the skate-system.

cheers
Dirk

[Eric]

Thanks for reminding me...

I've let my version of that sit on the backburner. I've had the parts for ages, but I got distracted with other projects.

[Jimustanguitar]

Have you seen the prototype for the new carriages that SeeMeCNC is working on?

http://imgur.com/a/zmGoE

[Mac The Knife]

Have you seen the prototype for the new carriages that SeeMeCNC is working on?

http://imgur.com/a/zmGoE

They need to include some led holders, those would light up nice with leds.

[critical_limit]

New Cheapskates looks good. But no chance to modify for magnetic ball ends

[teoman]

I am sure you will be able to adapt it for magnetic arms. Just need an extra part.


The real question is why doesnt seemecnc start using magnetic arms.

[Polygonhell]

I am sure you will be able to adapt it for magnetic arms. Just need an extra part.


The real question is why doesnt seemecnc start using magnetic arms.

Go read through the threads on the delta bot google groups, from people who've done the modification, there are issues that can arise, the most common of which is the magnetic joint disconnecting while printing. Getting a reliable design with mass produce-able parts would probably take some iteration.

FWIW If I were looking to improve the current arm design, I think the biggest issue is the flexibility of the nylon arms not the UJoint design, the arm issue manifests itself at the edge of the build area if there is a twisting force on the platform. It's mostly a none issue on the later machines because the extruder was relocated to the top of the machine removing the most significant twisting force.

[teoman]

It seems to me that the existing design for the skates is ok. And it should perform better than anything printed. You cpuld look for an industrial ball bearing / slider type of solution if you wanted.

If the adjusters have gone lose try tightening the screws down.

If that did not work, I would be trying to print half of an adjustment wrench with a hook at the end and the using a light rubber bad to apply a constant torque. That way it would be as if they were constantly re tightened lightly.

[bdjohns1]

Go read through the threads on the delta bot google groups, from people who've done the modification, there are issues that can arise, the most common of which is the magnetic joint disconnecting while printing. Getting a reliable design with mass produce-able parts would probably take some iteration.

I've done tight octagonal infill at 120mm/sec on mine. Shakes something fierce, but no pop-offs. At some point, I need to do some dry run testing without worrying about pushing filament just to see how hard I can run mine before they pop off.

I think the problem a lot of people have is that they're using JB-welded balls on SHCS. Most of the failures I've read about were balls detaching. I'm using Traxxas ball studs, which eliminates that problem. I imagine with the volume of parts SeeMe is going through, they could contract out making a few thousand of those and do better than Traxxas's studs, which aren't perfectly spherical, but seem to be close enough.

There are plenty of linear industrial motion components out there - the problem is that most of them involve milling of parts with fairly tight tolerances to allow for 18-24" of travel. And most of those are overengineered for deltas because they're designed for heavier loads, like CNC milling machines. The Shapeoko CNC can do light milling in aluminum with its setup (delrin wheels on V-rail), but anything stouter needs steel.

The next step in deltas (in my mind) would be more like how most pick-and-place delta work that I've seen - the motors are all housed in a central location above your 0,0 point, and the arms are all hinged. At that point, you've eliminated the need for linear motion systems. It's probably a tougher arm setup to do, but you can just slap the head on a fixed surface above the build platform.

[critical_limit]

...
If that did not work, I would be trying to print half of an adjustment wrench with a hook at the end and the using a light rubber bad to apply a constant torque. That way it would be as if they were constantly re tightened lightly.

jo, that´s maybe the easiest way. preloaded by a spring or rubber. Will try that first. The prototype ones seems to be also put little preload to the wheels. Good Idea.

[Jimustanguitar]

The real question is why doesnt seemecnc start using magnetic arms.

There's not really an advantage to justify the added expense. Most of us bought their machine based in part on a price point, and injection molded u-joint mounts with common 6-32 screws are about as high of a value as you can get. And they do work extremely well, so why "fix" it, you know?

[Jimustanguitar]

The next step in deltas (in my mind) would be more like how most pick-and-place delta work that I've seen - the motors are all housed in a central location above your 0,0 point, and the arms are all hinged. At that point, you've eliminated the need for linear motion systems. It's probably a tougher arm setup to do, but you can just slap the head on a fixed surface above the build platform.

Already exists in the RepRap world. Probably one of Nick Seward's designs, but not sure.

[Polygonhell]

The next step in deltas (in my mind) would be more like how most pick-and-place delta work that I've seen - the motors are all housed in a central location above your 0,0 point, and the arms are all hinged. At that point, you've eliminated the need for linear motion systems. It's probably a tougher arm setup to do, but you can just slap the head on a fixed surface above the build platform.

Already exists in the RepRap world. Probably one of Nick Seward's designs, but not sure.

There have been several, they usually don't work well, because of the motion resolution required, the motors turn less than one revolution for every point in the print space. The solution is gearing the motor down, but any play in the system at the motor gets massively amplified by the arm levers. It's the same reason that scara bots haven't really been big in printers.
This is solved in industrial designs by using what's known as a harmonic drive, they are very high resolution, low backlash motors. And unfortunately very expensive.
You also lose he ability to scale the height separately, but that's not a huge issue.