Magnetic Arms Good or Bad

Any opinions on switching to magnetic arms on a mini kossel vs traxxas/carbon fiber arms? It seems effector tilt is my biggest problem i would be in hopes this switch would combat this problem.

Is there a lot of re configuring that needs to be done for this switch?

Traxxas/CF arms can be very good if there is no play in the joints and the rods are accurately made to the same length. As with any type of arm, you need to make sure that the bearing spacing is the same at top and bottom. With Traxxas joints this is easy to adjust by adding washers or shims.

I have just changed from using wheeled carriages to linear rails, and my auto calibration deviation halved. However, I intend to try magnetic joints later this week.

See [duet3d.com] for causes of effector tilt.

Edited 1 time(s). Last edit at 03/14/2017 03:05AM by dc42.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Thanks DC42! I always appreciate your feedback.

At this point in time i believe have have corrected all my mechanical flaws and my spacing on my rods are all very similar ( i recently rebuild effector and made sure top and bottom washed are the same for the set), i have added rubber bands to squeeze the 2 rods to remove play from joints which increased my deviation of auto cal significantly i can now stick prints usually only small prints in the center of the bed, i want to start moving into software compensation, using H parameter in code has failed for me because it seems to fix only that probe point which throws the others off when using H to compensate for different trigger heights, What i want to try but have just discovered is the mesh bed compensation....i know very little about this but i am getting promesing results when i print larger prints with a slicer Z off set.

But i am running out of mechanical things to tune... the last thing i cannot correct for is:

somewhat loose traxxas inserts which i was thinking about using magnetic rods if i see/hear good things

THanks again let me know what you think.

Over the past week I have upgraded my delta printer:

- from 2020 to 2040 extrusion
- from wheeled carriages to linear rails
- from Igus joints to magnetic arms with Delrin ends made by Haydn Huntley. I ordered 360mm arms. When they arrived, each was labelled one of 360.28mm, 360.29mm and 360.30mm. I can live with 20um difference!
- from Robotdigg metal carriages and a printed effector to PCB effector and carriage adapters, which provide precise separation of the ball studs. I still use​ the Robotdigg carriages to anchor the belts and adjust tension.

For the first time, I have no significant​ effector tilt. See [www.duet3d.com] for photos. I suspect that the change to linear rails had the most effect.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Really interesting...You think that the rails helped the most. I was thinking and pricing that same upgrade. I am curious what benefits you see for the 2040 upgrade? Does this just add weight?. I really want to upgrade to rails + magnetic arms but i might as well upgrade to 2020 because i am still using an older/smaller 1515 mini kossel. Not sure what the scope of work because other things would also need to be upgraded like build plate and heat pad in addition i would imagine longer belts also still need to collect more info... But i really would like to increase the build volume if i can be confidant the rails and mags are worth it i might as well size them for the upgrade.

Initially I used the old arms, effector and carriages after upgrading the frame and changing to linear rails. The auto calibration deviation (using 9 peripheral points, 3 points at half the build radius and the centre) halved from about 0.1mm to 0.05mm. But I hadn't yet adjusted the Z probe trigger height corrections to account for the effector tilt being lower than before. Had I done that then the deviation might have been lower still.

I'll report the new auto calibration deviation when I have a working Z probe again.

I upgraded from 2020 to 2040 extrusion because my delta is 1m high and I think the extra rigidity will probably be worthwhile, especially for tall prints. When I first built the printer, the metal corners were only available for 2020.

I have a draft blog page about this printer build along with a detailed and costed BoM.

[Edited to correct units]

Edited 2 time(s). Last edit at 03/15/2017 06:00PM by dc42.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Quote
dc42
Initially I used the old arms, effector and carriages after upgrading the frame and changing to linear rails. The auto calibration deviation (using 9 peripheral points, 3 points at half the build radius and the centre) halved from about 0.1um to 0.05um. But I hadn't yet adjusted the Z probe trigger height corrections to account for the effector tilt being lower than before. Had I done that then the deviation might have been lower still.

I'll report the new auto calibration deviation when I have a working Z probe again.

I upgraded from 2020 to 2040 extrusion because my delta is 1m high and I think the extra rigidity will probably be worthwhile, especially for tall prints. When I first built the printer, the metal corners were only available for 2020.

I have a draft blog page about this printer build along with a detailed and costed BoM.

Are you sure you're getting 0.05 microns? That's... 0.00005 mm. If you can measure that, then you're doing really well.... Are your steps/mm higher than 20,000?

Surely mm. But back to the op's question. Good magnetic joints are great, average or poor ones are awful.

A good magnetic system has the cylindrical magnets on/in the arms, this means the balls are on the effector and carriages, this ensures the magnetic force is maintained at practically any angle. If the magnets are in the effector/carriages (which don't tilt) the magnetic force on the ball is far lower at low arm-to-effector angles. The result is disconnections especially at higher jerk/acceleration. An example is the k800 system, it works as long as your arms are long and your bed small.

Also the system is optimal when the balls are angled at around 30-40 degrees, this gives the widest range of movement. It will work with the balls at 0 degrees to the bed on the effector and 90 degrees to the bed on the carriages but with slightly more limited movement.

The lowest friction possible between the cups surrounding the magnets and the balls also helps. This can be achieved with grease, or use of delrin/acetal cups. In a word Haydn Huntley makes just such a system to a very high standard and for quite reasonable cost. If there is only one thing to say I'd use thicker carbon tubes than 6mm for 350mm+ rods.

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions

Great feedback from you all and i thank everyone again these discussions help me better prep for this work and helps me price the right quality stuff.

It sounds like hayden huntly is the place to go for these mag arms. I am curious about some effector designs as well in order to maintain the balls at 45 like you are saying. I would also need special carrages because the spacing would need to be the same as on the effector i think.

So basically the 2040 extrusion is just double wide 2020 and adds weight and stability? If i was to upgrade i am thinking of using metal/aluminum corners instead of printed parts to improve the build quality and squareness...then i would use rails and mag arms specced for the XL.

Please let me know of your progress DC very interesting and promising!

Yes the spacing at the carriages has to be the same as the spacing at the effector.

2040 is 2x wider than 2020

Metal corners are one way, printed ones and lots of cross bracing is the other.

Enjoy.

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions

Quote
DjDemonD
The lowest friction possible between the cups surrounding the magnets and the balls also helps. This can be achieved with grease, or use of delrin/acetal cups.

A bit of a noob question, but I'm trying to get a grip on exactly how this works... So, the balls rotate inside some kind of spherical cup? The magnet doesn't slide across the face of the ball? And the magnets are rigidly attached (glued?) to the rod? In which case, what's the benefit of the magnet over rigidly attaching the ball to the rod?

Hi, yes the magnet sits in a delrin (in this case) cylindrical end piece on the carbon rod. There is a half cup (its not retentive in its own right), which sits on the ball. I think the idea is its a zero backlash joint as the magnet holds to the ball firmly during all operations. If you made it a ball and socket joint there would be some play in it. The k800 system has the balls on the arms which sit inside cup magnets, but I am not sure how much wear they can take even greased.

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions

Yes of course I meant mm not um.

With the magnets in the rods, I don't believe it matters what the angle between the rods and the ball studs is, provided that there is enough movement before the rod ends foul on something.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Does anyone know the build volume increase for a 2020? its looking like just upgrading my mini 1515 is a better idea because an upgrade to 2020 might not be very helpful if i could just build a whole new printer and have 2 because it might only be a few hundred dollars more for a new larger build.

The build volume is whatever you design it to be. If you want to increase build volume, you need new extrusions, rods, bed, bed heater, and belts. The 2020 delta I built had a print area of about 280mm diameter X 400mm high, however most people would use 2040 extrusion for a printer as tall as this (1m).

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Okay! thanks for the info DC....basically a small build volume upgrade at the cost of a new printer minus new electronics. In my opinion a scale up effort sounds like i will build and entirely new 2040 all aluminum with mag arms and carriages when i think i have got my small one perfect and the need for the investment... I really like your PCB board and how you mounted them behind your carriages and your PCB effector. Very neat idea i have been considering having aluminum custom machined but PCB could be cheaper and more readily available and for what sounds like the same stiffness.

Are there any good recommendations for printable magnetic carriages and effector to match for a kossel mini 1515, E3DV6 hot end,mini IR Z probe.

Yeah Haydn's system has some effectors which might suit. [www.thingiverse.com]
Lykle/Zesty's piezo integrated effectors are designed for Haydn's system also, Though are a bit larger so might suit an XL better than a mini. [www.thingiverse.com]
Carriages are available from Haydn's youmagine [www.youmagine.com]

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions