A different multi-extrusion approach - UM tool / printhead changer

[ultiarjan 1,223]

I still need to experiment more.... Curent setup is a 13cm rubber flap mounted in an aluminium strip at the back of the bed. To avoid too much pollution I use the whole length of the strip by setting a random X position on the wipe blade for every change. Only issue now is the rubber used is a bit to soft so sometimes some pla will be draged over it... i'll try to make the wiper sturdier by cutting some mm's of the top....

[macua85 52]

Tree frog success. Printed at 80% size, 0.2 layer height, 50mm/s.

[ultiarjan 1,223]

very nice ! did you use wipe towers or something else to keep nozzles clean? what (retract) settings did you use for the tool change?

[neotko 1,417]

Just a small update of what I'm trying to do with my mod. A small parking dock.

Also I ordered a flexible part on a material that reaches 80C and it's flexible, so my idea it's to use that on one of the layers to have a easy way to adjust the Z. Anyway the more I think of the @macua85 design, if it doesn't have any heat deformation or tilt problems, then his solution it's the way to go with the stock head. My solution needs aluminum laser cut, wood laser cut etc etc. (well the wood can be avoided on the last version that has fan but still I prefer to be safe). Also @macua85 approach for what I see has more print area and a smaller footprint.

[ultiarjan 1,223]

It probably all depends on the materials you want to print. My PLA head deformed after printing XT at 240 ... But printing the heads in XT could already be enough..

[macua85 52]

very nice !  did you use wipe towers or something else to keep nozzles clean? what (retract) settings did you use for the tool change?

I used the standard retract settings, not sure what that is in cura for the nozzle swap. Used and wipe and prime tower thing only, not the ooze shield.

[KevinMakes 22]

I have a random question. What are you doing as far as making sure that the head stays accurate?

Do you add limiter switches to the front, left, and right to re-zero the head when you pick those heads up?

Also are you cooling the unused head after a switch?

Do you think that you could build a wipe plate at the nozzles "resting location" to wipe the nozzle after pickup and before drop off?

Have you considered using that heat resistant rubber for a wiper on those...hmmm...i don't know what these are called.

http://www.amazon.com/Rubbermaid-Commercial-High-Heat-Spoon-Scraper/dp/B00Q8KIMC4

I made something similar for a previous employer for a tool switcher for a multi-axis cnc mill except it was a quick high pressure spray of coolant, dragging it through a wire brush with a slow backspin and dipping it in its rest in oil so I am very interested in how yours will turn out.

[KevinMakes 22]

Tree frog success. Printed at 80% size, 0.2 layer height, 50mm/s.

As much as I hates to ask this:(... but in the name of science...? can you dissect that...? or a closer up picture? or make a simple test square with a diagonal separation line to dissect? I am curious to see the layer bonds between the white and orange.

Are they the same type of material?

[macua85 52]

Haha, I've given the frog to my brother! The accuracy is taken care of my the magnets and the locating cones idea that foehnsturm came up with. The head offsets are set in cura until they are perfectly aligned (although I think I can do more on that, will print more calibration items off).

By using a wipe and prime tower, the oozing nozzle isn't a problem, as it just gets caught on that before moving to the print. I would still like to try to incorporate a wipe anyway though to neaten up the build area a bit. Here's a photo of a finished robot before cleaning up:

The White and orange are both pla. My next print I'm going to try abs and pla. I've done a njnjaflex and pla print which turned out ok, but pla doesn't adhere to ninjaflex that well.

[KevinMakes 22]

The White and orange are both pla. My next print I'm going to try abs and pla. I've done a njnjaflex and pla print which turned out ok, but pla doesn't adhere to ninjaflex that well.

 

Would you give your recommendations for ninja flex? (durability? Breathable? does the color rub off on cloth? ) Also what temperatures did YOU find best to use with it? Did you notice any quirks you would find mention-able?

[macua85 52]

I've not printed too much with ninjaflex. Some feet for my ultimaker and some test robots! I find though that I get good layer adhesion and no feed problems at 240 degrees, flow 110%, 30mm/s and no fan. It's amazing stuff, pretty much rubber, uber flexible and bonds so well impossible to tear layers apart. Not tried rubbing it on anything, but it's stopped my ultimaker making marks on the table its sat on.

I love the stuff, and have no problems printing it with a Bowden setup. I do have a dust filter with oil in it on the filament before the extruder, not sure if that's makes a difference or not, but reduces friction in Bowden a bit I'm sure.

[neotko 1,417]

Well I think the addon to level the head it's finished. Based on the @macua85 idea of one screw. Mine has 2 to control the tilt and I had to reinforce the inside to remove any lateral shaking and to avoid needing 4 screws. The beta I print yesterday worked almost nice, and I think this will be ok. By removing one of the wood pieces compensates the z loss and stays almost the same. This design allows only 3mm calibration, but it can be easily stretch to any almost any height if one head uses ultimaker nozzles and the other e3d.

[ultiarjan 1,223]

I finally replaced my tape mounted docks with something that clamps to the frame... and made some tiny size adjustments along the way...

[neotko 1,417]

Well the z align addon works well so far, it's quite stable to lateral moves and gives 3mm of adjustment. But I had to use x2 springs (I had a toon of them from the many beta designs). I could just go back to the original idea of using x3 little springs on the lower area and change the upper zone of the big screws but I'm too busy. With this will just work and allow a fine adjustment.

[ultiarjan 1,223]

The new docks work like a charm ...

 

[neotko 1,417]

That's just sweet men. Used the z stuff from s3d for that or excel?

I bough from amazon some silicone molt for cookies to cut it and make a wiper.

Your um2 looks sexy!

[ultiarjan 1,223]

Still use the excel template... only takes a simple copy paste.... and I would not know how to set a randomised X position, to use the whole wipe blade, in s3d...

[neotko 1,417]

Well the z adjustment addon works great. It's a bit bigger and I'll try to make it smaller but for now I need to focus on the wiper and fixing 'again' my umo+ board since freaking again the transistor died (even with the pwm bow outside). So right now I'm a bit tired of this board and I'll just have to take out two of the 4 fans and connect them on an outside pwm box to 'hopefully' avoid killing this weak board transistor again. I just can't get it but the only factor that's common to all the events it's that I did turn off the machine while having the fan at max.

Anyway this are the photos of the z addon. Works great and it's big for now to have a really really stable sollution (without just changing the full design).

Edited September 16, 2015 by Guest

[neotko 1,417]

Now much better from 14mm to 7.8 so by replacing one of the woods it won't affect the z and will be less bulky.

[neotko 1,417]

Well it works but because there's only two springs the prints losses precision and the y shift just a tinny bit enough to kill the quality.

So unless I redesign it completely I will just focus on z soft alignment with Simplify3D.

The wiper using silicon base for cookies works great to hold the heat and it's easy to attach to the metal clips of the base but it will need a printed base like ultiarjan did with his wipe system. My pal it's to place them on the laterals to make the wipe part of the load/unload movement. But that will be another day. I need to start making an external pcb to control this fans reliablely without burning any other transistor again.

[neotko 1,417]

I did this wiper a-la @ultiarjan design.

The think it's that it uses quite a lot of print area since the fans can hit it.

I was thinking on doing a simple mechanism that when pressed (by the head or the shafts) would self pivot 90degrees or more and make a 'wipe' on the nozzle. Anyone knows any design of something that pivots 90degress or more? It's to have a statt point. My idea it's to make it part of the dock or to attach it to the far bottom of the bed so it only pivots when each head it's on a corner saving print area.

[neotko 1,417]

This is my idea 'basic' with some legos. A mechanical return than when pushed brings the wipe. To save area. Ofc this it's just the idea I not even sure if I will be able to make it work since I never done something like this (except with lego technics).

[ultiarjan 1,223]

I would love it if you would actually make it in Lego

[jens 9]

I'm sure it's already been discussed extensively - so forgive me for asking, but I didn't follow too closely ...

Is the UM(2) electronics capable of handling another motor that you could use to lift the head parking area when in printing mode and only bring it back down for the tool change?

[neotko 1,417]

I'm sure it's already been discussed extensively - so forgive me for asking, but I didn't follow too closely ...

Is the UM(2) electronics capable of handling another motor that you could use to lift the head parking area when in printing mode and only bring it back down for the tool change?

 

No idea. But I would love that. A mini servo to clean like the one a guy did (posted a few pages before). I suppose that would require editing the marlin? No idea