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

[foehnsturm 969]

In the end I drilled through the whole heatsink with a 4.2 mm drill and then made a new heatbreak with 4.1 mm inner diameter. I then lined it with PTFE tube all the way down to the nozzle, the same way the E3D Lite6 is designed:

 

I had kind of the same thoughts when I ordered the E3DLite for the new project ...

All metal and PLA, especially with bowden, I'm still not convinced.

This is a subject you know way more than me: Why did no one try to apply a teflon coating to the metal parts like the famous non-stick pans?

[neotko 1,417]

There's a coating that's almost like that. It's tugsten disulfide. I tested it on titanium heat break and while it works it prone to scratch so it might be a problem with carbon-like filaments? Anyhow I never tested it on common bronce parts. I got a contact on uk that do coating for 20 heat breaks for 72€. Other company's I searched charged 250€ for the bath.

That reminds me that I need to retest the titanium heat break with the cooling I have now a-la um2

Edited August 14, 2015 by Guest

[neotko 1,417]

Also. On my test I found out that a mix of titanium and broce could work. But I did runnout of cash for this tests. There's a company that sells heads with ws2 coating but they go for 250€ a pop.

And just to share more info. The think that 100% worked was that the heat going to the peek area was cut by 30C (steady). So I suppose a head with ttitanium near the heat block with ws2 and the upper area with cooling or peek/brass should work. But that was the tests I wasn't able to test.

Edited August 14, 2015 by Guest

[Labern 774]

The Ultimaker team have always said that PLA doesn't work on all metal hot ends, I guess you now know why.

Edited August 15, 2015 by Guest

[ultiarjan 1,223]

Build my Dual head machine back to 2x UM2 head setup.

During the installation I was thinking of the high friction I sometimes have in adjusting the hight by turning the hot-end-isolator. I think the issue is high friction due to the screw holding the PT100/Heater into the Aluminium frame. It's only an issue when it needs to be rather far in.

I made some measurements, and there was a little difference in the Olsson block vs an UM2 nozzle. Of course I only measured one of each so it does not say much, I have no idea whats the normal tolerance on this .. ( @swordriff ? what do you think? )

Just to be clear, this is totally irrelevant for normal use and installation of an Olson block, It's just an issue when you need to change hight many times.

To make the adjustment easier I just drilled out the alu block a bit, now the hight adjustment is smooth... but I do have to be carefull as the heaterblock can also twist a bit more. Lucky thing is my head is mounted 180 degrees turned, so I have a good view on the heater block from the front..

Up and running again... and I used a hot-end-isolator (steel coupler) from http://3dsolex.com/other-parts

I did not check it by throwing it on the floor .... but I'm printing nicely with it, will see how it holds up in the long run, but the UM one's did not last me long so I'm hopeful these are better.

[ultiarjan 1,223]

Anybody (@neotko) tried to get separate ooze shields or prime towers in Simplify3D for the different extruders? Is it even possible?

[neotko 1,417]

You can. But for each process. Ummm now I get what you mean. I'll check it. Lately I'm focusing on assembling the other 2 umo and finishing the fan cap. Atm testing with x4 1.6W fans. So far it's all ok. I'll check s3d for that.

[ultiarjan 1,223]

When you look at what they do in the dual setup wizzard (I tried it with the Felix dual profile as UM is not supported in the wizzard) they turn on the ooze shield for both processes with the same settings, and "both extruders". I don't like the nozzles to get contaminated by the other material...

Anyhow now I'm back to 2x UM2 heads I'll give the default ooze shield a try tonight.

[neotko 1,417]

Well I have not touch the wizzard, since well, I don't like wizards. But for each process you can change what the ooze shield does.

Ok builded a compatible profile for the wizard, omg I hate wizards. But yes it's so nice that the program finally reads a object as a solid object even if each part it's done by each head. And yes, the options above do work perfectly if you set on each process of each head which head does what.

Sorry the img it's of what I'm printing atm, hopefully I'll finnaaaaly will have my own carbonfill eyeglasses (the part of how to order the glasses it's already undercontrol).

[ultiarjan 1,223]

don't like wizards myself, but thought it could be useful to see the intended use and settings of the ooze shield. If you set the ooze shield on both processes/heads to a different nozzle and different offset you'll only get one iso two .....

and totally off-topic... curious to see how you mount the glass (made some myself more than a year ago ..) and you can adjust the shape of the glasses legs easy under warm water when you print them in pla...

[neotko 1,417]

Oh I see what you mean. Maybe just use the prim towers? Weird they didn't let use the shield that way...

Oh the glasses the shop has a machine that measures the hole and they made the glasses so they can snap-on. Works quite well but it needs a solid extructure to hold the snap-on/off preassure. On pla at 60% infill it was almost ok. I want to avoid heat because they deform just a bit and I'm quite a bit peculiar for my eyeglasses 8-)

[neotko 1,417]

Ok now I get how this works. The options for each process are under just one process once the wizard shorts the stls. So, if you change supports, ooze shield, etc. For each subjob, it changes for both. So you need to choose what extruder will do each of the 'additions'. Imo this sucks because the prime towers, ooze shield, work like just 1 job. So the s3d wizard sucks big time.

So ...

Tryed playing a bit. It sucks and it's really really stupid. Even if you make two process, the ooze shield it's somehow 'shared'. Well... It works fine this part. The really really stupid part it's the prime tower. They are somehow connected even being two separated process, so... You can only prime at the same place, no mater what you choose. Try to play with it, it sucks.

But, the ooze shield, can be done by, changing it to all extruders on both process... It's really really the wrong way to think... If they are separated, they really should be separated... /sulk...

But, the ooze shield, can be somehow solved by not using the wizard.

Prime tower it's just... pfff... You can do it on the same place or just one extruder, so unless there's some automatic layer height adjustment (like they do for rafts) then it will fall down like with old cura prime towers. Well at least we can do it bigger so it might old in place...

Anyway, s3d developers should think about what means 'separated' process...

[neotko 1,417]

Umm I think your problem can be fixed by just using the ooze shield on one extruder without the wizard. Wizard seems to lock that to both?

Anyhow prime towers lock for any process it's kinda a downer.

Edit. Yeah it works on the wizard, just choose the first or second extruder for the ooze shield. Also it does it 'closer' and seems more accurate than the img up that was using the separated process. Too bad they are trully mix, less options to play with.

Edited August 18, 2015 by Guest

[ultiarjan 1,223]

The combined ooze shield seems not too bad on the color mixing, think I'll give it a few more tries. It does detach from the bed too easy, so I'll advice to combine it with a cm of skirt. Although the software tells you the skirt is offset from the part... it's actually offset from the ooze shield, so set offset of the skirt to 0.

Without the skirt it will detach;

Then there's another bug... part of the first layer (blue open square) does not print of my test cube, played around a little but have not yet found a reason (it's fine in cura..)

[neotko 1,417]

Cura eats stl almost without problems. S3D can't fix a think. I use to netlabb cloud them almost everytime if its not solid object. It's imo the worse think of s3d.

[macua85 52]

Hey guys, quick questions. I'm trying out the new printer motors to check all is ok. I can't get the extruder motor to activate...however there is no heater installed yet, just the thermistor. Even if i set print temperature to 25 degrees (to test extrusion), it doesn't do anything. Is there a failsafe here which stops it extruding when cold?

Also, another annoying problem which i doubt anyone can help with...i've wired up my heated bed kit (not oem, but a bang bang one eBay used to sell), and just blown two power blocks (19v, 230w). No idea whats causing it, wiring is all ok. weird.

[Labern 774]

@macua85 Yes there is.

If you put M302 in gcode near the start to allow for cold extrusion

Edited August 20, 2015 by Guest

[tinkergnome 927]

Is there a failsafe here which stops it extruding when cold?

 

It's called "PREVENT_DANGEROUS_EXTRUDE"

(follow the link for an easy workaround)

[macua85 52]

Great thanks for the help guys. Also found another problem. The chinese 1.5.7 board i have got is an updated version, so its all sms components...but crucially, comes with resistors already soldered in to r21, r24, r4. Thats good for my heated be kit, but now I've decided to use k type thermocoupes, it left me scratching my head when error max temp and shutdown kept happening. All sorted now. Now I just need to figure out this blowing power supply for the hbk.

[macua85 52]

Ok, so it turns out the Chinese power supplies I got ("genuine HP"), are utter balls. I used a 24v adjustable 200w supply, and its working fine.

One more questions, the z stage when homing is really slow. I have used the same stepper as the UM2, and adjusted the steps/mm to 200. Is there anything else I can do to make it home as quick as my other UMO which has the genuine HBK?

[gr5 2,172]

Yes. It needs to be 170C. You can either the nozzle to 175C or easier send an M302 gcode "allow cold extrudes".

Do you use pronterface/printrun? If not you should use that to test your motors - it's very very simple to use and powerful for this kind of testing (e.g. it's trivial to send an M302). Download it here:

http://koti.kapsi.fi/~kliment/printrun/

[gr5 2,172]

Ok, so it turns out the Chinese power supplies I got ("genuine HP"), are utter balls. I used a 24v adjustable 200w supply, and its working fine.

One more questions, the z stage when homing is really slow. I have used the same stepper as the UM2, and adjusted the steps/mm to 200. Is there anything else I can do to make it home as quick as my other UMO which has the genuine HBK?

 

What firmware are you using? There is for example the zmax parameter (max z speed) but I think there is a homing speed somewhere in the Configuration.h file. You could compare that speed between UM2 and UMO.

[jonnybischof 60]

Ok, so it turns out the Chinese power supplies I got ("genuine HP"), are utter balls. I used a 24v adjustable 200w supply, and its working fine.

...

 

If it says "genuine" on a chinese product, then run as far away from it as you can!

This is the number one indicator that you're getting bullsh.. stuff / are in a fishy shop.

Of course there are great chinese products (I proudly own a Huawei P8 smartphone), but in this case I'd say you bought a hunk of junk

Edited August 20, 2015 by Guest

[macua85 52]

Damn, just as I was about to test everything, I shorted out the fan wires when checking the voltage (controller fan wasn't coming on, probably another piece of chinese junk). So the printer immediately clicked and cut out. The question is, is there a thermal cutout on these, or have I just fried the board?

[macua85 52]

Well, I've buggered it I think. The PSU cut out on thermal overload. So thats reset now, but when I try and power up the ultimaker, it comes on momentarily and then trips again. Shame, was so close!