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

[neotko 1,417]

Yesterday tryied to connect my pwm little board from china to the 30x30x10 fans and it does work at some degree. So now that I'm using them to cool the head I might just print a box for this small pcb and left it connected to the 19-24v auxiliary.

The interesting think it's that really really almost nothing drip from the heads now that they are cooled. Also it can keep 260C (so no problem there) and on the good side it can cool the head muuch faster from 240 to 200. So also good for materials that can burn if left there without moving.

Edited July 31, 2015 by Guest

[neotko 1,417]

Today experiment with the fan cap.

Bough some Nomex heat insulator paper and kapton. The plan was to cover the large area of the heat block that (imo) releases a lot of heat over the print area (like a oven) and make's the right fan unable to properly cool down the pla.

So far no bad at all. The overhangs have improve, and now the right side it's a little tiny bit better than the left. I think this it's because the fan design must be 1-2mm lower and the open area makes the heat of the block to follow the air. Anyhow, I'm pretty happy with my robot, the best I have been able to print so far. 0.1 layer 40mm/s 195C Faberdashery pla Perimeters simulating 0.35 nozzle (s3d extrusion width).

I clearly have backlash from the belts, specially one that woobles, the quality of the stock pulleys it's really really bad. I been having some problems with one full axis 'woobling' until I changed the pulleys of order. I have a full set of pulleys from aliexpress and the quality it's just almost perfect but I'll wait for my gt2 pulley/belt set to arrive to make the change.

Just as a side note, check my woobling X (the motor it's tight, I did dismantle everything and mounted it back, it's still there).

 

 

 

Edited July 31, 2015 by Guest

[gr5 2,167]

That's the X motor but you do have a point about the pulley.

[macua85 52]

Looking good. I would say that the biggest problem there seems to be cooling, s the overhangs are not coming out as good as they could. Maybe some fan ducts which point more towards the printing nozzle (or just below it)?

As for the wobble, go for GT2 and direct drive for sure. Its so cheap from robotdigg, no belt tensioners required, and the mount for direct drive is easy to print. You don't even need to get longer axis if you don't want to, as someone has designed an adapter to house the bearing and mount to the bearing cap holes.

Edited July 31, 2015 by Guest

[neotko 1,417]

Indeed, edited the post just after posting :DI bough the machine assembled but this past weekend when I did check the pulleys they are easily 2-3 way offcenter, way way off. If I change one of order the full axis woobles.

Check this other video (this I was able to fix it by changing pulleys order)

 

 

 

Edited July 31, 2015 by Guest

[neotko 1,417]

Indeed @macua85 I got a pretty good set from aliexpress (the the same shop I got very nice mxl pulleys) and they are on the way.

[seipia 1]

Hi Ultiarjan,

Sorry to come back to page 27 of this forum but each of us advance at its own speed!

I've got some problem with heater.

On the UM2 card, I can see 3 connectors:

Temp3 with heater of head 1

Temp2 : free

Temps1 with PT100 of head1

How do you control the second head?

After modification and compilation, I read the "ERROR-STOPPED, Temp sensor" on the UM2 screen.

It's probably because there is no sensor on TEMP2. But ther is no plug for the second heater.

I understand you connect the second heater by an external power supply. How do you control it? and make connection with UM2.

Thank's

PH

[neotko 1,417]

If I understand you. The temp 2 it's free. Then there you connect the second head pt100 connector. Unless I didn't understand it and the physical connector isn't there.

The temp error pops when there's no second temp sensor connected (since you are usig extruders=2 in marlin.

[tinkergnome 927]

On the UM2 card, I can see 3 connectors:

Temp3 with heater of head 1

Temp2 : free

Temps1 with PT100 of head1

How do you control the second head?

 

This is only almost correct...

Temp3 should be connected to the temperature sensor of the heated bed.

Temp2 is intended for the PT100 sensor of a second nozzle.

The nozzle heater is connected to the screw connector marked with "heater1" - "heater2" is apparently intended for the second heater (consider the provided voltage (24V AFAIK).

If you're in doubt - take a look at the board description or at the assembly manual

Edited August 1, 2015 by Guest

[amedee 349]

Great thanks for that amedee. Its not such a pressing issue for me now though.

 

Just updated my builder, you can now select the 'MAXTEMP' for the hot-end...

[seipia 1]

Thanks Tinker for your help.

OK! after a look at the board, it's clear : screw connector J21 with Heater Bed, Heater1 and Heater2.

I use TinkerGnome-MarlinUM2 for my machine and it's work perfectly.

Do you know what modification I can do in order to use it with two heads ? and have you the project to do it (it would be great!)

first, I have compiled sources of MarlinUltimaker2-master and it's OK.

After I try some modifications like :

#ifndef EXTRUDERS

#define EXTRUDERS 2

#endif

in different files, compiles again and get ERROR TEMP due to there is no sensor on the connector TEMP2

I try to modify :

#define TEMP_SENSOR_0 20

#define TEMP_SENSOR_1 20

#define TEMP_SENSOR_2 0

in order to disabled sensor_2 but apparently, it's not the right way...I'm not sure sensor_2 is the sensor of the second head !

 

I'm not a specialist in firmware, as you can see! but my project is : use à second extruder for ceramic paste. I have mechanical solution and provide the steep motor but I am stopped by the soft...my second motor is well plug on the board but don't move...

Thanks again,

Bill

[neotko 1,417]

If you active 2 extruders and the sensor isn't connected the error will appear no matter what sensor you choose.

The temp-sensor it's just to define the kind of sensor you connect. In UM2 the default (20) it's a pt100

[tinkergnome 927]

in order to disabled sensor_2 but apparently, it's not the right way...I'm not sure sensor_2 is the sensor of the second head !

 

As you mentioned - it isn't... TEMP_SENSOR_2 is not relevant (not used at all for the UM2 AFAIK). The counting begins at 0 (TEMP_SENSOR_0 for the first etc.)

Only two different defines are needed to compile the stock firmware either for one or for two extruders. And this should work similarly for the tinker version:

Single Extruder:

 

TEMP_SENSOR_1=0EXTRUDERS=1

 

Dual Extruder:

 

TEMP_SENSOR_1=20EXTRUDERS=2

 

I've just added the (compiled) dual version to the recent tinker release.

I'm afraid that you will have to make some other changes - but this depends on the type of your hardware modifications.

Just be aware that you probably have to struggle with the compilation anyway (sooner or later)...

[Skimmy 0]

Hello!

I am a long time lurker on this thread and I am excited of the whole idea to change the hotends with magnetic holders! <3 I appreciate your work so much, that I want to share my knowledge on magnets with you and I only registered here in this forum because of this thread.

So, the knowledge I want to share:

First:

You are absolutely right to use types of magnets, which withstand higher ambient temperature! I did measurements at another project of mine at the hotend-assembly after long prints and my hotend-carriage gets heated to around 65°C only from passive heating through the heated bed. Warmth is a really bad distortion factor for magnets.

Second:

You _really_ should use round magnets. There are only benefits when using round ones. Rectangular shape is only "for the eye".

E.G.: They are magnetically stronger. They are tougher. They are up to 10% more long-lasting when being "permanently" hit (I am referring to the shock from snapping the carriage) and much more.

Third:

You should further use magnets with optimal measurements for maximum magnetic pull. As a rule of thumb for round magnets its: "diameter is twice the thickness" - e.g. a 10mm diameter magnet should be 5mm thick.

Feel free to read some VERY good explained information at http://www.ibsmagnet.de/knowledge/index.php

and also feel free to get numbers with an interactive web-calculator for magnetic flux density at http://www.ibsmagnet.de/knowledge/flussdichte_berechnung.php?f=rd

With this calculator you can easily compare your rectangular magnets to round ones.

I know about this, because I needed to to scale several really big and expensive magnets and therefor learned about it.

[neotko 1,417]

I thought also that round magnets could work. But on my first versions they tend to work worse imo. The reason it's that the way the movement to clamp the head and the need for it to be smooth makes 3 kilo force just the right spot. Tested with two round magnets with m3 holes and was okey but the force was too big for the clamp to be released smoothly since the change of angle when parking the head creates a distance that allows to park it. With round magnet I got troubles to release it properly. A flat magnet (specially on the current orientation) makes the magnet to produce a sliding effect that helps quite a lot when doing the park/clamp movement. And the upper/lower forces are compensated by the tiny female/male cones.

[macua85 52]

Thanks for making that update to the builder amedee

[neotko 1,417]

Finally after all this time yesterday got a good idea to fix 2 main problems of my setup:

1) Z alignment

2) Ptfe isolator to stay always on its place

Finally what I'm doing its to keep the design but adding a spring spacer that works independently. With this:

Also I'm changing the head holder to make the fan caps something that could be unplugged fast to replace, etc.

[neotko 1,417]

Well the design it's great now. Changed the fans from 24v to 5v and connected them in parallel to the 5v fan output of the umo+ board. The heat on the upper level never goes higher than 55-59C while printing at 260C for 2h 45m. So my new head assembly printed on carbonfil will fit there perfect (carbonfil crystallization point it's at 85C).

Heat printing at 260C 2:45h 0.2 layer 60mm/s carbonfil formfutura. Yellow means higher than 60C

Btw guys I'm testing an @anders-olsson new nozzle (can't reveal much details) and even when it's not the final version I must say it it's pretty impressive! Perfect overhangs on my print so far. And that with my old fans design (carbon doesn't need fans but it helps when doing hard overhangs). Amazing

[seipia 1]

Hi All,

Thanks for all, Go ahead.

But I need some help again ;-)

OK for temp sensor, it's work and with Tinker Firmware (it's great!) it's easy to control that.

As I have not yet the second extruder installed, I 'v simulate the second PT100 sensor by plugging a 150 ohms resistance on TEMP2 and it's OK on the screen.

Of course to allow proper operation, I modified the set point of Extruder2 at 130°C.

My second step motor is plug on connector E2, and I had test it before on E1 ( print something and it moves)

I created à GCode based on a CAO of two very simple pieces (one above the other) with CURA.

When I read GCode, I can see the codes T0 and T1 at the right place.

The first piece is well print....Youppi !!! BUT

The head move at the offset position for the second print.

Head move to print but step-mottor2 remains stationary.

I precise that it is powered, it is not possible to move it manually.

I also added the G10P1....instruction just to see, without any result.

What I have forgotten again ?

[neotko 1,417]

Maybe minimum temperature. Marlin has a minimum to be able to extrude. I think it's around 170C, if the head it's below that number the motor wont move. It's for safety reasons.

[neotko 1,417]

My final head holder and fan. Now the fancap can be easily replaced/redesigned without having to dismantle the head. I did some tests on the head z adjustment and it works nice, but it need's the latest design of the upper part to make a correct 4 sides grip, the 3 screws works but it's a bit more unstable to do a proper tight adjustment. With the 5v fan the area near the upper peek/ptfe never goes beyond 55C (at 260C) at 210-220 it stays around 42-46C. Also I can confirm that there's a big difference when cooling down the head, it goes down fast (and also can reach 260C without a problem) so it's nice to cool down the head when it's parked if the filament it's heavy (like carbon).

I been doing heat photos and it's all nice now. Also ofc the sides of the fans never go beyond 60C, but since now I have them in carbonfil they should not deform in anyway. Ofc, I think for a final version I would just print the central block in stainless steel, I checked the price on imaterialise and the central head block it's around 50-60 bucks, so I will consider this for the future but I suppose it would be a bit heavy on it's current design.

Edited August 5, 2015 by Guest

[seipia 1]

Maybe minimum temperature. Marlin has a minimum to be able to extrude. I think it's around 170C, if the head it's below that number the motor wont move. It's for safety reasons.

Thanks, I try tomorrow, but if it's true, I have an other problem because I want use my UM2 with other type au material (ceramics) which need less than 100°C.

So I have to modify the firmware in order to accept low temp and the limit is without any heating.

May be you or TinkerGnome know exactly where I can modify firmware

It will be a great help.

PHH

[neotko 1,417]

 

//this prevents dangerous Extruder moves, i.e. if the temperature is under the limit//can be software-disabled for whatever purposes by#define PREVENT_DANGEROUS_EXTRUDE//if PREVENT_DANGEROUS_EXTRUDE is on, you can still disable (uncomment) very long bits of extrusion separately.#define PREVENT_LENGTHY_EXTRUDE#define EXTRUDE_MINTEMP 170#define EXTRUDE_MAXLENGTH (X_MAX_LENGTH+Y_MAX_LENGTH) //prevent extrusion of very large distances.

 

On Configuration.h

[tinkergnome 927]

or add an M302 gcode command somewhere to allow cold extrusions

[neotko 1,417]

or add an M302 gcode command somewhere to allow cold extrusions

 

Now that's cool