jonnybischof

HOLY SH!.. *cough* Dude, that's some awesome work! I've noticed the problems with Nick's fanduct. It won't fit my printer because I use foldback-clips which collide with it. However, it's not possible to raise it much, because it would touch the heater block. So, I'm actually in need of a more suitable fanduct design, which is a problem for me and my Sketchup-skills. Is there anyone willing to help me on this? I'd also like to switch to a different mounting style for the fanduct, something simpler and more rigid than the current solution in my design.

Afaik the nozzle and heater block are one part on the UM2, right? That makes it a bit more complicated to drill out (since most 0.6mm drills would be only a few mm short). Ideally, this should be done on a mill with a precision x-y-table. With this, you could drill through the nozzle from the tip instead of going through the whole block. Should also work with a drill press, but it takes a bit more finesse to align the nozzle correctly. I have a mill, but I'm not even in Europe :(

Please keep us posted! This sounds very interesting!! They claim their plastic bearings to be better (longer-life) than sintered bronze bushings...

How difficult is it to set up an online wiki for this? Couldn't we use the UM wiki for this purpose? I've been thinking of writing a guide for some time as well. But a wiki would be great, too...

I've seen something similar with Colorfabb XT (PET-G filament). By printing a bit hotter and reducing fan speed, I could solve the problems in my case. Also, I've enclosed my UM1 so that it prints in a heated environment, about 40°C "cold" air temperature. My guess is that my cooling fan would lead to a cooler temperature than what the sensor measures. After all, the nozzle is made of brass which is a rather slow thermal conductor. So it can well be 10K cooler than the heater block if it is cooled too much by the fan. PLA is usually not too picky about extrusion temperature, but other filaments (like XT) seem not to like it. /edit: According to the taulman "features" page, Bridge shouldn't be as moisture sensitive as their other nylons:

Impressive list of exhibitors. Wish I could be there...

What country do you live in? Maybe there's a FabLab somewhere near you. You can get the panels cut there for a good price. Sometimes you can also have acrylic parts lasercut directly from the suppliers. I'd recommend getting an Ultimaker 1 instead of a 2. The UM2 is not made to be self-sourced. (Note that Open Source has nothing to do with self-sourcing parts). You should also consider getting a different hotend right away. This makes finding the right parts easier, and you can get something more reliable from the beginning. Have a look at the E3Dv6 hotend for example. You'll need to have someone print a mount for you, but that can easily be arranged through www.3dhubs.com. There's also a "please print this for me" section here in the forum. Do you have the possibility to order stuff through your company / workplace? There is an industrial supplier called Misumi where you can get most of the metal hardware parts (shafts, bearings, leadscrew and so on) in awesome quality. Not comparable to the stuff you get from reprap stores! But you can't make a customer account as a private person, only as a corporate customer... /edit: You'll find dozens of reprap shops from all around the world, so we'll need to know where you live before we can make any recommendations...

It depends on the physical distance that is made with an "open" switch. 24V actually arcs pretty awfully - I've arc-welded some test probes together before (24V, 10mA max.)

Lol, well that's one way to go I had a very busy week at work, and one more following :(. But I hope I find the time to complete the v6 hotend mount soon. At least it's usable by now, only some minor improvements I had in mind, and some additional parts for clean and tidy operation that still need to be finished. And then the documentation of course.

How do you mean, add cm3? If you mean the volume of the models, I don't know how to do that... Sketchup won't calculate that for me (it's just a drawing program...). You could derive it from Cura, use 100% infill, then take the length of filament value you get from Cura and multiply it by the cicular area of the filament.

Arcing happens when you put two physical electrical contacts together (for example when you switch a mechanical relay on). With AC, the arc will be extinguished when the AC sine wave passes zero. DC is an "always on" current that doesn't periodically change it's direction, so it never passes zero. This means that an Arc can remain much longer without being extinguished. Personally, I wouldn't run a DC through any physical switching element. Any solution which works around that would be preferable imho

That could work, but it's a lot easier to just solder the resistor into the PCB

180°C or 200°C is too high. 150°C sounds like a good point for "single use" fuses. I'd even lower it to 130°C for a solution that can be used continuously. Seems like there are multiple possibilities (no particular order): 1. Thermal "melting" fuse. Simple to use, but needs replacement when triggered (which should never occur, so no big deal). 2. Bi-metal thermostat switch. Should be simple to use, too, and doesn't get damaged when triggered. Probably the best solution.. 3. Integrated thermal switch (LM27 or similar). Needs circuit board design, but should give a nice result which does not interfere with the heated bed in any way (no serial connection). As an electronics guy, I prefer this one 4. Did I forget something? Would be nice to see the solutions compared against each other (price, ease of use, effectivity, reliability...).

Small update: Nick's fanduct also works when made of Colorfabb XT (post Feb-14). It doesn't melt...

I'd rather just throw all the relays away and simply use MosFETs As long as you're not talking 110-240VAC, then there shouldn't be any problem with using AC. Just please don't use 240V heaters, this is dangerous...

What would such a solution cost? The beauty about the LM27 is, it costs pretty much nothing Sure, you need a PCB and some small components, so in the end you're somewhere near 20 Euros. It depends on how many PCBs you order...

There's a tooltip when you mouse over these settings which describes what they mean. Gantry height is the vertical clearance between the build plate and your x-y gantry (on the Ultimaker, these are the 6mm shafts which hold the head). If you print two objects - one after another - then the first object must be shorter in height than the gantry height. Otherwise, the gantry would crash into the first part while printing the second part.

Könnte es an einer verbogenen Z-Spindel liegen? Ich habe schon eine gesehen, welche um einen guten Millimeter krumm war und sich an bestimmten Stellen kaum mehr von Hand drehen liess. Versuche einmal, in der Nähe der betroffenen Höhe die Spindel von Hand zu drehen. Wenn du einen Unterschied in der Belastung spürst, ist das ein Hinweis darauf. Vielleicht hat es auch eine Kerbe oder sonstwas im Gewinde, die dort zu einer Unregelmässig führt?

Magnalube-G wurde in meiner Lehrfirma (heute Besi Switzerland) für Spindeln und ähnliches verwendet. Keine Ahnung wo die das her hatten. Man kriegt's auf Amazon, keine Ahnung ob die auch nach Deutschland liefern...

It's not the last one! The last one uses an 1k pull-up resistor (no clue why...). I made the same mistake... The correct one for the E3D thermistor is the one which is 4 places lower than the one you selected in the picture above: 100K thermistor - ATC Semitec 104GT-2 (Used in ParCan) (4.7k pullup) No, it doesn't matter for both the thermistor and the resistor (they are non-polarized components). You should be good to go like this. Now you just need a mount, right? I'm in a bit of a rush right now, got a lot to do at work. But I'll get to it asap if you still want me to print a mount for you! Right now, I'm trying to print Nick's fanduct with Colorfabb XT and see how well it goes. But I think ABS is preferable for that part.

MosFETs can generally become unpredictable if they fail. Generally, there shouldn't be an electrical problem here, because even if the MosFET fails and is 100% on, you still have a fixed resistance (the heater) that will prevent short-circuits. Plus, pretty much all switching power supplies have over-current protection. But there can be a thermal problem if your heater is left 100% on all the time. The ideal thing would be something that does not rely on a mictrocontroller or computer to work, for example a bi-metal switch operating a thyristor (mains switch). But I don't know if there are bi-metal switches for these temperatures that you can mount to the heated bed (and that are affordable). Another thing I would put a certain trust into would be an OP-Amp switch reading a thermistor. Purely analog circuit that will, if done correctly, switch off the mains in every case of failure. It takes some time to design something like that, but I may have a look at it (I'm designing an UM1 replacement mainboard anyways...). /edit: TI has an interesting device: LM27 Guess I'll have to look into that one. It's basically a thermal switch that will activate an output when it measures a temperature above 140°C (there are mulltiple temperature options).

Thanks for the medal It is a lot of work to do, but if you're used to doing it, it takes less and less time with every design. Imho, even the best product is worthless if you don't know what to do with it. And my mount isn't that simple to work with. Maybe even a bit too complex, but the instructions should make up for it. I'll work on this some more, and when it's done, I have a good starting point for future designs.

XT requires a heated bed as well... Did you print the parts with XT and 20% infill? My observation was more "general", that XT becomes extremely strong when printed with 100% infill, whereas it's rather brittle when printed with the usual "2-pass, 20%" configuration. I've uploaded my "test shape" I use to determine part strength: https://www.youmagine.com/designs/material-stability-test I haven't been able to destroy a part made of XT, printed at 100% infill... It's ridiculously strong. I'd probably rather break my fingers instead of the part... Haven't actually done a comparison with 20%, I might print one out tonight.. /edit: my 100% infill piece is actually scaled to 5mm width instead of 6. I used this as a "pause" model for a small part and noticed I couldn't crack it, no matter what /edit: Print is done. Cracked the 20% infill piece with two flat pliers, no problem... What this means for the mount: You have a much higher risk of splitting the part where the screws grip into the material, because the layer-bonding just isn't strong enough with only 20% material. 100% material means 100% layer-bonding strength.

Update! Just added the PDF documentation. It's still work-in-progress, but it does at least help finding the needed hardware, print and assemble it. Wiring and electronics will follow asap. My free time is limited... By the way, source files will be uploaded as well, but they're a bit messy at the moment as I'm still adding some small parts (bowden coupler fixing ring, TC amp board fixation...)

Don't forget the 4.7k Ohms resistor you need to solder into the mainboard! This would be "R4" for Sensor "Temp3". R4 is one of the three empty spaces just above the temperature sensor connectors on the mainboard (UM electronics 1.5.7), it is marked on the board. Any 4.7k Ohms, 1/4 Watts, 1% tolerance resistor will do, such as this one. You can usually find these in most RepRap shops, too. /edit: George already mentioned that, sorry...