jonnybischof

If you get the usual chinese or "reprap quality" stuff, then yes you will most likely have problems, like heavy ringing and wavy "straight" lines. Imagine your shafts heavily bent like bananas, then imagine what that would do to the "linear" movements. Now, in reality the bend will be much smaller, but the effect remains the same. If you have access to above-industry-standard quality like Misumi shafts, then it might work well. Also the platform weight will be a big issue. You might need a stronger motor - or one motor for every leadscrew (be that two or three) to move the platform reliably. As mentioned before - we're going waaay off-topic here. Please open a new topic if you have more questions about your project.

So PLA-HT is probably very similar to Orbi-Tech's PLA90. I hated that PLA because it was pretty much useless. It was so soft that printed parts would actually deform by their own weight after some time... @Stefan787 If you want to make your build area that much bigger, you should take into account that the longer you make your gantry shafts, the more difficulties you'll have to get precise prints. Shafts are never 100% straight, and the longer they are, the worse they "may" get (depends heavily on the quality of the shafts!). Not wanting to go too off-topic here, but you'll find more discussions about that somewhere in the forum... Also about the direct extruder vs bowden.

13:50 is ~ 1:3.85, not 1:4.6? Just to make sure I'm not messing up my calculations Ordered one of these 13 tooth gears. Misumi will have to wait a bit, though. I don't want to bother them with a 1-item-order.. /edit: Gotta love Misumi You can directly download STL files of their parts from them. Will make a quick mockup gear set now

Interesting! Misumi only goes down to 15 teeth minimum. Makes quite the difference for the large gear when you need 1:6 transmission... The large gear looks pretty heavy though. Why didn't they include spokes in a steel gear o.O Do these gears match with any other module 0.5 gears, e.g. the ones from Misumi? (GEABM0.5-50-3-B-5) /edit: Oh, wait. 10 teeth only has a 3mm bore. :( Well, 13 (smallest with 5mm bore) is still better than 15. Will try one

Nice drive wheel Maybe you could save some more weight by cutting out spokes? What drivers and current did you use? You can probably get it to run cooler with a TMC2100 stepper driver. At least if you use it in a dual toolhead scenario where there are large portions of "waiting time" - the TMC2100 can automatically lower the motor current when it is not in use. Maybe a thin aluminum sheet can act as a passive heatsink?

This is a very common question that most people don't know. But the idea is very simple: A 300W switching power supply unit can deliver up to 300 Watts of power. But it will only deliver as much power as the "load" draws. So, you can connect a 200W heatbed (generally speaking: a 200W electrical load) to a 300W power supply unit which will then deliver the 200W drawn by the load. Now, every power supply unit has an efficiency rating (e.g. 85%), so the power supply will draw a little more from the mains socket than it will effectively deliver at it's output. But the power rating counts for the output - so don't worry about that. So, why would you want to use a 300W PSU for a 200W load? 1. Some cheap power supplies will not actually be able to deliver the full rated load continuously. They may become unstable or die after a few months of use. 2. The closer you are to the maximum rating of any electrical device, the more "stress" you put on it. It will run hotter and generally wear out sooner. 3. You don't lose anything in terms of power. On the contrary: Most power supply units have their optimum efficiency at around 50 - 75 % of their rated load. The only downside is that a 300W PSU costs more than a 200W one. --> You should generally dimension your power supply units to be able to deliver around 30% more power than what your application actually draws. That is a rule of thumb, not a mandate. But it has it's reasons. About the relay vs transistor: the "switch" (relay or mosfet) is put in series to the load (heatbed). The switch has a certain electrical resistance, which means it will actually "steal" a small portion of power from the heatbed. Simply put: A mosfet has a much lower resistance than a mechanical relay. The one that I use in my hack is virtually invisible (practically 0 ohms). Now, if you compare the datasheet values, you will find that the contact resistance of a 10A is around 30 mOhms (0.03 Ohms). My mosfet is around 3 mOhms, which is only a factor of 10 better, not 100. But there are several other factors that make the mosfet better: Relay contacts are generally NOT intended to switch DC, but AC. You can see that beautifully in the picture neotko showed above: It says "10A 277 VAC". The thing about alternative current (AC) is, that it alternates between a positive and a negative voltage in a sine form. That means, the current will periodically hit 0 every time it alternates from positive to negative or back. Direct current (DC) doesn't do that. You have 12V at all times, period. Now, if you switch a mechanical relay contact while there is a voltage present, there will be a spark between the contacts. An alternating current will always extinguish that spark when it hits 0V (which is 100 times per second on a 50Hz signal). A DC however never hits that 0V mark, so the spark won't be extinguished. Of course the spark disappears by itself at some point, BUT as a matter of fact, running a mechanical relay on DC will lead to burnt out contacts eventually (much sooner than AC). And, a "10A AC" rating means NOT that you can run it at 10A DC reliably and for a long time. Now, a mosfet doesn't have that problem, because there is no mechanical contact "opening" but a semiconductor "switch" that changes its resistance. No spark - no problems! So, what does "contact burn" mean on a mechanical relay? Simply put: It means that over time, the contact resistance will increase (and therefore steal an increasing amount of power from your heatbed)! The bad thing about this: The more it increases, the faster it will continue to increase. This is why many manufacturers write "initial contact resistance" in their datasheets. Note that a mosfet doesn't have that problem, either. Now I've used a lot of time for writing... gotta get back to work o.O

You'll need a cartridge sensor that fits into the hotend. Not any PT100 sensor will do. While they all measure temperature the same way, you still need one in the correct shape and size. https://shop.ultimaker.com/product/62/PT100-B-sensor The original one is very expensive. You might find a compatible part (for around 25$ - estimate) and also the correct connectors on Mouser, but you'll need to put the wiring together yourself.

The UMO+ uses the v2 electronics mainboard, which is made much better than the 1.5.7. You can compare the power consumption between the two - but not how their power supplies are organised. So, if you get the sainsmart package you will have 1.5.7. Theoretically, the best solution is to use a single - powerful and good quality - power supply for everything. However, I wouldn't run the UMO 1.5.7 electronics from 24V - it gets hot enough already at 18V. The fan PWM transistor might also blow out sooner (it's a crappy design that even fails at 18V some times..). Also, if you're really going to run the UMO1.5.7 at 24V, you should switch out the UMO's linear regulator for a (pin compatible) switching regulator. Maybe that is actually what the youtube mod does, but I can't watch a video now while I'm at work... Yes, it will probably work. No, I don't recommend it because the 1.5.7 wasn't designed for this and might die sooner when powered from 24V. The simplest and probably best thing when you have an 1.5.7. board is to just go with the dual PSUs. Take the standard UMO 18V power supply plus a good one for the heated bed. There is no drawback to this solution - it is the ideal thing to do when you have an 1.5.7. board. My UMO has been running like this for over two years now - never had any problems since I've installed my relay hack (which was well before I posted the design). If your customer doesn't want this, then I'd recommend you to get a better mainboard in the first place. For example an UM2, or even a smoothieboard. My UMO "relay" hack uses a transistor (mosfet) instead of a relay because that transistor is 100 times better than any mechanical relay (or solid state relay). You can of course use a mechanical relay, but that solution is very crude and inefficient. Not recommended... The hack is documented in detail. You should understand the difference when you read the information.

Tried conducting that experiment. Sadly, I couldn't get it to work because I don't have a suitable PWM generator. PicoScopes are really useless for these kinds of things :( I'll need a battery powered PWM generator. Makes taking measurements and experiment set-up a lot easier... I'm ordering one from ELV, but it will take a few days for it to be delivered.

Yes, it means the fan should start turning when you power it with 8V DC. BUT, if you have an adjustable DC voltage (for example a laboratory power supply), then you can "kick-start" the fan by supplying 12V DC for a few seconds. Then, you can lower the voltage which will also lower the fan's rpm. At some point it will obviously stop. But that should be more like 5V DC, maybe even less. The fan needs a certain amount of power in order to overcome the rotor inertia. Once the rotor is moving, you can lower the power. Now, an adjustable DC voltage is not the simplest thing to achieve. Mathematically, a PWM'd voltage is the same as adjusting a DC voltage. 12V DC, with 50% PWM, is effectively (mathematically) 6V DC. But in reality the voltage switches between 12V and 0V rapidly. Many fans don't like that and may be damaged when you apply a PWM'd voltage to them. But there is a way to "actually" convert a PWM'd voltage into an adjusted DC voltage: By adding a low-pass RC-filter that flattens the PWM signal away, leaving a "more or less" DC voltage. There will be some ripple left in the voltage, but that won't matter to the fan. This is the simple part. Now you want to know the correct values for the resistor and capacitor. I have to admit I've never actually done this before, and there are quite a few things to consider even for something as simple as an RC filter. I can try putting together a little experiment at lunchtime. I'm not really the type to try and find a mathematical solution. I'd rather just put it together and see what happens. Then (maybe) derive a formula from that...

Have you tried running the non-PWM crossflow fan on a PWM control + RC-filter? With the right components you can make a pretty good adjustable DC source with only two components (an R and a C)

If you need a cheap, but very good relay board for the Ultimaker 1.5.7. board: https://www.youmagine.com/designs/ultimaker-heated-bed-mosfet-relay-hack-v2 Good power supplies don't come cheap, period... I usually get mine from mouser.com, where you'll also find the necessary components for the relay hack. Free delivery above 60 EUR! Mean Well PSUs have always worked fine for me. They are "medium cheap" chinese stuff, BUT industrial certified. And Mouser is not Ebay, so you don't get the really crappy / surplus stuff but only the "real" products. The link above lists some possible power supplies. /edit: You'll also find your C14 sockets at mouser, but at better (actually about the same) prices than Conrad And we're talking about the good quality "Schurter" sockets. (Note that you need to get the fuse holders for some of these separately - read the datasheets if you're interested!)

In the description it says That makes sense. For one, there is a huge tolerance because there are large differences on low quality fans. Around 100 mA is the usual case for fans with sleeve bearings. It's also nothing unusual that chinese cheapo-ware has useless labels...

Hate being fussy - but my electronics brain is making me do it! 0.08 A, not mA. There is quite a difference (factor 1000...)

0.8 A is 800 mA (milli-Ampère means a thousandth of an Ampère). If your fan really draws 800 mA, then it will quickly destroy the PWM output of your UM2 board. I don't believe you really have 12V, 0.8A fans. That would mean that this fan draws 10 Watts. A 10W, 12VDC fan is more like a wind turbine than a regular fan. For example, have a look at this one: http://www.mouser.com/ProductDetail/Delta-Electronics/FFB1212VH-F00/?qs=%2fW4LtXOBxKsvsU4J0zilXQ== This is a Delta Electronics high power 120mm fan with 126 CFM airflow, at a noise level of 52 dBA (that is A LOT). You might be able to use that one to blow finished prints right out of the printer Are you sure it doesn't say 0.08A ? (that would be 80 mA, which would make much more sense...)

Mein Farbensortiment ist noch vollständig Aber ich bin halt ein notorischer Schwarzmaler (oder sagt man Schwarzdrucker?)

True, true... I've been working on a new printer for.. 2 years? 3 years? I don't even know anymore...

@Dim3nsioneer, gehen weitere Bestellungen über deinen neuen Shop (http://shop.dim3nsions.ch)? Ich bräuchte bald Nachschub an schwarzem PLA....

@neotko guess what I just ordered yesterday noon Btw, that motor probably weighs 80g, not 50g. It says 50g on the page, but 80 on the datasheet... Still I think it's a good model, and comes at a very attractive price, too. You will need some gear ratio - direct drive with this motor will not deliver enough torque. I'll get a set of Misumi's polyacetal gears (#GEAB) to try with this setup.

Oh, and one other thing. Brass nozzles "could" be an issue, too (I heard that, need more details to confirm..). Probably not a problem for home use, but especially if you want to do anything commercial you should definitely use a stainless steel nozzle.

@foehnsturm, that design looks very nice Is it a big improvement over the NEMA8 stepper with worm gears? How much does that flat motor weigh?

Afaik, Taulman's Nylon filaments are certified food safe. Check their homepage to be sure if you're interested. The problem with PLA filaments not being food safe is not the PLA (which is a food safe plastic), but the additives such as the color pigments. I read somewhere that especially neon green and orange colors are pretty nasty.. /edit: I'd recommend Colorfabb's XT for the coffee mug. Should be able to withstand the temperature. It might be a problem if you put boiling hot water into the mug.. (interesting experiment though ) I'm not sure if the food safety certificate only applies for the clear XT, or also the colored ones. Make sure before you buy!

The Cura version numbers correspond to their release date: Version 15.04 is from April 2015 and is commonly called "legacy" Cura Version 15.10 is from October 2015 and is the current "all new" Cura Beta version Note that 15.10 is not available through their official site yet (closed Beta), but you can download and use it from the link xisle posted above. There was a dispute about the new Cura not being labelled as a beta version initially. You probably got that new Cura then. Now it's labelled as beta. Actually, version 15.04.2 is from July 15, but they have a problem with their version numbers because they first released the "all new" Cura as verison 15.06 in June. Now they need a way to keep legacy and new apart... https://ultimaker.com/en/cura-software/list This one explains it a bit.

I doubt that the source of the problem is the transistor. It may be dead, ok. But it's pretty much impossible to get multiple faulty parts in succession - there must be another culprit tat lead to this failure. So, did your external PWM board fail, or was it the Ultimainboard again? Btw, I'm doing military service atm so I'll be slow to respond...

Very nice print quality on that box! Just to note: The Ultimainboard does have a flyback diode on the PWM fan output, so there should be no problems. Yeah, the "npn transistor instead of mosfet" solution is kinda crappy, but it's what they chose to do... The diode you got there could take on 1000 PWM fans