anon4321

The videos were helpful First observation that you already know is that the feeder is running WAY too fast and WAY too far during both the Pronterface UI moves and during the print. A 10mm feed done through the Pronterface UI should take about about 6 seconds and rotate about 2/5ths of a turn maybe a little more but less than half. Yours goes like 6.75 revolutions!!!!! A .1mm move is like barely 3 degrees and from a distance it's hard to tell it is moving. In the pronterface UI, there is a text box on the bottom right below the output from the printer. In that box, enter this as shown (case-significant and no spaces) and press return: M503 You should get output back from the printer in the box above like: > M503 < echo:Steps per unit: < echo: M92 X78.74 Y78.74 Z533.33 E836.00 < echo:Maximum feedrates (mm/s): < echo: M203 X500.00 Y500.00 Z30.00 E25.00 < echo:Maximum Acceleration (mm/s2): < echo: M201 X9000 Y9000 Z800 E10000 < echo:Acceleration: S=acceleration, T=retract acceleration < echo: M204 S2800.00 T3000.00 < echo:Advanced variables: S=Min feedrate (mm/s), T=Min travel feedrate (mm/s), B=minimum segment time (ms), X=maximum XY jerk (mm/s), Z=maximum Z jerk (mm/s), E=maximum E jerk (mm/s) < echo: M205 S0.00 T0.00 B20000 X20.00 Z0.40 E5.00 < echo:Home offset (mm): < echo: M206 X0.00 Y0.00 Z0.00 < echo:PID settings: < echo: M301 P27.00 I1.05 D91.41 Copy and paste the output back to this thread. The rest might be important but the critical line is this: < echo:Steps per unit: < echo: M92 X78.74 Y78.74 Z533.33 E836.00 As for the noise, that isn't normal either. Do you have the black plastic clip on the other side not shown in the video? Shown on the left here: http://wiki.ultimaker.com/images/Drive_bolt_v3_assy.jpg If you don't that's your problem causing the noise. If you do, can you move the gear from side to side by pushing in and pulling out on the large nut shown on the right off the above image? If so by how much? If it is a lot, load some filament to bind up the driver bolt a little. Now while holding the gear, removing the big nut. With the nut off, rotate the big gear clockwise while holding the filament. This will screw the big gear on a little. Check the axial play and repeat until only a small amount of play exists. Now while holding the gear in position, screw the big nut back on without letting the gear move. Tighten and check the play. Remove the nut and repeat if needed. See this for more information on assembling the drive bolt: http://wiki.ultimaker.com/Ultimaker_rev.4_assembly:_Material_feed_mechanism#Step_2:_Drive_bolt_assembly

I have found that XT is MORE flexible in thinner prints. I printed a fan duct (camel toe!) in XT using a thickness that was good for PLA. However it was too flexible in XT so I had to thicken it. The thicker part doesn't seem brittle at all and remains quite pliable.

Just be aware that the GT2 pulleys are slightly different in diameter and pitch. This prevents you from getting the exact right length of belts for the long belt. Thus you may run into tensioning issues.

I'm glad you stayed away from the pot. Hmmm, drugs are bad... mmmkay.... Best not to futz with it... In addition to the things GR5 mentioned. Two other things you can do in the cura UI that will cause this behavior. First would be setting the filament size much too small. It should be 2.85. Actually, it should be set to whatever you measure the filament diameter as using calipers. The second and I've done this and I know the freakin extruder goes crazy is to choose the volumetric flavor of gcode. Quickly, that mode generates gcode where the e feeds are in 1mm3. In other words if the gcode needs 1mm3 of material during a move, the feed will be +1. However, if the printer isn't in volumetric mode, a feed of 1 means that the printer should feed a linear amount of 1mm. However, this will produce 6.4mm3 of material (pi * ((2.85/2) squared)). Basically, the printer is feeding at a rate that is 6.4 times faster than it should. Marlin supports volumetric mode but you need to send M200 D<millimeters> command to get it into that mode. This is all somewhat experimental so it's best just to change the mode to the correct setting. I don't think the value is saved in settings. By the way, if someone is wondering why this was supported, it is because the gcode becomes portable to other printers using different filament sizes. The gcode specified feeds in common terms of mm3 and the printer decides how much of a linear feed is required based in the size of the filament configured in the firmware to extrude the requested volume.

There is a machine settings menu item that will allow you to define the machine's maximum. However, brim and skirt settings reduce this a little so you need to play with both to get the full max volume.

Limiting is required because most steppers are specified at 2-3V and we are cramming 18V into them. Without the limiting, the motors would burn up Nope, trust me, one of the fundamental functions of the the driver is to limit current. The skipping occurs when the force required to move to the next step is more that the force generated by the coils at the set current. There is no reversing due to a missed step The controller has no way of knowing the step was missed. I checked the electronics and there is NO way for the controller to measure the current. These are used in the UM1 and use the same chip as the UM2. See the section on current limiting. http://www.pololu.com/product/1182 There are sense resistors on both the pololu carrier board and the UM2 (which have the AD4988s built in). The sense resistors are NOT monitored by the UM1 (they aren't even available off the carrier) or the UM2 where they are only connected to ground. Go here: https://github.com/Ultimaker/Ultimaker2/blob/master/1091_Main_board_v2.1.1_(x1)/Main%20Board%20V2.1.1.pdf?raw=true go to page ten. It shows the 5 drivers for X, Y, Z E1 and extra (E2) Notice that the only pins that go off the page are STEP, DIR, ENABLE and VREF. The sense lines on the right side only go to ground. The STEP, DIR ENABLE are self explanatory. VREF is the voltage FROM the processor and is what sets the current.

I believe the following... Oh and yes, the driver chops the current to the set value. Because of holding torque, the motor always consumes the same amount of current whether stepping or not. Basically, there is either enough current to overcome the force to step or the step is missed. It won't consume more current under load.

Just looked, the controller can only set the current but not sense it.

I don't have a UM2 and haven't looked at the PCB in detail but I suspect that the controller only sets the current but doesn't monitor. However, I could be completely wrong.

Thanks, I meant to check on that and got distracted. That meanwell supply wasn't cheap (90USD) but worth the money!

Daid, are the accel settings different because of that Z motor change? Also, if you remember the thread about the quality difference with 14.06 version vs previous versions, could they have flashed the heated bed version and be running at a higher speed? That was the discussion where you did a double blind test and posted a dimly lit photo of the result.

Most people can't use the firmware from Cura because it is set for a different temp sensor type than the currently available 3rd party beds. What bed are your using?

PeggyB, I don't think the bubbles are normal. Maybe you got a bad batch. Contact colorfabb and ask. http://colorfabb.com/contact-us/

Well, in the US, we are more self involved so we have a multitude of hair care products. For reference it is all weather Aqua Net professional hairspray extra super hold, unscented. UPC 6799060011 http://www.pharmapacks.com/product_images/g/093/111755__73463_zoom.jpg I'm surprised. I was getting XT to almost stick good enough to a clean glass plate. I dusted it with hairspray and had trouble removing it !!!!

Quick note, it is often overlooked but Cura has a Pronterface like UI build it. Go into File -> Preferences and change the printer window type to Pronterface. The new UI will appear when you print a model. You can enter the UI without a model loaded but selecting File -> Print even when the screen icon is dimmed.

I think the added resistance in the wiring is saving me at start up. I'm using a consumer type PS with overcurrent protection. It is 280 watts so on start up, without the added wiring resistance, I think the PS would shut down. This is what I'm using : http://www.meanwell.com/search/GS280/default.htm

Could you make a video with sound and post showing the feeder moving through the printer UI and during a print?? It seems like it is working and I hate to tell you to do something that has risk of breaking something or making it worse. The wooden parts on the feeder are known to resonant making it fairly noisy. When moving with the UI, the steps occur regularly and at a high, consistent rate. During a print, the rate is much slower and might just be hitting the resonant frequency of the wooden parts. PROCEED AT YOUR OWN RISK !!! Here are some suggestions but I really suggest that you post the video with sound so we can tell if the noise is normal. One low risk thing to try is changing the meshing of the small and big gear by loosing the feeder stepper motor and pushing it so the small gear is forced more into the bigger gear or less. Just note that too much pressure could cause the feeder to stall and the motor to run hot. Too little will cause "slop" and the material flow might not be accurate especially after a retract is recovered. A second option would be a different gear set. This is a double helical set which has the benefit of a constant mesh. I've used it for a few months now and it works well. Printed in UM Blue PLA. It really quieted the retracts (when the feeder reverses to relieve the pressure in the nozzle as the head moves to another position). The knob broke off when the printer was knocked over and if you read the comments it is common for it to break off so you might want to not print it. Print with at least 80% infill. I used 100% http://www.thingiverse.com/thing:40334 The one note is that it was VERY difficult to get the small gear pressed onto the motor shaft. And due to the double helix, the center line of the gears must align. Other option would be to try and decouple the feed from the back of the printer as the back serves to amplify the sound. See things like (read the comments though): http://www.thingiverse.com/thing:53690 and: http://www.thingiverse.com/thing:131534 This is really the point where you PROCEED AT YOUR OWN RISK !!! The more risky way is to reduce the current the driver uses for the motor. However, I and others have damaged the drivers messing with the current. See http://wiki.ultimaker.com/Electronics_build_guide Use very small changes like 1/20th of a turn. Reduce it too much and the motor miss steps and not feed when it is supposed to. In theory, reducing the current should not damage the driver but other have said that too low can cause damage to it. By reducing the current, the motor "steps" with less force and less force will produce less noise. Good luck.

Hey Zoev, I thought about the temp coefficient which is why I measured twice, once at room temp and once at 60C. I didn't see much of a change.1.9 and just a little over 2 at 60C. At 100C you might see power drop. Using this http://www.endmemo.com/physics/resistt.php Resistivity, Temperature Coeffient: 0.0039 Original Resistivity: 1.9 Original Temperature: 24 Final Temperature: 60 Final Resistivity: 2.15992 At 100C: Final Resistivity: 2.44872

Yeah, Jason should really update the PS requirements. I guess I got lucky as mine is a little higher in resistance and I went with a 280w PS. Only with the drop in the wires, I'm just barely within the PS spec. An 18V - 20V is probably what should be used.

This one? http://www.ebay.com/itm/Heated-Bed-Kit-for-Ultimaker-Reprap-3d-Printers-MK2-PCB-Aluminium-plate-/161157442118?pt=LH_DefaultDomain_0&hash=item2585bb6a46 I believe they are shipping the third version. I have the 2nd version with the thinner 2mm alum. It measures 1.9 (room temp) - 2.0+ (60C) ohms. There is enough drop in the wiring to measure 23v at the connector.

Hairspray on glass works well with XT too.

Well if you want to just turn something on and off like a single speed screw driven extruder, it think there are many unused pins that are brought to the top shield and these can be controlled with another gcode command. [*]M42 - Change pin status via gcode Use M42 Px Sy to set pin x to value y, when omitting Px the onboard led will be used. However, be aware that the pins can't drive much so you would need some electronics like a MOSFET and diodes to drive a motor load. You certainly could use the heater outputs for this too and they already have the additional electronics to drive an inductive load. The big issue you would have is obviously the firmware is set up to run the temp control loop on them. Also, unfortunately, the M42 command has protection code for the "special" pins like those that control the heater MOSFETs, stepper drivers and the other already wired pins. You could probably comment out some things in the firmware to free up the MOSFET pins so you could use the M42 to turn the MOSFETs on and off in gcode. Definitely doable...

Hey Jem, you are always doing cool stuff. Your question is a little unclear. If you want to control something through the printer using PWM, I believe the UM1 board and firmware actually support two PWM outputs for the type of servos used in radio control models (although you might need to rebuild the firmware). If it is PWM output you want, that would be a good way to go because the firmware has commands that allow you to set the position via gcode. The primary use of these outputs for automated probing that uses a small servo to flip the probe down, measure the bed and flip it back out. If you look closely at this pic at just about the center of the board, you will see a 6 pin header location (the headers are populated) where up to two servos would connect. Looks like the topmost pin is gnd, middle is +5 and the bottom is the PWM signal. Search for the word servo here: https://github.com/ErikZalm/Marlin

Check the accel settings. They might default to higher values with custom firmware. So even though you set 50mm/s, the default accel settings means it never gets to that speed. However, increased accel settings will allow the speed to be reached.

Daid might be a little gruff but as a developer I can understand. He is working on a lot of things and probably a lot of "uncool" mundane stuff. In commercial settings, you work on broken things first and new features that marketing wants to sell then sorely needed features then cools stuff (which you never get to). Here is some more info: http://umforum.ultimaker.com/index.php?/topic/6183-why-cura-so-slow/