gr5

To remove the sensor someone said to put a small wood screw in there that is the right diameter to get a good grip then pull it out with pliers. The sensor is repairable. Split it down the side with dremel, locate the tiny temp sensor (it's very very very small - like grain of sand) and attach it to fresh wires and pack it back in there with maybe kapton tape. Beware normal solder melts well below 230C so you need solder that is mostly lead. Check melting temp of the solder before using it. Good article on melting points of different types of solder in wikipedia. Kapton tape can handle 300C no problem.

What kind of printer is this? Do you have UMO board? UM2 board? Ramps board? Were you fiddling with wires on controller board or on head?

When I get new materials (some day) I will test them. I'd love to do PET. Very disappointing in the protopasta. I have put PLA on something outdoors in the sun and it has lasted 15 months now (over a year) so it's not the sunlight - it's the temperature.

Welcome. I suggest you start here: http://support.3dverkstan.se/article/23-a-visual-ultimaker-troubleshooting-guide And then post a picture of one particular problem you've been having. Preferably one issue per posting.

The skinny wires go to the temp sensor (you can see the traces on the board going over to the tiny temp sensor part) and the thick wires go to the heater (you can see those traces also - they cover the whole board). Both objects are pure resistors and can be wired up either way (no polarity - it works backwards and forwards).

@twistx my heart goes out to you! You've tried so many fixes! In some of your posts I think there is a tiny drop of sand in your nozzle but you've cleaned it out so many times! Please update your lcoation. If you are in North America you can get a replacement isolator within a day or two. But much slower in other countries. UM has a USA factory now. I'm hoping the problem is your isolator. The new one came out roughly March 2014. A lot of recent printers have a faulty temp sensor that works fine at 20C and 100C but not at 220C. I created this short video to help you test your temp sensor without needing any equipment:

Cold pull is very simple. Heat head to 180C or warmer and push in the PLA until a little is coming out the nozzle. Then let it cool to 90C (130C for ABS). Then pull hard back at the feeder. Maybe 10kg of force. If you need more force it's too cold. If you need less than 1KG force it was too hot. Here are 4 successive cold pulls - the first one was slightly too warm but good enough:

My fine is basically silent. Someone else came over to my house and didn't believe me when I said it was on. He had to touch it. He has an UM2 and his is louder. If I turn off everything in the house (televisions, fans, dishwasher) and no cars are driving by outside and if I close all the windows (noisy birds and insects!) I can hear it from about 3 feet away. Any farther and the blood going through my ears drowns out the noise. So... If you can hear the fan at all then I consider it defective. We are talking about the "third fan". Behind the print head. I recommend buying another one. Or remove the decal and add a drop of oil (I've done this to other fans but don't know if this trick works with this fan).

Many many people have underextrusion just like this. It is amazing how many things people try that actually fix the problem. I wish I could say there is only one problem and one fix. Just today someone put the filament on the floor and that fixed all their problems! Because the filament entered the extruder straight. That's my fix by the way. The most common problem for new printers is the temperature sensor seems to be off such that the print head is too cold. For older printers like yours the most common problem seems to be the isolator. Fortunately 90% of the printers out there are fine. Clogged nozzles are also common. You can remove the nozzle and put the entire thing in a flame to carbonize anything left in there.

If you don't care how pretty the parts are you can print them quite fast. The best method whether you print fast or slow is to print it once and then measure all the errors and then adjust your CAD model by those amounts. For example all your vertical holes will be around .5mm too small in diameter but your outside surfaces will be only shrunk by around .3%. Horizontal holes will be almost perfect. Bottom layers might exhibit swelling inwards or outwards depending on heat bed temperature and height from heated bed and fan speed for a given layer. All of these can be compensated for in CAD such that you get a perfect part. Usually if a part has 50 dimensions only 5 or so are critical for fitting with other parts. However if you change your print speed, bed temp, fan speeds, nozzle temp the part might change a little bit. Even if you change printers. This may sound like a pain in the neck but usually I only have to print a few layers to get a measurement as the Z dimensions are always the most accurate.

I think it's going clockwise around the cylindrical part some times and counter clockwise other times. Combine that with high print speed changes where the nozzle pressures vary greatly and you get this pattern. You might be able to reduce this if you reduce the polygon count around the cylinder. Marlin can only look 20 or so line segments ahead and if you have 20 line segments in that cylindrical portion it has to slow down a lot (be ready to stop because it doesn't know what's coming up and there might be a sharp corner coming up and not just this smooth curve). You can almost surely fix this by printing at the jerk value which is 20mm/sec. That way the printer will not speed up or slow down at these spots. Or you can get it close enough at 35mm/sec to greatly reduce this. 1) What speed where you printing at? 2) How many line segments are in each of those curved areas? (try to keep it under 10 line segments) In other words, for a given slice, how many polygons does the printer head pass across? What CAD software did you use? Some of these have settings when you save the STL (e.g. solidworks or DSM), other's have settings when you create the Curve (e.g. sketchup).

Regarding makerbots - everyone who I've talked to and has one and also a UM (I've met maybe 20 people who have both) hate the makerbot compared to the UM. They say the quality is much worse and it keeps breaking and currently doesn't work at all but their UM is working great. 100% of the people who have both have told me this. Of course someone will now probably post below saying their makerbot is better and works fine, lol. Also I am told that their customer support is completely backed up for months now. I totally agree with your perception about people buying UMO versus UM2 and makerbot. Part of the problem is the phrase "3D printing" which people and the press make it sound as simple as 2D printing. It should be called "additive manufacturing". 3D printing is just as difficult and complicated as using a milling machine yet my neighbors are much more scared of getting started on buying a milling machine. Isolator UM has improved the teflon isolator so that it will last longer and they are working on an all metal head I believe. Unfortunately that might be the UM3. Atomic Method (aka cold pull) I now do this on EVERY filament change. In fact I don't use the menu to change filament anymore. At the end of every print I set the temp to 90C and push the filament in a bit while cooling (unless I know for sure I'm printing again with the same color) and when it gets to 90C I pull the filament out the back. After it is almost to the feeder I let it cool another 20 seconds then pull fast and hard so a piece doesn't break off in the feeder. I then cut off the tip of the filament (two cuts at different angle to make a point) before storing the filament. When I change colors I no longer have to wait to get all the old plastic out.

Of course if you feeder is slipping or skipping then pushing can make a huge difference.

@mevander 1) Pushing on the filament should only help the extruder not slip/skip. It shouldn't change the flow by more than 1% which is not visible. Better to just raise the temperature to 240C as this reduces viscosity by a huge amount. But just as a test. 2) I recommend you calibrate Z and E axes. 2a) To calibrate Z move the Z axis by 10mm or even 100mm and see if it really moved that far with a ruler. Maybe lay a flat pencil or ruler on the bed hanging over the edge and tape a ruler vertically to the corner and watch where the flat lines up with the ruler as you move it up and down. 2b) To calibrate E axis tell the UM to retract 10mm of filament, mark the filament with a marker. Retract another 100mm and make sure it really moves 100mm. To have control like this on the axes you should connect a computer through USB to the UM2 and issue move commands. I like to use pronterface here: http://koti.kapsi.fi/~kliment/printrun/ but you can use Cura in pronterface mode (I've never used cura in pronterface mode so don't know exactly how to turn that mode on) which simulates a few (but enough) of the pronterface features. Do Cura print window.

Worked great for me - no holes - 220C:

Sketchup is difficult with Cura but I've used it many times with success. YOu can't have any duplicate walls or intersecting walls or missing walls or holes in your walls or interior walls. Any of those will show up red in Cura xray view. Even if you do you can often just check "fix horrible A" or B and that often fixes it. You can tell in layer view as labern says above. Don't try to print it until you look at it in layer view. Check *every* layer - slide through the layers one at a time.

That's normal. You can best fix that with fan. Lots of fan was it at 100%? That's in expert settings I think. Also if you want good quality I would print cooler (maybe 200C) and slower (maybe 25mm/sec). The slower you print, the better the quality. You have to have patience. But this is a very tricky part. You might need much more support material. The supports will help keep the overhangs form creeping up like that and getting hit by the nozzle. It's easier to wet sand after removing supports than to try to get it to print perfect without support but if you go for the latter method then print slow and cool.

Damn - youmagine is buggy! In Chrome there is no way to see the red/blue markings. I just uploaded a new version of the diagram with tons of whitespace above and below to make youmagine happier about the aspect ratio.

Here is the exact picture - you have to click on it to see all of it - youmagine bug!

In the icons, youmagine only shows the center of the picture. When you click on photo arrows over to the "checkboard" photo you can see the red line and blue arrow.

Sorry! I see your post now. It fails at 3mm/sec. You have underextrusion problem. There are about 15 possible causes. Many of them shouldn't have changed since the printer was new. The most likely problems where something changed are: 1) clogged or partially clogged nozzle 2) tiny pieces of plastic in the Bowden causing it to be tight 3) Deformed isolator 4) Friction form filament at an angle or tangled filament #3 is most likely Testing #1 and #3 properly both require you to take the head completely apart. Testing #2 is best done by sliding filament in and out of bowden with power off. Through the feeder. The feeder does not provide much resistance. Testing #4 - simply put the filament on the floor so that the filament arrives at the feeder more vertically. If you take the head apart, take the white teflon isolator and push some cold filament through it and check for resistance. Especially curved filament. Consider drilling the hole with 3mm drill. Consider ordering a new one (the new ones have glass fibers and are more resistant to deformation). 240C and hotter tend to deform the isolator over many weeks. I often print at 260C but try to avoid these temperatures if at all possible. Lower temperatures (210C) can possibly also deform the isolator. Especially if your temp sensor is inaccurate.

Something is strange. It could be: - flow is not set to 100% (check tune menu while printing) on UM2 - diameter of filament doesn't match diameter set on UM2 (e.g. you set 3mm but filament is 2.85mm) - You didn't use ulticode mode in Cura and set the flow or filament diameter wrong in Cura - No infill to support top and not enough layers (with no infill you need maybe 1.0mm top/bottom thickness) - 20% infill selected but not enough top layers (with 20% infill you need maybe .6mm top/bottom thickness) - infill set to 80% but "solid infill top" or "spiralize" or "only follow mesh" unchecked in Cura expert settings - infill speed higher than print speed (should be 0 - advanced settings) - underextrusion (try printing at 240C, .2mm layer, and at 20mm/sec - even a half clogged nozzle should be able to handle this) - top/bottom thickness is not an integral multiple of layer height (e.g. top layer thickness is 1mm and layer height is .21mm) - Z stage is moving too far on each move causing mild underextrusion (nozzle too high off part for last layer) - Z stage steps/mm miscalibrated - Extruder steps/mm miscalibrated That's all I can think of right now. If you send me gcode for 20mm cube (keep it small! - preferably only 10mm tall try to make it under 20 minute print) and post photo of your problems I will print it on my UM2. Post gcode on some web service somewhere and provide a link.

I think you should do this test at 230C: http://umforum.ultimaker.com/index.php?/topic/4586-can-your-um2-printer-achieve-10mm3s-test-it-here/

Consider printedSolid.com. They have great filament and ship from USA. Not sure if they ship to Canada but the owner is on this list somewhere.

Try 85C then. You should need at least 1kg of pull and It should look like this: