gr5

In the past people have been all posting questions in this thread below - your answer might already be on one of those 6 pages of questions and answers. I'm not sure if @foehnsturm is more likely to see a post there or if you'll get a reply from him here. But I just mentioned his name so hopefully he'll see your question.

It's hard to tell from the photo but I think they *are* all connected together (by brim) and the whole sheet slide over. But it could be that the original poster purposely removed the "sheet" and looked at it and then put it back on the glass for a photograph and that the sheet did not slide around.

I think you want to set "combing mode" to "off"

In general though, printing slower and making sure ALL printing speeds are the same value (and non printing speeds are as fast as possible - at least 150mm/sec if not 300mm/sec). If you have changing printing speeds then the pressure in the nozzle doesn't match the speed right after a speed change and you can get too much pressure and leaking nozzle when it shouldn't be leaking.

I find the default retraction distance is best. You want to retract just enough so that the filament is genly laying on the inner curve of the bowden at the top. you don't want it to actually retract out of the head because then air can get in the nozzle causing all kinds of issues (including stringing). The printer doesn't normally retract for inner moves so that last picture - that's normal. I mean the model for this part shows a solid inside, right? Cura will not retract for moves inside the model - you can look at the part in cura layer view and the darker blue lines are non-retracting moves and the lighter blue lines are retracting moves. You might be able to force retraction on for interior moves if you disable all combing. Make sure I'm correct in layer view first.

No. Wash the tape. 🙂 Masking tape has a layer of some kind of waxy substance to keep the non sticking side of the tape from sticking to the sticky side of the tape (I suspect). You need to remove that layer. It just takes a few seconds with a tissue and some rubbing alcohol. I can wash a UM2 or UM3 bed in about 10 seconds. So it's not a lot of work but makes a huge difference to how well things stick. Even better, if this is PLA then use the heated bed without any tape. Clean it off very well and then put a layer of PVA glue mixed with water. There's 3 easy ways to get diluted PVA described in the video.

It would be good if you saw the failure in action however... It looks like the whole mess slid to the right at some point - does it go off the right edge of the glass? So your basic problem is that your parts are not sticking well. What are you printing? PLA? Is that masking tape? If you use PLA with masking tape it's important to clean the waxy non-stick surface off first - use rubbing alcohol. If you are printing PP (as opposed to PLA or ABS) then tape is indeed good to use. Also wash with rubbing alcohol aka isopropyl alcohol. In USA this is found in any drugstore next to where they sell bandages. In europe it's a little harder to find. But you can use ethyl alcohol instead (also known as vodka). However you have a heated bed! Okay you need to watch this whole video and become an expert at getting your parts to stick to the bed every time:

I would print in PLA first and try it but if you want to jump to ABS which is more difficult... In cura just select ABS as the material and 0.25mm for the nozzle size. Set all the printing speeds to be 30mm/sec. The rest of the parameters should be okay in cura. If your part is small enough you can skip the brim and if the teeth are on the glass maybe set "initial layer horizontal expansion" to -0.1mm or so to get rid of some of the elephants foot on the bottom layer. Look at this in layer view so you can see what it is doing. On the printer make sure the ABS material settings have the bed at 110C and the fan at 30%. This will not likely be default. Also since you are using a small nozzle and small layer heights you are printing slower so you can lower the default temp by 5C. ABS clogs easily if you leave the nozzle at temperature without printing. Within a few minutes it turns to a clog. So keep things moving. Put some gluestick on the bed and then spread it around with a wet tissue so it is super thin. Or even better use abs juice (google it) on the glass. Finally cover the top and front of the printer. You won't be able to get to 110C bed temp anyway without covering the top of the printer. 100C is not hot enough most likely and your part will come loose from the bed. 105C is probably hot enough but 110C is safer. For the front I usually use a 1 gallon zip lock bag or some saran wrap and blue tape. For the top I usually use a large box. The boxes that hold photocopier paper are perfect for um2. It fits on top perfectly with no cutting needed and no tape needed to hold it on. After you are on the 5th layer or higher - double check the fan speed in the tune menu or just look at it. It should be very quiet but still spinning a little. If ALL sides of the part are vertical you might be able to get away with 0% fan but probably not. If you have any overhangs or bridging or if your parts is smaller than 2cm you will probably need a tiny bit of fan.

The firmware is identical for the UM3 and UM3 extended (and I think maybe S5 also). There is a file on the UM3 (and extended) that says which printer it is. This file is not touched during upgrades. Funny story: There was one customer that got a replacement PCB for their printer and it had the wrong value in the json file that said which type of printer it was so the printer was a bit messed up because it thought it was a regular but it was an extended. Updating the firmware did not help. It was easily fixed by the customer with detailed instructions from Ultimaker.

It's not usb to serial. it's "3 pins on the olimex circuit board" to usb serial connector: https://thegr5store.com/store/index.php/catalogsearch/result/?q=olimex I stock it for convenience for USA people since Olimex only seems to sell it in Europe. I mean you can buy it from USA but the cables are shipped from Europe so it takes a while and your printer may be bricked while you wait. Plus I seem to remember shipping to USA costs much more than the cable. I plan to sell the flow sensor kit here in the USA when it is for sale.

I think for mechanical you need a picture so I'd look at that video posted above. For the electrical - a simplistic explanation. There are two coils in the stepper. Each coil has 2 wires going to it. Call the coils A and B. A is on the outside and B is on the inside part that rotates. We put sine waves of voltage and current through these coils. Not sure the frequency typically. If you shift the phase of the two sine waves of electricity it starts rotating the stepper. Once you move completely back into phase again that is one "step". UM steppers are I believe 200 steps per full rotation. Because of the gearing they are 5 steps/mm. There is something called "microstepping" or substepping. The stepper driver chip takes care of setting the frequency and the phase between the A and B signals. If you have 16 steps microstepping (as UM printers use for X and Y - I think Z is usually 8 microsteps) then that's 80 steps/mm. The driver shifts the phase by 16 steps. Since there are 360 degrees of possible phase shift put out between the A and B signals, 360/16 is 22.5 degrees so each time you do a microstep, the stepper chip increments by 22.5 degrees the phase between A and B. The arduino only sends 2 digital signals to the stepper driver: direction, step. If direction is high it moves the servo one way, if direction is low it moves the servo the other way. Each pulse on the step pin moves one microstep. So 80 pulses would be 1mm. This is very simplified explanation. I'll leave it to you to summarize even more simply.

Correction to Sander's above post - the origin is always at the front left corner although some of the UM printers do their homing at the rear left corner but that corner is considered Y value around e.g. 225mm.

The version of Marlin that is installed on Ultimaker printers doesn't implement the Linear Advance feature (someone correct me if I am wrong). It's a pretty cool feature but a bit of a hack (since the shape of the amount of extra filament extruded over time when graphed is not the shape of what is needed) but it is certainly an improvement over no linear advance.

There's a good chance your reseller will fix this for free, or if you damage the printer repairing it (after they give you permission) they may give you parts for free. This is usually called a "head flood". To clean it yourself (again I strongly recommend you at least contact your reseller before doing anything) you need a heat gun and lots of patience. Also heat up the nozzles to 200C. Expect this cleanup to take 2 to 4 hours of work and you may still damage one or both cores and or part of the print head. Everything is fixable though. To prevent it from happening again you might want to watch my video:

You will get less elephants foot. First some background: Let's say you are printing a 10mm-on-a-side cube. The diameter of the nozzle is let's say 0.4mm. When you look at the cube the X and Y lines on the outer edge of the cube will move 9.6mm because it pulls inward by the nozzle radius. Look at it another way - when "drawing the line on the outer edge" of the cylijnder it is NOT infinitely thin - it is 0.4mm thick. So you have to "draw" the line inward by 0.2mm (actually it pulls in by half the "line width" parameter - not the nozzle radius - but typically these are the same value or close to it). This is also true for example for a circular hole inside your part - say there is a vertical cylindrical hole going up your cube. Now the edge of that circle is also pulled away from the surface by 0.2mm so the diameter of the circle if it was supposed to be 3mm will now be 3.4mm (radius 1.7mm). Now HORIZONTAL EXPANSION. If set to negative value it will "pull inward" even more by that amount. So if horizontal expansion is -1mm that cube edge will have the servo moving 8.6mm instead of 9.6mm. autolevel is unrelated to any of this other than that UM purposely levels things a bit low (for better adhesion). They do this such that where nominally you would expect the nozzle to just barely kiss the glass at Z=0, instead it kisses the glass around Z=0.03 or Z=0.05mm. Personally I like to squish more than that for even better adhesion.