SYMPTOMS

[untitled86]

So, I've been having a hell of a time calibrating my Rostock MAX. Everything is a learning process. Part of my trouble is that I don't know what to look for. Help me help the community. With this thread let's list all the symptoms of common problems.

Print nozzle size (what happens with it's too small or too large)

Print nozzle temp (too hot vs too cool, I'm burning ABS at random heats. I don't know what to look for to tune this.)

Bed temp (again, what should a beginner look for here? My Oynx never gets above 60, so I just leave it on constantly when I'm using the printer).

My 4-walled square pieces are thinner in the middle, wider at the top and bottom. I have no idea what that means.

I just printed out these pieces, I have no idea what I'm looking at.

[foshon]

I believe the hotends ship with a standard .5 nozzle. This is the number you want you to put into whichever flavor of slicer you are using. Not sure what happens if you have something in there that is incorrect, although I can imagine everything from overpressure nozzle destruction to really crappy filament lacking prints.

ABS runs well from 220-240 degrees or so. You should check your hot end temp with a known good temp sensor to verify that you are in that range. To cold and you'll strip your filament in the extruder, to hot and you could melt the PEEK insulator in your hotend.

If you are running ABS without ABS juice 60 degrees is to cold IMHO. The goal of heating the bed, among other things, is to help the first layer stick to the bed. If your not having problems with that, I would leave it alone for the time being.

The "rotted" cube looks starved for filament. Polygonhell has an excellent extruder calibration post stickied around here somewhere. Find that and follow his instructions. Without knowing exactly what you are extruding and being confident that every rotation of the extruder equals a certain amount of plastic being deposited you'll beat yourself silly chasing down issues.

Usually, my order of calibration for a new printer is, extruder, mechs, then tweaking the motion. I calibrate the extruder/nozzle/temps, sort out any mechanical issues effecting print quality, and then worry about the subtle issues that can screw a print (sticky motion, backlash, uneven temps, fans).

Hope it helps.


EDIT: Polygonhell's calibration post is in this topic http://forum.seemecnc.com/viewtopic.php?f=54&t=1163

[cassetti]

Actually, from what understand, your extrusion widths and nozzle size should be 10% larger than the actual physical size - so you should enter 0.55 for the nozzle size. Incorrect settings will change your single wall thickness - if you set your nozzle too large (or let Slic3r guess automatically) - you'll get really thick walls (I measured almost 0.94mm on my first single wall print test!)

You really need to test your hotend temperature with an external thermocoupler to determine how far off the real temperature is from the temperature reported by the thermistor, and adjust accordingly. I have found that when your temperature is too low, your single-wall prints will 'crack' as the plastic cools (sometimes hours after the print is completed I'll hear a loud CRACK). Increasing temperature by 5C when you see cracking helps. If you have the temperature too high, especially with PLA, you'll have issues with the PLA jamming up and not extruding.

60 is way too low for ABS printing. I suggest you check your wiring, or add an extra set of wires to the power input for the hotbed, or between the Rambo and the bed (more wires means more juice can flow to the bed) - the onyx is a huge bed, by comparison, my work's Makerbot Replicator bed is only 9x6"! I suggest throwing a towel over your onyx and pre-heat manually to 80c before starting a print (i pre-heat to 80, hit print, and quickly remove towel, works great). 60C is what I use for printing PLA

Your single walls are thinner in the middle because not enough plastic is flowing - you probably have the wrong extrusion width settings, OR your steps per mm on the extruder is incorrect (depending on what board version you've got).

I personally went out and bought a 1" micrometer purely for the purpose of accurately measuring my single wall thicknesses to calibrate my prints (since I have 3 3D printers, i figured it was time to finally get one).

I've personally found it's extremely important to eliminate as many hardware related bugs as possible first before even attempting to calibrate, because trust me, configuring the slicing software is an artform itself. Even a few small changes to the slicer profile can have a major impact in print quality.

[untitled86]

I suggest you check your wiring, or add an extra set of wires to the power input for the hotbed

So, add another set of wires to the Hot End 2 plug? Do I simply mate up the positive and negative wires?

Also, I was printing directly on the glass. It worked 1 out of 5 times (most the time it wouldn't stick). Then I started using that blue tape. ...Wow, I should have done that before. It sticks every time now.

Thanks for all the help, but it would be amazing if we could trouble-shoot by symptom. I keep hearing that configuring the printer, Slic3r, etc. is an artform. A formal list of symptoms would take a lot of the technique out of this. I've even considered, once I have mine working properly, changing the settings on purpose and taking pictures of what happens when each setting is wrong.

[untitled86]

Likely connections are bad or motor is bad. This just happened to me.

[Polygonhell]

Is that PLA or ABS.
The holes on the walls are indicative of inconsistent filament feed, most likely the filament is slipping in the extruder, possibly after a retract, hard to tell without watching the print.
If you haven't already you should turn off retract/resume and get everything else dialed in first, then you can play with it and know what it's responsible for.
The other thing to do is start slow and increase the speed when you get decent prints again to be able to trace issue to a cause.

The general approach is start as simple as possible, resolve the invariable mechanical/settings issues then change one thing at a time.

[untitled86]

What causes edges to curl and be misformed like this...

[Polygonhell]

What causes edges to curl and be misformed like this...

To hot, too fast, not enough cooling.
The single walled test print take very little time per layer, it can easily be less than the time for the layer to cool below the plastic transition temperature.
If your worried about it try printing 4 at the same time, or run a bigger fan on the part.

[untitled86]

The .5mm wall is printing fine now. At least, I printed two that looked great. I'm printing the Hollow_Cube now. ...And I'm aiming the fan away from the model. I'm guessing I should aim it the other direction. I'm using a Feedrate of 50 and Flowrate of 100. My temp is set to 240 (probably high). I'll make the changes you suggested and try again. Thanks.

[cassetti]

So, add another set of wires to the Hot End 2 plug? Do I simply mate up the positive and negative wires?

No, I meant add an extra pair of wires to the Onyx hotbed. Your hotend should be sufficient with only 1 pair of wires. The onyx takes a LOT of juice to heat up to 80C - adding an extra set of wires to it will allow it to heat up faster (note, you may need to add another yellow and black pair of wires from the power supply to the hotbed power input)

I know you're trying to create a single thread for all problems and how to fix them - but trust me, there's just too many to fit into 1 thread. The same problem can be caused by 10 or 20 factors. It's tough to say here's the problem, and here's the fix.

Getting the plastic to 'stick' to the bed is only one part of the equation (I'm a big fan of Blue tape as well). In order to keep it 'stuck' to the build surface throughout the temp, you need to make sure you have a hot enough bed. For PLA, I have the bed around 50 or 60c, for ABS, I start with 75C for the first layer, and set it to 80C for all other layers (faster heatup times for my setup).

FYI - the point of the fan is to blow air onto the middle of the hotend (cooling the middle just enough so heat doesn't creep up and melt the filament before it enters the nozzle). This is important for printing with PLA.

[MSURunner]

Concerning nozzle sizing... Enter the size of the nozzle. Going larger or smaller is rather inconsequential to a point. The amount of plastic extruded is dependent upon the nozzle size entered, not the nozzle itself, so if you enter something too small or too large, you basically lose the accuracy of the nozzle. Going larger will shoot more plastic out and create wider line because the nozzle will essentially push the plastic outward after it has pushed it down on the print. Going smaller will push a small amount out, somewhere in the .5 nozzle area (it's still pretty darn accurate and you can often get into those hard to reach places on a tight print). There is a limit to how much larger you can go, though, and if you have a properly built hotend, the consequence will be filament grinding (when the extruder starts chewing through the plastic) and eventually a clog as the grind will stop filament flow, which allows heat to travel up the nozzle into the "cold zone". The reasons for the different nozzles are allowing a much more controlled extrusion, and in the case of the .7 mm nozzle, allowing more material through, thus effectively increasing your maximum nozzle size. SO, moral of the story, if you want to slightly (and I really do mean slightly) decrease the time of the print, you can bump the nozzle size up. If you have a corner that's not getting filled or a very thin wall and the slicer has recognized that the nozzle is larger than the area, you can bump it down slightly. Again, there are limits, but good software is flexible...

[Polygonhell]

The amount of plastic extruded is dependent upon the nozzle size entered

This is only true if you don't set the extrusion width manually, which I would recommend everyone do.

[MSURunner]

The amount of plastic extruded is dependent upon the nozzle size entered

This is only true if you don't set the extrusion width manually, which I would recommend everyone do.

Some slicers don't allow you to do this, namely Cura. Also, if you are setting an extrusion width, then it is essentially ignoring the nozzle size as the amount of plastic needed to create a line of a set width when forced through a .35 mm nozzle is the exact same amount of plastic needed to be forced through a .5 or .7 mm nozzle (simple math). So setting a value for one ignores the other. Pick one to adjust as needed if there are two available to adjust in your slicer of choice. Point being, nozzle size isn't as important as trying to calculate the amount of cooling contraction or extrusion width percentage based on nozzle size. If you are seeing an over extrusion it is the result of the slicer either not having the correct steps per e or incorrect filament diameters.

[Polygonhell]

While Cura doesn't allow you to set extrusion width directly, it more or less sets the extrusion width to the nozzle diameter.
Unfortunately Slic3r does some extremely dubious math and picks a width much larger than the nozzle diameter, as high as 0.9mm for a 0.5 mm hotend, IME this is much too large for a high quality print with a 0.5mm nozzle, which is why I suggest people set the extrusion width directly.
KisSlicer has no nozzle diameter setting, you have to explicitly set the extrusion width.

[MSURunner]

Part of what Slic3r is trying to achieve with its math in picking a size extrusion width is filling in those areas where a .9 mm wall exists and a .5 mm nozzle cannot make two passes through the area. While I would agree that going nearly twice the nozzle thickness is probably not ideal, it's a nice thing to make sure the print is solid. FWIW, KS and Cura (at least the limited release experimental engine that Daid has created to replace the SF backend) accomplish this by placing smaller amounts of plastic in those areas after the regular width extrusions are done. What you referenced with Cura and KS each not having the other measurement is exactly what I was trying to say. When your slicer calculates the math associated with volume of plastic, it doesn't care what the nozzle diameter is if it's trying to achieve a specific width, nor does it care what the width is if it's assuming it's the width of the nozzle. Setting your nozzle or width setting to the correct opening of the nozzle is the best way to achieve the most consistent extrusion. You can go up or down a little bit with either one and be okay, you just lose some of the consistency that the nozzle was designed and machined for.

[untitled86]

So, add another set of wires to the Hot End 2 plug? Do I simply mate up the positive and negative wires?

No, I meant add an extra pair of wires to the Onyx hotbed. Your hotend should be sufficient with only 1 pair of wires. The onyx takes a LOT of juice to heat up to 80C - adding an extra set of wires to it will allow it to heat up faster (note, you may need to add another yellow and black pair of wires from the power supply to the hotbed power input)

Following the instructions I received there are already two wires per pole (2 negative, 2 positive) already. I just double-checked my bed and board.

I know you're trying to create a single thread for all problems and how to fix them - but trust me, there's just too many to fit into 1 thread. The same problem can be caused by 10 or 20 factors. It's tough to say here's the problem, and here's the fix.

I know it's taunting, and my efforts may be wasted here, but eventually this will happen somewhere. I'd be happy if I only knew the 10-20 factors and how to rule them out.

Getting the plastic to 'stick' to the bed is only one part of the equation (I'm a big fan of Blue tape as well). In order to keep it 'stuck' to the build surface throughout the temp, you need to make sure you have a hot enough bed. For PLA, I have the bed around 50 or 60c, for ABS, I start with 75C for the first layer, and set it to 80C for all other layers (faster heatup times for my setup).

Good to know.

FYI - the point of the fan is to blow air onto the middle of the hotend (cooling the middle just enough so heat doesn't creep up and melt the filament before it enters the nozzle). This is important for printing with PLA.

So glad you told me this. I had no idea. Since I'm just doing ABS for now, no fan.

[cassetti]

I'd still suggest a fan - just activate cooling in slic3r. The point is to cool down the middle of the hotend so you have firm enough filament to properly extrude the melted filament out of the nozzle. It's very important for PLa, but still beneficial for ABS.

I personally plan on using a fan for 100% of my prints (Note, I'm currently using a 40x40mm fan I had from an old computer, not optimal, but still drops temp by 10-15c. Although I am going to purchase a 25x25mm fan next week (these kits really should ship with one!).

How thick is the wire used for your Onyx? I'm using 3 wires per pole of 18awg speaker wire and I still have trouble maintaining 80c over long prints. The supplied PSU is far too weak for this large onyx bed, I'm thinking about installing a second ATX PSU dedicated to the Onyx.