Perfectly Printing a Vase, detailed instructions

[swordriff 119]

Updated!

Dear reader, these instructions are weak in the sense that they will not print much else than this vase correctly, and certainly not the water pump from simcity2000 in thingiverse.. Wich I recommend for beginners.. Try and see!

However, it is fun to see this ultra thin item print so quickly and look so good, it is much nicer irl than my photos! and it consumes very very little material too.

It has been pointed out to me by much more experienced and "older" members, that many of the settings here are extreme, and close to impossible.

It is therefore important to understand that these settings will not work for most other objects.

While printing this vase, feel free to fiddle with TEMP, SPEED, MATERIAL FLOW, while it is printing.. and see (and hear-- read article) how this influences the print.

Once it prints perfectly by maybe tweaking my settings, try another material, try increasing outer shell to 2 mm, try changing fills.. etc..

All the comments here are right, but the vase still prints.

Perfectly printing the “vase”. http://www.thingiver...6/#instructions

This is my second article for the beginner. The first is about levelling the build bed

correctly, which you can find in this forum. (DON’T do it yet, we need you to upgrade the firmware first) : , https://ultimaker.co...es/our-software

This vase is not water proof—but will give you a sense of accomplishment, and it really

looks good too! You can put a led light inside.. Guarantee! – that this print will work for you on UM2, at room temperature, Gold or stock Bue PLA filament.

Again, if you are already an expert, you don`t need this article, and there are no new “secrets” here. Here are detailed instructions and photo examples on how to print this vase:

 

Step-by-step

CURA

-2. Upgrade to latest CURA by downloading from here, https://ultimaker.co...es/our-software

and install it (if you have not already)

-1. Connect UM2 to computer w/ supplied USB cable, and turn UM2 on

0. Update UM2 firmware, In Cura: Machine, Install default firmware. Detach cable.

1. Level build board. (follow instructions in UM2 Manual or more explanatory, but not better, in this forum. http://umforum.ultim...for-best-print/

2. Download the vase, and open it in Cura.

3. Set Basic, Expert, Advanced settings in CURA according to my screenshots..

FIRST MAKE a note to your existing settings, especially in the expert section.

Cura basic:

Quality

Layer height 0.2

Shell thickness 1

Enable retraction ON

Fill

Bottom/top 1.2

Fill dens 20

Speed 50

Support type None

Platform adhesion Brim

Advanced:

Nozzle 0.4

Initial Layer 0

Cut off 0.0

Dual extr 0.2

Speed

Travel speed 150

Bottom Layer speed 40

Infill 0.0

Cool

Minimal Layer time 5

Enable Cooling fan ON

EXPERT

Retraction

Minimum 0

Enable combing ON

Minimal extrustion etc 0

Skirt

Line Count 2

Start Dist 3

Min length 150

Cool

Fan full height 2

Fan speed min 50 (generally ignored by UM2)

Fan speed max 100

Min speed 10

Cool head lift OFF

Infill

Solid top OFF

Solid bottom ON

Infill overlap 15

Support

Fill amount 10

Dist 0.8

Dist 0.2

Spiralize ON ( REMEMBER TO REMOVE THIS WHEN FINSHED VASE)

Brim 8

Leave the rest

Fix Horrible All OFF

4. Save to SD card like this: File,Save Gcode –choose destination, or: Press “save to SD” icon on screen..

Remove card after Cura replies "Saved to ..{card destination]

5. Insert SD card with this file into UM2. (Cura is done now. It has divided your vase into a large number of slices, telling the head to move from here to there, lower the bed, move at this speed, etc, with or without printing, etc.)

 

ULTIMAKER 2:

(the SD card is now inserted)

0. Clean bed glass ! (In Norway, we use water, a little “Zalo” (same as Fairy Ultra, a green dishwasher agent you can probably drink or at least mix with something,) pluss a dash “Salmiak”.

This will remove nearly everything from the glass, leaving it dry. Or use the other potions recommended in these fora…). Alcohol is not a particularly good fat solvent, but may work too.

0.5 No glue or tape this time..

1. Load Blue PLA (came with UM2) or Gold (from https://ultimaker.co...oducts/pla-gold)

(these detailed instructions are not guaranteed to work with other makes/colours, more on that in another article not yet published..)

2. Select PRINT, then vase (or whatever its name, from SD card, the USB cable MUST be removed already)

3. Wait ca 2 mins for heating of bed and nozzle, follow indicator..

4. AFTER and WHILE it makes noises to start printing (UM2 has now received instructions from CURA via SD card vase file). Select TUNE, TEMPERATURE. 230.

Select Material 100%, Speed stays at 100%.

5. While it starts printing, remove long string decending from nozzle, use finger.

 

The first couple’o turns may not stick on the glass, because the hot nozzle let the front of the filament pour out on its way to the print area, but with correct bed height it will after 2 “turns”. The high Temps are to help the very first layer… PRICELESS INFO!... haha.. Dieser Jägermeister ist sehr gut)

 

If, at this stage, it really %/)-ed UP, then your BED is too high or too low.

== ABORT, Adjust Bed and try again.

 

http://umforum.ultimaker.com/index.php?/gallery/image/2027-blue-vase/

The photo show the BRIM, automatically added by Cura to help making the

print stick. As you can see, it didnt stick too well in the beginning, because of slightly

incorrect bed height.. But managed to save this one! So Even if the “start is bad” it can be nice at the end!

6. LOOK: The Brim or “star” may look almost perfectly like mine, or be a little corrupted, due to bed height fault. Wait a few turns and see.. If the Middle turns out well; nice even lines in all directions, it will still be very good.

 

7. After 2 layers ( 2 minutes ca) go to Tune, and make the following change:

Temp 220 . This is your setting for the next small hour until finished

 

Except: Hold a finger on the filament in the back under the feeder. If the filament quickly “jumps back” with a “choff” noise, then there is too much feed for this print.

Slowly back off Material %% until the choff noise goes away at least a minute.

WARNING: These settings will not work for a building or many other prints, like the experienced moderators are pointing out !

 

Failure to adjust M will give uneven print, like lower part of yellow vase..

 

8. Watch the vase in the making!

9. When Finished, wait, Wait, WAIT 15 minutes for bed to cool down. Then even a fly can remove your vase! (PLA..)

About the settings and AVOIDING THE GAPS

Increadibly, printed at 0.2mm

http://umforum.ultimaker.com/index.php?/gallery/image/2019-blueyellow-vase/

I will not explain everything, only that when you look at my shots, the blue is nearly perfect, and the gold (ehemm.. yellow?) has more faults, especially lower down, and I will share why, and how you can copy the mistake:

The yellow vase, where there is trouble, is trying to over-feed, pushing more material from the feeder in the back than is actually printed. (thanks IrobertI)

 

If you, while printing the vase put Material to much more than 53 (lets say 100) (this could be another figure on your UM2 depending on the settings of your filament feeder 40-70?, sorry),then you get the gaps when the filament jumps back. Try it, you can set M at 150 and it jumps all the time. Remember to put it back!!

**********************

You need to learn to hear the difference between the retraction (mechanical brrrrrr noise) and the jumping back of the overfed filament (chuff-noise).

**********************

Put a gentle finger on the filament where it enters the feeder. Feel what it does while printing. As long as there is a CHUFF noise and the filament jumps back, reduce Material by 5, until the skips stop.

 

The vase will amost never retract, so it’s a poor example of just that ( retraction is pulling the filament back 45 mm very quickly when the head makes a jump, so not to get bridge or string, perversely in this particlar vase model there are almost no jumps..).

 

Now.. as long as you every 10-20 secs or more often get the “chuff” and jumping back of the filament, YOU CAN TURN DOWN THE MATERIAL FLOW, and eventually there will be no jumps and your vase will print perfectly like my blue one!

End.

The following changes have been made after feedback received.

I leave the text for completeness.. it has no impact on the print and is already emplemented in the text.

Very important to understand, after feedback I have received, that speeds are relative.

Settings in Cura are modified by multipliers in UM2. (example: Cura speed 50mm/s, UM2 Speed 50%= effective speed 25mm/s. Thanks gr5, illuminarti, iroberti and more..

Great eye-opener for me! Solved many problems I had.

Actually setting the speeds directly in CURA makes sense, It will then be saved inside each gcode file and saved for next time! Thanks, illuminarti.

If that's what you want,for this print, then in Cura set Print speed to 50, and bottom layer speed to 20. Then in UM2, leave the speed at 100%.

In many other prints you want a lower speed. If you set it in Cura, it is saved in each individual gcode (model) file, smart. If you set it in UM2, by changing the percentage, it is much quicker.. but not so smart, it is not "remembered"...

How quickly can you print the vase? 100mm/s? Try playing with speed, Temp, Material.. etc.

Thanks for reading and lets know how it goes

..

 

[IRobertI 520]

I'm confused... are you saying that you are printing the vase at half the normal flow? And that you are increasing the flow for the bottom layer? If you have to turn down your flow by 50% to avoid extruder skips there's something seriously wrong with your settings in the first place. Ideally you should never have to have flow set to anything but 100% (personally I never, ever, touch it).

You forgot to include pictures btw but I did take a look at your gallery and the bottom layer doesn't look perfect to me at all. Are you using a raft or something? Why aren't your lines connecting?

 

[swordriff 119]

This works. Bergis!

If you enter the settings I provide, the Vase will print perfectly, unless

the feeder is adjusted differently.

As you can see from the blue vase it is nice, and the blue provided by UM is a much more "difficult" material than some other colors. (The color agent greatly influences "printability").

Now.. the base is not perfect, but the vase is, and printing very quickly at 0.2mm, an unbeatable value.

Keeping Material flow at 100% makes the material jump back every 5 secs, making trouble as indicated in the yellow vase.

I am printing (according to settings) with only 1mm walls (very fragile, but nice and transparent).. and these are the correct settings.

By manipulating Temperature and Flow in the beginning, we are helping the fil to stick, avoiding the glue, and when its cool its dead easy to remove the item.

 

[illuminarti 18]

Thanks for writing this up, swordriff! I sent you a PM with some thoughts.

 

[gr5 2,189]

Thanks for the post.

If the extruder is making the "chuff" noise and slipping backwards, you should instead slow down the print. The nozzle can only extrude a certain amount of PLA per second. If you are getting the "chuff" sound you can fix this by either slowing down the printing, or raising the temperature (but that might cause other problems) or print thinner layers (less plastic per second). Decreasing flow is usually not a good idea but I'm sure it worked for you! Still if you could print at 100% and print slower you might get a better result.

HOWEVER I also note you have spiralize checked and you have the shell thickness set to 1. This causes a similar setting as if you set the material flow to 1mm/.4mm or 250%. So setting material flow to 40% then brings this back to a .4mm bead with 100% flow.

What you posted makes sense and it will work but it's a little strange to set the thickness such that flow is 250% and then compensate by setting flow to 52%. You could have just set the wall width to .52 instead. I think it's a little easier to understand if you keep the wall at .4mm or you keep the flow at 100% and only change one of these parameters at a time - not both at once.

 

[swordriff 119]

Thanks for all the feedback.

I did make some mistakes,and am editing the article.

Agree to most, but everything is clear if you PRINT the thing.

I do not concur on slowing down print speed to avoid the chuff jump of filament.

Slowing down will (if anything) increase the jumping back of the material.

I am really interested in the calculations of the speeds and other settings, and still have a lot

to learn, but I feel a little like Im on a steamer upwind of the Mississippi.

This just works.

I did make some changes, especially wrgt the SD card usage, speed settings (amended the article).

The thickness is set ententionally, so that any misalignment is really easy to see while it is printing

and you can make adjustments all the time!

Thanks for all the feedback guys, Ill tread more carefully next time.

 

[SandervG 1,521]

Hi Swordriff,

Thank you for your post and the efforts you put in writing this down.

It is very helpful to beginners to read such detailed instructions on how to make certain prints.

It is also good to read that you are willing to think outside the box and create your own workflow.

But I think, it is important, especially in how you communicate, that it is clear that it is not the most common way of creating a profile (even though it works for you). And not tell it as an absolute truth, but more like a possibility / alternative. The guys who replied to you are very knowledgeable and really know what they are talking about

Have a great day, looking forward reading your next post!

 

[IRobertI 520]

I do not concur on slowing down print speed to avoid the chuff jump of filament.

Slowing down will (if anything) increase the jumping back of the material.

I am really interested in the calculations of the speeds and other settings, and still have a lot

to learn, but I feel a little like Im on a steamer upwind of the Mississippi.

No, slowing down a print will not cause an increase in extruder skips. It will lessen skips. Skipping happens when you're trying to squirt too much material too quickly. Re-read gr5's post again and I think it will make sense to you. Avoiding skipping is all about keeping volume/second reasonable.

 

[gr5 2,189]

Slowing down will (if anything) increase the jumping back of the material.

 

I believe you. This is possible and makes sense only if flow is too high or if cura is extruding for more than a .4mm nozzle (1mm in your case which is like 250% flow).

But when spiralize is off and flow is close to 100%, you actually get less "jumping back" with lower speeds (or higher temperatures). Decreasing feed rate slows down all 4 Axes: X,Y,Z, and E. So it slows down the E, or extruder as well as X,Y. This in theory shouldn't reduce the "jumping back" except that the problem that causes "jumping back" in normal prints is that the pressure in the nozzle gets too high and the nozzle can only extrude a certain amount of PLA in a given time.

I have to admit though that the "jumping back" feature of the UM2 extruder is still new for me and I've only been using it for a few weeks and I'm still learning much that was not so obvious on the UM1. When the UM1 under-extrudes it is not so obvious! The jumping back is instant feedback that something is wrong.

Did you seem my curve that showed speed versus temperature on the UM2 that finds the limit where it does the "jumping back"?

http://umforum.ultimaker.com/index.php?/topic/3418-um2-extrusion-rates/

I think I will have to completely rewrite that post some day as I've learned some new things since then but overall I think it is still useful.

 

[swordriff 119]

Im avoiding the skipping by turning down the M, or turning up the speed.

Just tried it. There could be something wrong with my UM2, of course!

Today, my feeder assembly loosened... I fail to see the logic of

less speed-- less jumping back of material.. Im my UM2 it is the opposite..

"Lend Ears of You in Heaven!"--

 

[illuminarti 18]

The extruder clicking happens when the extruder can't move the filament when it tries to. The reason it can't move it is because the pressure in the head is stronger than the force the extruder can apply to the filament. The reason the pressure in the head is so high is because you are trying to force a very large amount of plastic out of the nozzle every second. If you slow down the print, then less plastic is being forced through the nozzle every second. This by itself requires lower pressure, and it gives the plastic more time to melt in the head, lowering its viscosity, also reducing the needed pressure.

So as you slow down the print speed, it becomes easier to extrude, and the extruder will click less often. The only reason that I can think why it would increase is if originally you were just grinding away the filament at the high speed, and as you slow down the teeth grip better, but stall more often. But that's a long shot.

George - what was your thinking as to how the clicks rate would increase with slower print speeds?

 

[swordriff 119]

I have the complete opposite finding!

When it clicks and shoots back the filament because of too high pressure in

the tube, it helps either increasing print speed and or Temperature or lowering M.

Puh.... its tough!

Anyway.. it prints. If you print at M100, then you get

the skipping back every 5-10 secs. Lowering M changes it gradually intil

the skipping stops, and the vase has no faults in the sides!

Thank you for your combined help, and I will study the Diagrams very carefully, gr5.

Im signing off for the weekend..

 

[gr5 2,189]

Illuminarti - if for example flow is at 200% and you print something normal, you are likely to get those kick backs no matter how slow you go.

 

[illuminarti 18]

Yes, but if you slow down the print while keeping everything else the same, you should get the same or fewer clicks, no - not more?

 

Illuminarti - if for example flow is at 200% and you print something normal, you are likely to get those kick backs no matter how slow you go.

 

[gr5 2,189]

you should get the same or fewer clicks, no - not more?

 

I would have predicted same quantity of clicks but since everything is slower it would happen less often in a given minute. But swordriff experienced something different.

You're going to have to experiment yourself some time I guess. The UM2 is new for all of us and it is making the "clicks" more obvious so we are learning new things about under extrusion.

 

[illuminarti 18]

The only effect the print (i.e., linear, head) speed should have on the extruder is in determining the rate of plastic to be extruded per second. From the extruder's perspective, lowering the flow rate directly, and lowering the head speed amount to exactly the same thing. The extruder motor doesn't really care where the print head goes to deposit the plastic that it is squeezing out. (Unless the UM2 firmware is trying to do some sort of pressure management calculation during printing?? Or there are unintended mechanical feedback loops via the Bowden tube?).

So, all things being equal, lowering the speed should reduce the volume per second, and hence the pressure, and so reduce the number of clicks. Theoretically, to the point where they stop happening altogether if you go slow enough. So if the clicks decrease with higher print speeds, something very strange is happening.

Which is not to say that it isn't happening in swordriff's experience - just that it doesn't match any of the standard models we have of how the extruder behaves. More tests are indeed in order!

 

[gr5 2,189]

If the head is .2mm above the layer below and you are extruding at 100% then there is just enough room for a .4mm wide bead. But if you try to extrude 200% flow or .8mm worth of plastic for a .4mm bead then you have to squeeze that plastic out through a .2mm crack which means higher pressures in the nozzle. The plastic has to go through a .4mm hole, and than an additional resistance getting out through the .2mm crack on either side.

Typically on my first layer both my UMs won't output more than a .4mm wide bead no matter how much pressure there is. This is more true if the first layer is only .1mm above the print bed. This tells me you need quite a bit more pressure to get that .8mm wide trace over a .4mm wide trace. If you are printing with .1mm layers it's much worse (much higher pressures).

 

[3dcase 37]

pfffffffff......

 

[illuminarti 18]

When you're extruding a nozzle-width bead, the tightest constriction that you are forcing the plastic through is the nozzle opening itself. If the nozzle diameter is n, then the area of the opening is pi(n²/4).

If you're extruding a bead wider than the nozzle opening, then you then have to force the plastic out between the nozzle tip and the the layer/bed below. If you consider that space as the surface of a cylinder, then the opening that you are extruding into is the surface of a cylinder that is n across and h high (where h is the layer height).

I.e., the area is going to be pi x n x h.

This size is going to be greater than the nozzle opening until you get down to h = n/4. In other words, for a 0.4mm nozzle width, if you have a 0.2mm layer height, any 'extra extrusion' width is being squeezed through an area that is twice the size of the nozzle opening that it has already gone through. So it's probably not adding greatly to the pressure on the nozzle, compared to the initial challenge of getting out of the nozzle.

However, once you get down to a layer height below 0.4/4 = 0.1mm, then the space that you're trying to spread the extra filament into is smaller than the nozzle opening, and so at that point i would expect extra head pressure to be required.

Of course in practice there are probably various issues of turbulence, cooling at the boundaries, capillary effects etc, which introduce extra resistance at layer heights somewhere thicker than n/4. But it seems like a useful rough guide - certainly when going below 0.1mm layers, using an extrusion bead that is wider than the nozzle opening is probably a really bad idea.

 

[gr5 2,189]

Well you can't include the enitre piXdiameter area because the UM prints in straight lines and you can't include the area you just printed. And you probably shouldn't include the area up ahead that you will be printing soon. So I think it's more like two hXn rectangular openings or 2 x h x n instead of 3.14 x h x n. But there is "more" resistance near the edges - I don't know what this effect is called. Anyway I agree that by the time you are at .1mm layers, printing more than nozzle width is going to be tough and so you have to print slow or just don't do it.

 

[jml31347 0]

I am not new to 3D printing , and I have discovered that a flow rate of 110 to 120 % solves the problem of smooth , or lack of it , sides.

I seems to depend on the type of filament and print temp.

What I am needing to find out is the best way to increase the filament feed without increasing bed speed.

I am experimenting with NON plastic filaments and paste, and need a better way to control the extruder speed without changing , say 10 t0 50 mm/s bed speed.

 

Joel

jml31347

[jml31347 0]

I solved the issue I was inquiring about.

I had forgot that the compiled FEED RATE is not affected by the control panel speed FR% knob.

In Cura , they do not want you go go higher than 150%, but it will take and compile up to 350%.

So I recompiled my test , and it worked ?

 

Joel

jml31347