How to increase nozzle height?

[René @3dnerd]

Dear all,

I am not sure how is the procedure to increase the nozzle height, may you can help me.
Here my problem: On the first layer (which is created by the 0.8 mm nozzle) the nozzle is too close to the PEI printbed and sometimes the head wipes away the already printed filament. Nozzle opening and nozzle height is already calibrated by the wizard.
Now I wanted to increase the nozzle height to lift up the printhead a littlebit to avoid this.
But I am not really sure how to do this.
Is this the Z height value under head settings? If yes there are two values one for the 0.3 mm and the other one for the 0.8 mm nozzle.
Do I have to increase only the value vor the 0.8 mm nozzle now?
Lets say the Z value for the 0.3 mm nozzle is 0.35 and for the 0.8 mm nozzle it is 0.45.
Now I want to lift up the print head. What values I have to set now for the Z value of both nozzles?

Thanks for any help!

 

[biscuitlad @biscuitlad]

Well, frankly this appears to be a thorny issue… but you can change those values for each nozzle and see if it makes a better first layer. That’s what I do, and it works for me. The thing is to make sure you experiment very cautiously, i.e. change 0.45 to 0.46 and so on until you get your first layer perfectly smooth. Beware that if you raise it too much, then you’ll start to have problems with the first layer adhering properly to the bed. Of course, you can raise the bed temperature a little to see if that helps, but a higher initial bed temperature also lowers the viscosity of the filament, so it “spreads” or “flattens” more as it leaves the nozzle, but I guess that also helps it stick… ;-)

There are quite a few parameters to play with to try and get it right, and I find I have to do it separately for both PLA and ABS. But this is the only way I have found to get really good flat surfaces (nearly perfectly smooth to touch) - just using the standard nozzle height calibration wasn’t accurate enough for me.

[René @3dnerd]

Hi iscuitlad,

thanks for reply.
But since I did the nozzle height calibration already I worry that if I change any of the Z vlaues of the head this nozzle calibration gets obsolete.

So what does nozzle height calibration exactly?
And what happens to this calibration if I change the value of only one nozzle?

[gid @gid]

You know, all of this — including what the routine is actually calibrating; the relevance of the Z values; and so forth — is covered in the wiki calibration pages I’ve linked to a whole bunch of times now: http://roboxing.com/calibrate_nozzle_height

If you understand what the calibration routine is doing, it makes a lot more sense. Admittedly, the calibration routine isn’t very well worded, but it does actually work if you know what you’re doing. It’s the routine CEL themselves follow, and I’ve verified that they all seem to use different test sheets, ranging from business cards to scraps of copier paper. I’ve successfully calibrated it using a 5mm thick piece of pressed steel.

The routine itself is not subjective… the only thing that’s unclear is the description of the routine.

Tom Gidden -- Bristol, UK -- New Roboxer? Check out the wiki, and add yourself to the map! http://roboxing.com/user_locations

[Clicky @clicky]

@gid “The routine itself is not subjective… the only thing that’s unclear is the description of the routine.”

It doesn’t change perception it being random to the extent. What we would really need is explanation of:

1. why and where randomness is coming in

2. what is Nozzle Z offset really about and why it changes after re-calibration, turning PEI bed, changing filament, etc…

I know some of the questions were answered here - maybe this is good time to repeat answer’s on our wiki?

[René @3dnerd]

Hi,

thanks, and thanks Clicky ;-).

It is very simple, I just want to lift up the nozzle to get a better result for the 1st layer - nothing more.

Here my clear question:

Can I just increase the Z value for the 0.8 mm nozzle to get this? Or do I destroy the ‘nozzle height calibration’

[gid @gid]

If you do it correctly (ie. you know what you’re trying to detect), and if the hardware itself is deterministic (eg. nothing’s loose or worn, such as the belts, the motors, the bed clips, the head fixing screw) then it’s not going to be random. The random element comes in partly from the opinion of the user combined with misunderstanding the woolly terms “Too tight”, “Too loose” and “Just right”. The write-up in the wiki hopes to dispel these notions, especially the myth that the calibration routine is measuring the thickness of the card, and from there the height of the first layer, which still seems to do the rounds!

I didn’t just magick the explanation in the wiki out of thin air… I asked @Pete what the calibration was doing, and wrote it up on the forum. Then that was confirmed by @chrisyt, the guy that designed the thing!

My post, his confirmation, and the wiki page explain what Nozzle Z Offset is about.

http://www.cel-robox.com/forums/topic/robox-3050-cannot-print-a-single-object-from-the-complementary-flashdisk/#post-6134

CEL have agreed since the beta that the calibration routine is unclear — which is why I endeavoured to produce a better write-up — but they haven’t released a better one yet.

As far as the randomness is concerned… there could be a lot of different reasons that the variation happens, but those aren’t limitations of the calibration routine. Nozzle Z Offset shouldn’t need to change; however, if other things are variable then re-calibrating the Z offset can compensate for them, but it’s not a good solution. The better solution is to stop those things being variable.

Tom Gidden -- Bristol, UK -- New Roboxer? Check out the wiki, and add yourself to the map! http://roboxing.com/user_locations

[Clicky @clicky]

@gid That post you referred to is exactly what I was thinking of as start of a wiki page explaining what is happening in the calibration process. That post is going to be lost - buried under tons of other noise (like this one) and lost over time.

As for randomness I mentioned earlier - it is strange behaviour I observed on Saturday when spent almost 30 goes on one print - each time abandoning it around first layer or so. Over there I achieved almost perfect first layer (exaggeration) and touching manual Z offset of a 0.8 nozzle one way completely screw it. Returning to the same value where layer was acceptable produced non sticking of filament. At the end I ended with nozzle quite low… Almost as something has changed from the beginning of the process.

Anyway - more importantly - it is good to (try to) understand what is happening and how values in manual/semi-manual process of calibration affect the result. My understanding of tight/loose was similar to yours in the post you linked - but had no confirmation until now. Thanks for (re-)clarifying it for me.

[gid @gid]

@clicky: the fact that posts like that get lost is precisely why I set up the wiki. Forum posts are not a good form of canonical documentation; wiki articles such as http://roboxing.com/calibrate_nozzle_height that summarise and link to those posts are!

The forum post I linked to is pretty enlightening on other matters too… it talks about the UI disparity in the manual and the reasons why; the matter of flipping the PEI bed and what they’re doing to fix that; the staffing issues; and a few other things still being mentioned on these forums. All of that is under a post from August with a subject implying a problem with the complementary flashdisk: not even slightly relevant to that subject.

My point about randomness is that changing the Z offset might seem to fix the problem, but it’s actually just temporarily compensating for the problem. The idea that it’s the calibration routine that’s to blame because it’s somehow imprecise is missing the point that the calibration routine is just showing up a symptom of another problem. By dwelling on the calibration point, it distracts from investigating and fixing the cause of the problem permanently. It’s just wallpapering over the cracks.

Tom Gidden -- Bristol, UK -- New Roboxer? Check out the wiki, and add yourself to the map! http://roboxing.com/user_locations

[Clicky @clicky]

@gid sorry - didn’t pay attention to changes in the wiki. That page has changed a bit since long time ago when I followed it! O: )

And - yes - that’s the reason I mentioned it being captured there!

“My point about randomness is that changing the Z offset might seem to fix the problem, but it’s actually just temporarily compensating for the problem. The idea that it’s the calibration routine that’s to blame because it’s somehow imprecise is missing the point that the calibration routine is just showing up a symptom of another problem. By dwelling on the calibration point, it distracts from investigating and fixing the cause of the problem permanently. It’s just wallpapering over the cracks.”

If I understand the original problem (calibration of Z offset sensor) then I am not sure what would be ideal way to calibrate it.

[René @3dnerd]

Hi,

I still did not get a clear answer, may someone can assist? Here (again) my clear question:

Can I just increase the Z value for the 0.8 mm nozzle (the Z value vor 0.3 mm nozzle keeps untouched!) to increase nozzle height?
Or do I destroy the ‘nozzle height calibration’

• This reply was modified 1 week, 5 days ago by  René.

[gid @gid]

@3dnerd : yes, probably. Adjusting the value does have the effect of raising the nozzle to some extent. While expressed absolutely, the value for the 0.8mm nozzle is really relative to the 0.3mm nozzle, as that’s the one that does the height probe at the start of the print.

Just don’t adjust it too low: you don’t want to engrave the PEI bed. Best to only move the numbers up from the calibration point.

Tom Gidden -- Bristol, UK -- New Roboxer? Check out the wiki, and add yourself to the map! http://roboxing.com/user_locations

[René @3dnerd]

Hi gid,

thanks. Moving the numbers up will go into the wrong direction for me since the filament does not stick to the bed I want to lower the nozzle now.
The calibration gives a value of 0.57 for the 0.8mm nozzle, I reduced it to 0.37 and now it sticks better (at least on some regions of my PEI bed since it is warped/twisted).

The default value 0f 0.3 mm nozzle was 0.85, this value I leave untouched now - is that correct?

Does the Z value only effect the first layer?

[gid @gid]

In principle it affects the whole print, but with far less significance per subsequent layer. The quality of the first layer is probably the most important thing, though.

Incidentally, there’s a possibility that a warped bed might be correctable in a future version of AM… it doesn’t matter what profile the bed has if the software can move the head up and down accordingly. Right now it just considers a flat but not necessarily level plain, calibrating the height at four points. If it calibrated on more than four points, it could compensate.

Tom Gidden -- Bristol, UK -- New Roboxer? Check out the wiki, and add yourself to the map! http://roboxing.com/user_locations

[René @3dnerd]

Hi gid,

> The default value 0f 0.3 mm nozzle was 0.85, this value I leave untouched now – is that correct?

What about this question? Thanks again…

[Aimdy @mistsoul]

@3dnerd, after calibrating, if you need to manually change the .3 and .8 values, typically it is better to increase or decrease them together in increment and by the same amount to keep their offset consistent. Check step 10 of the first layer wiki which tries to explain this. Other members have reiterated this in other posts as well:

http://roboxing.com/first_layer_wizard

• This reply was modified 1 week, 5 days ago by  Aimdy.
• This reply was modified 1 week, 5 days ago by  Aimdy.

[Author]

Posts