Optimum nozzle/bed distance

[yellowshark 153]

So, you have levelled your bed, set your z-offset and start the print. What is the optimum distance between the nozzle and the bed for adhesion?

George posted on the “Printing on glass” thread that …”The software assumes your paper is .1mm thick which is typical” - which I assume is referring to Cura. Does this mean the answer is .1mm?

Slic3r adds .2mm to the z-offset. I do not see any adjustment to z-offset in the Cura g-code.

Does the distance change depending on filament type? I.E. PLA vs ABS vs Nylon?

I am suffering a little bit with adhesion at the moment and whilst I am using a setup process that has always worked for me, I would like to be sure that I have the target distance correct.

 

 

[Dim3nsioneer 557]

I use 0.06mm distance with clean and cold nozzles. I measure it with a feeler or with the eye. This distance gives a very nice first 0.2mm layer with optimal adhesion. At least on my UM1.

Approx. 0.1mm ist the distance you should have with a cold nozzle as the thermal expansion of an UM1 hotend between room temperature and 200°C is 0.1mm.

 

[illuminarti 18]

Yellowshark, I think you're misunderstanding how this works.

When z=0, the nozzle should be exactly touching the bed. That's what it means.

A 'z-offset' in e.g., Slic3r, is a way to compensate if you have a physical z-min endstop that is hard to adjust, and leaves the nozzle not touching the bed. I'm pretty sure that Slic3r doesn't add 0.2mm to the z-offset. That makes no sense (although Slic3r does plenty of things that make no sense, imho). Rather, the 0.2mm that you are seeing is the probably just the first layer height that you set Slic3r to use.

When you start printing, the nozzle will move to the correct first layer height, and then extrude enough plastic to fill that gap. In theory that should be plenty, if the bed is clean, and the nozzle really is touching the bed at z=0.

What George was referring to was that when you use the bed leveling wizard in the UM2 controls, it asks you to set the bed 'one paper-thickness' away from the nozzle. The firmware assumes that you have then set the bed at the z=0.1mm height, and calibrates itself accordingly so that its z=0 position has the bed 0.1mm closer to the nozzle - i.e., just touching it.

 

[gr5 2,172]

What George was referring to was...

Yes. Exactly.

After printing the first layer (or the whole part) if you pull it off and look at the bottom layers carefully and there are thick gaps between the infill pattern then less of the part is touching the glass and so it will stick (half?) as well. If the volume of air in the gaps is say 20% of the volume of a stripe then you are probably levelling the plate 20% too far away as a percentage of the first layer height. So if you bottom layer height is .3mm and gaps are 20% then you should get closer to the glass by .06mm.

This isn't how I measure it. Usually I just use the paper, use a thick first layer (.3mm or at least .2mm) and it usually works out fine. If it is something very critical due to strong lifting forces I always use brim and make sure there are no gaps in the brim and if there are I abort the print and re-level.

If you are too close to the bed then the extruder tends to click a lot. Usually I don't care but sometimes I abort and re-level. Once you get levelling close enough it tends to stay good for many weeks.

 

[yellowshark 153]

Thanks guys for your help, much appreciated. Yes Illuminarti I think it is fair that I have not fully understood what goes on and this has been caused by me using Slic3r when I started, as I have never known what this “0.2mm” line of code is and therefore assumed that slic3r wants to be 0.2 above the bed before accounting for the 1^st layer width.

I have just realised I still have slic3r on this laptop and I can now see that I am wrong. My reading now is that it is lowering the bed by 0.2mm to move the extruder from one location to another and then raising it back to the original position prior to extrusion. Sorry for the red herring.

I use the z-offset to account for the glass tray that I lay on top of the print bed. IE I set the bed levelling by…

 

• 
  Remove glass tray
   
  
• 
  Heat bed to 60
   
  
• 
  Send code G1 z0.15 to the printer
   
  
• 
  Use a sheet of paper 0.15mm thick
   
  
• 
  Set the bed levelling so there is just the slightest amount of friction felt at the three setting points.
   
  
• 
  Set the z-offset with the depth of the glass tray (now using Cura that is a line of start g-code)
   
  
• 
  I then print the first layer of a test cube to ensure all is well and if necessary will tweak the z-offset marginally until I get the desired result.
   
  

 

So I know what I am looking for in end result and I know how to achieve it but I did not know what the target gap was because I had been confused, in error, by what Slic3r was doing.

I now know that theoretically is should be 0 J.

Dim3nsioneer makes the point about heat expansion of the hot end. So with my average glass plate depth being 7.06 should I be setting my z-offset to between 7.12 and 7.16?

This is a bit strange to me as I have always got my best results with a z-offset of around 6.91-6.95, which seems to be a significant difference. I have only just swapped over to Cura so I need to look at that carefully, which is why I posted.

 

 

 

 

[illuminarti 18]

What type of printer is it?

 

[yellowshark 153]

3ntr - pretty similar to the UM apart from dual extruders, all cnc machined alloys and the filament feeder drive is on the external rear of the casing whereas I get the impression on the UM, maybe wrongly, that the drive is located near the extruder assembly.

 

[cecilievf 0]

This is a bit late but I'm trying to do the same as yellowshark and printing on top of a plate on the build plate. Could you please show me what you start gcode looks like as I can't get it to work.

Thank you.