Author Topic: Chasing finer detail in 3D printing  (Read 2974 times)

Offline Joules

  • Hero Member
  • *****
  • Posts: 1271
  • Country: gb
Chasing finer detail in 3D printing
« on: August 14, 2018, 04:34:21 PM »
Moving on from the 1000hr service on one of the newer printers, I found the nozzle was badly worn, so ordered a couple of replacements.   Going right back to the 3D printed Quorn I had wanted to try a 0.3mm nozzle to see if detail could be improved, especially in the area of ringing where you get echos in the print process and small text or details carry ripple or shadowing.  I serviced the other printer prior to fitting it with the 0.3mm nozzle and started doing some testing.  Just 10x10x6mm cubes, to get the filament feeds and speeds right, make sure everything is running concentric.  It is now dialled in to about 0.015mm all round.  I needed something I would normally use, and that is my inlaid text on prints.  The 0.4mm nozzle really starts to struggle with fonts smaller than 4mm, down to the 0.48mm blob size, that gets even worse as a nozzle wears and you get ooze creeping out in random places.  The font used here is 4mm tall, but should print well even at 3mm, looking how many lines I have in the thinest portions.

Test 1 below was unfortunatly done using (old) translucent filament, this made it a bugger to photograph.  I then further cocked it up by trying to use sharpie to darken the front and show off the 0.2mm deep text.  Well a few hours later after wiping the Sharpie off with IPA solvent I added some red paint to the text with a toothpick and left it over night.  Careful use of a blade scraped away the dried paint, but I still had bleed between the layers, but this time the bleed was only a couple of thou wide and not very deep unlike the 0.4mm where you get quite a deep ridge between layers.  By test 3 (same part) I had scraped the plastic back to a smooth surface, maybe 0.08mm off.  The text is now much sharper and you can just make out the ripple on the vertical edges, that you can't see with the naked eye.

Last image shows you the scale of this part, I am hoping to try some small gears now I have a blob size of 0.36mm, that being my finest detail.  I might be able to go to a 0.25mm nozzle, but you are really pushing the limits on low cost printers as they struggle to hold the precision needed at these scales.   Another factor that also comes into play is nozzle pressure, as you go smaller the pressure increases and becomes more difficult to meter, that in turn will give you random blob size and difficulty in managing changes of direction where you may need to lower the pressure with a retraction or coast, change direction and reapply the pressure.  It all starts to get chaotic at higher pressures and the viscocity of PLA is very tricky below 0.2mm, so not worth the trouble in a commercial setup.  This is really the territory of resin printers, that being said I am very pleased with the results I am getting with the 0.3mm nozzle and I can print all my design up to 25% smaller without any problems.   We shall see if it goes out of tolerance quicker than the 1000hrs on a 0.4mm nozzle as wear will also be increased with both size and pressure.

Honour your mentors, and pay it forward.

Offline WeldingRod

  • Sr. Member
  • ****
  • Posts: 400
Re: Chasing finer detail in 3D printing
« Reply #1 on: August 14, 2018, 05:16:27 PM »
Just.... wow!
I had to do a high hours service on my printer a few months back too; huge improvement!

Sent from my SAMSUNG-SM-G891A using Tapatalk


Offline Joules

  • Hero Member
  • *****
  • Posts: 1271
  • Country: gb
Re: Chasing finer detail in 3D printing
« Reply #2 on: August 14, 2018, 06:31:04 PM »
Interesting to note, looking at my blown up picture of the text, all the ripple is on the left side verticals.  This indicates I have an issue with the Y axis causing distortion as it moves backwards, it doesn’t occur as the Y moves forward.  The error is however so small I probably couldn’t fix it, as we are at the limit of this belt driven mechanism.  Another indication it is probably not worth perusing a smaller nozzle on this printer.   I do have another printer I am working on with linear rails, but still belt driven that may have the improved resolution needed.  This printer is being rebuilt with a 32bit controller giving quite a few benefits in stepper control for smoother and more accurate motion, excluding the belts.

I should also point out I have the higher resolution image here which shows more clearly the bias to the left on distortion.
Honour your mentors, and pay it forward.

Offline Joules

  • Hero Member
  • *****
  • Posts: 1271
  • Country: gb
Re: Chasing finer detail in 3D printing
« Reply #3 on: August 28, 2018, 07:39:33 AM »
Chasing the detail yet again.   A few years ago I was given some IGUS bearings to try on one of the printers and give my impressions.  At the time it was hoped they would be good replacements for the XY axis, as it turned out they had too much friction and were detrimental to accuracy as the backlash became far worse.  With the recent two part hull I printed and welded, I still have quite a bit of post processing to get it smooth.  I had already noticed the very small ripple I got using the 0.3mm nozzle equipped printer above, whilst chasing that I found very slight notching in the Z bearings as they moved.  Some balls in the recirculating bearing engaging the rod had a perseptible feel, it's tiny movement, but when you move that out on the cantilevered table it becomes a noticeable surface trait in the prints.

With little to loose I swapped the Z bearings out for the IGUS, two on each bar.  The Z carriage has locking grub screws for the bearings and I was able to use these to tune the bearings for smooth motion with minimum play.  As they support the weight of the table, it overcomes the friction I had issue with before on X and Y.

Below on the edge comparison you can see the very obvious outer surface ripple on the lower hull, and this is undergoing levelling.  The upper is a noticeable improvement,  The inner surface comparison is even more dramatic as that area has had little attention to the surface.

It would seem to be the IGUS bearing does have a place, but not where you need high acceleration/decelleration.  I would say they currently out perform a bronze sleeve bearing as the IGUS is deformable, it also damps down much of the vibration I was seeing, possible due to the recirculating bearings ringing as they need clearance to operate.  Long term, we will have to wait and see how they wear over time, but if you are chasing the detail consider them for the Z axis.
Honour your mentors, and pay it forward.

Offline tom osselton

  • Hero Member
  • *****
  • Posts: 1255
  • Country: ca
Re: Chasing finer detail in 3D printing
« Reply #4 on: August 28, 2018, 04:59:13 PM »
That looks a lot better!

Offline Joules

  • Hero Member
  • *****
  • Posts: 1271
  • Country: gb
Re: Chasing finer detail in 3D printing
« Reply #5 on: August 28, 2018, 05:58:12 PM »
Thanks Tom, I just spent 2hrs welding this second hull together.  The surface still has some faint lines and a couple of faults in the same place as the first, so that indicates rounding errors in the layer height.  Another reason to move to 32bit controllers as they can process much higher resolution stepping and forward planning for g-code moves.

This second hull will require much less post processing to get it smooth.
Honour your mentors, and pay it forward.