How to modify / machine a UM2 nozzle to 0.8mm

[gadgetfreak 7]

With the help of the very skilled maker Erik (Meduza on the forums), we got his help in machining two UM2 nozzles to 0.8mm hole and with a bigger shoulder to match the bigger hole. The shoulder is 1.6mm across, so 0.4mm on each side of the hole.

See his excellent instructions on Youmagine here.

 

the vase (Propellor from here) is printed with 0.5mm layer height at 40mm/s and a single 0.8mm wall. Which calculates to 16mm3/s.

It is not perfect so some tuning is in order. Probably the layer height is a bit too height for the speed/cooling possible with standard fans.

 

We tried to find the extrusion limit and used Illuminartis standard test cylinder "UM2-noret-ExtrusionTest-3-10mm3" and tuned the flow to 200%.

We used Colorfabb Ultramarine Blue at 230C and it passed the "8mm3/s" fine which should then be equivalent to 16mm3/s with 200% flow.

At "9mm3/s" i.e. 18mm3/s it started to underextrude.

 

Daniel

 

 

 

[meduza 191]

I am really happy with how these turned out, the 1/8" (3.175mm) shank tools is perfect for reaching down into the 3.2mm barrel and drilling the hole from the inside to minimize the risk of damage to the nozzle and in combination with a nice precision lathe (courtesy of Stockholm Makerspace, www.makerspace.se) it was fairly easy to get a perfectly centered hole.

The enlarged shoulder only shortens the end barrel by approx 0.3mm,so there is still more than enough material thickness left.

[gr5 2,238]

I also had someone modify a UM2 nozzle but to .7mm instead of .8mm. Here is the only post I wrote about it. I also was able to get up to 18.2mm^3/sec before it started skipping - but man does that extruder spin fast! I made a video:

http://umforum.ultimaker.com/index.php?/topic/4127-um2-extrusion-rates-revisited/?p=38869

 

[gr5 2,238]

That was 230C by the way. Note also that there was no skipping at that speed but there was some minor, visible underextrusion. I had to slow down to 10.5mm^3/sec to have no visible gaps in the infill.

 

[peggyb 171]

i used this method:

It is the other way around and no difficult tools/skills needed.

Especially a nozzle will center on the hole that is already there, just hold it and let the drill do the work.

After mounting it back you can put sanding paper on the build platform and move the nozzle by hand.

 

[meduza 191]

It is a good way of doing a one-off nozzle without tools, sure thing, and the technique has been used in the reprap community for years, there is some problems tough.

Firstly you do not want to clamp the sensitive brass thread on a UM2 nozzle in a drill chuck, and since there is about 20mm sticking out in front of the chuck you can not be sure that it actually rotates around the nozzle center (drill chucks are not precision tools), it will probably wobble a little which leads to a risk of either not finding the center or finding the hole but getting drill wobble = either a too large hole drilled or a broken carbide drill.

This is quite a problem when doing UM2 nozzles since they are quite expensive.

There is also a second problem due to the fact that the UM2 nozzle assembly is hugely unbalanced and will introduce vibrations in itself if your drill setup is not very sturdy.

It would actually be better to do the above technique in reverse, and mounting the round part of the nozzle in the chuck, then it would run closer to true to the drills rotational axis, the problem then is that you do not see what you are doing, but since the internal hole is such a tight fit between a 1/8" drill and the 3.2mm barrel (about 0.025mm slop) you have a better chance at doing it with a good result if you do it carefully, but you would still have the problem with vibrations.

I Stand by that the method i decribe at YouMagine is the better way of doing it if you have access to a lathe or a mill.

 

[IRobertI 521]

*waves at Meduza* 'Bout time you joined the forum

 

[meduza 191]

*waves back at IRobertI* Well it was propably time for me to post some stuff here, been a long time reader tough