The Flex3Drive

Talk about your experience with Ultimaker printers
User avatar
gr5
Reactions:
Posts: 28
Joined: October 20th, 2016, 11:14 am
3D Printer(s): umo um2 um3

Re: The Flex3Drive

Post by gr5 »

Well my favorite test is very simple. Start with a 30mm cube. Slice in (old) cura with no walls, 100% diagonal infill, and speed=100mm/sec to make math easier. When you print it, don't start testing until about 3rd or 4th layer when it is at full speed. Now pick a temperature and then start messing with the feedrate %. Since you sliced at 100mm/sec then feedrate of 50% gives you 50mm/sec. Increase and decrease until you are at the limits where you start to see gaps in the infill. Use a loupe or reading glasses and look carefully while it's printing. Once you determine the max speed, change the temperature by at least 10C and repeat the test after the temperature settles. In 10 minutes and a single cube you can get results for 4 different temperatures. Take notes as you go. Also you can further experiment with flow rate. I found you can safely increase flow to 110% or 120% but if you have to increase to 140% to close the gaps then you should slow down as you are likely to grind up the filament at these pressures (140% and perfect extrusion was only possible on a bondtech for me). I've done this test with many filaments, nozzles, temperatures, and feeders. I'll try to post a picture (I'm new at this forum).
User avatar
gr5
Reactions:
Posts: 28
Joined: October 20th, 2016, 11:14 am
3D Printer(s): umo um2 um3

Re: The Flex3Drive

Post by gr5 »

Here's an old example. I have dozens of these cubes. This one was for the old black feeder that skips. But for the new feeder it's trickier - you have to look very carefully at gaps between the lines as it doesn't do sudden major skips but instead does underextrusion (filament slipping 10% or so in feeder)

I have newer photos if you want to see. I also have a bunch of these "cubes" sitting around.
testing underextrusion
testing underextrusion
User avatar
Izzy
Reactions:
Posts: 604
Joined: February 12th, 2016, 2:29 am
Location: England
3D Printer(s): Ultimaker 2

Re: The Flex3Drive

Post by Izzy »

Yes please, do you also have the Gcode file available? That way we are all running tests on the same Gcode.
User avatar
gr5
Reactions:
Posts: 28
Joined: October 20th, 2016, 11:14 am
3D Printer(s): umo um2 um3

Re: The Flex3Drive

Post by gr5 »

I'll get you tonight for um2 or um2+. But really it's trivial. 30mm cube. Sliced as above. The point is you look at the gaps between traces with magnifying glass.

Layer height matters of course - I mean you have to multiply layer height times speed times nozzle width to determine the volume that your printer can do. So I do 0.2mm layers for .4mm nozzle and .4mm layer height for 1mm nozzle and .6mm layer height for 2mm nozzle.

I tried attaching gcode but this forum does not allow.

Here's a dropbox link:
https://www.dropbox.com/s/4wn8eaubyyj2g ... gcode?dl=0
User avatar
LePaul
Reactions:
Posts: 3966
Joined: February 7th, 2016, 10:26 pm
Location: Bangor, Maine USA
3D Printer(s): 24 - Yes I have a problem!
Contact:

Re: The Flex3Drive

Post by LePaul »

I should add an extension (.gcode?) and have it show as text...thanks for mentioning that.
MTVDNA
Reactions:
Posts: 138
Joined: August 7th, 2016, 10:29 pm
Location: Netherlands
3D Printer(s): Ultimaker 2

Re: The Flex3Drive

Post by MTVDNA »

By feedrate I assume you mean speed?

I agree this is a good way to test what speed you can reach, but what I like about the cylinders is that you get an easily reproducible result that you can compare to others.

In my experience so far, you do indeed need to increase the flow rate when you print faster to make sure the lines connect, but this does sometimes cause the edges of the solid layers' infill to be a bit overextruded.
User avatar
LePaul
Reactions:
Posts: 3966
Joined: February 7th, 2016, 10:26 pm
Location: Bangor, Maine USA
3D Printer(s): 24 - Yes I have a problem!
Contact:

Re: The Flex3Drive

Post by LePaul »

Side note...added ability to attach .gcode and .stl
User avatar
ivan.akapulko
Reactions:
Posts: 389
Joined: February 11th, 2016, 11:27 am
Location: Ust-Kamenogorsk, Kazakhstan
3D Printer(s): Ultimaker 2
Contact:

Re: The Flex3Drive

Post by ivan.akapulko »

LePaul wrote:Side note...added ability to attach .gcode and .stl
But still forum engine cannot upload large stl and gcode files. Can you remove this limitations?
User avatar
LePaul
Reactions:
Posts: 3966
Joined: February 7th, 2016, 10:26 pm
Location: Bangor, Maine USA
3D Printer(s): 24 - Yes I have a problem!
Contact:

Re: The Flex3Drive

Post by LePaul »

How large do we need?

I set it at 25mb
User avatar
ivan.akapulko
Reactions:
Posts: 389
Joined: February 11th, 2016, 11:27 am
Location: Ust-Kamenogorsk, Kazakhstan
3D Printer(s): Ultimaker 2
Contact:

Re: The Flex3Drive

Post by ivan.akapulko »

Just about couple minutes before was trying to upload 14 mb gcode file, and forum engine told that is too big. Then i zipped it, and 4.35 mb upload smootly. 25 is enough, i think. In any case, for extraordinary situations we can use some cloud like Dropbox.
User avatar
ivan.akapulko
Reactions:
Posts: 389
Joined: February 11th, 2016, 11:27 am
Location: Ust-Kamenogorsk, Kazakhstan
3D Printer(s): Ultimaker 2
Contact:

Re: The Flex3Drive

Post by ivan.akapulko »

My bad luck. Or just slow connection, dont know )))
Attachments
Безымянный.png
User avatar
LePaul
Reactions:
Posts: 3966
Joined: February 7th, 2016, 10:26 pm
Location: Bangor, Maine USA
3D Printer(s): 24 - Yes I have a problem!
Contact:

Re: The Flex3Drive

Post by LePaul »

ivan.akapulko wrote:Just about couple minutes before was trying to upload 14 mb gcode file, and forum engine told that is too big. Then i zipped it, and 4.35 mb upload smootly. 25 is enough, i think. In any case, for extraordinary situations we can use some cloud like Dropbox.
Oh....you'll laugh. I wasn't sure on .gcode size...so I made it 12mb! stl is 25!

But you got it zipped up. I can up the .gcode size to match the stl

Hijack over!
User avatar
gr5
Reactions:
Posts: 28
Joined: October 20th, 2016, 11:14 am
3D Printer(s): umo um2 um3

Re: The Flex3Drive

Post by gr5 »

But you can have 50% underextrusion and the cylinders look great still. So it definitely can't detect 10% underextrusion.
Plus you can test many different temperatures in less time than it takes to test one temperature with cylinder.

If you have a UM2 with the original black feeder than the cylinders are excellent. But that's it.
User avatar
gr5
Reactions:
Posts: 28
Joined: October 20th, 2016, 11:14 am
3D Printer(s): umo um2 um3

Re: The Flex3Drive

Post by gr5 »

Here are a few test "cubes". On the top row 2nd from left is .4mm nozzle. Also here are .8 1mm 1.5mm and 2mm nozzles I believe. The center most square you can see about 20% underextrusion on the upper right corner/half. But the layer below is pretty good. The bottom most square has pretty severe underextrusion but still good enough to look great on a test cylinder - maybe around 30% underextrusion (30% gap, 70% fill).

When actually testing it takes seconds to change the speed and examine the print. Within a few minutes you know EXACTLY where underextrusion starts, and how much flow rate to add to fix it. Then you can test at a second and third temperature. So you get several results in the same time it takes to make one cylinder. You just have to take good notes.
Attachments
extrusion test cubes
extrusion test cubes
korneel
Reactions:
Posts: 13
Joined: February 14th, 2016, 10:44 am
3D Printer(s): UM2, Um2+extended

Re: The Flex3Drive

Post by korneel »

Nevermind :)
Last edited by korneel on December 12th, 2016, 11:00 am, edited 2 times in total.
Post Reply

Return to “General Discussion”