[Solved: Bad knurled sleeve] Under extrusion help

[rhino_aus 2]

EDIT: I fixed the problem! I bought a replacement for my knurled sleeve (the Aurarum Mk7 part) then I recalibrated the e-steps for the larger gear and now there is 0 slippage of the filament extruding with no load or at high speed! Excellent!

Okay so today I have been trying to get my printer dialed in and I am coming across a problem with the extrusion that my searches cannot resolve. My printer is an Ultimaker 2 that I have replaced the hot end with an Olsson block with 30 watt heater, and have replaced the extruder with the one designed by |Robert|.

Here is an example picture of the second layer. I level my bed with a dial indicator and deliberately did it too low (as you can clearly see) with the first layer to ensure the spacing between print and nozzle was right for the second layer.

It's pretty clearly under extruding. My filament is fine. My slicer settings are fine. Printing an object in vase mode I get wall thicknesses between 0.37 and 0.40mm. Taking advice I decided to calibrate my e-steps with the normal 100mm method. This is where things get confusing for me:

- If I extrude with the bowden tube disconnected (ie not through the hot end = no load) the extrusion is perfect.

- If I extrude with the bowden attached, the extrusion distance is function of the extrusion speed.

Extruding at 90mm/min extrudes 68.3% of the expected distance, 60mm/min give 86%, and 30mm/min gives 90.4%.

As far as I can tell, the filament is not slipping (EDIT: FORESHADOWING!) and I would not expect it to with this extruder and hot end. I am watching it up close and there is no sign of slipping or grinding. This tells me that the motor is not turning as much as it is is being commanded to under load, with less movement as load increases

If this is the expected result, what speed do I use to calibrate my e-steps? If this is not the expected result, what do I do to make the stepper motor move as commanded when extruding? It doesn't feel right to use any of these values since then the extrusion under no load will be too much.

Edited February 13, 2017 by Guest

[owen 19]

I think this is typical characteristics of the the um2 feeder and umo even though there is no noticeable slippage or skipped steps. The upgrade to um2+ greatly alleviates this problem. Other things to do is make sure you have a clean nozzle, have an easy unrestricted entry into the extruder, have the tension adjusted right on the extruder, check the ptfe coupler in your hot end and then always print at a moderate speed like 25 to 35mm/sec.

[rhino_aus 2]

The upgrade to um2+ greatly alleviates this problem.

 

I have already upgraded my printer with the |Robert| feeder, an Olsson block hot end with a 30 watt cartridge, and 3dsolex high temp PTFE coupler. Is the included stepper motor different to the UM2 motor?

My setup should be able to print at much higher speeds. Even at 35mm/s effective extrusion rate (20mm/min extrusion -> 35mm/s @ 0.4 width, 0.15 layer height) there is under extrusion in the 100mm command.

Edited February 12, 2017 by Guest

[owen 19]

 

The upgrade to um2+ greatly alleviates this problem.

 

I have already upgraded my printer with the |Robert| feeder, an Olsson block hot end with a 30 watt cartridge, and 3dsolex high temp PTFE coupler. Is the included stepper motor different to the UM2 motor?

My setup should be able to print at much higher speeds. Even at 35mm/s effective extrusion rate (20mm/min extrusion -> 35mm/s @ 0.4 width, 0.15 layer height) there is under extrusion in the 100mm command.

 

The UM2+ feeder is geared.

[rhino_aus 2]

Okay so I did some more testing and it seems that the motor is not the issue. Commanding an extrusion with the filament feeding through the hot end or not causes the stepper motor to rotate the exact same distance, as such it is not a motor problem, but the filament slipping on the knurled gear.

I shall investigate solutions to this

[tomnagel 126]

You mention that extrusion is a function of extrusion speed. While that is true, it is more accurate to say that the transported amount of filament is dependent on the force that the feeder has to overcome when pushing the filament forward.

The force is dependent on a lot of things, such as speed, hot end temperature, nozzle size, material etcetera.

The counterforce at the exit side of the feeder causes sort of micro slip. You can actually see that the teeth marks in the filament get a different distance.

The working point where you calibrate is up to you. I think it should depend on what you want to do with your printer. If you plan on pushing the limits on printing with high flows, maybe you should calibrate at a high flow.

[Torgeir 264]

Hi rhino_aus,

I have to say I'd been looking at this interesting post a few times.

And you're absolutely right in your assumptions, -but as owen said; this is kind of normal for an Ultimaker.  Well, all kind of bowden printers suffer more or less of this problem, this is one thing I've tried to address many times in this forum.

When we increase the flow we might increase the feeder speed or/and the nozzle temperature. So for the feeder, more speed and higher pressure on the bowden tube, right? Lets analyze a little here; "the pressure from the filament running through the bowden tube is trying to push the tube upward". So when the filament is pressed/forced to follow the upper part of the inside of the bowden tube, this will act as an old drum brake and create a "drag force" due to the friction. The higher the feeder speed is, the higher the bowden pressure will be as well as the drag forces. This can be increased until the feeder wheel start grinding into the filament -or one of the bowden lock popp out.

The rest of the story we all remember..

However, there is one worse thing to note in here; this is that the track line, made by the knurled or the hubbed wheel on the filament is most of the time on the upper side of the loop inside the bowden tube. This factor actually increase the friction factor and make the bowden pressure (the filament feeding) varying all over the building platform as the extruder changes position. (When this occur the (EDIT) filament turn back and forth inside the bowden tube making the friction factor changes).

Soo...

There is ways to avoid this and one of the best I've seen so far is:

https://ultimaker.com/en/community/35608-zero-gravity-direct-drive-extruder

This is an extraordinary way of doing smart things and you have no pressure difference here due to movement around the building platform.

I havent go that far at the moment, but I did changed the trackline to go on the underside of the filament inside the bowden tube (mostly) during printing.

Hope this help.

Good luck.

Thanks.

Torgeir.

Edited February 13, 2017 by Guest

[rhino_aus 2]

Okay so I fixed the problem! I bought a replacement for my knurled sleeve (the Aurarum Mk7 part) then I recalibrated the e-steps for the larger gear and now there is 0 slippage of the filament extruding with no load or at high speed! Excellent!

I believe my stock gear was somehow either not knurled properly or had become worn down over time such that the grip on the filament was too low to be effective.

I can now easily extrude at 60mm/min which is the equivalent of printing 0.15mm layers with a 0.4mm nozzle at 105mm/s at 230C. Thats about 6.3mm^3/s. Fairly happy with that. I can reach 10mm^3/s extrusion but there is ticking of the extruder motor, although the actual length that is extruded isn't terribly affected. I will retry the extrusion test vase at some point.

Edited February 13, 2017 by Guest