Jump to content
Sign in to follow this  
vincentp

Lack of extrusion in certain part... of a part

Recommended Posts

Posted · Lack of extrusion in certain part... of a part

Your filament settings are probably fine (packing density and diameter). As for flow rate, that is a holdover value. It's derived from steps per e, but it isn't steps per e, if that makes any sense. The transition to step based volumetric values allows for greater accuracy and some hardware is lagging a bit. With regards to figuring out your steps per e, this discussion went about detailing the process you can do to figure it out. You can then go to File>Preferences>Steps per E and change your value there. http://umforum.ultimaker.com/index.php?/topic/1647-extruder-calibration/

 

As far as a "safe" value of steps per e, mine's at 855.

My next suggestion would be to partially disassemble your hotend. This is just from my experience, so take it for what it's worth, but I also haven't had any plug issues. When I say partially disassemble the hotend, this is what I would suggest doing. Unscrew the four long containment screws that go into the aluminum block. This will allow the actual nozzle assembly to move away from the bowden tube. Then simply push another mm or two of the bowden tube through the coupler and reassemble it. When you reassemble the hotend, your aluminum will have a slight gap from the wood. You don't want a massive gap, but a small one is good. In a nut shell, this just allows for a bit more of a pressure fit from bowden tube to PEEK/PFTE, hopefully preventing the oozing to start a donut, as the donut turns into a plug.

 

Share this post


Link to post
Share on other sites
Posted · Lack of extrusion in certain part... of a part

very good tips thank you both of you.

firmware : great minds etc. I decided to build my own custom firmware from that website that has all the options online (webpage). Does it have the latest firmware / source code ?

that new firmware fixed the E step thing in the controller, however, I still have weird valures for the temp PID (D and C IIRC).

I did my update before reading your message, so the UM is currently printing. I'm going the latest version of cura though.

My thoughts about pressure and over extrusion, ATM : I had 868 (ish) in cura after calibrating with the old scrip of cura (v10 ?). But then I moved to the new extruder drive, found discussion about the new OD of the knurled screw and read that illuminarti was using 833.

But then my UM was still set to 868 (in the firmware). I don't know if it has some impact but I think that if the Gcode does not export the E step value in the code itself, the UM is using its system value, isn't it ?

that could be the source of my issue. I'll keep you posted and thanks again.

I'm also going to re assemble the hot end with your advice I totally forgot abou the tightening of that area. Things will change with the hot end V2 but I can at least keep experimenting with this first.

 

Share this post


Link to post
Share on other sites
Posted · Lack of extrusion in certain part... of a part

Flow in Skeinforge is an older concept, maybe... you can adjust the amount of plastic requested to be extruded by Cura by altering the packing density - which is really just increasing or decreasing the effective diameter of the filament. In general though, you should leave that at 1 for PLA. In the newer firmware, the flow rate setting is a percentage where >100 means extrude that much more plastic than the nominal amount requested in the gcode.

Yes, if the gcode file doesn't explicitly set a steps-per-e it will use whatever is saved in the EEPROM of your machine (which you can view and change with the UltiController and then re-save). That said, a difference of 860 v 830 isn't that much in percentage terms. It would be better to have it right, but unless you're printing really, really fast, it's unlikely to cause blockages etc.

Di you have minimum layer time turned on in Cura; that will cause bits like the horns to slow down... but has the opposite effect to what is wanted on very small parts, because the head just sits over them - you can counter that by setting a minimum feedrate in the expert settings to keep the head moving in cases like that.

You've probably seen this thread about the bugs in the PID values...

http://umforum.ultimaker.com/index.php?/topic/1762-fyi-ulticontroller-bugs-with-pid-and-max-acceleration/

They are now fixed, and Daid has merged them into the UM Marlin source, but I think the builder tool is using a slightly older version of the source code, from Daid's personal fork. You could build from the my latest version, mentioned on that link above, if you want. On the other hand, while you're still troubleshooting it might be best to go with the 'known to be reasonably good' firmware from the latest Cura.

 

Share this post


Link to post
Share on other sites
Posted · Lack of extrusion in certain part... of a part

Having a steps/mm value that is off by a bit is unlikely to be the problem. More likely reassembling the hot end will fix some of this, or checking your extruder. But since you are going on and on about the steps/mm, look at your gcode. If I understand correctly, he very first line overwrites the steps/mm regardless of what is in your firmware e.g.:

M92 E865.888000

- George

 

Share this post


Link to post
Share on other sites
Posted · Lack of extrusion in certain part... of a part

thanks george + illu. Re-installed the extruder, changed the minimum rate (set it at 25 mm sec instead of 10) installed cura 13, new firmware and noted for the erractic values in the controller. After tweaking and not being sure I reinstalled the default firmware via cura and it's all good. I'll just have to build my own firmware once I have the heated bed installed.

printing another robot. Found the new firmware faster (illusory ? ) maybe just the new min rate. Print looks good so far. Of course it's going to fail once I stop looking at it. Fate :mrgreen:

 

Share this post


Link to post
Share on other sites
Posted · Lack of extrusion in certain part... of a part

Yes, if your gcode has an M92 line, that will take precedence over what is in the firmware - more precisely, it will update the current setting on the machine (until the machine is reset), and that will be the value that now displays in the Ulticontroller.

Cura gets the value that it uses from the preferences pane. I think it gets added in the starting gcode? I have custom start gcode, so I'm not sure if it adds it there, or if it gets hacked in at some other point by Cura. I think it must be part of the start gcode - because it never shows up in my gcode files.

 

Share this post


Link to post
Share on other sites
Posted · Lack of extrusion in certain part... of a part

correct, I saw that too. Still, I was running at 869+ for a long time since I upgraded to the new drive + knurled screw. IT's not a big deal for a quick print, but I could see pressure accumulating after a long time (1 or 2 hours).

I got some better results after re assembling my hot end, tightening the bodwentube etc. 2 weeks off from now, I'll test more when I'm back.

thanks everyone !

 

Share this post


Link to post
Share on other sites
Posted · Lack of extrusion in certain part... of a part

Resurrecting this old thread, I was looking at the gcode that was posted for the sphere...

While there may have been a number of mechanical tweaks that might help with the print, I also found something in the settings that were really messing things up. 'Joris' was turned on, and the gcode was a mess. It's really only intended for single wall, continuous prints. The thicker wall, with cutouts, was really confusing it, resulting in layers getting printed that don't touch, and just general messiness.

I resliced it with the same settings, but no Joris - and as you can see it turned out much cleaner (click on it to zoom in):

With and Without

So that may not be all of your problems, but it's certainly some of what was going wrong :smile:

 

Share this post


Link to post
Share on other sites
Posted · Lack of extrusion in certain part... of a part

whoa, thanks a lot. Indeed, I thought that since it was a "vase-like" (understand thin object) I would base my first print settings on cura "thin walled object" settings in which joris was set.

very good tip, I'll try that on monday when I come back to the lab.

What did you use to simulate and get that visual result ? I'm impressed to see the EXACT aspect I was getting on certain layers (the ones with many holes and extrusion stops).

 

Share this post


Link to post
Share on other sites
Posted · Lack of extrusion in certain part... of a part

It's Repetier Host. It does a really nice job of visualizing gcode - it actually calculates the thickness of the bead of plastic, based on the volume of extrusion for each segment, rather than just showing the path of the head with no sense of the amount of plastic being deposited.

Joris is really intended for where you have a solid stl - say a cyclinder. And you print it with 0% infill, and no top, and wall thickness = nozzle width, so that you have just a single width perimeter that spirals all the way up to the top.

 

Share this post


Link to post
Share on other sites
Posted · Lack of extrusion in certain part... of a part

Greetings,

finally a moment to give it a go. First results with the new hot end were a catastrophy, lack of extrusion.

I had to crank up the temp to 250° (!) with like no reason as... the alu block, heater, thermoccouple etc, are the same.

I started investigating about the temp itself, then though about the mod I did last time, in order to prepare for the hotbed : the additional 4.7k resistor on the controller.

bingo. I removed it, I have my "normal" temp at 220 or 225°C.

This is a side (and ANOTHER) issue but I started to think about the temp sensor and the readings. Looking at the controller, there's simply NO opamp on the analog inputs to read the temp, right ? So no impedance adaptation. Not surprised then that the results vary depending on the load on the arduino MUX inputs.

I still have to confirm this but it could explain why people have big temp offsets when trying to print ABS (therefore, after installing the heating bed temp sensor)

does it make sense ?

trying the new firmware + cura 13 with the half sphere tonight, I'll see the result on thurday.

 

Share this post


Link to post
Share on other sites
Posted · Lack of extrusion in certain part... of a part

I strongly suspect the arduino has op amps on all inputs but I didn't look it up.

I doubt that 4.7K resistor made any difference either - you probably had a clog and fixed it at the higher temp and now the lower temp is working because the clog is gone.

Share this post


Link to post
Share on other sites
Posted · Lack of extrusion in certain part... of a part

nope. After the filament started to flow a bit, and a good 30cm has passed thru, I tuned the temp down with the ulticontroller and extrusion stopped working properly at 245°. I tried 3 times, filament was barely extruding properly at 250°. I disconnected the resistor and extrusion started to work again at 215-225° like before.

there's no op amps on the analog inputs of the arduino. Analog inputs go straight to the internal mux of the ADC, which has the same defect as ALL multiplexers, internal resistance and capacitance, which equate a low pass and increases switching time (or you get wrong readings). Floating inputs have less impact on the muxing as they are hiZ, but the 4.7k could definitely change the reading.

I think there's a serious design flaw with this particular point. I've been designing sensor front ends and MIDI converters for the past 15 years and at least a voltage follower is required if you care about what you're reading & digitizing.

 

Share this post


Link to post
Share on other sites
Posted · Lack of extrusion in certain part... of a part

update time. now printing correctly. First off temperature : sticking to 220° which is becoming my "number" for PLA I guess. At 215°, the bonding is less quality and it's possible to tear the part off (the half-sphere is thin).

I solutioned a lot of things in the past 2 days. I reduced my wall to 0.4 mm instead of 0.8 which allows a real filling in the thickness of the half-sphere wall.

I also printed with no support and I changed the design of the part. Having a continuous flow exclusively on the part gives really good results. Now testing with high quality (0.1 mm layers), so far so good.

I'll move to heated bed and ABS right after that.

 

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this  

×
×
  • Create New...

Important Information

Welcome to the Ultimaker Community of 3D printing experts. Visit the following links to read more about our Terms of Use or our Privacy Policy. Thank you!