commercial extruder vs conventional

Are there any high quality commercial extruders out there for the repraps. And if there are, is their print quality really any better than the standard Wades or Greggs extruder most of us make ourselves?

Main reason i ask is i am fighting very light banding and i just cant seem to nail the problem down. I am thinking it is my Wades extruder causing the problem but i have no way to verify it without switching out to another extruder.

Did you optimize the z-steps/mm in relation to the mm/rotation of your z-axis threaded rod?

When I couldn't solve banding, this helped me get rid of it.

Not sure if changing extruder might work. You can try just printin another extruder design before buying one.

---

http://www.marinusdebeer.nl/

show us a picture that clarifies the type of banding you have. if it is sinusoidal waves bumping throughout the print then it is z height mismatch from the z step movement, if it is only where there is heavy infill then it is flow control and speed of extruder being exceeded, if the banding is up/down and bands left to right and only where the infill intersects then the flow is too high.


try reducing flow rate, and see if it improves, if it does not improve that much, then round z steps in firmware to 1 place past decimal 0000.0 whatever number is in firmware, the reason is the inaccuracy of float of arduino, [en.wikipedia.org], I should forward a patch to the marlin team to fix this issue with z height banding.

this will resolve most of banding issues
#define DEFAULT_AXIS_STEPS_PER_UNIT {78.7402,78.7402,int ((200.0*8/3)*10)/10,760*1.1} // default steps per unit for Ultimaker

this is now issue #1119 on github for programmer of marlin firmware

change
#define DEFAULT_AXIS_STEPS_PER_UNIT {78.7402,78.7402,200.0*8/3,760*1.1}
to
#define DEFAULT_AXIS_STEPS_PER_UNIT {78.7402,78.7402,int ((200.0*8/3)*10)/10,760*1.1} // default steps per unit for Ultimaker

Edited 4 time(s). Last edit at 10/24/2014 02:22PM by jamesdanielv.

Ok. Im a little lost with all of the math equation for the z steps. Wouldnt it be easier to just write in the actual steps per mm instead of using a formula? Or is that what you are saying is the problem? For instance, my Z steps in Marlin are (10000/1.25) . I have 5mm x .8 threaded rods. My actual steps are 8000 but when i changed it to read (8000.) it went a little whacky so i left the /1.25 in place set it to 10000. So can you give me an example of what my Z steps should read in Marlin?

Here is what i have right now
#define DEFAULT_AXIS_STEPS_PER_UNIT {105.3411 , 104.7836 , 10000./1.25,943.444}

I attached a photo of the banding i am talking about. This part is .75" square in X and Y (19.05mm) printed with .1mm layer height and no infill. I get the same results printing at .2 or .25 layers. To me it looks like extruder problems.

Edited 1 time(s). Last edit at 10/24/2014 04:27PM by GITRDUN.

Yeah, setting the z-steps to an integer should avoid any kind of floating point error. Anyway, it looks like your banding isn't consistent across layers, so inconsistent extrusion seems a likely cause.

The most important factor that influence print quality, when it comes to the extruder, is the hobbed bolt. If the hobbing isn't consistent, or the diameter varies, it will affect the extrusion rate. So unless you have tweaked every other nitty gritty detail on the printer, and the hobbed bolt is of decent quality, going commercial on the extruder, probably won't make a huge difference.

But before you start tweaking too much on the extruder, you should check if you are using PID or BangBang for controlling the bed temperature?

I had a similar issue on my printer, because I had the bed temperature control set to BangBang, and every time the heat turned on of off, the bed moved up and down as much as 0.1mm.

Quote
Ralf
The most important factor that influence print quality, when it comes to the extruder, is the hobbed bolt. If the hobbing isn't consistent, or the diameter varies, it will affect the extrusion rate. So unless you have tweaked every other nitty gritty detail on the printer, and the hobbed bolt is of decent quality, going commercial on the extruder, probably won't make a huge difference.

But before you start tweaking too much on the extruder, you should check if you are using PID or BangBang for controlling the bed temperature?

I had a similar issue on my printer, because I had the bed temperature control set to BangBang, and every time the heat turned on of off, the bed moved up and down as much as 0.1mm.

How do you tell?

To diagnose it, I got a hint from Nophead here on the forum, and when I put a dial indicator on, and it was obvious... Without a dial indicator, I guess adjusting the print head to a height where you have slight drag with a piece of paper, and the heat bed up to temperature, and then see if the drag is constant. If it changes every 10 seconds or so, depending on the setup, heat loss PSU etc.. then you have a problem.

The PID setting for the bed is in either configuration.h or configuration_adv.h, if it's Marlin/Sprinter.