Extrusion multiplier and overextrusion?

[Jeff Tapia]

Guys & Gals,

Do you feel that our Bob's RP9's generally overextrude and perhaps require extrusion <1.0 extrusion multiplier all the time?

I ask because I picked up a second printer this recently (Prusa i3 Mk2).

I printed a Maker's Muse tolerance test on the Prusa and, same as Angus, I was able to get 0.2 tolerance to free up on the Prusa without much trouble but not the 0.15.

https://www.youtube.com/watch?v=TYuLVN3YHw8

I thought it would be interesting to try with BobsCNC RP9 but found that I couldn't free up ANY (not even the 0.5) despite using the same settings, same 1.0 extrusion multiplier, same filament, etc.

After studying the print a bit I realize that Bob's was extruding much more material than the Prusa.

I then went back and looked at some past prints and also feel my past prints may all have been overextruded.

I printed some darts (tall, thin) with a 1.0 extrusion multiplier but the head kept knocking them off the bed mid-print, so...

I printed the 0.9 extrusion multiplier on the Bobs and they came out fine.

So I'm thinking I need to either always use an extrusion multiplier (or perhaps adjust firmware but that might be trouble).

Anyone else?

[Dan Keller]

I know that I've often heard that PLA should have an extrusion multiplier of 0.9 off the bat. Due to its hardness relative to other plastics, the burrs of the hobbed gear of the cold end don't bite as deep, effectively meaning the radius of the pushing element is larger (hence more is pushed at one go). I think I've even seen an older video either from Angus or Thomas regarding this.

It's entirely possible that if you are using Prusa's edition of Slic3r or even using a Prusa profile in another slicer that these things are already taken into consideration and the end result being that things are better "tuned" when printing on the Prusa.

I've definitely found that owning a printer such as the RP9 requires a bit more manual tuning given there are no profiles ready for the machine out there. That said, I now get pretty consistent results after much trial and error.

I really hope to get a Prusa i3 MK2 myself - it looks like a fantastic machine and would be a great next step into 3D printing. I love my RP9 and it has forced me to learn so much through using it (including how to maintain just about every part) that I think putting together a Prusa would now be a breeze.

Did you buy the kit or an assembled unit?

[Jeff Tapia]

Hi Dan,

I got the kit as a b-day present.

Ordered end of February and received mid-May.

Shipping cost hurts - like $75 for the printer and $28 for any replacement parts (after several weeks of printing my PINDA (prox sensor) wire failed and I ordered a replacement).

Building took a weekend, similar to RP9.

Sincerely, Jeff Tapia

--
You received this message because you are subscribed to the Google Groups "BobsCNC" group.
To unsubscribe from this group and stop receiving emails from it, send an email to bobscnc+u...@googlegroups.com.
To post to this group, send email to bob...@googlegroups.com.
Visit this group at https://groups.google.com/group/bobscnc.
For more options, visit https://groups.google.com/d/optout.

[Michael Platt]

I used a single walled cube to calibrate my extruder. I have a 0.5mm nozzle and I'm printing a 0.57mm layer width. After measuring the thickness of the wall I concluded that a 0.97 extrusion multiplier gave me exactly 0.57mm wall thickness for polymaker pla. For petg it's like 0.87 or something. This article might help: http://forum.makergear.com/viewtopic.php?f=3&t=1964
If you're getting something wildly different from 1.0 then you should go back and calibrate Esteps/mm before changing the extrusion multiplier from 1.0. Here's an article on that: http://zennmaster.com/makingstuff/reprap-101-calibrating-your-extruder-part-1-e-steps
I have Angus's calibration file, I just haven't gotten around to printing it yet. When I do I'll post my results here.

[Jeff Tapia]

I think Bob used 0.4 nozzle so it is interesting that you moved to 0.5 nozzle.

I will try some of the calibration recommendations.

I also think Dan was on to something with respect to the gear not biting as deeply. I have probably exacerbated that quite a bit because I moved to regular rubber bands on the extruder simply because they seemed to work and were less tense. There is perhaps some irony in that putting a "looser" band might cause MORE extrusion.

Thx guys.

[Michael Platt]

Ya I bought an E3D Chimera and installed it on my Printer. It came with a 0.5 mm nozzle so that's what I've been using. I really haven't noticed that much difference, It can print 0.1mm layer heights just fine. Unfortunately, I failed at getting the second nozzle to extrude so I have a Chimera that is just using 1 nozzle :|  

[Jeff Tapia]

Oh yeah, I now recall you mentioning the Chimera.

I just printed a 0.8 perimeter calibration block with my typical black PLA

With two black compression bands (meaning tight) on extruder: averaged about 0.82 thickness

With a single, rubber band (loose, gentle): averaged about 0.81 thickness

So I'm thinking this is insignificant (my ability to use a caliper is easily responsible for more than 0.02 error.

[Dan Keller]

So this is an interesting experience I had...

I had never printed a Benchy benchmark (https://www.thingiverse.com/thing:763622). I thought I would try it out to check some cheap filament I got.

The result was more than acceptable:

One thing I noticed, however, was some bulging/blobbing on the hull. Also, you can see shadows/echos around some of the fine details which I figured I could minimize by lowering my print speed from 50mm/s to around 30mm/s. I thought I would use the opportunity to check what would happen if I reduced my flow to 90% as well. So, I took the exact gcode, used my graphic controller to set flow to 90% and speed to about 70% (original gcode was set at 40mm/s and I had sped it up to 130% for my faster print) and this is what I got:

A few things I noticed: blobbing on the sides of the Benchy had all but disappeared. Fine details came out better with fewer shadows. And, this was probably the most surprising: surface finish completely changed. The original had something of a matte finish in most places, while the second has a glossy appearance. Most of my past prints, regardless of speed, have tended to have a matte finish which I have to wonder if it is due to the slight over-extrusion.

In any case, I wonder if I should try Angus' tolerance test. I'd be curious what I get with a 0.9 flow rate. All I know is from now on, my PLA is printing at 0.9. (Though I'll keep my flexible at 1.0.)

[Jeff Tapia]

Dan, you did a lot of stringing clean up before the photo's right??!

I'm going to print a couple this weekend to compare.

[Dan Keller]

I did absolutely no post-processing. There was no stringing at all - this was straight off the build plate. :-)

I use Cura and I play with settings all the time. My retraction is 6mm, travel speed is around 120mm/s and there are a number of other settings I've played with. One of the key settings is that for certain travels, the slicer tries to find a path for the extruder that goes over a printed surface. Any ooze will just get wiped off on the printed surface which will likely be covered in the next layer anyway. This helps a LOT with flexible filaments, by the way.

I was quite pleased with my results on this - I was totally expecting some stringing despite my settings.

[Jeff Tapia]

Wow, I'm going to do this and I'm learn something - I'm sure