Going with the Flow

Discuss general topics that don't suit another sub-forum
nuggetz
Posts: 17
Joined: 18 Oct 2017, 16:15

Going with the Flow

Post by nuggetz »

I know Michael Hackney is working on a flow wizard tutorial but I'm kinda impatient and just wanted to run a few things by you all here and solicit some feedback. I'm trying to tune my nylon filament. E-steps are calibrated (Thanks Michael for the jig), and I'm working thru the Flow wizard now. With regard to starting and ending values, I don't really have a good understanding on what values are ideal here. I took a stab and went with 2.5 for the high end and 0.5 for the low and ran the test print via the wizard. Not quite sure how to interpret the print but I do see areas where there seems to be some surface "warpage" (see attachment) so I measured up to the point where that seemed to fall off and threw that into the Wizard.

Plugging in the measurement says that 0.58 is what my flow should be set to. OK, so I haven't run a test print with this setting but I can confirm that using 1 seemed a bit over-extruded on a print I ran earlier. Anyway, I jumped over to the Preload wizard and I need to populate the Viscoelasticity wizard with some flow values. It says to use a value that is in the middle of the values I have set for the material under min and max (mm^3/s) for flow. Problem is, assuming my correct flow rate should be 0.58 like I calculated via the wizard, should that be my max or should I still use a range for Min and Max? This is as far as I ever get. Every time I get to this point I'm not sure where to go so I figured I'd ask. While we're at it, I'd like to know what value I should use for Cool (mm^3/s).
Attachments
GUI.PNG
FlowWizard.PNG
User avatar
mhackney
Posts: 147
Joined: 25 Jan 2015, 10:48
Contact:

Re: Going with the Flow

Post by mhackney »

nuggetz, the range of flows are somewhat dictated by your extrusion system/hot end. For example, an E3D V6 hot end (stock) maxes out at around 10mm^3/s flow rate for filaments like PLA at 195°C. That is a good upper flow rate to use. On the low end, slow is not always better! Some filaments just won't print at a very low flow rate - but that's why we want to do this calibration to understand how your particular filament actually does behave.

A good starting point is the range 10mm^3s to .5mm^3/s if you don't know anything about your hot end. If you have a Volcano, you can increase the max flow rate to 15m^3/s.

NOTE: as per the help in the flow rate wizard, put the FASTEST flow rate at Z=0.
Screen Shot 2017-11-01 at 11.50.05 AM.png
Screen Shot 2017-11-01 at 11.50.05 AM.png (126.45 KiB) Viewed 3291 times
''

(I know you did this, just wanted to point it out for others.)

So in your case, I think you can safely increase the max to 10mm^3/s and rerun the test print.

And you have the right idea, "flow rate" is actually going to be a range of acceptable flow rates specified by a minimum and maximum. You set those in the Material tab:
Screen Shot 2017-11-01 at 11.54.26 AM.png
For most filaments and purposes, you really only care about the max flow rate because if you attempt to print faster than that, you get filament starvation since the extrusion system/hot end can't deliver enough molten filament. However, there are cases where you also need a minimum flow rate for filaments and situations where the extrusion system can't deliver or the filament just doesn't like low flows (a good example is a material that shear thins, at low flow rates its viscosity might be too great to extrude but as you speed up, sheer thining decreases its viscosity so it can flow). If you've ever printed PLA at really slow speeds (which translates to low flow rates) you might see little "dashes" of PLA rather than a nice, solid extrusion path. Well, that's what's going on!

You pick your max flow rate from the printed model layer height. Min flow rate may be obvious and if it is, use the measured value. If not, a safe value to use is .5mm^3/s until you get some experience with that filament and print conditions (slicing parameters like layer height and width for example).

In your test print, the ripples you see at the base are printer artifacts from printing fast! Backlash/slop in the mechanical parts of your printer contributes to that. So you get double information from this wizard! Now you know that if you want really clean corners and surfaces, you either need to print slow enough to keep flow rate to below about 1m^3/s OR you can start tuning acceleration (in firmware ideally) and/or work to eliminate or minimize the slop (tightening belts, tightening pulley's etc).

I am confused by your comment "I haven't run a test print with this setting but I can confirm that using 1 seemed a bit over-extruded on a print I ran earlier" because flow rate does not in itself generate over-extrusion – that is a function of extruder calibration. (on the flip side, filament starvation - i.e. under extrusion - can be caused by exceeding the max flow rate or not being above the min flow rate as I mentioned in the PLA example above.

So, to recap for your specific example:
Rerun the wizard over the range of 10mm^3/s (Z=0) to .5mm^3/s
If the printed part has a "rip" (and I mean literally, there is an obvious rip in the part like the following photo) find the point where the rip closes and use that measurement to calculate your max flow rate. Set min flow rate to .5mm^3/s or if there is an obvious rip or defect in the print at low flows, use that value.
IMG_8866.JPG
Cool! This one is really tricky and depends so much on your specific printer/extrusion system/part cooling efficiency/filament/print speeds and even the model geometry. Basically, if you attempted to print at a faster flow than this, the overhang/bridge would not be sufficiently cooled and would slump. The best way to dtermine it is through experiments – lots and lots of experiments (and I am working on that tutorial too).

cheers,
Michael
User avatar
mhackney
Posts: 147
Joined: 25 Jan 2015, 10:48
Contact:

Re: Going with the Flow

Post by mhackney »

E-step Calibration Tutorial

The description has a link to the printable model file on Thingiverse.
User avatar
mhackney
Posts: 147
Joined: 25 Jan 2015, 10:48
Contact:

HOLD THE PRESSES

Post by mhackney »

nuggetz, I was looking at your screen shot on a small screen. Now that I'm in front of my computer I see you entered .58 into the Flow Tweak field. DON'T DO THAT! Flow Tweak is a fudge factor to compensate for over or under extrusion and - as a sweeping generalization - you should not use it. That's the purpose of calibrating your E-steps - to make sure your extrusion system is delivering 1mm of filament when it's told to. Set Flow Tweak to "1" and calibrate your extruder.
nuggetz
Posts: 17
Joined: 18 Oct 2017, 16:15

Re: HOLD THE PRESSES

Post by nuggetz »

mhackney wrote:nuggetz, I was looking at your screen shot on a small screen. Now that I'm in front of my computer I see you entered .58 into the Flow Tweak field. DON'T DO THAT! Flow Tweak is a fudge factor to compensate for over or under extrusion and - as a sweeping generalization - you should not use it. That's the purpose of calibrating your E-steps - to make sure your extrusion system is delivering 1mm of filament when it's told to. Set Flow Tweak to "1" and calibrate your extruder.
OK, so flow tweak in this case isn't the extrusion multiplier like I originally thought. Thanks for clarifying that. I will rerun the flow calibration using 10 as the max and see what the print looks like. However, since I have those wrinkly artifacts at 2.5 as the max I hope that I'm still able to interpret the results on the test print. I think I understand the max setting now and based on the print I should be able to determine the new max by seeing where the "rips" stop but then do I just guess at the low as I dont think I will be able to visually see any difference on the low end.

Once I get to tuning Preload, can you tell me what I should put in the "flow rate" box? If I ascertain that a certain value to be my max, and something else is my "low", with regard to the Preload wizard, what would my normal "flow rate" be?
User avatar
mhackney
Posts: 147
Joined: 25 Jan 2015, 10:48
Contact:

Re: Going with the Flow

Post by mhackney »

When the flow rate is too high, it is VERY obvious in the printed part as shown in my photo of the rip line.

On the low end, as I said, it is not quite an exact science. If you don't see an obvious rip at the low flow then go with a realistic minimum like .5mm^3/s.

In the viscoelasticity experimental dialog, click Load Values and it will fill in all that stuff including flow rate for you.
nuggetz
Posts: 17
Joined: 18 Oct 2017, 16:15

Re: Going with the Flow

Post by nuggetz »

Ok, so I re-ran flow and this is what my print looks like. Does this look normal? Only the very top section of the print looks "good" and if I measure it the best Z turns out to by 20mm which is 2.4 mm^3/s.
Attachments
WIN_20171102_11_08_17_Pro (2).jpg
User avatar
mhackney
Posts: 147
Joined: 25 Jan 2015, 10:48
Contact:

Re: Going with the Flow

Post by mhackney »

It looks good from the perspective that you have enough range to see negative effects up to good effects - which is exactly what you want.

So that tells you anything faster than 2.4mm^3/s flow rate is not going to give good print results. On the low end, what ever the top flow rate you used (.5?) should be fine.

Next, you could home in on the range and run another test from say 3mm^3/s at the bottom to something really low like .1m^3/s at the top. This will help refine the low range - maybe.

Out of curiosity, what filament and extruder/hot end do you have?
nuggetz
Posts: 17
Joined: 18 Oct 2017, 16:15

Re: Going with the Flow

Post by nuggetz »

Running a Prusa MK2 with E3D v6. I have a hardened nozzle and a copper heat block plus a nozzle cam hanging off of it all so my hotend is a bit on the heavy side :-)

In terms of filament, the red I posted initially was Pheonix Nylon and the white in the new pic is PolyMax PLA.
Last edited by nuggetz on 02 Nov 2017, 15:56, edited 1 time in total.
User avatar
mhackney
Posts: 147
Joined: 25 Jan 2015, 10:48
Contact:

Re: Going with the Flow

Post by mhackney »

What is the white filament?
Post Reply