What about calibrating filament by the weight?

Lately, PLA filament ordered from a couple of different sources features quite an oval section (let's say, 3.01 x 2.78 mm), and of course this brings back some part of guesswork when dialing in the diameter in the slicer software.

Two main options:
- calculate an approximation of the equivalent round filament section by using the largest x smallest measured diameters
- calculate volume by weighing a known length of filament (let's say, 100 or 500 mm) and using an estimation of filament density

The second option needs the purchase of a relatively accurate balance.

Which method, in your view, would be less error-prone?



Of course there is a third way, to ask the sellers to provide accurate measurements of the stuff they are offering, but I do not see much of this emerging in a still crude and undiscriminating market.

Easiest to measure min and max diameters and use the area of an ellipse to work out an equivalent circle diameter.

Having said that oval filament is useless because a) you need to make the extruder bore oversized and that leads to excessive back-flow, which increases the force needed to extrude, and hence reduces the max feed rate and b) the feed rate varies depending on how the oval aligned with the hobbed bolt, giving reduced quality prints.

---

[www.hydraraptor.blogspot.com]

Reject it back to your supplier, or the problem will never get better, it's been going on for years and most of it is caused by lack of care in the production process (at the source filament manufacturer) - the warm filament gets pulled and stretched and (sometimes very tightly) wound on to reels while it's still warm enough to deform into thinner sections or oval shapes. This also leads to cracking and breakage of the filament coils- all of which is a real pain.

I would not try to use oval filament again even if it was given to me for free.

And if you are unlucky enough to get oval ABS it's almost impossible to use, your hobbed bolt will get constantly clogged, and prints will be very poor.

The only real solution is to use good dimensionally controlled filament, you may pay a little more but it saves time, mess and wastage along with actually being nice and reliable to use. with oval or varying dimensions of filament you will just constantly get frustrated.

---

[richrap.blogspot.com]

The situation with the providers seems excessively fluid. I ordered from source A some PLA some months ago, it was just gorgeous in color and a genuine pleasure to print. I ordered from the same provider this year, quality, color and even the resistance of the resulting printed parts took a serious plunge. I didn't send the stuff back, I took the time to test it, but source A has lost me as a customer.

I know at least this PLA comes from Esun and there are people from Esun reading the groups so for them: do improve your process and provide filament with a round section, or your products wll become better known for poor quality and unsuitability to the task. According to the label, this batch dates back to late october 2011.

-----------------

Very seldom are sources indicating the filament's properties and measurements. You do not know what they really sent you until you open the postal package. This is a domain where some sort of labelization process might help. Standards, minimal standards, or the law of the jungle...
Which sources do you guys consider as serious and quality providers?


----------------
Meanwhile: to finish the roll, I will be upgrading the extruder bolt to one CNC-hobbed with an abrasive disc (provides increased grip), and start calculating the equivalent filament section.

Given the exacting standards we need to get good prints, I don't hold out much hope that something like filabot will produce anything close to usable.

Buback Wrote:
-------------------------------------------------------
> Given the exacting standards we need to get good
> prints, I don't hold out much hope that something
> like filabot will produce anything close to
> usable.


And here we are mass-sourcing PLA from China like almost everything else while this was supposed to promote local production and recycling (which, you are right to point out, remain to be solved).

Since -in spite of the high expectations- we are at risk of seeing oval filament for quite a while, here enclosed is an equivalent diameter calculator for oval filament...

... time for some selective sintering exploration?

I've had nothing but trouble from suppliers in china. In fact, I think I've had every single problem there is over the last year; Brittle filament, out of round, non-continuous (pieces taped together mid-spool), large blobs on the filament, colors changing shades partway through the spools, too small, too big, full of air bubbles, full of moisture, getting the wrong type of plastic...

I've had good experiences with Ultimachine and Makerbot filament in the US.

---

www.Fablicator.com

Last year Makerbot sent me some red ABS that was a nightmare to print - still don't know why, probably too floppy for the kind of bolt I am using.
Ultimachine always sent nice samples - Looking inside Urope because of the pre-eminence of custom duties.

Right now I am looking at reprapsource's "extruding in germany" rolls. I have not yet ordered from them. Might be time to try.

reprapsource filament is always round and very consistent diameter in my experience, 4 day delivery to the UK and no duty in Europe.

---

[www.hydraraptor.blogspot.com]

For PLA then Faberdashery has excellent dimensional stability across all colours, it's the best PLA I have used to date.

I have had good results with Orbi-tech and GRRF generally very happy, sometimes the Orbi-tech filament has had a rippled surface and a +/- of more than 0.3mm diameter deviation.

I don't use much ABS, but Reprapsource has always supplied very good rolls to me, and they have has some very great colours in the past.

---

[richrap.blogspot.com]

Thx for the information!
Yes I definitely love the colors of farbshadery, but they sell by the meter (or 100m), while I am buying by the spool
I ordered one from reprapsource. Next to test will be GRRF (when they have the color I want...)

Meanwhile, using the equivalent diameter calculation somewhat improved the printability of the esun pla. The bolt tends to catch the filament perpendicular to its longer diameter.
But it is like it lacked something: pieces tend to be brittle and delaminate easily. Solid fill over sparse layers tends to fail with gaping holes. Upped the temp from 184 to 190, I'm seeing less delamination, but still holes in the top solid layers. I'd hate to still decrease printing speed...