You are here:

Plastics/Hobby Extruder

Advertisement


Question
Extruder
Extruder  
QUESTION: I am helping with the development of a simple extruder to make filament for 3D printers at home-

It uses an auger to push pellets through a pipe into a short melt chamber, and out through a nozzle which is a brass hex plug with a 1.6mm hole drilled through it.  It is slow, the auger turns about 1 rpm, and it extrudes about 6"/minute but with the right temperature it can extrude 1.75mm filament +/- .03mm which is close enough tolerance for 3D printing.

I would like to improve the rate of extrusion, but without running the motor faster.  Higher speeds would require gearing, and a major goal of this kit is simplicity and low cost.  It only costs $150-$200 to put together and doesn't require any machine tools beyond a saw, a drill, and a dremel (though a drill press helps).

How do you balance the nozzle opening, length, shape, temperature etc to achieve a target filament diameter at the highest rate? Are there any formulas or rules of thumb that would help this process be something other than simple hit and miss?

Given only off the shelf parts from the hardware store (or McMaster) and typical garage tools, do you have any ideas about how to improve its performance?

ANSWER: Ian

Output of the extruder is determined by the melting or plasticizing ability of the machine.  That is the simple answer.  
Your concern is how to balance the nozzle opening, length, shape, temperature etc to achieve a target filament diameter.
Letís address your concerns one at a time.
By nozzle opening I assume you mean the opening at the end of your extruder that the plastic flows from, it is commonly called a die. The opening of the die has little affect on the output of the machine unless you go to extremes, too small creates too much back pressure and your motor may stall, too large and you get no back pressure and no melted plastic.  A typical extruder gets 80% of the energy needed to melt the plastic from the mechanical shear in the barrel, in you device, I suspect that 100% is coming from resistance heaters.
Die size (nozzle) can affect size as most plastics prefer some drawdown (a larger die than the finished product).  You might shoot for something twice the size of the finished product.
Nozzle length ((Land length in a die) has an affect on backpressure.  You need back pressure to create a stable output.  In extrusion the Land Length Ratio (Ration of land length to die opening) is generally 3 to 1 or greater.
The shape of the nozzle should be the shape of the finished product - I assume round. Be aware that the hotter the plastic is when it leaves your machine, the more it will tend to sag and go out of round.
Temperature of the plastic (extrudate)  should be hot enough to for a homogeneous mix, but not so hot that it sags to a flat or oval shape as it cool.  One note on cooling, running a solid rod into water to cool it will cool the outside of the rod before the inside.  The cooling of the inside will then shrink up to the outside causing voids in the center of the rod.  Roundness is best achieved in a vertical extruder where the rod can air cool evenly with no pressure on one side or the other from supports.

The target filament diameter will depend on synchronizing two things, the output of the extruder, and the haul off speed of your puller.  I did not see a mention of a puller or caterpillar.  Letís assume that you have a machine that is producing a 1/32 inch diameter rod through a 0.100Ē die at 5 FPM.  Doubling the die size to 0.200Ē while keeping the same RPM on the extruder and the same haul off speed WILL NOT double your output or change the rod size.  They will both remain the same.  That is a difficult concept for some to get their heads around.  

That should give you some ideas.  Give me more details and perhaps I could help more.


---------- FOLLOW-UP ----------

QUESTION: Thank you for answering.

I am working on a spool winder that will have a set of rollers for haul off.  Until then, haul off is accomplished by gravity.  Where does die swell factor in?  Is there a minimum amount of haul off required to keep the extrudate from filling the die?

Currently the die is about 1.6mm.  The plug drilled through with a bit of countersink on the inside gives a land length of around 25mm.  The plastic swells to about 1.73mm at 175c with the weight of the filament dropping to the floor providing the haul off.  I've made some dies by drilling a hole through the length of a bolt and threading that into the plug, so I can try out some different land lengths.  A 10mm land yields a diameter of about 2mm. I can get that down to 1.9 by increasing the temperature.

If use a 3.5mm die (double the desired 1.75mm extrudate) 3x the die would be a 10.5mm land.  Does that formula assume a certain amount of haul off?  Without any haul off beyond the pull of gravity, would it still extrude a diameter that is smaller than the die, or would it fill the die and maybe swell to some amount dependent on land length and temperature?

In making the die, I can simply drill a hole at the die size all the way through the brass plug.  Or I can drill that hole, and then drill the inside of the plug with increasingly larger bits to create a funnel shape leading into the land.  How important is that funnel, and what effect would it have on the performance of the extruder?

ANSWER: I am working on a spool winder that will have a set of rollers for haul off.  
Until then, haul off is accomplished by gravity.  (This means that the size coming out of the die is the size that you will ultimately get.  It will be bigger at the start and draw dwon as the weight of the rod increases.  Not the best way to control size.)
Where does die swell factor in?  (Die swell is not a factor unless you are NOT pulling it off.  Longr lands will reduce die swell as will warmer temperatures  of the extrudate)
Is there a minimum amount of haul off required to keep the extrudate from filling the die? (Not sure I understand this question, the die is filled by the plastic being lushed into it by the feed screw)

Currently the die is about 1.6mm.  The plug drilled through with a bit of countersink on the inside gives a land length of around 25mm.  The plastic swells to about 1.73mm at 175c with the weight of the filament dropping to the floor providing the haul off.  I've made some dies by drilling a hole through the length of a bolt and threading that into the plug, so I can try out some different land lengths.  A 10mm land yields a diameter of about 2mm. I can get that down to 1.9 by increasing the temperature.

If use a 3.5mm die (double the desired 1.75mm extrudate) 3x the die would be a 10.5mm land.  Does that formula assume a certain amount of haul off?  Without any haul off beyond the pull of gravity, would it still extrude a diameter that is smaller than the die, or would it fill the die and maybe swell to some amount dependent on land length and temperature?
(Haul off is what you are missing.  I doubt seriously if one can control the output of the screw and gravity to the extent required to maintain size)

In making the die, I can simply drill a hole at the die size all the way through the brass plug.  Or I can drill that hole, and then drill the inside of the plug with increasingly larger bits to create a funnel shape leading into the land.  How important is that funnel, and what effect would it have on the performance of the extruder? (A funnel or taper leading into the die will reduce die swell for a give size die.  It will also prevent degradation of the material as it hangs up in that dead spot and overheats. )


---------- FOLLOW-UP ----------

Spooler
Spooler  
QUESTION: I have a roller made now to haul off and spool the filament, however I can't seem to get tolerances below .1mm.  It might be due to slight variations in the speed of the DC motor powering the roller.  I can try changing to a stepper which would be more accurate.  However I want to see if there are other factors as well.  

I think the rate of the motor is pretty constant.  I was getting closer tolerance with a smaller die, letting it swell and then be pulled off by gravity.  You mentioned that more back pressure creates a more stable output, and using a die 2x the diameter creates less back pressure than a die smaller than the desired diameter.  Is it likely to help if I increased back pressure with the larger die?  How can I achieve that?  I know reducing temperature will, but is there a minimum temperature where drawdown is ineffective?  I know that once the ABS cools to a certain point, it will tend to stretch and snap back.  I can also change the distance between the end of the auger and the die.  What effect does the length of the melt zone have?

Could the tolerance be affected by cooling?  Is there any significant shrinkage with ABS as it cools, or is it overwhelmed by the stretching of the haul off roller?

Answer
Letís take a step back to see what changed.  Iíd suggest going back to the smaller die and run conditions that held better the better tolerances, save samples to prove how well it is holding size, and then put your puller in line.  If the size varies more with the puller, then you have found the issue Ė gravity is more stable than your puller. If the size variation is the same with and without the puller, it is the extruder output that is varying.

Back pressure for the extruder can be varied with a given die size by using what is called a breaker plate with Screens.  The screens act to filter out contamination, but more importantly they create backpressure.  Typical screen packs start with a course screen against the breaker, 20 or 40 mesh, and go up as high as 240 mesh.  
Alternately, the back pressure for a larger did may be increased by increasing the land length of the die.

As to cooling affecting the size, yes, anytime the die is larger than the finished product, or the extrudate coming out of the die due to die swell is larger than the product, any draft or change in cooling will cause a size change.  Generally, those are what we call short surges or sizing surge, as the machine and the puller may both be very consistent, yet the size fluctuates.  

By the way, are you saving the ABS that you extrude and recycling it back into the extruder?  You chould be able to have several heat cycles before it discolors.  Just make sure that the rod is chopped to a uniform pellet size before introducing it into the machine.  

Plastics

All Answers


Answers by Expert:


Ask Experts

Volunteer


Louis DePaul

Expertise

I can answer questions on the extrusion of Medical grade thermoplastics, compounding, braiding, multi-lumen extrusion, radiopaque fillers, tooling and machinery, as well as general management and operations of plastic extrusion facilities. I am familure with most secondary assembly operations such as bonding and tipping.

Experience

My background includes management of several small custom medical extrusion houses, and I am currently the manager of the Plastics Extrusion Business Unit at a major medical U.S. device manufacturer.

Education/Credentials
Details available by written request

©2016 About.com. All rights reserved.