Wednesday, July 28, 2010

Another Rambling thought on why heat sinks are a bad idea...

A lot of people have trouble with the filament getting "stuck" and not pushing through.  The filament pushing mechanism doesn't have a lot of push before it fails, temporarily, or permanently.

what happens, from what I can see, is that the filament melter opening gets hot, and melts the filament as it's going into the opening, and that allows the filament to mushroom around the barrel go down the threads of the PTFE which expands with heat anyway, and then contracts when it gets cold.

If we let the whole thing heat up a while, I'm pretty sure the filament would probably move through ok.

But then people started sinking the barrel to pull the heat away from the top end to prevent the mushrooming in the first place.   good idea, but I think it adds another problem.  --- metal, brass, is HIGHLY conductive, and ABS softens at even relatively cool temperatures.  Not to mention that it's going to soften up the length of the filament even without help from the brass.

well -
stop trying to sink the heat away, in fact, get as much heat as you can near the filament opening
stop trying to use PTFE which is a bad insulator, not dimensionally stable, and all around silly.

WOOD is a much better insulator - and with a smoke point well above (please someone correct me if I'm wrong) the 250°C that we're looking at, shouldn't be a risk.
WOOD is a much more stable material - takes threads easily, and holds them without stripping out so easily.

So - as an experiment, I'm about to try and use the makerbot heater barrel in a WOOD barrier.  It will eliminate the retainer rings, and hopefully insulate the filament pusher completely from any heat, and keep all the heat inside the block so that the filament will simply soften and push through.

-------

will let you know how things progress.... personally I still like the barrel I'm currently using that is PTFE lined, so there's not a single seam in the pathway.  That's right I said PTFE - but it's not structural, it's simply a lining.

There is one other experimenter I've seen using a PTFE lined heater barrel - and the only problem I had with their set up was that it was overly complicated and required a lathe - which I don't have.

anyway - onward and upward!

don't sink the heat - it's wasteful!
insulate the crap outta the barrel and make it EASY for the extruder controller to get it as hot as needed.

Friday, July 23, 2010

Designing the new extruder



In order to minimize the design of the extruder, it is necessary to outline what is needed.

A Supply of plastic
A Device to transport the plastic to the extruder
A Device to liquefy the plastic
A Device to direct the molten plastic into a fine stream in a precise location.

- I have ideas swimming around my head as to how to solve thiese problems, so it's hard to prevent what I BELIEVE to be an intriguing design path from influencing the bare-bones minimum description of what's needed.

- Let's look at the extruder component list above and think about it - is everything needed? is there anything missing?

supply of plastic - seems pretty general to me. could it be something else? perhaps its not plastic, ok so "SUPPLY OF MATERIAL"....

A Supply of Material
A Device to transport the plastic to the extruder
A Device to liquefy the plastic
A Device to direct the molten plastic into a fine stream in a precise location.

Perhaps this is too specific still... if we change to "Material" then perhaps there is no need to liquefy the material... what if it is liquid already (e.g. epoxy, or silicone, or frosting)...

OK so -

Simplest version is

A Supply of Material and a way to direct the material to a Specific Location.

I think something's missing...

A Supply of Material and a Pathway to Direct a Specific Volume of Said Material to a Specific Location at a Specific and Variable Rate.


--- I like that....

One last thing - that's optional is processing the raw material.

A Supply of Raw Material(s) and a Pathway that includes any Processing, and Directs a Specific Volume of the Processed Materials at a Specific Rate to a Specific Location...

So - interesting - Specific - means that's an important aspect to control, either through design or actually on the fly needs to be controllable.

1) Raw Material Supply (Hopper for Pellet Plastic, a Spindle for a Spool of Filament, a Chamber for Frosting-Like Materials, a Pair of Chambers for a 2-part epoxy)

2) Processing
  • Pellets need to be dried, melted, degassed, and homogenized, possibly colored, then if they are thermoplastic, then they need to be made molten by heating or perhaps by u/s vibration, or pressure. Injection molding uses a combination of pre-heating and pressure to inject the plastic and retain the best mechanical properties. Melting plastics with heat is not necessarily the ideal way to liquify them, but it's working - so we can use it for now... but consider there may be better ways that produce superior output.
  • Filament might need to be dusted/cleaned, possibly pre-heated, possibly oiled (for materials like PLA) - similar to above the filament needs to be liquefied, but we don't have the opportunity to compress it and use pressure the way we can with a screw. However, nothing says we can't use u/s pressure. we'll see... ( I just had a thought of sandwiching the filament between an ultrasonic plate, and a a stationary plate that contains the nozzle-outlet... feed the filament into the u/s pressing chamber and vibrationally smash it through the nozzle a droplet at a time. -

  • u/s extruder head - just a thought

  • Frosting-Like materials might need to be mixed to a smooth consistency, and perhaps degassed as well, perhaps coloring as well, but that would essentially make it a 2-Part material - see below.
  • 2-Part materials would need thorough mixing, possibly coloring, degassing?
3) Specific Volume of Material
  • Pellets would need some drive mechanism that accounts for their granular nature, and the fact that they will need to be degassed... the "specific" part would mean it would probably need to be a mechanism AFTER the processing, so either the hopper is gravity feeding into the dryer aparatus, etc... or there are 2 feed mechanisms... possible more
  • Filament needs to be driven either by a pinch-wheel, pax-style, or some other mechanism. Perhaps though, there is another method that is more controllable - this could use more thinking - keeping in mind tolerances and avoiding situations where precise construction is required. Best design = Design that anyone with half a brain and limited tools can make it.
  • Frosting - I'm going to stop considering frosting, it is a much larger task than I care to take on right now.
  • 2-Part Materials, same as above - both of which I may come back to in a later post.
ok time to get back to work - will work on this post later - feel free to leave comments if you think of anything to include... certainly I'm open to suggestions.