Thursday, November 3, 2011

Ideas for Skeinforge - Multiple Zones

There's work being done for multiple materials, which is really quite interesting... here's an extension and/or generalization of that thought:

what's being done:
  volumes get tagged in OBJ file, or use multiple STL files to define separate regions... tagging in OBJ is better because you can have arbitrary tags, e.g. color per object, color per vertex, etc...

  tags that match settings in Skeinforge would override the SF value.  e.g. a tag: infill_30 would be 30% infill.  if infill was absent, don't do infil.

  alternatively it might be simpler to have multiple profiles, however, the reason that's a bad idea, is because SOME settings should be accessible as adjustments to SF settings, but some are material-stock specific.  e.g. infill is an preference variable, a situation variable.  I might want multiple infill% on the same model, e.g. on a big flat boxy area, I can handle a 20% infill, but on a vertical tower, I might want it at 80%... or maybe on that vertical tower I want NO SPARSE layers?.... that should all be my choice... and those things should be definable within the OBJ.

  Another region could be support.  maybe I want support material under a specific bridge, but don't want it anywhere else.  why not just make regions that are filled with no perimeter, and only support material? Would be just better if an object can be designated as support material (in which case each layer from ground up could be essentially like a shadow of all the layers above it not looking at the other regions, only the region designated as support.   setting: region=support only, has support, no support.

ok that's a few thoughts, but the main pleading here is, to anyone developing multiple region capabilities, make sure to keep it generalized.  we are moving toward a need for more control over the entirety of the model, like variable infill% etc.

thanks for listening.

Tuesday, October 25, 2011

Another Idea for Skeinforge...


Minimum Layer Time...  isn't very useful.  It slows everything down unnecessarily... or doesn't slow it down enough.

What is REALLY needed is a time to cool parameter for the material, and a cooling temp to achieve before printing over the same spot.

In other words, imagine if you lay down the plastic, at 185° C, and as soon as you lay it down it begins to cool... it shouldn't be exceedingly difficult to map the estimated temperature across the surface area, and then give the start position for the next layer priority to print on the opposite side.

alternatively, how about just adding a setting to the "fill" plug to flip-flop the start at from lower left only, to alternate lower left, upper right.

anyway... just a few thoughts.

Sunday, October 23, 2011

IDEAS for Skeinforge

for Cool - it would be "cool" if you could have the extruder  head move such that the fan was blowing on some region - or orbiting some region, or criss-crossing over the whole model... but with the nozzle high enough it doesn't hit anything....

wonder if I could just add a relative hop-up in the cool start, have the print head start at 0,0 go to max,0 then max,30, then 0,30, then o, 60... etc....

silly for large layers I mean for tiny objects, all you need to do is have the fan blow directly on the top layer for a couple seconds, literally, and then do the next layer.

orbit sucks, and slow down doesn't do much at all.

Prusa Mendel is up and running - nicely.

OK I've been away for a while...

Prusa Mendel is up and running and doing so quite reliably - so far.

I've been broadcasting on

Anyway ;) just thought I'd share what I was up to.

Sunday, August 22, 2010

Finally – The MK5 Plastruder… well, almost.

(almost meaning it’s going to ship in 1-3 weeks :( Uh, hurry up!… NOW PLEASE!!!! ) OK, so they’ve announced it!!! YAY this looks very promising, in a bunch of ways! nice… stainless steel.  NO STUPID BUTT SEAMS!!! nice.  Paxtruder inspired, without any STUPID IDLER wheel!!! NICE.  … I didn't see where the thermistor would go, and I’m very curious about using a relay board with the heater.

One problem I can really see with this is – since its still using the gear motor, and if it’s using the relay board, will it be able to hold a nice constant temp, and will it be able to cool itself down? – perhaps a modification of this design will be easy to hack.

I could see using forced liquid or air, to drop the temp of the hot end, when the temperatures are shifting.  And using insulation to reduce the load required to keep it at temperature.

Whatever the issues that may have room for improvement, the MK5 is SURE to start flying off the shelf!!! GO MBI, I’m psyched.

Cost, $185 – worth it? (comparatively… HELL YEAH!)

edit: finally got the page to load (unavailable earlier… too much traffic? maybe) and there’s the thermistor! COOL.



PS and for what it’s worth – their latest design is doing the single most critical smart thing possible – guiding the filament all the way through with PTFE tubing INSIDE the metal structure!!! kick ass… been working great for me in my rigged up design for a few months now as well and it sounds like they’ve been testing it for at least as long! SWEET!

BUY? --->


Thursday, August 19, 2010

It’s all developing nicely, but… just catch up when a great path is found, try to keep up at the head of the pack, or try to forge our own path?

I don’t know why I am this way, but I am.  I’m not outwardly competitive.  Logically I simply love the idea of the entire community moving forward as quickly as possible, and benefiting everyone.  However, on more than one occasion I read someone’s blog post and I get that twinge of '”crap, are they going to post the idea I had before I get around to it?”  I get that competitive urge of I need to make sure my idea is out there first!…  but really, I have to remind myself, as I do, that this endeavor is NOT so any one person can go down in history as having discovered “the best way to print”, it’s for ALL of us to be able to print reliable useful items.  Coming from a traditional product design background, the idea that we’re all in it together is a little foreign, so it takes some getting used to :)
Anyway – it was just an idea that I keep coming back to, and that is – you don’t have to be first with the idea and it doesn’t matter if you end up re-doing someone else’s work.  Re-doing the work, validates it.  Posting about it as well gives others more experimental data to work with/from.  And the fact is, even the best of us aren’t necessarily documenting every single step of what we do.  (sure some are)
The overall point here is a reminder (perhaps only to myself) to enjoy “MAKING”, it’s why I started this in the first place.
So, whether you want to lead the pack or follow the well worn path – make sure your end goal isn’t about others, but rather about what makes you excited.  That excitement is what will have you producing your best work.  Competition is a good thing, as long as it’s in the spirit of fun and sharing… at least as far as this project goes.
Big thanks goes out to the pioneers of RepRap.  And a big Kudos goes out to those that have built their lives around bringing it to people! Hats off.

It’s about time I post a photo of my setup! (yes, it’s with me in a hotel, thanks to my handy Pelican Case!


Sunday, August 8, 2010

Sounds like the MK5 actually will be a lot like this - without the long barrel

sounds like the MK5 hot end will have a seamless pathway all the way to the melter... even if there's still a PTFE  insulator - which I'm not sure I get yet... WHY use PTFE - WOOD is cheap-and hard-dimensionally stable - and unless I'm mistaken, won't burn at the temps we're working at.

Though, please, someone correct me if I'm wrong here.

So there's the PTFE tubing that guides the filament all the way down to the hot end from the supply and filament driver - cool, but hey - what about reel changing? do we have to un-thread the entire length? hmmm... sounds not so great.  do they hope we'll spend $100 each for a spindle + another $100 for each filament driver + another $50 for a controller for each color? or are we expected to push 2 feet of filament through the tubing each time we want to change colors.

if you aren't familiar with the mk5 concept - or the bowden extruder, it's basically pushing the filament driver away from the nozzle - which is cool in some ways.  It makes the nozzle end very light-weight, and smaller by a lot.  however it's a long travel from the driver to the hot end.  so there's compression of the filament (although I dont' think that's a huge problem) but material changes is more problematic.

I'm thinking it'd be better to have the filament driver as close to the hot end as possible - and just insulate it better... again - I SAY use wood, not PTFE.  It's serving me well, and I've NEVER had even a hit of burning wood smell... so ***DISCLAIMER*** I personally have never noticed any risk of fire hazzard from the wood, and it doesn't expand or fail like the PTFE - BUT!!! I also have no possibility of leaking plastic into the wood threads or anything - there are no seams in my set-up.  I have a thin (1/32" wall) PTFE tubing that runs the length of my wood insulator and down into my 5/16" - 18tpi hollow tubing that has a 3/16" bore - which fits this 3/16" OD, 1/8" ID PTFE tube like a glove... in fact I had to shave the tubing down on the outside with a perpendicular exacto blade (like how you'd scrap the top of your nail to check the sharpness of a freshly sharpened knife)

anyway - .... I really need to post a drawing of my hot end.  picture = 1000 words, and somehow my 1000 words is only worth 3 pictures of around 25% clarity.

can't wait to see the final mock-ups of the MBI Heater Barrel... Although I think there are still issues to solve, and I think they are not 100% goal oriented on ideal for end user, but they are trying to run a business... so it has to be cost effective (for them) to sell this.... so some of their decisions make more sense when you look at it from a business point of view - and I can in NO WAY fault them for that.  Doesn't mean that I think it's pointless to design my own hot end either. :)