interchangeable pressure nozzles

In a recent post I make a offhand comment about using pressure nozzles. But it occurs to me thats not that bad an idea. Cake decorating nozzles are stainless steel and built to be easily interchangeable. Its possible that using a slightly star shaped nozzle would increase interlayer adhesion since the bottom layer would have a much higher coeffient of friction even when the top layer was completely liquid. Currently we are attempting to extrude the new layer on top of the crest of the hill formed by the previous layers, which is of course the reason that most of the infill is done diagonally. Taking that to its ultimate conclusion brings me to the cake decorating nozzles that are used for making flat ribbons of icing. I would imagine that the sag resistance of those shapes is excellent. Also given the very high surface area of those shapes the cooling rate is going to be much better as well, further reducing sag. Another interesting possibility is that you could move a quarter or half of the distance of the ribbon and build a layer that overlapped the lower layer by only 90, 80, or 70 percent thus allowing for the construction of much more severe overhangs then would otherwise be possible without support materials.

Mike

Mike,

Ah, that's great! I have a friend here at the university who has been wondering about different shaped extruder nozzles for a couple of weeks now. I never did say much of anything about it cause I didn't see the use of it. He will be vindicated to know that it could have some merit! Very funny.

Demeneted

There are 2 issues with shaped nozzles:

1. building them- remember we want layers to be as thin as possible for more precise building, so at least one dimension needs to be as close to .5 mm as possible. It is tough to get and use drill bits in that size, much less the milling heads you need to create slots, stars, etc. Also, the material at the nozzle tip needs to be very thin to reduce the pressure needed to extrude the plastic. Thin material needs to be strong if it is to hold complex shapes (star or arc) against extrusion pressure.

2. orientation- a pinhole leaves the same trail whichever direction you move it. A slot leaves very different deposition patterns depending on it's orientation to it's direction of travel; a wide thin band if the slot is 90 degrees to the direction of travel, a narrow heavy line if it is aligned to the direction of travel. While this trait offers certain benefits it also imposes significant costs. As i frequently argue, I think RepRap needs to maintain laser like focus on reliability, simplicity, and user empowerment. If a feature adds to those goals great, if not I am VERY suspicious of it.

Hi BDolge,

... i already milled in plastic, aluminium and brass with a 0,3mm-diameter milling head, so for milling freefrom-holes in thin plates or aluminium/brass-heads it should be no problem too ...

For the orientation: - think on 'tangential'-cutter-plotters, here the cutting knife is turned with a motor in the driving direction, so you have a clean cut exactly in the moving line, not a bit delayed, as in the cheaper passive cutting heads.

For the pre-orientation the PIC (or Arduino) can calculate the angular position from the next MoveToXY-coordinates easily ...

Viktor

Hi Victor,

I was not arguing that creating shaped nozzles was impossible or even inadvisable, just that it might not be advisable to build the basic machine around doing so. While I am sure that you and lots of others are comfortable with .3 mm milling, lots more folks are not. The simpler the base machine is the farther/faster it will travel; therefor I ask what is the benefit of this complexity. There is another thread about speeding up deposition where this sort of thing is being discussed and raising print speed is a legit goal and ribbon dep is a possibility. I was more worried about the drive toward more complex "cake decorator" nozzles for which I don't see any benefit.

I have also thought about orientable nozzles and I agree that a printhead could be made to orient a linear nozzle fairly simply but only with 2 provisos: 1. only one orientation (probably 90 degrees to travel) is wanted, and 2. moves are straight lines. Multiple orientations are probably not that tough but I don't see much benefit. Maintaining orientation on a changing path would almost certainly require interactive communication with the motion controller and I can't think of anything that would make that worth while.

BDolge,

Ah, I see your point. Simplicity in the first machine is necessary. Future machines--which could require the first machine in order to make them--can be as complex or as simple as the given branch of development choose to make them.

Demented

Demented,

True

(See I can be concise)
(but I think I am doomed to be parenthetical, self-referential, and over-analytical)

... i didn't have any problem with simple setups

For me it's a good point too, to give a not so skilled individual the ability to build it without expensive tools and material in his basement in two weekend-sessions ...

But as good way to evolving faster, i hope to seed some regional competence-centers - for example at scools, institutes or at home by some enthusiasts as me

And here the higher complexity/accuracy is no fault, but a chance for rapid-evolution ...

And why not support the rrf-store with highly sophisticated tools?

Viktor

Just a thought on how to get that ribbon extruder.
If you make it along the lines of a wood plane, you could probably machine it easier.
Make a shallow "U" shaped "shoe", and a blade that fits tightly against it. Against it, so you can tap the sides of the shoe for screws to hold the blade. Preferably at least two screws per side. The shoe would be wedge shaped, basically a wedge with a wedge shaped volume removed from one plane, leaving the bottom surface a consistent thickness, but with tapering walls. The blade should have oval mounting holes, so it can be adjusted relative to the shoe, by loosening the screws, and tapping with a screwdriver and hammer, then retightening the screws.

As for advisability. I don't know.

Vicktor,

I am sure that if RepRap gets any traction at all one of the first things to happen will be the occurrence of "regional competence-centers" or just plain community accessible machines (in libraries, churches, and community centers) which will naturally develop their own experts. Certainly in less affluent areas it would make sense for several families to share a machine and I expect that most apartment dwellers would rather not give up space to such a machine but would love to use one. And yes, lots of innovation will spring from these centers, including lots of new printheads. Also such centers would be excellent for machines that do rarely needed but critical operations like creating pieces from titanium or making replacement parts for auto brakes. There may also be machines that should not be around the house for operational reasons such as noise or hazardous wastes. Certainly machines in these "regional competence-centers" can be arbitrarily complex (or large, or expensive).

I believe however it is important to maintain a branch of the RepRap tree which is comprehensible to and buildable by an individual of very limited means (skills and material resources). Such a machine would prevent RepRap from becoming just another corporate product. (The open nature of the RepRap license allows me to make one, put it in a box and sell it to you as long as I give you the plans. If there are one or more pieces that I can make economically that you cannot, then I effectively own the product). It also means that RepRap could become the answer to the complaint by Less Developed Countries that Developed Countries are trying to deprive them of the benefits of an industrial revolution in the name of Environmentalism. A functional, highly capable RepRap would make it possible for an individual or community to walk away from the mass economic system without sacrificing a decent lifestyle, but only if there is a way to access its power apart from that system. Until the goal of %100 self-reproduction is reached RepStrap will be more important than RepRap. This is one reason why modularity is so important. If I can start with something I built at home from mostly scrap for little money and incrementally add capability here and there until I wind up with something that can build a computer or an airplane, then I'm free (in one important way). The fewer sources there are for any critical part the less likely that vision is to be realized. That's why I hope more people start casting parts and offering them up on the net. Yes I think the RRF should sell parts for higher end RepRaps, but also for lower end ones. Another reason I stress modularity is that it encourages creativity. If I think of a way to print felted wool I am much more likely to go forward with my idea if I know that I can count on a large base of support, that I don't have to worry about how to calculate Cartesian coordinates for a pattern piece, all I have to figure out how to do is lay down a strip of good cloth and thousands of people can pick that capability up an use it. Similarly, if I hear that someone has created a great new printhead that allows me to print a new material I can start designing around that material without worrying about the tech specs of the printhead, I can rely on it working in any vaguely appropriate machine.

Hi BDolge,

... i agree fully - this is my personal path in developing:

First bring to work my 3Axis-CNC-Repstrap (mostly for milling the complexer parts for the next developing steps)
- then make some extruders and printheads (as for the CNC too)
- and play around with different materials ...
- then activate/realize the Tripod (only software left to do) ...
- then activate the diode-laser for sheet-cutting and paste-sintering,
- and experiment with LOM- and lathe-fabbing ...

And then it's time to think about real self-reproducing ...

Viktor

something that will increase the area of the extrusion seems like a genious idea, then use a motor to set it in the right angle. Great brainstorming everyone!!!

In response to the remark that the first repraps should be simple: I find that making things simple, and making parts interchangeable, often seem to go hand in hand. Besides, all that I understand is required for an interchangeable nozzle is the twist off the old one, and screw on a new one. Don't we have that already? Maybe the nozzle is glued on or something, I don't know. But it seems to me that being able to take the nozzles off is a good thing anyway. If the plastic pools and gets clogged, it would be easy to unscrew the nozzle, and scrape the barrel clean.

Our nozzles are currently just acorn nuts, so why shouldn't they be interchangeable?

I probably just missed the whole point, didn't I.

-Samuel

Hi Samuel,

Yes, currently changing out nozzles is a fairly simple process. The difficultty we are discussing (at least at the foundation of the thread) is that the currrent nozzle is fixed relative to the print head which maintains its' facing relative to the x/y table regardless of its' direction of travel. In the case of a circular nozzle this is irrelevant. But if the nozzle is slotted so that it produces a flat ribbon (in order to increase the area covered per pass in order to speed up printing) THEN the orientation of the nozzle/printhead relative to the printhead's direction of travel is important. Varying the nozzles orientation is mechanically trivial BUT the software which drives that machinery and integrates it into the overall RepRap machine offers lots of choices as to how to accomplish this task; choices which reflect deep choices regarding what sort of machine we are making. Thus it has become a proxy for discussing these choices.

Hope this helps,
Brian

I am insuffiently familiar with the software and firmware concerns to really judge what is easier. If having a ribbon like pattern removes the necessity for a support material extruder because curves can be built "igloo" style, then it might be easier to have a stepper motor attached to the nozzle mount to control the ribbons position rather than add another extuder. That was the point of the suggestion to facilitate a discussion with the goal being simplification of the end design.

Personally I am interested in the paste extuder because of its greater simplicity and versitility. I want to work with multi-part pastes that can be piped into a extremely streamlined head. No heating coil, no screw necessary to feed the filiment. Pressurized pots holding the materials and a simple pinch valve on the extruder head. Pour the pressure pots out of concrete and use a bicylce pump to pressurize them. Embed rebar to mount the rest of the system and use the resultant concrete table as the build surface. The result would be a very light head mounted on a very heavy base, which should greatly improve the tolerances.

Mike

The view in this post are my own, not my employers. Any attempts at implementation are at your own risk.