Generational Degradation

Hello. I'm an accounting student at WSU Tri-Cities with an interest in open-source and emerging technologies. 3d printing is of particular interest to me. Your project seems like a workable approach to decreasing the cost and increasing the popularity of this technology.

RepRap machines cannot produce all of the parts for a new machine, so each 3d printer gets an infusion of new parts, right? I was wondering:

1) Could degradation in print quality over successive generations of machines from printed parts become a significant problem?

2) Would the new parts in each generation of machines compensate for the problem?

Thanks for your consideration.

>Could degradation in print quality over successive generations of machines from printed parts become a significant problem?

Its not an issue. The machine has been designed completely adjustable and replicated parts are only required to lie within a certain tolerance. One of the requirements of operating the machine is that the size of produced parts *should* be measured to ensure their geometries lie within tolerance, if it doesn't some one of the machine parameters can be adjusted so that it does....

This fits in with traditional machine theory. Lathes have been making parts for other Lathes since the industrial revolution, and their capability has been getting better and better over time....

That makes sense. Tools are used to make tools, and all that. I suppose quality would even get better over time with design improvements and all that. Thanks!

Actually I think the printed pulleys are an exception to this. I think they will propagate any inaccuracy in the parent's printing to the child and get worse over each successive generation.

It would be an interesting experiment to print a pulley on a Mendel and then swap that pulley in and repeat many times to see if it gets worse.

---

[www.hydraraptor.blogspot.com]

I wonder if a 2:1 pulley rig would help preserve accuracy over generations? The accuracy wouldn't be 2x, but it might be enough to keep things on the positive side.

This could be an issue, but is unlikely if multi point contact is employed. Two gears together could compound any errors, but pulleys and belts tend not to as the toothed pulley averages its error across several teeth as it wraps around. This is further assisted as has been mentioned by gearing in the mechanism improving accuracy.

If I remember my history correctly, the first metal lathes were developed from wooden ones. Such things as lead screws were made by hand from wood. By averaging the errors across several teeth in the lead screw nut, the errors were reduced with each succesive generation.

Slightly off topic has anyone discussed the fundamental stumbling block of extruding a part that will stand the temperature of extruding the same material. How are we ever going to make an extrusion nozzle?!

I improved the precision of my wooden mendel by printing more precise pulleys with less precise ones. I simply printed a few pulleys, measured them and took the most precise ones. Even the badest ones, were more precise than the original ones.
It looks like there's some sort of convergence at work.
So if some of the more precise printed parts are converging, I stronlgy assume the less precise ones are also converging.
The most critical parts for precision are (currently) not printed like bars (straigth), bearings(round), extruder nozzle(hole), print bed(flat). It's possible that some of these parts would accumulate manufacturing errors (I have a hard time imaging printing a more precise extursion nozzle with a less prcise one)

[yetanothermendel.blogspot.com]

I agree with Nophead, after some generations you would have print quality isses, but I think it would very quickly drop off the cliff from print quality issues to failed print issues.

I usually take a jeweler's files to my printer gears.

---

repraplogphase.blogspot.com

Quote
Markus Amsler
Even the badest ones, ...

Is that proper English English because it's certainly not proper American English!

Edited 1 time(s). Last edit at 04/27/2010 08:36AM by rhmorrison.

---

Bob Morrison
Wörth am Rhein, Germany
"Luke, use the source!"
BLOG - PHOTOS - Thingiverse

Hi Bob,

rhmorrison Wrote:
-------------------------------------------------------
> > Even the badest ones, ...
>
>
> Is that proper English English because it's
> certainly not proper American English!

... this is more like "German English" - sometimes i have similar issues

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

rhmorrison Wrote:
-------------------------------------------------------
> > Even the badest ones, ...
>
>
> Is that proper English English because it's
> certainly not proper American English!
It's proper Standardized Helvetic English.

Bad, Bad Leroy Brown

Well the South side of Chicago
Is the baddest part of town
And if you go down there
You better just beware
Of a man named Leroy Brown

Now Leroy more than trouble
You see he stand 'bout six foot four
All the downtown ladies call him "Treetop Lover"
All the mens just call him "Sir"

And it's bad, bad Leroy Brown
The baddest man in the whole damn town
Badder than old King Kong
And meaner than a junkyard dog

Now Leroy he a gambler
And he like his fancy clothes
And he like to wave his diamond rings
In front of everybody's nose
He got a custom Continental
He got an Eldorado too
He got a 32 gun in his pocket for fun
He got a razor in his shoe

And it's bad, bad Leroy Brown
The baddest man in the whole damn town
Badder than old King Kong
And meaner than a junkyard dog

Well Friday bout a week ago
Leroy shootin' dice
And at the edge of the bar
Sat a girl named Doris
And ooh that girl looked nice
Well he cast his eyes upon her
And the trouble soon began
And Leroy Brown learned a lesson
'Bout messin' with the wife of a jealous man

And it's bad, bad Leroy Brown
The baddest man in the whole damned town
Badder than old King Kong
And meaner than a junkyard dog,

Well the two men took to fightin'
And when they pulled them from the floor
Leroy looked like a jigsaw puzzle
With a couple of pieces gone

And it's bad, bad Leroy Brown
The baddest man in the whole damn town
Badder than old King Kong
And meaner than a junkyard dog

---

Bob Morrison
Wörth am Rhein, Germany
"Luke, use the source!"
BLOG - PHOTOS - Thingiverse

> Slightly off topic has anyone discussed the
> fundamental stumbling block of extruding a part
> that will stand the temperature of extruding the
> same material. How are we ever going to make an
> extrusion nozzle?!


Well, single-material isn't a design goal. So - just extrude something that isn't a thermoplastic.

1) Could degradation in print quality over successive generations of machines from printed parts become a significant problem?

If the printed parts was used as a patterns for successive parts there could be degradaton. As the parts are printed from a digital file, I see the only variation would be the the print quality of the indivigual machines.

2) Would the new parts in each generation of machines compensate for the problem?

Hopefully, lessons learned from each printing would be used to improve the print quality of later printings. Thus, better quality parts.

I also remember that "Moire Fringes" also called interference fringes is a method where you can use less accurate scales to measure more accurately than the scale itself. I think this is the way digital calipers work. Vernier scales do this also as used on an old micrometer or a slide rule. (Am I revealing how old I am here!)

The side topic of extrusion nozzle manufacture. As far as I am aware its not possible to extrude a material of a higher melting point through one of a lower melting point.
It is also highly difficult to extrude a harder material through a softer one for any length of time. Perhaps a hard coating of some sort might overcome this.
I am glad there are other rapid prototyping methods being developed in parallel as this might be the baddest problem we encounter for total replication using FDM.

Edited 3 time(s). Last edit at 04/27/2010 04:16PM by martinprice2004.

Nyarlathotep Wrote:
-------------------------------------------------------
> > Slightly off topic has anyone discussed the
> > fundamental stumbling block of extruding a part
> > that will stand the temperature of extruding
> the
> > same material. How are we ever going to make an
> > extrusion nozzle?!
>
>
> Well, single-material isn't a design goal. So -
> just extrude something that isn't a thermoplastic.

... that should be easy: - print a dispenser and then dispense ceramic slurry ... was mentioned several times

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

We are also looking at adding milling and other subtractive technologies. But I guess then we end up with the problem of making new drill/mill bits.

As for harder materials from softer ones, many tools start off soft and become hardened by a secondary process, like tempering, case hardening, peening, work hardening. It may mean that our machines become complicated enough to heat treat and surface harden parts, and possibly cast parts as well. As mentioned above, ceramics can be extruded as a paste, baked until solid, filled with molten metal, then opened to release the part. Even metal can be melted by using electric current to localize the part that melts. MIG welding melts steel wire as it exits a copper tip, even though copper melts at a lower temperature than steel.

With the full range of techniques, additive and subtractive, thermoplastics, ceramics, and molten metal, tempering and surface hardening, it should be possible to make all the mechanical parts. Remember, a black smith shop is a self-replicating fabricator.

Mike

Foundry crucibles are sometimes made of a material with a material of a lower melting point than it's contents. They use water cooled plumbing to wick heat away from the crucible. Same concept as lighting a paper cup on fire.

We are also looking at adding milling and other subtractive technologies. But I guess then we end up with the problem of making new drill/mill bits.

Step 1: Bolt a dremel or large router on the machine.
Step 2: Worry about fabricating vitamins.

Ditto with laser tubes or laser diodes:
[reprap.org]

I'm not saying that I don't want us to design a cnc endmill grinder, (hypothetical) sodium silicate cement or epoxy-granite flycutter with carbide inserts, or printable diode laser in the wiki.

Edited 1 time(s). Last edit at 04/28/2010 01:39AM by SebastienBailard.

---

-Sebastien, RepRap.org library gnome.

---

Remember, you're all RepRap developers (once you've joined the super-secret developer mailing list), and the wiki, RepRap.org, [reprap.org] is for everyone and everything!

... the most promising omnipotent technology is laser-sintering - here you haven't mechanical forces, so your mechanics can be lightweight, but have to be really accurate.

And you can make any shape from every material you can melt/sinter with the laser.

The only problem is the fabrication of the laser source, here only CO2-tubes with extrenal mirrors and a high-volt PS are in the DIY-range, all other systems need high precise/cleanroom-equipment ...

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

As you have all suggested FDM would seem to require other processes like milling or paste extrusion in order to make a machine capable of total self replication. This makes the machine more and more complex and raises ever increasing difficulties (The technology pyramid is the wrong way up!).

I agree with Viktor that laser sintering would seem to be an easier route. I am surprised that this technology doesn't recieve much more attention than FDM. Perhaps this is natural as the Reprap Machine design has pointed people down the FDM route. The laser is a problem, but perhaps focussed light may be a possibility to overcome this.

A guiding philosophy of the reprap project was to make the machine capable of bieng built in a third world country, growing the raw materials. Dremels and milling cutters would be quite hard to source. Heat treatment of metals would however be possible in a simple forge.

By the way MIG welding doesn't melt metal wire near the copper tip but some distance away from the tip. This is also why TIG welding has to use a Tungsten tip.

Edited 1 time(s). Last edit at 04/28/2010 08:01AM by martinprice2004.

... beside SLS there is another sintering technique called "Selective Mask-Sintering" - here you rise your powder-bed as with SLS but then you place an optical mask on the powder-surface and activate an areal IR-heater above the mask, so only the powder not covered by the mask melts.

I would atach the PDF-file, but it's 3MB, so i cant upload ... here the (maybe temporary) link to the (German) PDF ...

This masks can be made with the same process as in Skeinforge - essentially you can use the created SVG-file of the cuttings directly without any change

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

There's another variation on the mask, where you use a print head to spray water or a cooling mix then bring a heater over the bed to fuse other areas.

... not water, it's called "Inhibition Sintering" and you print with a salt-water mixture - the water evaporates, the salt remains and "inhibits" the melting/sintering of the 'contaminated' area.

When heating the surface with an IR-heater, the clean powder melts and the salt crystals prevent the fusing in the printed lines ...

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

Sorry fgrams we seem to have completely hijacked this post!

I didn't understand how the mask is made? Does it mean that you can only make what are essentially extruded parts or do you have to remake the mask every layer as the section changes? What is the mask made out of.

Is it possible to focus infra red heat to a beam or through a small hole slot? Effectively producing a poor mans laser?

I did like the look of this process where the heat seems to be bounced off a reflective printed mask.

Sintermask

I liked this video also

Inhibition sintering video

These technologies to me seem to have much more going for them than FDM. Particularly the way they overcome the support material issue, have natural higher resolution and also would tend to shorten the time to produce parts.

Its finding a good source of black plastic powder that seems to be the first hurdle.

That inhibition sintering sounds good, you could print salt water using a conventional inkjet head without issue.

What about a reflective powder and you overprint a dark ink to change the infra red absortion where you need it. Or vica versa

Another thing that puts me off lasers is the inherent danger of losing ones eyesight whilst experimenting...am i just a coward!

Edited 2 time(s). Last edit at 04/28/2010 03:13PM by martinprice2004.

... inkjet printimg and areal sintering is much faster than fabbing with FDM or laserspot, but you are limited by possible materials and final stability.

Masks are made from heatresistant and/or reflective material and you have to cut every changing layer or combine single elements as in oldstyle Gerber photoplotters.

With laser you have much higher temperatures, so you can melt/fuse any material to solid, what's much more interesting with ceramics, metals and hard-alloys.

So you have to select your methode corresponding for your needs - my needs point me to micromechanical structures made fom resins, ceramics and metals ... sometimes even combined in interfusing 3D-structures

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

Martin,

The sintering process, whether laser, mask, or inhibition, does have the advantage of support material so that any shape can be made. But it also has the problem of big piles of left over sintering material that must be cleaned out of the work area before the next piece can be started. This makes any degree of automation beyond single parts difficult, and the process in general rather messy. The inhibition process also wasted a great deal of material by fusing almost all of the negative space, too. This increases per-part costs. And the mask based sintering is fast, but requires a huge number of distinct masks to make a thick part. It is not clear if the masks are easy to generate and the material reusable. FDM and laser spot sintering are slower, but less wasteful.

Mike

... by the way - if you already have a cutting/sintering laser, you can combine the mask-sintering method by applying an IR-heater and cutting the masks with the laser.

But there are some really hard to realize preconditions: - for the masks you need a material which you can cut with the laser, that withstand the heat from the IR-source and is cheap enough for highcount masks per part.

For plastic powder you can use paper (dark coloured with diode- or SS-laser, any colour with CO2-lasers), as the ignition temperature of paper is much higher than melting temperature sof most plastics ... but for materials with higher melting temperatures you need 'green' ceramic sheets, what's really expensive (i have some blue-coloured samples from Dupont Chemical with 0.1 to 0.5mm thickness)

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

rocket_scientist Wrote:
-------------------------------------------------------
> ...Remember, a black smith shop is a self-replicating fabricator.

That brings up a good question: Is the goal of this project to make a rapid prototyping machine that a human can use to make another rapid prototyping machine? That seems to be what the RepRap project is aiming for. I wasn't sure if articulated robot arms for assembly and so on were part of its scope.

... i think the idea behind RepRap was to create a system capable of building most (if not all) of its own parts with an affordable cost, so anyone interested in the technique could participate.

The idea is not unique (most CNC-routers can do the same), but it was based around FDM with DIY-plastic (PLA) for reducing the basic costs and an ecologic meaning ... but you need some 'vitamines' as electronics, motors, cables and bearings anyhow ...

With time there will (and have) evolved and forked other designs and technologies, but this is not a problem and will help to find better optimized solutions

Edited 1 time(s). Last edit at 04/29/2010 04:06AM by VDX.

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]