Cost and quality

hi,
i hope i'm in the right forum. please don't feal bothered about my questions, but this reprap thing is kind of confusing to me, because there are so many different informations.

firstly, what would a reprap cost? i could have all parts printed you would have to / could print. this price shourly is hard to calculate and i'm more thinking of the step-motors etc (everything you couldn't print). i've read something from 300 - 2500 USD... *confused*

secondly, i'm a product design student and would like to use the reprap for building models. how exact does the reprap print?

thanks, best regards.
dave

Ok, I'm sure that those more experienced and knowledgeable than I will jump in here and be able to answer your questions more fully, but here's a start, anyway.

As for the cost - the goal of the project is to be able to produce a RepRap for US $400 including the (material for the) printed parts. So to buy everything *except* the printed parts, I'd imagine you'd be looking at $300-$400, depending on how much of that cost is budgeted for printing costs.

The RepRap produces parts using FDM (Fused Deposition Modelling). More information on this technology can be found here [en.wikipedia.org]. I think the "standard" design has a resolution (extrusion size) of 0.5mm and an accuracy of ~0.1mm (don't quote me on those numbers). The material used to extrude is variable by application, for structural models the current favorite seems to be HDPE which is pretty cheap, common, and strong. Overall, this machine has been designed to capably fabricate functional objects in addition to non-functional prototypes.

Kyle

a good summary.

an important thing to note is that these numbers are currently goals. the extrusion size number is accurate (as we've tested the extruder) and now we're in the stage of testing the cartesian robot to get the practical resolution of the positioning.

using a reprap to print print models is one of the ideas we have in mind. currently its a large undertaking to assemble a working machine, but hopefully we will be lowering that barrier in the near future.

> The material used to extrude is
> variable by application, for structural models the
> current favorite seems to be HDPE which is pretty
> cheap, common, and strong.

LOL! And it's very difficult to work with.

I think the 2500 USD quote you saw was for the plastic parts to be printed by a commercial RP company at their standard costs. Our goal is to be able to print our own parts with a RepRap machine, which will drop the costs down to somewhere around 10-30 USD. We are unfortunately not able to print our own parts quite yet, though we are certainly getting much closer, and will hopefully be able to in the next several months. The non-printable parts will amount to around 300-350 USD or so.

So basically if you want to order all the parts RIGHT NOW and get extruding, 2500 USD is a reasonable cost estimate, but if you wait til we've actually reached the self-replication goals of the project, it will cost you around 300-400 USD, but you will have to wait until the end of the year or (hopefully not) later.

I hope this clears it up for ya.

thanks for the fast replies. sorry, that i have to ask again (kind of checking, if i understood everthing correct). my motherlanguage isn't english, so...

at the current moment - all parts (except for the printed ones - i can do those in university) is about 300-400 USD. the reprap is actually functional to build for non-developers?

is it a software "problem" not to be able to print own parts? how will the reprap be "upgradeable" and "updateable" (e.g. to version 1.0 ...)

sorry, again for maybe asking the almost same question still somewhat confused, and a little language-barrier.

dave

Yes, you are correct on the price. From what I can tell, the parts will create an machine that will extrude pretty well (my own machine is a modified version, so I can't say for sure about the 'real' one). The main problems with printing the parts is partially software, partially firmware, and partially figuring out which settings to use. Once we get more time experimenting to find the right values and work out the software bugs, we should be able to print parts fine. Also, Adrian is currently working on an extruder for support material, which is needed for some of the parts.

> The main problems with printing
> the parts is partially software, partially
> firmware, and partially figuring out which
> settings to use.

It's also a problem of doing FDM printing at room temperature rather than in a temperature controlled oven kept a few degrees below the melting point of the plastic being printed.

We're the first guys that have been seriously trying to do that and it's not clear yet whether it can be done successfully for a wide variety of plastics or whether it can be done in the same way that oven-baked Stratasys machines work.

don't the repraps work in normal room-temperature? maybe i missunderstood s.th. fundamental?!
are there any modified projects an the net Eric? Did you come up with all of this by your own?

I believe repraps could be interessting for many more students, who have to do prototyping. this could help the, what i think very interessting project i've been kind of excited for months now, and finally have some time during semester-brake

dave

Dave Wrote:
-------------------------------------------------------
> don't the repraps work in normal room-temperature?
> maybe i missunderstood s.th. fundamental?!

That's the idea. Basically we are pretty sure it is possible to print HDPE at room temperature and end up with a usable object, but so far (I think) Forrest is the only person who has been able to achieve this successfully (of course, so few people have machines that Forrest has been the only one actively working on it for a while). However, several people have been able to print more forgiving plastics such as CAPA at room temperature, so even if HDPE ends up not working out, there are alternatives. As jmarsen would say, "This is a research project!"

> are there any modified projects an the net Eric?
> Did you come up with all of this by your own?

There are several people working on modified RepRaps (referred to here as RepStraps). Off the top of my head there's Forrest [3dreplicators.com], who may have successfully "self-replicated" depending on how you define it, EricM (see the builder blog) who is printing *almost* accurately (not *almost* printing, right Eric? ), Joost (see the builder blog) who had a fully-built modified Darwin last I heard but who I haven't seen around here for a while, Nophead [hydraraptor.blogspot.com], who has just recently completed his home-baked CNC milling machine and is now using it to produce parts to convert it into a RepStrap, and emf, who I believe (?) has just started trying to convert a semi-automatic drill press into a RepStrap. And I'm sure there are many more who I've forgotten, and many more beyond that who haven't yet introduced themselves on the forums but who are quietly chugging away behind the scenes on their own variants.

> I believe repraps could be interessting for many
> more students, who have to do prototyping. this
> could help the, what i think very interessting
> project i've been kind of excited for months
> now, and finally have some time during
> semester-brake

Amen, brother! I'm planning on getting a good start on a RepStrap over the summer as well. I don't have access to a 3D printer myself, so I'm planning on following in the footsteps of EricM/Lamarck and modifying a commercial CNC milling machine into a RepStrap. Wish me luck!

Kyle

What Forrest was referring to is that the FDM machines sold commercially keep everything at a very high temperature until the entire extrusion is done, then the entire piece cools down at once. With the RepRap project, we are trying to extrude while only heating the plastic that is being placed, not the entire part. This can cause problems when the part contracts while it cools.

My modified machine uses the RepRap extruder attached to a CNC milling machine. There are no actual plans for this, but I have worked out how to do it with a lot of help from the forums and the IRC channel. A number of us are doing similar things, since the regular parts are not readily available.

If I relocated the Z limit and made circular adapters to fit into the hoop clamps that fit the dremel tool, I have everything I need, except an extruder. Why aren't we trying for a group buy of RP extruder parts! There are two extruder plans on the web site, one using a plastic torque driver and another using a steel cable. This is very confusing to someone trying to follow these plans. We need RP extruder parts!
Uniform heating was to limit thermal stress and shrinkage I think.

englewood: check out the reprappers forum. i found someone to do CNC manufactured extruder parts. in 3-4 weeks i should have an extruder kit with everything for sale in the RRRF store.

as for a group buy of RP parts, that might be do-able, however i think it might be hard to make that happen. however, you are free to go around and get quotes. all the required files are up on sourceforge. like forrest says, make sure you're sitting down when you get the quote back =)

as for the extruder version to use, it should be pointing to the new one. my number one priority this week is getting us migrated to mediawiki, at which point i'm going to try and solicit help in getting the wiki documentation whipped into shape. we're close, but there is also a lot of improvement to be made (like linking depreciated pages to the newest version, etc.) once we're on mediawiki, it will be easier for people to get in there and help out.

What is included in the kit? I have the nozzle and 3/8" threaded rod and I was planning on attaching the nichrome wire and thermistor to the near end, (so when I change nozzles, I don't need another thermistor). I have the motor and the teflon insulating tube, should I machine this pieces?

the kit will include all the mechanical parts you need to construct an extruder. i'll provide a full list later, but it will probably include everything.

however, i'll also offer the individual parts for sale on the site as well, so no worries there. if i end up not offering exactly what you need... i basically *AM* the RRRF store, so feel free to send me an email and we can work something out.

No you missed the point, should I do work on the threaded rod? If the kit doesn't include this they I can go ahead and drill it out and wire it up and mount the thermistor. Is the kit going to follow the basic extruder design in dimensions and materials? These questions probably can't be answered until the quote comes back!

if you want to work on the threaded rod, go ahead. its easy to make... put the rod in a drill press and use a dremel or other abrasive tool to smooth out the right areas.

as for the kit, heres a list of things the kit will include (basically all the mechanical stuf you need)

* CNC'ed equivalents of the printed parts
* drilled / tapped PTFE insulator
* drilled heater barrel
* machined drive screw
* a 0.55mm nozzle
* the half bearings
* the braided steel for the drive
* 1 foot nichrome wire
* the motor
* all the fasteners needed to assemble it

of course all of these will be available separately. some may not be listed on the site, but if you email me personally, i'd be glad to work something out.

i confirmed with Luis that I want to order an initial set of 10, so hopefully they'll be done in 3-4 weeks. we're getting closer and closer, guys!

englewood,
I might be wrong as I haven't got that far yet but I don't think attaching the heater to the wrong end of the nozzle will work very well. I am not sure how big the temperature gradient down the metal is but the incoming plastic has a very high specific heat capacity compared to metal so it will cool it more than say solder cools a soldering iron bit. Normally the temperature gradient is such that stiffer plastic is pushing on softer plastic, it may jam or back up if the gradient is the other way.

Chris

---

[www.hydraraptor.blogspot.com]

WHAT! I just didn't want to waste my time drilling, mounting a thermistor and nichrome wire on a heating tube that is already provided in the kit (I already bought the 3/8" 16 threaded rod). I want to mount the thermistor near the nozzle instead of on the nozzle, because I don't want to put a thermistor on each on the different nozzle sizes. I have the Teflon insulator, drilled out by an 1/8" through hole and tapped 3/8" 16 TO A 1/2" for the heater tube. I just needed the plastic parts. Now it looks like one version of the extruder uses a soldered in steel rope to transmit torque, and another version uses a RP adapter and filed flats on the drive screw to transmit torque. I personally prefer the adapter, since if it wears out, it can be replaced. The industrial extruders that I've seen have a hopper to feed pellets into a chamber where twin screws, (Heated by internal hot oil flowin the screws) move the PVC and partially heat it through conduction and friction. The melting plastic exits from an extrusion die in a continuous forming process, (Body side molding for cars). Dr. Higgs seems to be making progress towards extruder parts, his blogs are fascinating. IS the kit going to be metric or standard?

I understand the confusion, (the thermistor near the nozzle and the nichrome over the threads, in the drawing only a small fraction of the length has nichrome, but it is brass and small so it has a biot number less than 1, so it is uniform temperature, especially with the insulation!) I just didn't want to waste my time drilling, mounting a thermistor and nichrome wire on a heating tube that is already provided in the kit (I already bought the 3/8" 16 threaded rod). I want to mount the thermistor near the nozzle instead of on the nozzle, because I don't want to put a thermistor on each on the different nozzle sizes. I have the Teflon insulator, drilled out by an 1/8" through hole and tapped 3/8" 16 TO A 1/2" for the heater tube. I just needed the plastic parts. Now it looks like one version of the extruder uses a soldered in steel rope to transmit torque, and another version uses a RP adapter and filed flats on the drive screw to transmit torque. I personally prefer the adapter, since if it wears out, it can be replaced. The industrial extruders that I've seen have a hopper to feed pellets into a chamber where twin screws, (Heated by internal hot oil flowin the screws) move the PVC and partially heat it through conduction and friction. The melting plastic exits from an extrusion die in a continuous forming process, (Body side molding for cars). Dr. Higgs seems to be making progress towards extruder parts, his blogs are fascinating. IS the kit going to be metric or standard?

The version with the steel rope is the newer "improved" version that has a straight path for the filament. It is to allow stiffer filaments to be used.

---

[www.hydraraptor.blogspot.com]

whoa, englewood, chill out dude.

re-read my post. if you want to only order the printed parts equivalents, thats fine! once the parts are in the store, send me an email and i'll setup a kit just for you thats only printed parts and nothing else.

i'm guessing that most people will want a kit with everything that they can just assemble into a working extruder in one sitting.

I just don't want to redo anything needlessly. As far as the thermistor and cap go, you have to get the beta and Rz determined from reading the resistance at two extreme temperatures. I used room temperature and 383F and got a Beta, Rz and a cap value. This is entered in the preference box, (the Rz and the beta). I sent Jonathon a duplicate thermistor and I'm hoping he will make the measurements and like TOTALLY AGREE!! with my beta and Rz. Then I'll know I'm right.

I'd be wary of putting the thermistor near the nozzle, but not on the nozzle. I could be wrong, but I suspect you won't get the accuracy, or consistency, of readings if there isn't a good union between the thermistor and something on there that's metal and close to the actual extrusion.

Hmm...
Has anyone considered burying the thermistor between the threads of the nozzle and the extruder? Is there enough material there to cut a groove large enough to accommodate the thermistor?

LOL! A few tips from an old instrumentation wallah. The big problem with control of the Mk II vis a vis temperature and that heater coil is the time constant of the extruder barrel. It needs to be very low if you are really interested in getting fast response. Not to put too fine a point on it, it isn't. Now that Zach has come up with that drilled out aluminum bolt for the barrel the time constant is a lot better than it was when it was brass, but it's still not wonderful.

You really need to be measuring the temperature of your molten polymer stream in the center of the barrel very near the orifice. Figuring out how to design an extruder barrel and head that will let you do that, however, is likely to be a nasty problem.

The further away you are from the center of your molten polymer stream, the more effect your time constant is going to have on your controls.

Use the biot number hDl/K is less than 0.1, the temperature is uniform throughout the pipe wall. The fiberglass definitely reduces the h factor and K for Aluminum is fairly large, since it is a conductor. I would say for steady heat transfer, a uniform temperature field exists in the pipe wall, so location of the thermistor behind the acorn nut nozzle should make much difference, Lets not forget to mention the accuracy of our 8 bit measurement from 0 to 240C,240/255=0.95C/count or so. On my ship we had RTD installed in welded wells in the pipe wall.

Edited 1 time(s). Last edit at 07/18/2007 05:10AM by englewood.