RepRap / RepStrap for making Aerodynamic Models

Hello,
I would like to ask you for a piece of advice.

I am looking for a non-commercial printer 3D with very good surface quality.
It will be used mainly for the production of aerodynamic models.
What matters is the accuracy of any external suffices of casing and internal surfaces of channels and nozzles, which runs inside the casing, as a 3D shapes.

Secondary matter is a time of elements accomplishment.
The size of the workspace: circa 500mm x 500mm x 500mm.
I would prefer buy similar RepRap / RepStrap. Can you indicate me the models that I should pay attention to?

I've also talked to my friends, who promised to help if necessary.
One of them has a lot of experience in constructing various types of mechanical equipment, and the other can link in electronics together.
My friends also have an access to such things as extruded profiles and CNC machines.
Therefore, in case of the absence of proper RepRap / RepStrap, I wanted to do, with the support of friends, a device dedicated to my needs.

If possible I would like to use the already existing and tested solutions.
The simplest for me would be building of the frame.
In the case of electronics, I prefer stay with the existing solutions, although I allowed minor exceptions as other motors or servo. Changes, however, must be small enough that the newly formed unit can be combined with existing drivers, software and other accessories.

That’s why I would like to ask you about general tips on what I should pay special attention, when I’d be building my own RepRap / RepStrap? Which models of Reps can be considered as a reference points?

- What extruder should I choose to get the best quality the final surface?
- Which plastic is in this case the most appropriate?

I know there are different design solutions. There are Reps in which:
- the beam moves in three axis,
- the beam moves only in one axis up – down, and table moves in two axis,
- the table moves in one axis and the beam moves in two.
Which of the design solutions is in your opinion the best and guarantees the highest accuracy of detail?
What advice could you give someone who is a beginner in the matter of Reps?

Regards,
Kuba

IMO with a reprap like machine, you will still need to do some surface cleaning and/or finishing (it maybe coating, sanding, using a solvent to polish...) for the kind of applications you are looking into.


ABS maybe be better than PLA to finish off as acetone is a known usable solvent that can polish it off (bath or vapours). On the other hand there might be some similar system for PLA but I didn't see as much experimentations done around the reprap community.
Still on the other hand, a lot of people whom are doing very fine layer prints use PLA - I'm not sure it's possible to do the same with ABS, but here again it may just be that people didn't experiment that possibility enough yet.


At that size (50x50x50 cm print volume) it's far better to have the head moving on all axes, because the weight of the table will be far greater and trying to make it move will trigger more inertia problems.

---

Most of my technical comments should be correct, but is THIS one ?
Anyway, as a rule of thumb, always double check what people write.

Some people have built large repraps, with varying degrees of success.

A printer layout like the Ultimaker and/or original Darwin is likely to be the only route for scaling up the print area to the size you want; certainly your bed is only going to be able to move in one axis, vertically. You will have huge problems with prints curling with prints this large, especially in ABS, without a heated bed, and a heated bed will be difficult/expensive to integrate - you'll have to heat a 500x500x5mm aluminium printbed to 120C minimum for ABS. Also, just the time it will take to make a print will be measured in days rather than hours, especially if you want high detail levels, which means small layer heights.

It is unlikely you will be able to double the print area without beefing up things like the linear rod. This in turn means increasing the size of the motors used, perhaps NEMA23 rather than NEMA17 motors, which means taking a different route with the electronics.

The advice I would give to you is to build a standard reprap, probably a MendelMax: [mendelmax.com] or Ultimaker: ultimaker.com which have the best accuracy.
You will learn a lot, which will impact and improve your chances of success with a large-scale machine.

From my experience building a wooden darwinish printer, here are a few thoughts:

Consider the possibility of printing in parts and fixing together - you'll get away with a far smaller printer, prints will individually take far less time and plastic - better if something screws up and you don't waste hours/lots of plastic. PLA warps so little that you can glue two prints together with a barely noticeable glue line - this you could always just sand/smooth. 50x50x50 is very big compared to (say) a prusas 20x20x16, however unless you plan on using the full volume of a bigger printer, you'd probably only need to glue a few parts together from a smaller, cheaper, better documented machine. You could very easily substitute 12mm rods for the 8mm in a prusa and push up the build volume a fair bit safely.

If you go for a bigger printer, darwin is probably the way to go - it's tried and tested. I used bike chain on mine, which turned out to be surprisingly good - you can tighten it like crazy so it droops less than a belt over a long distance. Whatever design you go for, you want to try and use as much of the length of the smooth rods as you can - better to use the whole thing than use a longer rod, waste some, and get more flex.

I would personally recommend building a prusa2, or other known machine, see if you can glue the parts together. If not, you can reuse lots of the electronics, extruder etc. I would go for beefier drivers than the standard ones maybe, a little extra but pololus won't drive nema23s without dying easily.

james glanville Wrote:
-------------------------------------------------------
pololus won't drive nema23s without
> dying easily.

I have been running nema 23's off of pololus for over a year and never have issues. The size of the motor does not matter it is the resistance of the coils that determine the current being drawn from the driver.

---

FFF Settings Calculator	Gcode post processors	Geometric Object Deposition Tool Blog
Tantillus.org	Mini Printable Lathe	How NOT to install a Pololu driver

The printer you describe is more in the "expert" class then the beginner class. You seem to be looking for professional quality without the cost. I think it will be a big challenge to make a printer that larger with a high accuracy (although you don't specify what accuracy is required). I am preety sure there are no designs around that would be suitable.

The idea of printing smaller parts is pretty good, at least you can start with of the sheld design. I would try to get some samples made by different printers to see if they can achieve the quality you require first.

you should consider cartesian. you can have option to mill wood with proxxon. i know my balsa wood planes use wood "D

i m going to make 1 like a cartesian ... i think it will take me 3-6 months

Sublime Wrote:
-------------------------------------------------------
> I have been running nema 23's off of pololus for
> over a year and never have issues. The size of the
> motor does not matter it is the resistance of the
> coils that determine the current being drawn from
> the driver.

I have done the same. NEMA 23's run just fine off of pololus, as long as the current isn't set too high.

Yes but when the current is limited by the Pololu you probably get as much torque with a high torque NEMA17 and have much less rotational inertia.

---

[www.hydraraptor.blogspot.com]

nophead Wrote:
-------------------------------------------------------
> Yes but when the current is limited by the Pololu
> you probably get as much torque with a high torque
> NEMA17 and have much less rotational inertia.

That's probably true.

ok so reading off the 4988 PDF ...

this little chip eats up to 35vdc and spews approx 2A of amperage for steppers. and it seems its best thermal dissipation is 4watts maxed driven or something to that measure.

ok now i have a trivial ...

if i have to series 4 of my steppers, do i specifically have a seperate stepper supply rail to improve the stepper drive? and in that case it would mean what for the single stepper channel (i can only hook 1 supply to a ramps)? assuming all the steppers are 10ohm/1.2A types (sanyo denki steppers), will it mean that the 4 series steppers are more sluggish? but retains torque and the 1 channel with a single stepper is nimble and may overheat?

is there such a thing as adding resistors so that all channels with diff numbers of motors perform similarly?

Edited 1 time(s). Last edit at 08/15/2012 02:33PM by redreprap.

There is no point adding resistors to hobble the single motor. It will perform better due to getting more voltage. It won't overheat if the current is set to the right value.

10R 1.2A motors are 12V motors designed for constant voltage drive. For constant current microstepping you want low voltage motors. The high voltage ones with 2 in series will need a supply in excess of 24V and since the absolute limit is 35V (in practice you should keep below 30V) will be slow. We normally use 2.8V motors at 12V or more.

---

[www.hydraraptor.blogspot.com]