Cylindercal printing idea

Hi, I've been thinking about how to print scaled tyres.

I've read there is, or will be soon, good rubbery plastics that can be printed. I've also been planning to design and print an RC car from scratch with as minimal bought parts as possible (just as a hobby project).

I came up with this idea to be able to print tyres on a barrel/cylinder instead of a heatbed. Basically the barrel replaces the Y axis and heat bed. Using the normal Y axis motor and belt there would be no need to change firmware.

Here are my thoughts so far...

* The pulley on the barrel would have to be the same size as the barrel to ensure if moves the surface just as the heatbed would move, therefor not requiring firmware calibration changes.

* The print barrel would need to match the inside diameter of the print you want to print.

* The barrel can easily be mounted in the normal printer frame, mostly only thinking about the Prusa here.

* For different sized inside diameters you would need different barrels. And the pulley size would change as well. I would want a pulley style belt tensioner as to take up the slack of a one size fits all belt.

* Belts would maybe need to be a one piece type and not the single length type used on the heat bed. Or maybe a way of clamping a belt to the barrel would be good as it would also avoid needing to attach or machine a toothed pulley into the barrel.

* Barrel would have to be heatable, and wrapped in teflon tape as I can't see a way of setting up a glass surface. Aluminum comes to mind for the barrel, machined out hollow and maybe even heater areas machined into it.

* This brings me to heating. I was thinking maybe a bunch of those heating resistors, or that wire which heats. Slotted through holes machined into the barrel/cylinder?

* The drawing used to print would need to be drawn flat as to trick the printer into thinking it is just printing as normal.
The drawing would be drawn with the bottom edge being the length of the circumference and tapering upward at the angle of arctan pi. I think it works out the be around 72 or 18 degrees depending what edge you go off of. This is because the length (converted from the circumference) gets longer at a ratio of 1:pi.

* The drawing would have to be drawn as though you're looking at the radius and not diameter.

With this idea, you could print round objects which usually require a lot of support but without that support. hollow tyres with tread would not work well on a heated bed. But depending on the shape of the inner carcass of the tyre it might be very easy to bridge just the tread instead of having to print support to hold the curved sidewall.


Does anyone have any thoughts about this idea?
I'm open to good and bad comments!

Edited 1 time(s). Last edit at 03/12/2013 10:52PM by Phizinza.

This does have potential especially in the jewelry business. I would guess you would want to work on quality of parts as the lathes that are used to form rings and other items that use subtractive processing are capable of fine detail beyond 0.02mm and typically in wax. These objects are then cast, and the created part is destroyed in the process. There would be need to reprint a part for one of the each lost cast molding processes

That is a great idea! I'm very interested in this however I don't have much to contribute except for a machinist's perspective. Do you have any links to info on this 'printable rubber'? Hope this thread grows! I'm going to follow along.
joe

Check this link for more discussion and further links to a finished additive lathe. [forums.reprap.org]

---

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

Phizinza Wrote:
-------------------------------------------------------
> Here are my thoughts so far...
>
> * The pulley on the barrel would have to be the
> same size as the barrel to ensure if moves the
> surface just as the heatbed would move, therefor
> not requiring firmware calibration changes.

Changing the Y steps in the firmware takes about 2 seconds now, there's no need to try to match the pulley and barrel sizes just to avoid that.

There's another problem though, you would only be able to print thin objects like this, because the print radius increases as the object grows and you will need to compensate for that or the object will end up having too little extrusion farther out from the barrel surface.

> * Barrel would have to be heatable, and wrapped in
> teflon tape as I can't see a way of setting up a
> glass surface. Aluminum comes to mind for the
> barrel, machined out hollow and maybe even heater
> areas machined into it.

Teflon tape on the outside? You know that nothing will stick to teflon...?

> * This brings me to heating. I was thinking maybe
> a bunch of those heating resistors, or that wire
> which heats. Slotted through holes machined into
> the barrel/cylinder?

If the barrel is hollow, I would set up a hot air blower to heat it. Just blow hot air from one end and direct it away at the other end. Setting up resistors on a rotating curved surface would be very annoying.

cncjoe Wrote:
-------------------------------------------------------
> That is a great idea! I'm very interested in this
> however I don't have much to contribute except for
> a machinist's perspective. Do you have any links
> to info on this 'printable rubber'? Hope this
> thread grows! I'm going to follow along.

Only printable rubber that I have seen for sale right now is TPE or Thermoplastic Elastomer sold by Orbi-Tech. I bought a roll of that about a year ago and printed some tests. It works.

ttsalo Wrote:
> There's another problem though, you would only be
> able to print thin objects like this, because the
> print radius increases as the object grows and you
> will need to compensate for that or the object
> will end up having too little extrusion farther
> out from the barrel surface.

... no problem with adjusting the 'tangential' speed, so regardless of the radius the extruding rate will be the same -- this means, that the turning speed will be faster with small radius and slowed down with rising radius to giv constant vector speed along the surface of the cylinder ...

---

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

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

Hmm? I don't see how could work right if you just map a cartesian print onto a cylinder without some sort of correction. The rectangular layers are being stretched longer and longer as the radius increases, and if they are from a cartesian print, they will have a constant volume of extrusion per layer (if we consider printing an uniform block for example).

... some years ago we had discussions about designing and fabbing in cylinder coordinates - here a thread about LOM-fabbing on a lathe-like setup with a sketch showing 'cartesian' objects embedded in a cylinder/spiral-tube : [forums.reprap.org]

On CNC-mills this is common too with swapping one linear axe for a turning one (X/Y/Z will then be A/B/C)

Some CAD/CAM's can convert the cartesian data into cylindrical, but all can engrave a 'plane' on a cylinder ... so with some tweaking (perspectivic distortion) you can make this for a turning axe with aRepRap too ...

---

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

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

This is how I imagined doing it...?

... yes, this is this 'perspectivic distortion'-trick for printing/engraving on a turning axe.

But for correct measures you'll need another opration, which 'distorts' cartesian coordinates (of let's say a box) to this virtual cylindrical volume, or a printed box with only 'perspectivic' distortion will come out as part of a ring!

Another needed modification is recalculating the vectorial speed for every radius, so the effective printing speed will be constant for all radii ...

---

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

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

I guess you just need to change the y steps per mm each layer. You would want the origin to be near the middle of the print and make sure that the extruder moved back to the origin when the steps per mm was changed. Then, provided that you distort the model in the way shown above, you should be able to print using standard firmware.

Edited 2 time(s). Last edit at 03/14/2013 07:50AM by Greg Frost.