Westlake Publishing Forums
May 23, 2013, 10:12:13 PM *
Welcome, Guest. Please login or register.

Login with username, password and session length
News:     REGARDING MEMBERSHIP ON THIS FORUM: Due to spam, our server has disabled the forum software to gain membership. The only way to become a new member is for you to send me a private e-mail with your preferred screen name (we prefer you use your real name, or some variant there-of), and email adress you would like to have associated with the account.  -- Send the information to:  Marc Reusser at rbadesign@earthlink.net -- Russ Smiley
 
   Home   Help Search Login  
Pages: 1 2 [3] 4 5 6 7
  Print  
Author Topic: 3D Printing  (Read 9994 times)
W.P. Rayner
Guest
« Reply #30 on: May 24, 2011, 11:33:23 PM »

Marc:

I figured there had to be an adjustable setting for that, but having tested SU only briefly (out of curiosity), I wasn't sure. It's a basic preference in any CAD program I've ever worked with.

Ed:

As mentioned above, I'm not familiar with SU, being perfectly content with the CAD applications I use. I always do my drawings full-size, 1:1 and then let the computer do the scaling for me. I've never had a problem with losing any detail in the drawing, but then that might be a limitation specific to SU. The 3D CAD app I use will let me draw radii of .0001" if I want, though I don't. For rapid prototyping, I keep in mind that the minimum feature size from the Invision printer is .003" so I don't try to replicate anything that would scale out to that size or smaller. High-resolution Viper machines will print features down to .0016" but the cost is much greater.

The key to smooth prototyped curves is the STL file resolution. The fewer geometry segments (flat triangles actually), the coarser the curve. Increase the number of triangles and the geometry becomes smoother but at the expense of file size. The more triangles in the STL file, the larger the file size. Most STL generators address this issue by calculating the optimum number of triangles to best approximate the surface geometry without creating an enormous file. This setting usually can also be adjusted for critical work.

Paul
Logged
eTraxx
Hero Member
*****
Posts: 801



WWW
« Reply #31 on: May 25, 2011, 09:06:55 PM »

Bored .. I played around with the gate valve. I completely reworked the flange to make it meet ANSI standards .. put in a little separation between where the bits bolt together. Paul .. I bumped up the segments of a circle from the default 24 to .. umm .. 40 I think I used.

I amused myself by rending them in brass in Kerkythea.

The pic on the left (No.3) is the 'cosmetic' gate valve. This one has a close to correct ID. I modified the thing so I could use Evergreen tubing which means I had to 'bend the rules' just a bit. A 'normal' 6" pipe (or valve) .. that's the inside measurement. Standard pipe (SCH 40) for a 6" pipe is about 6-5/8" OD. Oh well .. Evergreen has 1/8" tubing which equates to 6" OD in O scale. I can live with that.

The one on the right (No.4) is 'operational' .. no .. it doesn't actually work .. ha. I mean that it has the bolts on the inner flange .. and that little projection will fit another flange .. or the ID of that 1/8" Evergreen. Figure I might go ahead and create elbows etc. There's nothing I've found that I like .. some white metal castings. The Plastruct stuff I used in the pump house is very simple .. more of a generic shape of pipe I suppose. I'm thinking if I can supply all the 'plumbing' in this size piping I could fill a void in the hobby.

Logged

Ed Traxler

Lugoff, Camden & Northern RR

Socrates: "I drank WHAT?"
eTraxx
Hero Member
*****
Posts: 801



WWW
« Reply #32 on: May 27, 2011, 01:37:13 PM »

If you got the cash .. FineLine Prototyping does some very fine work .. but you would probably need to take a loan out to afford it .. ha

Logged

Ed Traxler

Lugoff, Camden & Northern RR

Socrates: "I drank WHAT?"
Design-HSB
Full Member
***
Posts: 214


Klosterstollen


WWW
« Reply #33 on: May 27, 2011, 02:52:37 PM »

I think all of us but it should be even more important, some even produce their own resources and tools.
CNC with us model builders will not help if you do not understand and the process of the craft.
Logged

Regards Helmut
the journey is the goal
DaKra
Hero Member
*****
Posts: 604



WWW
« Reply #34 on: May 27, 2011, 03:25:56 PM »

Ed, do you have any actual information on Fineline's pricing for smooth finishes? 
Logged

VectorCut.com
eTraxx
Hero Member
*****
Posts: 801



WWW
« Reply #35 on: May 27, 2011, 03:56:56 PM »

Dave,
Ok. Just uploaded this to Finelines



Quote for ONE in the In their Invision HR (set up kills you)


Quote for TEN in Invision HR


Quote for ONE in MicroFine Green


Quote for TEN in MicroFine Green
Logged

Ed Traxler

Lugoff, Camden & Northern RR

Socrates: "I drank WHAT?"
marc_reusser
Curmudgeon
Administrator
Hero Member
*****
Posts: 3884



WWW
« Reply #36 on: May 27, 2011, 05:40:55 PM »

Interesting info. thanks for taking the time Ed.

Marc
Logged

I am an unreliable witness to my own existence.

In the corners of my mind there is a circus....

M-Works
fspg2
Full Member
***
Posts: 120



« Reply #37 on: May 28, 2011, 01:34:30 AM »

There are various resolutions at the company where I let prototypes:

0.0127 mm (equivalent to approximately 78.7 shifts per millimeter)
0.0254 mm (with this stage were plotted my part) - that is about 39.4 shifts per millimeter)!
0.0381 mm (for large parts whose edges can easily be smoothed ) (about 26.3 layers per mm)
0.0508 mm (about 19.7 layers per mm).

So you get a thickness of 0.0127 mm more or less from stage to stage of solution. The more layers, the longer the machine runs and correspondingly expensive is such a part.

One example: The buffer for the Diema_DL8 (German light railway) cost € 60.00 in prototyping (3D-wax plotting). Then add up the cost of the first brass casting (€ 5.10) and the postage.



The picture shows the first prototyping part made from the wax-plott. Itīs raw, I have to plane the steps you can see in this backlight.

This way (wax - brass model - rubber) may be a bit expensive, but by the reworking of the first cast the steps may be smoothed. Likewise, further mounted thin parts, which were not able to be prototyped, for example because it were too thin.

From this part a rubber mold will be created for the serial casts (depending on the size of around € 25.00 - € 30.00).
Lastly, then the cost of the casts themselves can depending on material and weight between € 2.50 € to € 4.00.

I attached a small cost table in which each of you can gain an overview. To create less equal parts, then the cost are still very high, but the quality is accordingly.



For the way above, I have to push the size of the stl-file with a factor of about 1,04. The parts shrinkage in two stages during production. The first time during the molding of the wax model, then through the rubber mold. So I get my model in the right size at the end.

Here an other part for my Gmeinder 10-12PS:





I donīt know if a model builder would hammer out such a part of the same quality so ever - and then puts the price.
With each piece, the retail price relative to the bottom.

In my list above I have, however, included no costs for creating the drawings and which has been to realized in STL format.

Drawing is still a part of my hobby, we are rewarded by a precision I donīt want to miss.

Frithjof

« Last Edit: May 28, 2011, 02:07:08 AM by fspg2 » Logged

Frithjof
Hauk
Sr. Member
****
Posts: 442



WWW
« Reply #38 on: May 28, 2011, 04:01:22 AM »

There are various resolutions at the company where I let prototypes:


Frithjof, what company do you use, and what type of machine do they use?
It is a Solidscape wax-printer?

Those masters look really great, by the way!

Regards, Haavard H
Logged

Regards, Hauk
--
"I believe in the wasted years of work." -Wislawa Szymborska

Remembrance Of Trains Past
fspg2
Full Member
***
Posts: 120



« Reply #39 on: May 28, 2011, 05:03:01 AM »

@Hauk
I donīt know exactly what type of machine uses this company, sorry! Please ask Mr. Wild directly.


Frithjof
« Last Edit: May 29, 2011, 07:53:11 PM by marc_reusser » Logged

Frithjof
DaKra
Hero Member
*****
Posts: 604



WWW
« Reply #40 on: May 28, 2011, 07:41:43 AM »

Very interesting information, thanks!   The prices look OK in a commercial context, e.g. as a tooling investment that will be paid off. 
Logged

VectorCut.com
davej
Newbie
*
Posts: 32


WWW
« Reply #41 on: May 28, 2011, 08:15:53 AM »

The photos look like the output from a Solidscspe machine. I am also using these.

Frithjof, how big is the buffer block? I will have to send them some of my drawings to see how the prices compare to UK suppliers.

Logged
eTraxx
Hero Member
*****
Posts: 801



WWW
« Reply #42 on: May 28, 2011, 06:18:09 PM »

Free. Autodesk 123D .. FREE .. 3D Modeling Software. I found a link to this on the Shapeways forum. The one thing that they made a point of .. as it relates to uploading to Shapeways is that the program exports in cm. Since the choices at Shapeways for the measurement used for your upload is Meters/Inches/MM .. you have to scale by 10x before exporting. I haven't' tried it yet .. downloading as I type.

http://www.123dapp.com/downloads
Logged

Ed Traxler

Lugoff, Camden & Northern RR

Socrates: "I drank WHAT?"
fspg2
Full Member
***
Posts: 120



« Reply #43 on: May 29, 2011, 03:16:10 PM »

@ davej,

here the dimensions of the buffer: high: 11.33 mm, width: 24.44 mm, deep: 7,34 mm (scale: 1:22.5).
Let us know about the price in UK!

Frithjof
Logged

Frithjof
narrowgauger
Full Member
***
Posts: 105


« Reply #44 on: May 29, 2011, 09:08:15 PM »

Hi Guys,

this is quite en interesting debate with great outcomes.

It strikes me however, particularly when looking at the costing in the table, that we need to keep in mind that this is a case of "horses for courses".  For example the coupler / buffer block in the example is identical to one of my standard 1:35 parts, the original took about 40 minutes of machining & fabrication time, including the bolt heads, all absolutely to scale, ready for the rubber mould.  When this is compared to computer time, 3D printing lead times etc it would seem that the 3D approach is fabulous for complex parts, but is a bit over the top for simple components.

it also occurs to me that if we take the 3D application too far the model would inevitably loose the essential character that is imperative in replicating the original.  We go to the endth degree to weather and age our models, yet we seek perfection greater than the prototype in the part.  Not sure if I understand this trend line in the search for perfection, when I look at the prototype.

Bernard
Logged
Pages: 1 2 [3] 4 5 6 7
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.13 | SMF © 2006-2011, Simple Machines LLC Valid XHTML 1.0! Valid CSS!