Design and Re-Design
There was a very good discussion on the design of a turbine in a thread on the Railroad-Line forum back in 2011. You can find it here – The Caddo Creek Dam – I may repeat .. probably will repeat some of it as it becomes applicable.

Russell Green of “New England Brownstone” was a major contributor to the discussion – a lot of follows comes from him – modified as needed for a model.

Original Design
One thing to keep in mind is that this turbine isn’t very large when modeling i O Scale. To be true there were very large turbines that would be more substantial if modeled but this turbine I am working on is rather – small. The casing is only about an inch in dia. representing a 4-ft dia casing.

Here’s the design as it came out of that discussion on RRL and another on TWP. A casing about 4-ft in dia contains a turbine wheel about 16-in dia. Either end of the casing ends in a large metal cap with a bearing for the turbine shaft. In the drawing the near end also includes the mechanism for controlling the flow of water through the turbine .. a throttle I suppose. On the other end is the discharge pipe. In the center is the intake.

The intake I will freely admin is undersized. It’s sized to fit the Evergreen tubing I had on-hand at the time but in reality should be sized closer to the size of the turbine casing itself.

Problems with Ver I
The original design of a turbine as seen in the drawing above was cobbled together from books, photos and discussions on the RRL and TWP forums. This is all well and good and I think that the design ‘looks good’ on paper (or a computer screen) but has flaws when trying to model. A one-off model isn’t the problem .. but I was wanting to 3D print some of the parts with the intention of selling a small kit. That leads to a re-design. I’m going to talk about the problems as I see them and the ‘fix’.

  • In the drawing I have rivets holding the end castings to the casing. That would work fine if I was using resin decals to model this – but – not for a 3D print. Not at least .. a 3D print that I can afford. In this and all other references to 3D printing I will be using FUD (Frosted Ultra Detail) as the material. This Acrylic Polymer material is available from Shapeways and provides for me, the best cost/value for 3D printing.

    FUD has a minimum detail thickness of 0.1mm and a minimum wall thickness of 0.3mm. There are other design rules that have to be followed but those are IMO the important ones.

    I found it hard to get a ‘handle’ on just what “minimum detail thickness of 0.1mm” meant. I’m going to make a try though. If we zoom in on where the casing rivets to the end caps we see that the rivets are .437-in dia. This is nothing more then what ‘looked right’ at the time. It’s easy enough to do this in Sketchup and again could be replicated using resin decals such as those that Archer makes but will cause problems when trying to print in 3D – specifically FUD.

    To make this so I can understand it .. let’s see just what attempting to print that rivet in FUD would end up like. The rivet as I said, as created in Sketchup is .437-in dia. That is 0.009-in / .232-mm in 1:48.

    With that 7/16″ rivet in blue I then made a box in green representing that 0.1mm minimum detail that FUD can print. I think that what you would get would be a ‘blob’ of about four of those green boxes together. They would probably flow together actually to represent the rivet pretty well to the naked eye – but the limitations are obvious.

That got me to try and think ‘outside the box’. If FUD can’t reliably print rivet heads this size – then what else? That led to “why even print the casing at all? If instead of the casing sliding and riveting over the ends – I could instead have the casing slide inside the ends. I could then move the rivets/bolts to those end caps.


Turbine — 1 Comment

Leave a Reply

Your email address will not be published. Required fields are marked *

Heads up! You are attempting to upload an invalid image. If saved, this image will not display with your comment.