G Scale Model Train Forum banner

Stationary steam engine model

10K views 36 replies 10 participants last post by  CliffyJ 
#1 ·
As some of you know, I'm a total newbie to how steam engines work, but I've been modeling a stationary engine anyway. So thanks again to everyone who helped me with questions on prior threads.

Here's the state of the (computer) model design so far.

http://www.youtube.com/watch?v=OxZ361DxK50&feature=youtu.be

The Stephenson valve gear has been been quite the (interesting) challenge, but I think I've finally got something at least vaguely workable. It looks pretty chunky, but I need to make the parts 3D printable.

The intent is to make all the white bits with 3D printing, and cut the shafts from brass round bar. The screws are #0 and #1 lag screws from Micromark.

As for functionality, the gearmotor (beneath the "cylinder") is tiny; and if all goes well, it will drive a small grooved friction wheel that acts against the OD of the flywheel -- which is a square o-ring.

I've no idea if it will work at all, but I'll keep you posted.

===>Cliffy
 
See less See more
#3 ·
Your drawing is very nicely done... but my being a really, really nitpicky person... two things: (not that anybody but me would notice!)

The toothed gear and the flywheel appear to be on the same shaft, but seem to be going in opposite directions and at different speeds. I guess it might the stroboscopic effect that makes the gear appear to be going in the opposite direction than the flywheel, but the speeds should not vary if they are rigidly on the same shaft. (Yes, the car commercials on TV where the advertiser has spent the money on animation to make the cars morph from one model to another over and over, while they fly over, under, around and through various barriers, and yet they could not make the wheels turn in the direction the cars a moving, really disturbs my sensibilities!)

The eccentrics do not seem to be at the correct angles in relation to the crank pin. One should be 90° (plus/minus a "lead" amount) ahead of the crank pin and the other the same number of degrees behind the crank pin. (Yours appear to be at 0° and 180°.) The "lead" amount will be about 28° for a cutoff of about 80%, given nominal dimensions of valve ports and other parts. Whether the "lead" should be added or subtracted depends on some other factors of the valve gear design that I have not taken the time to work out in your particular construction, but normally the angles will be +/-62° from the crank pin.
 
#4 · (Edited)
Thanks Jerry, if i get it working, I'll bring it to our next meet. Maybe you'd like one for a mine or sawmill near you? ;0

Semper, I really appreciate your comments. First, yes, it's a strobe effect of the 15 fps video; they're locked to the same shaft. But I didn't consider your main point, and it's confirmed in the main valve gear diagram I've been using:

Text Technical drawing Drawing Diagram Parallel


The small circle labeled "crank pin" I'd overlooked. And indeed, the eccentrics appear to be +/- 62 from that. Great catch!

Cliff

[edit]

Here's a vid of the corrected angular positions:

http://www.youtube.com/watch?v=zvuZhVYhDNY&feature=youtu.be

The transparency thing didn't work out, but you get the idea.
 
#7 ·
Craig, I'm afraid I don't know Sketchup; I use SolidWorks at this point. But someday (when I retire and no longer have access to a SW license), I'll have to do Sketchup. So I'm hoping it can do as well.

Jerry: good deal, that gives me a good bit of motivation! But I won't hold you to anything until it actually works, ha ha!

Now, a real $$-maker would be that flanger... what was it, $700 from that store? And out of production to boot (Accucraft, I think?). I think I can get my hands on some reasonable plans... that would be another nice winter project I think.
 
#8 ·
I've been gradually editing this model for 3d printing and buildability (with screws, brass tubes, rubber friction tire for the flywheel, etc.), and here's what it looks like.

Architecture Urban design 3d modeling Arch


For getting a quote on the 3d print, all I did was turn off all the non-3DP bits, leaving the separate parts as they floated in space, and generate the STL file. However, even after thinning things up as much as I could, the model from Shapeways was gonna cost $112.

But, after reading more, it turns out that they have a "density discount" for models over 20 cubic cm. Long and short of it, I squeezed the parts together to lessen the "bounding box" and get my part density above 10% for that box. There's an article on this,

https://www.shapeways.com/blog/arch...ificant-price-reduction-on-dense-models_html=

Anyway, here's the more densely-packed kit:

Auto part 3d modeling


The result of getting those voids out (and allowing this "density discount" to kick in) was that the price dropped to $70 -- for the same parts!

===>Cliffy
 
#9 ·
Cliffy,

Beautiful model. I have a couple of questions for you about the files you sent Shapeways.

Below is my MasterCam model of the PE pneumatic trolley pole I want to make for my 1 1/2" Baldwin electric.

My model is all surfaces, not solids. Will Shapeways accept this file if converted to DXF or DWG?

 
#10 ·
Hey Gary, thanks.

According to Shapeways:
"Shapeways accepts designs in STL, OBJ, X3D, DAE, Collada or VRML97/2 (WRL). "

So you'll have to convert the file.... You can send it to me, and I'll have a whack (at converting to STL thru Solidworks), or maybe you'll find a free tool of some sort. Mexhlab doesn't take DXF; Sketchup will, but only the Pro version. But maybe we can find something else.

Cliff
 
#11 ·
Thanks for the info Cliffy. MasterCam has a converter included to convert to STL. I was more concerned about my drawing using nurbs surfaces as opposed to your drawing using "solids". I could possibly convert my surfaces to "solids". I have that capability in Mastercam. I know Solidworks always wanted "watertight" solids! :)

I'm awaiting photos that a friend took a couple of weeks ago at Orange Empire Museum of the pneumatic trolley pole on the top of #1624. With these photos, I can finish my drawing and get the files to Shapeways to get a quote. I'm thinking about getting these parts printed in brass! How cool is that?:)

Thanks for your help. I may need somemore feedback from you when I get my quote.
 
#12 ·
I'm glad to hear you can output STL easily, Gary. That way, if need be, you can tweak and re-upload quickly.

And that part will be way cool in brass, make sure you post pics!

It ain't cheap though, as I'm sure you know, but it's pretty neat that they offer that.

Something to consider though, a way to save money is to convert regular cylinders to simple brass rod or tube.

The way I did it on this project went sort of like this:

1. Make the model, get it all complete. Base model cylinders on tube / rod I can easily get a hold of or already have.
2. For each cylinder, make cuts that not only remove the cylindrical model surfaces, but create holes (pockets for the actual brass rod or tube).
3. Order the brass, and 3D print everything else.

Just a thought. Either way, have fun!
 
#13 ·
You read my mind.:)

For the cylinder parts, I will use brass rod and turn on my lathe to get the rings on the outside and drill for the pivot shaft and install a smaller brass rod for that. The pieces I really need printed are the odd shaped yoke supports for the trolley pole. Difficult to machine the contours. Easily printed though. This will be an interesting project. Thanks for the inspiration.:)
 
#14 ·
One other thing Gary,

Each material has it's own "fuse" distance. That's the minimum gap for parts to remain separate. With strong/flexible plastic (as they call it), that gap is 1/2mm.

So... folks can and do put gaps in, get the thing printed, and now you have parts that can move relative to one another, without further assembly.

Your mechanism made me think of that, though you may not want moving parts.
 
#17 · (Edited)
Hi Semper,

The gap applies in any dimension, as far as I know. This Shapeways "strong and flexible plastic" material is a powdered nylon that gets fused or doesn't; and the unfused stuff supports the rest. There doesn't need to be supports coming up from a base of some sort (unlike the process where the laser hardens a fluid). When the run is done, they shake all the loose powder out, and what's left is the part.

Here's a video, one of many on this (laser sintering) process:



This next one is a cool demonstration of interlocking moving parts, right out of the printer. The hinged parts that the woman is showing come out that way. But, if the gaps between the parts was less than the recommended clearance for the particular material, the bits would be locked together.



Here's a video of the analogous process for metal. It's easy to see why it's so much more expensive!



I hope that helps,
CJ
 
#19 ·
Hi Semper,


Here's a video of the analogous process for metal. It's easy to see why it's so much more expensive!

https://www.youtube.com/watch?v=i6Px6RSL9Ac

I hope that helps,
CJ
That's not the only metal technology out there though. There is at least one brand of system that works exactly like the plastic machine shown in the previous video. I got to look one over (an ARCAM) when I worked at ORNL. They were making titanium parts with no additional binder which were up to something like 95% solid, no brazing involved.

The technology is growing at an extremely rapid rate and really awesome stuff is coming out of these machines. :)

Trot, the technical, fox...
 
#18 · (Edited)
Monday, I took the plunge and ordered all the bits. Which means, I can't change the model anymore... and dang it, I found my first mistake a couple minutes ago...

Anyway, this evening I've been working on a color scheme. I've seen a few examples on the web of painting the main base casting, but leaving the working bits bare metal. Also, it seems like the flywheel was generally painted red -- I assume for safety purposes. I applied three shades of "iron" and "steel" to the remainder. Brass will stay unpainted, just because I like brass. I won't be able to do anything with the black rubber flywheel tire, so it'll stay black.

Steam engine Auto part Engine Wheel Vehicle


Auto part Product Engine Automotive tire Tire


For the foundation, though I've colored it as concrete, I'd prefer a brick-like finish, perhaps some Plastruct sheet as a veneer:

http://www.plastruct.com/pages/OnlineProductDetail.lasso?-op='eq'&CCode=PS-90

So that's the direction, but I'm all ears for further color logic on these kinds of engines. The only historical pics I have happen to be in black and white...!

===>Cliffy
 
#21 · (Edited)
:) Just got the parts back from Shapeways, here's what they look like.

[edit: trying to get the photos to work...]

Scale model Vehicle


The small (.03") holes are filled in, and I expected some of that. There are dimples though, and that's all that's needed to drill the holes out.

Though the washer & nut details are blunted, they're so small that I have to use a magnifying glass to really tell:

Automotive wheel system Rim Wheel


I figure, if I can't see the detail with the naked eye, then that's probably going far enough. But that's just me, ha ha! :D

===>Cliffy
 
#25 ·
Cliff,

Coincidentally, I am embarking on the same journey - albeit with SketchUp. Thanks for the info about separation tolerance for moving parts (.5mm for SFP). So for a 1mm rod, you need a 1.5mm hole? Likewise, a 2mm rod needs a 2.5mm hole? Also, Shapeway states that for parts that are to ahdere to one another (e.g. wheel on axel), a separation tolerance of .01mm is recommended. Is this your experience?

In advance, thank you, John
 
#26 · (Edited)
Hi John,

How cool, you're doing a stationary engine? What type? I'd love to see your progress, as you see fit.

For running shaft fits, It sounds like you're 3d printing both shaft and hole-part. And for that, if you want to leave your parts "in place" with the .5mm gap, they say it will print -- and come apart, without fusing. Not sure about long sliding fits, like a piston rod. For part adhesion, I've only modeled the shaft and hole to the same diameter.

If your'e interested in smoother or more predictable fits, you might want to "pull the parts apart" before printing (optional), and model the diameters for the results you intend. I say "might," but you might not. Anyway, to try to get a grip on SFP hole & shaft shrinkage / growth, check this current thread out,

http://forums.mylargescale.com/15-model-making/32609-sls-hole-shaft-gauge.html

For tolerance reasons, but also for cost reduction, I decided to make all my shafts and pins from brass tube (or rod). From prior parts, I knew that the SFP holes would come in slightly undersized, and didn't correct for that. Instead, I assumed drilling for all bores, at about 1/64" greater diameter if needed. So that's an option.

Good luck, and please start a thread a some point, I'd love to see your project!

===>Cliffy
 
#27 ·
Cliffy,

Thanks for the response. I have no immediate plans to create stationary steam engine models but I am interested in using my CAD skills to produce highly detailed steam locos and rolling stock.

For moving parts, I plan to use brass rod as well. Rather than it rotate against the SFP, I plan to insert a piece of brass tube between the rod and the SFP. My question is, how much larger than the tube should the hole be in the SFP? For parts that will lock, I will likely 3d print both parts. I want a tight fit but be able to reliably and repeatably dis-assemble and re-assemble those parts. In your experience, what should I use for the difference between 2 parts such as that?

In advance, thanks again.



john
 
#28 ·
Hi again John,

In that "gauge" thread, I posted today some findings that may come to bear on your design. Basically, the SFP holes & shafts are coming out like SW said, +/- .5mm. And it's not really predictable. So for any fit you care about, I'd go slightly undersized (.006 or more), and plan on drilling it. That way, at least you can count on one part being predictable.

So for a running fit, with brass rod, you'd probably drill to the nominal, and see how it feels. Then if needed, for looser, go 1/64 over that. At least, that's what I did yesterday. For locking, it's vary from batch to batch. So plan on drilling your hole, and make the shaft a bit more than that -- say, .008. So if the shaft comes in undersized, you'll still have some nominal interference.

===>Cliffy
 
#29 · (Edited)
Well, one year later, I'm working on this again. Here's what I'd hoped it would look like, after painting:



I wanted to have something to bring to the "drag and brag" at ECLSTS tomorrow evening. My initial choice was a fail (haven't received the Shapeways parts for my switch drive project). So, I turned to this unfinished project last weekend, and here is the result.





There are many adjustments to make to the design, *sigh*, but that's life.

Here was my big surprise though. I wasn't surprised that the parts screwed together. But it was almost magical when the motor ran, and the valve gear parts moved the way they were supposed to. Though I've researched Stephenson valve linkage a lot to do the modeling, seeing it work was very cool, and informative.

Ideally, the reversing lever should work, not just be modeled in statically. But, that seems too much to ask for...

Anyway, the motor is way too loud. So the main next thing on this is to redesign around a continuous-running servo.

Here are two animations from a year ago,





And here is the actual model. Please turn your audio off or put your fingers in your ears, haha!



Thanks for viewing,
===>Cliffy
 

Attachments

This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top