[Jack - Freshwater Models]

Please bear with me as this is my first post with a pic


This pattern was digitally modeled in Rhino4 and then 3D printed. Next this pattern will be molded in RTV rubber to produce Britannia metal production patterns. The Britannia metal patterns will be used to produce a vulcanized silicone rubber mold. The drivers will be cast in zinc-aluminum alloy. This is an experiment to come up with an easier and more cost effective way to produce wheels and drivers for G1. The prototype is an 0-6-0T shunting/goods loco. The model will be based on LBSCs Chingford Express but with some improvements I hope.

Jack

 

[Jeff Livingston]

Jack,

I'm not sure I understood much of your description but that's one fine looking wheel. Did the class have outside counterweights?

Jeff Livingston
Kaneohe, Hawaii

 

[Jack - Freshwater Models]

Jeff,

Thanks! No, this particular loco didn't have counterweights as is fairly common with locos like this in the UK. The passenger version of this loco had 10 spokes instead of 12 and did have counterweights.

To simplify, this plastic wheel was computer generated. It will be used to make molds to produce metal wheels. Hope this clarifies a bit.

Jack

 

[Dwight Ennis]

Today's technology is amazing! Good job!

 

[Gary Armitstead]

I saw this technology used at Cal State Los Angeles during one of my MasterCam seminars about ten years ago. Is this the unit that is "somewhat like" an ink-jet printer? Less expensive than Stereo Lithography. That will be a beautiful pattern. Thanks for sharing.

 

[Jack - Freshwater Models]

Gary,

Thanks for the compliment. The digital model, a 3dm file, is saved as a stereo lithography (.stl file) and sent to the printer. The printer is based on ink jet technology but laying down plastic in 3D. The goal is to test the viability of zinc-aluminum drivers or driver centers which are relatively cheap and easy to produce. I've given up on having small quantities of cast iron drivers made.

Jack

 

[Gary Armitstead]

Jack,

Does Rhino4 do surface modeling or solid modeling. Many years ago, at a local college, my professor and I built seven 1/8 scale Baldwin Electric locos. All the sheet metal parts for the cab and floor, the trucks and wheels were drawn in MasterCam. The wheels were all done using surfaces. We then made our patterns and had them cast. Lot of work!

 

[Jack - Freshwater Models]

Gary,

Rhino is a surface based, solidmodeling software. It's similar to Solidworks which is "feature" based. Rhino is very affordable and has almost a "cult following". Mastercam is awsome but increadibly expensive. I use Rhino with Rhinocam (Visual Mill) to accomplish almost the same stuff. I hope to someday get Visual Turn for CNC turning.

This driver was developed mostly from revolved surfaces and lofted surfaces. First the spoke was lofted. Then the various surfaces were revolved. Next the spoke was coppied and revolved. Finally the hub was lofted and all the surfaces were joined/trimmed. The last step was to make sure that the model is water tight using an analysis feature of Rhino and correcting any problems.

I have a small mill which I retrofitted with CNC but accuracy is limited due to acme screws rather than ball screws in the X-Y axis.

I did all this because because I have the capability and it is fast but the same could have been done with wood or plastic using manual machining/fabrication methods.

Jack

 

[Pete Thornton]

the viability of zinc-aluminum drivers

Jack,

The wear characteristics should be interesting, or are you planning to add a steel tire?

When you've done that, Kevin has a 48" EBT driver that we need in 1:20.3.

 

[Jerry Barnes]

Nice to see someone doing wheels. Where are you located?

 

[Jack - Freshwater Models]

Pete,

Yes, I will be interested in the wear characteristics as well. As you may know Aster used ZA on their little oscilator 0-6-0 for the wheels as are Mamod wheels though they are diecast. They had problems but I think they were mechanical fastening to the axle not the fault of the alloy. If the ZA wears too much I'll remake with CI tires. Interestingly, in industry ZA is used as a replacement for CI for all sorts of parts including bearings. The EBT wheels might be reasonable if you use tires for such a heavy loco. I would have to see a drawing.

Jack

 

[Jack - Freshwater Models]

Jerry,

Thanks! Actually this driver isn't exactly as it should be having fillets too small and spoke cross section not entirely correct but close.

Ideally I would like to make CI wheels but it is beyond my capability and small quantities are tough to get investment or shell cast. ZA may be a reasonable substitute. I am going to try to model some nice wheels for both British and American rolling stock.

At this time I still have a lot to learn about digital modeling. I especially need to learn more about how to join two surfaces together with proper fillet as done in full scale patternmaking. It all gets very complicated very fast and the learning curve is monumental for me.

I'm in Rochester, NY a couple of blocks from Lake Ontario and the Genesee River. Once a very important railroad area with the NYC, HoJack, and BR&P (B&O) not to mention a couple of trolley lines.

Jack

 

[Gary Armitstead]

Jack,

If you are having problems with the filleting, maybe you can send me the file in DFX and maybe I can import it into MasterCam, do the fillets for you and send it back. Does Rhino have the capability to import from MasterCam? It might be worth a try. Let me know. I have a buddy in Ventura, CA that is familiar with importing from Rhino to MasterCam.

 

[Jack - Freshwater Models]

Gary,

Thanks for the offer! If I get stuck I might take you up on it!

Actually I just have to learn more about Rhino! The fillet on the tire end of the spoke is there but too small (next time). The spoke/hub is the tough part and I will be playing with it on the next driver I design. I don't have any formal training and have been learning through tutorials and trial by fire! One reason the spoke cross section isn't the correct pear shape is that it is easier to cast the way I designed it. Drivers are nasty things to cast! Thick sections like the hub and tire next to thin sections of the spokes. Any cross section I can gain in the spokes makes it easier to cast. I will be adding a feeder on the back of the driver from the gate to both the tire and hub in an effort to feed both during solidification. Yes, I have to consider how the part will actually cast in designing the part. I am always trying to design taking into account progressive solidification.

Jack

 

[Jack - Freshwater Models]

This is the first try at a production pattern. Sadly it ended up about .010" out of round due to centrifugal force. Will try another method!




Jack

 

[Semper Vaporo]

Cast it too big and turn it true on a lathe.

 

[Jack - Freshwater Models]

Hi CT,

I wish it were so simple! Actually, it isn't far off and if it were cast in zinc-aluminum it could be used as a driver since the casting is oversized and .010" isn't too bad. Thickness was held within .003" which surprised me! The pattern has to be very accurate because it will be used to produce another mold that will be used to produce the actual driver castings. If I were to use this one the error would be multiplied in using the second mold. The idea is to do everything as accurate as is reasonably possible to produce quality driver castings.

My plan is to make another mold and gravity cast the pattern which will be more accurate.

Jack

 

[Jerry Barnes]

Jack,
If you get going on this, put an ad in Steam in the Garden magazine, one of the best places to contact builders. Published right there in New York!
Jerry

 

[Gary Armitstead]

Jack,

That "shot" of the wheel looks real nice! Is your process similar to die-casting? Is there a "shrink factor" used in your process. Is there a thin layer of "scale" on the surface of the casting after cooling? With my limited experience with zinc-aluminum and die-casting, you should be able to hold tolerances of plus-minus .001 without a problem. Thanks for the update on this.

 

[Jack - Freshwater Models]

Gary,

Not as precise as diecasting. The molds are silicone rubber. The casting is very clean as it comes from the mold. One advantage is that flaws that are usualy hidden in a diecasting are out in the open in spncasting. On the down side spincasting in rubber molds doesn't have the mold pressure as in diecasting. Though the molds don't hold up well they are cheap so they can be easily replaced once the mold shows signs of ware.

Jack