I subscribe to Kieth Rucker's You Tube channel and saw the video First thing this morning. It was a REALLY amazing demonstration. You could see the process in real time with the scanned image appearing as the laser unit was played over the part. And the process is applicable to nearly any sized object. including aircraft. Watch the video!

So, now that he has made a new flange, I guess the challenge would be to get it welded to the organic shaped tube in exactly the proper relationship to the original flange on the opposite end of the tube.

So next, he would convert the scan data into a Solidworks model, and the design a weld fixture in relation to that model. The weld fixture would locate and position the new flange in perfect relationship to the original flange which is still part of the tube. I suppose the weld fixture would just be a frame that presents four threaded studs with shoulders that fit the holes in the two flanges.

Actually, it would have to be a little more complex than that because the studs would have to be made to individually withdraw from the mounting holes in order to get the welded pipe off of the fixture.

The orientation of the new part is not super critical except for the alignment of the holes for the studs. The fit at the top and bottom is made with spherical rings so that a good seal can be maintained while providing sufficient flexibility between the saddle and the boiler steam header. Helped with steam pipe installation on the Saginaw Timber No. 2 and my only comment is that it is dirty, sweaty heavy work inside a cramped smokebox. And the 2 is much larger than the dinky 0-4-0 at Tipton.

I would love to know how the foundry made patterns and cast those snake-like steam pipes in the first place! Nowadays they would be welded up from sections of heavy steel pipe.

The alignment of the hole studs and steam passage openings is what I am referring to. Maybe they could just bolt in the lower flange and then bolt in the upper flange with the tube still attached. Then if the fit looks okay, just weld the new flange to the pipe end. But then why scan the part?

I think there are a couple of reasons for scanning the pipe.

1. Someone wants to show off their shiny scanning machine.

2. If the repair fails, you have a start on a new pipe.

3. Help to build a fixture to hold the parts in alignment for the brazing repair.
(weak idea, I would try to at least tack it together in the locomotive to help with the alignment.).

-Hudson

Yes, using the existing ports of the header and cylinder casting as the fixture seems like it might work. Otherwise, if a weld fixture were built from a Solidworks model derived from the scanning, that would be quite a large task to design and build the fixture.

According to the videos, the immediate goal is to repair the flange on the steam pipe to return the locomotive to operation. This is being done during an annual scheduled maintenance period. A local school wanted to tackle casting a replacement and a 3D drawing has already been prepared. The school is not in session now, so they aren't able to use the old part to develop a new pattern. The scanning outfit wanted to show off their capabilities and Keith thought it would make an interesting video, which it did.

I'm waiting for next week's video to show how the alignment will be done to braze on the new flange. My guess is that the pipe will be bolted in place and at least tacked together while it's mounted on the locomotive.

Im with you, Dick! Mr. Rucker is nothing if not practical.

It made me a little nervous watching that scanning process with the part so delicately propped up for support while reaching around it to paint it with the camera. But I suppose it would not make any difference if the part were moved during the scanning. Or would it?

I assume that the system must know when it has captured enough data to fully define the part. Does it communicate that fact to the operator?