Pennsylvania A3 Switcher boiler thickness issue

[mdavey]

I'm about three weeks into working on the Pennsylvania A3 Switcher boiler when I had a moment. I accidentally made the five bulkheads from 0.062" copper, not the 0.080" copper called for. I kept thinking the diameters were coming out a bit small, but wrote it off to my formers being a bit small. What are people's opinions, can I use the bulkheads as they are at 0.062" thick and the rest of the boiler shell at 0.080" as called for? Or, do I need to scrap the bulkheads and start over.
I see some boiler designs of this size with 0.062" thick material. My gut tells me the bulkheads are more reinforced and likely more overdesigned than the shell. Or maybe that's me not wanting to scrap a few weeks of work. I'd love to hear opinions.

[crueby]

Better to find out now rather than after soldering! I just looked it up at the back of Kozos New Shay book where he has boiler design tables. According to those tables, the thinner bulkheads cut the maximum distance between stay bolts and other supports nearly in half. That tells me that you likely won't have  enough strength in the bulkeads without changing a lot of other things, and should redo them out of the thicker sheet. Now, I  am making that statement without seeing the design, but thats a big change. Flat bulkheads will bow out without supports close enough together. Good thing is your formers are fine, and you now have the practiceof using them, but it means recutting and bending the bulkheads.


Chris

[Kim]

I'm sure you could figure out how to do the analysis and determine if the thickness will make an appreciable difference.  But as Chris said, it's better to find out now, rather than later.  Finding out later that the 1/16" bulkheads aren't strong enough could be disastrous!  And, like Chris also said, Kozo put in the minimal number of supports on the bulkheads.  It may be enough, then again it might not be.  I'd err on the side of caution and do them again, as daunting as that is.

If you're planning to run this on a club track, you should check with them to see if they're OK with the changes.  I don't know exactly how that works because I never plan to run mine on anyone else's track. But I still wouldn't want to find out the hard way that the thinner bulkheads weren't enough 

Kim

[mdavey]

Thank you for the input and a special thanks to Chris for pointing out the New Shay book, a book I have, and the charts at the back.  I had either missed or forgotten, with forgetting being the most likely, that all that boiler information was in the book. Now at least I can put some real calculations to the design and see one way or the other how big an issue it is. I am leaning toward redoing the bulkheads, it's just hard to accept when you first stumble on a problem like this. Sometimes the mind needs a day to settle into the idea of redoing a few weeks work.
Mike

[crueby]

Great  that you have the book and can do the calculations,  it would be interesting to see how much  leeway he built into the design. Either way its all good experience!

[Kim]

...I am leaning toward redoing the bulkheads, it's just hard to accept when you first stumble on a problem like this. Sometimes the mind needs a day to settle into the idea of redoing a few weeks work.

I can relate to that! 
 Sometimes I can get over it in a day... sometimes I've had to take a week or two off and do other things for a while before I finally come to grips with having to start a major restart like that!        It's like working through the seven stages of grief.  It can take some time.

Kim

[mdavey]

Thanks again for the comments. I have gone through the charts in Kozo's New Shay book and the results are interesting. I am going to make some generalizations here in my comments, so no one should take this as a recommendation as to what is acceptable in this boiler. Also, I will note, I am redoing all the bulkheads for my boiler in 0.080" thick copper.

My first observation is that the boiler could use 0.062" copper throughout. It would just reduce if the maximum design pressure to 70psi. Now this isn't something you'd likely want to do, but it's interesting just how close it is. The charts also would suggest that the boiler shell could be a few thousandths thicker than 0.062" and still work at 100psi. If I'm reading things right, Kozo picked 0.080" thickness throughout to simplify things, give a bit more margin, and reduce the number of stays. Likely a wise idea for a "first project".

For the bulkheads, it's a mixed bag. The front tube sheet would be fine at 0.062" at 100psi. Some of the other bulkheads would be marginal or a bit below requirements. Interestingly, some of them seem a bit marginal as they are designed at 0.080" material. Not markedly so, but a bit on the edge. Or, maybe I'm not reading things right.  My guess is that Kozo went with fewer stays and marginal spacing to reduce the number of stays and penetrations. I trust his design, he knows way more than I ever will. I also suspect some of this comes down to design margins. After reading through things, I feel that what I had with the 0.062" parts would have worked and been safe. But, after a few good nights to think about it, redoing the parts seemed worth it, both to be sure things were fine and to correct the diameter issues the thinner material caused.

Thanks again for everyone's input.
Mike

[crueby]

Thanks for the update,  and hope you will post progress pics here on the forum!

[internal_fire]

Kozo picked 0.080" thickness throughout to simplify things

Keep in mind that this engine was designed and first built in Japan, where everything is metric. Of course 0.080" is very close to 2 mm. That size would be readily available to Kozo, whereas a standard imperial size wouild be more difficult to procure.

Even though the A3 is stated to be designed for the western market, with all imperial dimensions in the book, it was originally drawn in metric.

As others have pointed out there are quite a few "unusual" imperial dimensions that are obviously translations from the original standard metric dimensions.

Gene