Flange design - economic hub thickness selection 1

[rww88]

I have a single page from what appears to be a very old untitled source (book?) that refers to an exhibit (a graph) within the source that can be used to determine the initial selection of an economic hub thickness (g1) of a tapered-hub integral-type flange. Not having the graph, it appears from the instructions that the horizontal axis of the chart is sqrt (B/g0) and the vertical axis is g1/g0. The only example given in the source for its use indicates that when the chart is entered at the bottom at sqrt(B/g0) = 7.75, then the value g1/g0 = 2.35 can be found along the left vertical axis. Of course, knowing this ratio and the value of g0 allows one to calculate g1. I have tried unsuccessfully to find a reference to this specific design parameter. Taylor Forge Modern Flange Design Bulletin 502 simply states: "For starting hub dimensions, g1 ~ 2g0 is suggested". Does anyone know this technical source or the underlying theory or math relationship mentioned?

 

[pmover]

Suggest conducting a search in this forum for further reference postings.

What size, type, material, class of flange? a ASME B16.5 standard flange may suit the needs.

may also start w/ UG-44, ASME Section VIII, Div-1 depending on flange data.

 

[JStephen]

I'd be curious how useful it actually is.
Seems like, if this gave you the least-material-flange (one possibility), you could accomplish the same thing with a spreadsheet pretty handily.
If you were actually looking for the minimum cost of a custom flange, then almost certainly material and labor rates, forging costs, machining costs, etc., have changed over the last 50 years so that what was the most economic 50 years ago wouldn't be now. In that case, a flange supplier would be the best bet for good information.

 

[BJI]

It sounds like you might be referring to a page from the original publication of the code formulas, which could refer to figure 14 or 16 (now included in Appendix 2 as Figures 2-7.2, 2-7.3 and 2-7.6). This is for the hub influence factor and the hub stress correction factor, but it doesn't resolve the hub thickness directly and the equation uses h/sqrt(Bg0). Try the reference below, there are two links to view the document.

https://catalog.hathitrust.org/Record/005752561

 

[rww88]

Thanks very much to everyone for your replies. Excellent link BJI! I will post to this thread if any additional research uncovers the mystery underlying the original post's query.

 

[SnTMan]

Interesting subject. I'd about have to think that rww88's source page is from original material, although I suppose an ambitious company standard (like you don't see anymore) is possible. I don't mind telling you, I am certain the math is beyond me.

It has always appeared to me that the most economic hub dimensions are ones that produce the flange with the thinnest overall thickness and smallest O.D., or in other words the lightest possible raw forging weight, the thinking being that all in all, machining costs will not vary greatly.

I have no idea how show that given hub dimensions produce that result. It would seem to favor a small g1 dimension to minimize the OD, and a smallish hub length to minimize total thickness, but it would take numerous runs with varied hub dimensions to show this.

The program I use most often that totally automates the flange design process actually has at least a couple of options for the hub logic. I have not tested one against the other.

Yet I feel there must be a method to it that the old guys knew somehow.

BJI's link is intriguing, the answer may be within somewhere. BJI, do you know how one becomes a "partner" so that the pdf may be downloaded?

Regards,

Mike


The problem with sloppy work is that the supply FAR EXCEEDS the demand

 

[Christine74]

I can't imagine a g1/g0 ratio of 2.0 or 2.35 being anywhere near optimal in terms of the resulting flange size. In my experience an economic flange design generally has a hub taper angle somewhere between 6° and 12°.


-Christine

 

[rww88]

Christine74
For what it is worth, and that maybe actually in fact nothing, the mean g1/g0 ratio for ASME B16.5 service rating 150 weld neck flanges sizes NPS 1/2 inch through NPS 24 inch is 3.17.

 

[Christine74]

Yes, B16.5 flanges typically have a much steeper hub angle than flanges that are designed in accordance with Appendix 2 of ASME Section VIII Div. 1, which normally results in a larger OD flange. Appendix 2 flanges for heat exchangers typically have g1/g0 ratios less than 1.5.


-Christine