To all
In mcad Roark's e-book 6th edition there is a worksheet to calculate the stress concentration (case 17b).
I just noted that the constants to calculate the values for C1, etc are different to the ones in the latest book (Petersons 3rd edition)
1. Am I correct?
2. I cannot find a 1974 copy 1st edition (I guess!) which appear to be the reference used. Has anyone got access and could confirm?
3. Is it the same in the latest ebook which I think is based on Roark's 7th edition but as far as I can tell in the printed Roark's 7th edition the numbers are from the 1974 edition too
Thanks
Regards
My printed 4th edition doesn't have a case 17 . . . (jumps from 16 to 18, ???)
My previous experience with the Roark ebook (under the original Mathcad owners) was that it had been farmed out to an intern as a project so they could improve their Mathcad skills; the information was (mostly) correct but the programming was stilted, plots were usually vectors not functions even though functions were used to calculate the vector values.
Given your choices I would choose (in preference order):
If you post your problem geometry (not just a table reference) I'll try to help further.
Good Luck!
Thanks. I set-up the Kt calc as per Petersons 3rd edition and for my geometry I got 2.65. the formula in Roark's ebook gave me 2.64. So not going to lose sleep
The geo is irrelevant (but thanks for the offer) I am more concerned about the factors and the discrepancy
Not my area of expertise at all, but the table in the seventh edition I found here
seems to correspond with the formula in the Mathcad sheet:
but as fas as I can see the constants in the calculation of the C values are not the one in Peterson 3rd edition
You may consider creating a variable name like "h/r"
😊
Yeah, I was in a rush, I don't think cntrl-shift-K works in Prime, so I'd have to type it in text, copy it, then paste it everywhere I needed it. so h_r is type-able. (Remember--I'm a lazy engineer!)
BTW, isn't it strange how the C's change so abruptly between regions? (Except for the first!)
BTW, isn't it strange how the C's change so abruptly between regions? (Except for the first!)
I am too far out of my comfort zone - even to make an educated guess. But at least I can say that its remarkable.
Empirical formulas ofton follow their own intrinsic logic and as along as we don't know how exactly they were arrived at ...
But i think that the values of the various C's are not that critical - but I would expect the cubic K.t to be nice and smooth without obvious crincles. But then, I may be wrong.
Empirical formulas often follow their own intrinsic logic and as along as we don't know how exactly they were arrived at ...
But i think that the values of the various C's are not that critical - but I would expect the cubic K.t to be nice and smooth without obvious crincles. But then, I may be wrong.
You're not wrong:
7th Edition:
8th Edition:
P.S. Years of edition (from 9-th to 1-st): 2020, 2012, 2002, 1989, 1975, 1965, 1954, 1943, 1938.
as per the original poster question I had a look at my copy of Petersons and indeed it seems that the constants defined in the C values are different - see attached Chart 3.10 from Peterson
The Roark's data seems to be pointing to the 1st edition of Peterson
@JXBWk The difference appears to be academic in this case - see graph attached (an "issue" at t/r = 2.0 possibly but I may have a made a mistake in typing the numbers!) but I get your point. One difference also is that the 1st set of C values are for 0.1 <= t/r <= 2.0 where in Roark's it is for 0.25 <=t/r<= 2.0
Peterson's also gives an equation for Kt bending, eq 3.7 - taken from Tipton et All (see attached), but if I "code" it get an imaginary number out of it !!
Tried with D = 200, d=100, r15 and I get 0.65+0.479i
Obviously does make sense - What I am doing wrong?
@JXBWk wrote:
Peterson's also gives an equation for Kt bending, eq 3.7 - taken from Tipton et All (see attached), but if I "code" it get an imaginary number out of it !!
Tried with D = 200, d=100, r15 and I get 0.65+0.479i
Obviously does make sense - What I am doing wrong?
How could we tell if you don't show what you did?
My best guess is that you typed 0.503*(d/D)^4 instead of 0.503*(D/d)^4.
The formula gives you real values as long as |d/D| is greater or equal to 1 and in your case D/d=2, so all seems to be OK.
Fair enough - Was still thinking about what I was writing in mcad when I asked the questing. Must have overlooked as I did attach the equation!
@JXBWk wrote:
Fair enough - Was still thinking about what I was writing in mcad when I asked the questing. Must have overlooked as I did attach the equation!
Wow! Looks like my freshly polished crystal ball did a good job 😉
Hmmm!
File of original Kt with your equation appended (Prime 4) attached.
Clearly I am an idiot who cannot make the difference between D/d and d/D !
I typed 0.503(d/D)^4 instead of 0.503(D/d)^4