Capacity of a round weld 2

[anchorengineer]

Hello All,
I'm looking for a formula to calculate the capacity of a fillet weld for a round post with a horizontal load at the top inducing a moment.
Thanks!

 

[CELinOttawa]

I really don't like the calculation of the weld area for the circular pipe... 2PI()r is the circumference alright, but the 6/8" assumption had to be known before the calculations began.

I prefer to use the more conservative assumption of d/2 as the r in the circumference calculation, and then assume a weld size to turn the linear value into an area.

Thoughts? Anyone disagree with this head case?

 

[Ron]

CEL...they are the same. (Pi)x 2r is the same as (Pi) x [2 (d/2)].

I think what you meant was to use only the leading 1/2 of the circumference in your resistance calcs. Then it becomes....

(d/2) x (Pi) x 0.71 x (assumed weld leg size) for the weld stress.

 

[desertfox]

I'm in agreement about stresses in welds being highly complicated and that calculations on these welds using assumptions and simplifying theories is done everyday but these assumptions have been verified over many years by research and experiment.
Here is a link to a welding paper that indicates why assuming the shear stress is distributed evenly in and around a weld.

http://www.ignou.ac.in/upload/Unit-4-60.pdf

Further links to weld design here

http://www.mitcalc.com/doc/welding/help/en/welding.htm

http://www.mitcalc.com/doc/help/en/C_safety.htm

The latter link talks about the factors of safety based on knowledge of loading materials etc

 

[KootK]

Here is a link to a welding paper that indicates why assuming the shear stress is distributed evenly in and around a weld.

Can you guide me to a particular page or paragraph Desertfox? There's a fair bit of information there and I couldn't spot the bit that you've referenced.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough that I want to either change it or adopt it.

The greatest trick that bond stress ever pulled was convincing the world it didn't exist.

 

[desertfox]

hi KootK

Start here

4.3 TYPES OF WELDED JOINTS, STRENGTH

 

[CELinOttawa]

No, I meant I like using 6" for the r, not 6&3/8. I know there is a minimum weld size assumption, but I like to exclude this...

 

[CELinOttawa]

Also: Am I the only one who thinks it is silly to use the additional 3/8" in concert with the crazy gross simplifying assumption of πr^3 for (π/4)(D^4 - d^4)?

 

[paddingtongreen]

@CELinOttawa, the 6 3/8" is the OD of the 12" pipe.

Michael.
"Science adjusts its views based on what's observed. Faith is the denial of observation so that belief can be preserved." ~ Tim Minchin

 

[CELinOttawa]

Thanks Paddington; This is hilarious - I had no idea that US standard 12" pipe was actually 12 3/4 inches... Funny what a big difference in reading a problem such information can make. I thought that the check was including a minimum weldment size to use a larger area and calculate within the centre of the weld; Obviously not!

Funny, I thought out HSS were weird sizes (273mm, 324mm, 35Xmm - Can't remember the 35something at the moment) because they were US pipe sizes as well... I've never seen anything around 319!

 

[KootK]

There's probably upwards of 100 collective years of structural engineering experience being brought to bear on this thread. At the risk of being offensive, I'm struck by the fact that not a one of us seems able to reconcile the form of Blodgett's design equation with our fundamental understanding of mechanics of materials. And this is an equation that we all use and espouse to our colleagues.

I don't feel that it's prudent to dismiss this discrepancy simply because of the limited computing power available to Blodgett or the fact that "knowing" weldment stresses accurately is hard. My intuition is that Blodgett's design equation does make theoretical sense, just like all of his other recommendations. Rather, I think that we are missing something here regarding the theoretical background to Blodgett's equation. This modern design document corroborates Blodgett's assumption of uniform shear stress (Link). For me, that is sufficient circumstantial evidence to conclude that we are somehow lacking in understanding.

I just don't want any EITs, or anyone else for that matter, thinking this actually matters.

This statement could not be more wrong if you'd written it with a gag-ball in your mouth. We answered OP's question sufficiently within minutes of his asking: do like the Romans do and defer to Blodgett. The rest of the discussion has been about trying to understand the basis a design equation that we all seem to be using blindly. I think that's important to us all and, especially, to junior engineers who need to think critically and to see their mentors for the fallible creatures that they are. As great an engineer as Blodgett is, even his work is not above scrutiny and validation. And, unless I miss my mark egregiously, Blodgett wouldn't have it any other way.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[KootK]

Here is a link to a welding paper that indicates why assuming the shear stress is distributed evenly in and around a weld.

Interesting links DF. The first paper provids justification for using average shear stress along the length of a longitudinally loaded weld despite localized effects due to starting and stopping and elasticity theory. In my opinion, it does not address the issue that I've been harping on; namely, why weld stresses can be assumed to be inconsistent with VQ/It shear stresses in the supported member.

The first website that you linked to did provide a nifty kernel though. This sketch is consistent with my original thoughts on the matter:

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[CELinOttawa]

We aren't being inconsistent, and the increased accuracy DOES NOT MATTER. I don't need a ball gag to know that if the profession has been doing something for a hundred years without failure and with close agreement from experimentation, refining it isn't necessary.

And this:

do like the Romans do and defer to Blodgett

is *NOT* at all what I have advocated. I didn't even use Blodgett in my defense of the standard approach, but rather one of his predecessors (while also methioning a preceeding great who's sum total of Weld comment is "Welding of steel shall not be permitted"). I have advocated that the standard formulae of f=P/A, f=Mc/I, and S=Tr/J.

So, for the record, I'll get all sensitive and point out that you are falsely misconstruing the argument as an argument from authority. This coulnd't be more false if you'd borrowed the ball gag to make it...

Finding a more accurante answer is a whole lot of potential fun. It is something we can keep pounding away at. It is something I'll be here in the thread dedicated to trying to help us all understand better. It is also, NOT NEEDED.

Now if you'll all excuse me, I have a meeting with a shower... And I'm really a little creeped out about Kootk's prescient comment about this family's local application of Wooten's first rule...

Also, in English, does falsely misconstruing constitute a double negative in English? Je ne comprends pas...

 

[CELinOttawa]

Some thoughts on the discussion...

As I see it, the refining of this issue is interesting, but will not change the answer. This is because we already use a combined strength equation on the components of bending and shear, which naturally pro-rates the strength of the weld into an amount to handle each type of load. Thus even if the weldment is behaving in a manner we do not explicitly address, our solution takes care of the behaviour in a manner accurate enough to permit safe practice.

Am I making sense here?

 

[desertfox]

Hi CEL

There is nothing in the attachtment

 

[CELinOttawa]

It seems that an ampersand (&) in the name breaks the upload... One last try.

 

[CELinOttawa]

And it would help if I squared the terms under the square root... *sigh*

 

[KootK]

Thanks for the sketch and the ideas CEL. I'll consider that my Xmas present.

I'm afraid that it is you who has misconstrued my purpose. I'm not trying to improve accuracy, shave down weld sizes, or advocate an alternative design method. Not at all. What I am attempting to do is to understand the theoretical basis for Blodgett's method and any simplifications that he may have taken. This is now the sixth time in this thread that I've described my goal as developing understanding. Not refinement. Not replacement. Understanding.

Based on this thread, it has become clear that none of us understands the theoretical basis of Blodgett`s shear stress simplification. So how is it that we`re fit to use it? Or to adapt it to other circumstances? And how do we know that it's conservative? I'm not looking to replace or refine Blodgett's method. I'm seeking to understand Blodgett's method. Surely that is a worthy goal and something that 'matters' to any engineer who cares to actually know their craft.

The M/Sx form of Blodgett's equation suggests that we're sticking to the elastic domain. Given that, and ignoring all olf the messy stuff like residual weld stresses, restraint etc, it seems to me that determining the analytically "correct" stress distribution is relatively straight forward. It's just VQ/It. See the sketch below which is also attached as a PDF file.

It is instructive to note that, for short tubes where bending stresses do not dominate, the predicted maximum shear stress of P/A would be out by a factor of about two.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[anchorengineer]

I love asking tough questions but never in my wildest dreams did I think it would lead to this kind of discussion... Wow!

 

[CELinOttawa]

Kootk: I get that you're looking for a better understanding. I simply think that the understanding we have is sufficient, and I am quite happy with the level of understanding as sketched out in my attachment.

The level of understanding you seek is interesting, but not necessary.

There, I've said it enough times, so let's just get busy trying to find the answer you want...

More PDFs if I have time tomorrow.

 

[KootK]

@anchorengineer: yeah, you never know what will capture the imagination. My first comment linked to a thread where I thought that this was already settled.

@everybody: some late breaking news. I misread the AISC provisions for shear strength of round HSS. AISC section G6 and commentary use Ag/2 which is exactly consistent with the derivation that I just posted. So My status is no longer "confused".

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.