Engineering Ethics 2

[Jerehmy]

Here's the situation:

NAAMM 510 says to use PROJECTED stair stringer length for calculating deflection. Well, doing it this way gives overall deflection (Perpendicular to beam) that is 44% less than actual and 20% less vertical deflection than actual.

When I say actual I mean using the members actual length and adjusting projected live loads accordingly.

Now the stringers I'm designing are supposed to comply with NAAMM AMP 510 and the stringers a deflection limit of 0.25". If I use NAAMMs estimated deflection calculation, I meet the deflection limit. If I use the right deflection calculation I'm over and need to use a larger size.

Ethically, what's the right answer here? Use bad deflection equations and a smaller size beam because that's what they want, or use correct engineering deflection equations and a higher size?

I prefer the latter but I got attitude from a drafter about how to design stair stringers that made me want to go through the phone...


And how did NAAMM become a standard when it's deflection estimate is so blatantly wrong. I sent them an email asking.

 

[Jerehmy]

I did them by hand and in RAM and got the same answer. I mustn't be describing it right. I'll draw it and post a picture if I have time tomorrow.

 

[CANPRO]

Jerehmy, see attached link. I'm not familiar with NAAMM but I found this on their website. In section 5 they have tables for stringers at different spans showing the allowable load and deflection. The deflection limits they seem to be working with in the tables are much less stringent than you're trying to deal with.

See table 5.26. At 19' they allow around 3/4" deflection (L/300).

https://www.naamm.org/landing_pages/AMP_510-92.pdf

 

[Jerehmy]

https://www.dropbox.com/s/u20apllkqz91dtp/calc.pdf

Link of calcs to what I'm talking about. I hope it's clear enough, my scanner here didn't seem to pick up my pencil very well...

Here's a graphic from ram. The 200plf for the diagonal beam is a "projected load" not an "actual load" just to make sure there wasn't any confusion. The nodes are at midheight. Visual2 shows the max local beam deflection w/ the global deflections.

https://www.dropbox.com/s/8c1euwds762dvov/Visual.jpg

https://www.dropbox.com/s/l8ztjkral2brici/Visual2.jpg

 

[SlideRuleEra]

Jerehmy - You appear to refering to two documents. One is the "spec", which you have not mentioned where it came from. This spec seems to require a maximum deflection of 0.25" or L/240.

The other document is NAAMM 510, to which CANEIT gave a link. I cannot find anywhere in NAAMM 510 that limits deflection in any way.

If so, your client may have special (unstated) requirements to limit deflections, therefore 0.25", per the spec. The solution to your problem is now easy. Select structural members that meet the client's "spec", using the NAAMM calculations. The fact that these members EXCEED the NAAMM recommendations is NOT a problem.

http://www.SlideRuleEra.net

http://www.VacuumTubeEra.net

 

[Jerehmy]

I just don't understand NAAMMM's deflection calculation because it doesn't accurately estimate either the member deflection in it's local coordinates nor the vertical deflection in global coordinates.

An engineering company hires a manufacturer to do its stairs. The manufacturer hires us to do the stair calcs. The spec comes from the engineering company.

 

[SlideRuleEra]

Jerehmy - I'll offer a possible reason for your confusion - this is just something to consider.

The NAAMM 510 document is from 1992. On page 5-2, NAAMM 510 mentions that everything in it is calculated using Allowable Stress Design (ASD). The way ASD was practiced in 1992 is different from the ASD of today you are most likely familiar with.

Could this discrepancy be a reason for your inconsistent results? Just a question, not an answer.

http://www.SlideRuleEra.net

http://www.VacuumTubeEra.net

 

[Jerehmy]

Slide - I don't think so. If you go to page 5-22, you see an example of ship's ladders. Here they think that the angle is large enough to consider its ACTUAL length. For stair stringers they seem to think the angle is too low to make a difference, but it does.

 

[SlideRuleEra]

Jerehemy - Believe that I am beginning to understand your confusion... and confusion appears to be justified. This NAAMM 510 manual has evolved beginning in 1959 (see the Foreword, on an unnumbered page just before page 1-1). The example problems most likely were prepared by different individuals, each of whom had there on "spin" on how it should be done. Note that the Foreword even mentions that the "Ship's Ladder" example is new for the 1992 edition. Frankly, I prefer the way the ship's ladder problem is addressed.

The solutions to the example problems certainly need to be coordinated by NAAMM. A general update seems to be justified to modernized the ASD / LRFD references. Am sure there are other issues, too.

One of your original questions concerned why NAAMM is the standard for stairs. A quick answer is because the American National Standards Institute (ASNI) says so. See this link on the NAAMM page:

http://www.naamm.org/ansi/

ANSI is considered to be the final authority on setting technical standards in the USA.

Since the NAAMM 510 manual makes it clear that its contends are "recommendations", suggest that you treat them as such. IMHO, perform your calculations in accordance with good engineering practice and general conformance with NAAMM's intent. Meet your clients "spec" requirements. However, make sure you go over your calcs with your mentor (find one, if you don't have one), or at least someone with more experience on this subject.

http://www.SlideRuleEra.net

http://www.VacuumTubeEra.net

 

[BAretired]

Yes I broke it up into components. If we have a 6-8-10 triangle, and a projected live load of W, breaking the load into actual length yields 8/10*W, and for perpendicular component gives 8^2/10^2*W = 0.64W.

So we have:

a) L = 8ft with W load and
b) L = 10ft and 0.64W load

8^4 * W *5/384EI = 4096*W *5/384EI

10^4 * 0.64W *5/384EI = 6400*W *5/384EI

Check in RAM if you don't believe me that NAAMM is wrong.

NAAMM may be wrong, but not as wrong as you suggest. You are comparing vertical deflection in the 8' projected span to sloping deflection in the 10' span. The vertical deflection in the 10' span is 0.8*6400*W *5/384EI = 5120*W *5/384EI or approximately 1.25 times the deflection in the 8' span.

BA

 

[woodman88]

You read and understand the calculations you are using.
Per the "Design of Wood Structures 3rd ed." by D. E. Breyer page 26 "NOTE: The horizontal plane method is commonly used in practice to calculate design values for inclined beams and rafters. The approach is convenient and gives equivalent design moments and conservative values for shear compared with the sloping beam analysis..."
Thus, you are comparing an equivalent method to an exact method. Of course the answers are difference. If you want them to be the same you need to fix the horizontal plane method before using it.

Garth Dreger PE - AZ Phoenix area
As EOR's we should take the responsibility to design our structures to support the components we allow in our design per that industry standards.

 

[Jerehmy]

woodman - The moments come out the same yes, but Lateral unbraced length will be larger obviously for the inclined beam. I guess it doesn't matter for roof rafters because of sheathing. And in NAAMM it doesn't matter because they count the stringers as constantly braced from the welded treads. But I'm not using welded stair treads and I have no lateral support from them. So if I used NAAMM's approximate method I would have an unconservative design. But I shouldn't be using it anyways since the situations are different I'm guessing would be your rebuttal, which is true.

BAretired - Yes you are correct, I noted that in my attached calcs. But shouldn't we be looking at overall deflection not vertical anyways? idk-


The guy I emailed at NAAMM wasn't very nice to me lol. I think he purposefully misspelled my name in his response mail...

 

[nutte]

Woodman's explanation states that the moment and shear values are the same. We're talking about deflection values, which are not the same.

Jerehmy, I'm curious as to what NAAMM's respoinse was..?

I still maintain that this is much ado about nothing.

 

[Jerehmy]

"Sir,

The gravity loads act vertically, not on the actual sloped length. If you
vector the load into components one of which is perpendicular to the actual
length and one parallel to the actual length you will find the results are
the same.

so and so"

my response

"so and so,

They are not the same. Deflection changes with a 4th power to length, the gravity loads only change to the 2nd power when you distribute them across the proper length, then change them into their parallel and perpendicular components. Check in RAM Elements or some other analysis software if you don't believe me."

his response

"Gerehmy,

I understand the problem. I will stick with my original recommendation that you should work with the projected span."

^^ was sent from a phone

That was frustrating because he didn't even look at the stuff I sent him as attachments, or analyze it himself. He kind of just blew me off. I was going to email him back that he clearly doesn't understand the problem if he thinks they are the same deflection but whatever, I give up.

And come on, the G and J are two spaces apart lol.

 

[woodman88]

Attached are the "Design of Wood Structures 3rd ed." by D. E. Breyer pages 25 and 26. Please read them to understand the equations so you can use them correctly.


Garth Dreger PE - AZ Phoenix area
As EOR's we should take the responsibility to design our structures to support the components we allow in our design per that industry standards.

 

[Jerehmy]

woodman - I do the exact same thing in my analysis above on page 2..

 

[Jerehmy]

and the fact that shear is conservative makes perfect sense

Shear varies with a power of 1 to length

moment is 2

deflection is 4 (for distributed)

the loading changes with a power of 2 for a sloped beam, thus moments are the same, calculated deflection is less than actual and calculated shear is more than actual.

 

[woodman88]

Jerehmy Yes, but you are assuming that all the answers for both methods will be the same. That is not the intent of the methods. The one method will only give the moment correct, all other answers with that method must be used knowing that they will not match the actual answers.

Garth Dreger PE - AZ Phoenix area
As EOR's we should take the responsibility to design our structures to support the components we allow in our design per that industry standards.

 

[Jerehmy]

Woodman88 - That's exactly my point. But NAAMM uses the horizontal beam method for deflection when clearly we both agree that it's wrong, and not in a conservative way.

 

[nutte]

It's not "wrong," it's just not calculating the same deflection you're after. It is reasonable to think the limiting deflection should me measured in the vertical direction, not parallel to the sloped stringer. So the ratio of deflections you have is 10/8 = 1.25. How does this compare to L/240 or L/360? After all, that is what the NAAMM examples check. For the 10' span, you have 1.25/10 = 0.125. For the 8' span, you have 1/8 = 0.125. The ratio of deflection to span length is the same. That is what NAAMM is checking, so their method gives you accurate results when comparing to a deflection ratio like L/XXX.

I won't disagree that the measured, actual numeric deflection in inches, is more for the sloped case than the horizontal projection case. But nobody checks it that way.

 

[Jerehmy]

I agree with you there. The only issue now is that I'm not limited by L/240, I'm limited by 0.25in. So now using the horizontal vs the actual has an effect. But whatever I'm just going to use NAAMM's deflection calc and be done with it.

I just find the way NAAMM does it to be unintuitive, misleading, and confusing when it seems like they're trying to simplify things. If I just blindly followed NAAMM it'd be OK, but when I try to understand what they're doing and why they're doing it, it takes more time and over-complicates things. Especially for someone, like myself, who has never used NAAMM before.