gerry,
OK i tried that experiment you suggested. It burnt my hand you tricked me. But i do agree with you that steel conducts much better than concrete does.
You're not there yet.
Steel conducts heat so fast that, over the course of just a couple of minutes, the steel will be fully thermally expanded while the BULK of the concrete (all except the thin layer directly exposed to the flames) remains at room temperature.
Therefore, for the dynamic situation just after heat is applied, a worst case stress analysis of the situation can be ACCURATELY modeled by considering the steel to have its proper heat capacity, thermal conductivity & coefficient of thermal expansion, while the concrete is modeled at a constant room temperature.
This provides the accurate thermal-expansion generated stress in the beam, shear studs and concrete during the heating transient.
Once the concrete finally comes up to temperature, then its expansion will be approximately equal to the steel's, but by then, the damage to the shear stud & its connection to the concrete and the steel beam has long since occurred.
You get the right answer in 1/100th the time by doing the analysis this way, rather than applying the "correct" heat capacity, thermal conductivity, thermal expansion to each element.
So why did NIST not let it conduct in their FEA model?
Who says that NIST didn't allow their steel elements to conduct heat?
Here's what NIST says about the concrete slabs & metal deck:
(pg. 350)
NIST said:
"The metal deck and concrete slab were modeled, as were the headed shear studs on the floor beams. … A single layer of shell elements was used to represent the concrete slab and metal deck, using a material model that allowed for different stress-strain behavior in tension and compression. The contribution of the welded wire fabric to the tensile capacity of the concrete slab was accounted for in this model using a smearing approach."
and (pg. 543)
NIST said:
"The primary feature required in modeling the composite slab was the different slab behavior in tension and compression, as concrete cracks under tensile loads at relatively low stresses. By comparison, the tensile strength of the metal reinforcement was an order of magnitude higher than the concrete compressive strength and two orders of magnitude higher than the concrete tensile strength.
The composite slabs in the WTC 7 global model were modeled with shell elements. The constitutive model used in LS-DYNA for the composite floor slab was material type 124 (Mat_Plasticity_Compression_Tension). This material model allowed the specification of separate tabular plastic stress-strain curves in compression and tension. The separate tension and compression curves for the composite slab were obtained by using the rule of mixtures, which provided average properties of the composite slab based on the volume fraction of the concrete, metal deck, and reinforcement (Chapter 2, NIST NCSTAR 1-9A). The calculated tensile and compressive behaviors for the composite slab are shown in Figure 12–5.
Doesn't look to me like they ignored any aspect of the metal deck's mechanical or thermal properties. It looks like they created custom elements that reproduced the concrete & deck's composite behavior.
What info do you have that says otherwise?
tom
PS. Gerry, it's a shame that it's gotten to this. I think that there's a fair chance that we'd have gotten along great over a pool table or a couple of beers.
Believe it or not, I am NOT in this for any "biggest swinging slide rule" competition. I am about 6 months from retiring, to spend the rest of my life doing whatever engineering projects I choose, self funded & extremely comfortable. There is precisely zero ego-points to be gained by beating up on truthers who have no background in the field of mechanical engineering.
You say that "you're in it to learn. You don't care if you lose a debate."
Easy to say. Let's see you step up & actually DO it.
Do it here.
But, for some bewildering reason, you won't even answer a couple of simple questions???
I have an innate inclination to teach. I've taught college level engineering students, plus 3 full generations of baby engineers over the years.
It's a knee-jerk reaction of mine. (& extra light on the "jerk", if the kids show the slightest inclination to put out an effort & learn.)
Buried inside of each of the questions that I've asked you are revealing facts about the issues under discussion. If you weren't so concerned with ego, you'd take a crack at answering some of those questions.
you tricked me. But i do agree with you that steel conducts much better than concrete does. So why did NIST not let it conduct in their FEA model?
