carrier beam design

huangyhg

carrier beam design
i am designing a steel structure that will be shipped as a "whole structure" using transporter. since i cannot use uniform support in staad i used pinned supports. but it is not acting the way i wanted.
anaylysis shows support reaction from 2 to 5 is under tension.
staad is neglecting the stiffness/rigidity of beam to make
all supports reaction under compression.
how do i do model it to simulate "actual" reaction where
all supports reaction are compression?
i think that staad is correct here.  with large loads on the tips of the cantilever sections, uplift is not only possible, but very likely in my experience from this particular loading situation.   
mike mccann
mmc engineering
with your current model, you should get tension in joints 2 thru 5.  your model has your transporter as infinitely rigid.  it might help to include the flexibility of the transporter in your model.  also, by using point loads, you are neglecting the stiffness of the frame above.  it might help to include the frame in the model.  if you still get tension at joints 2 thru 5, eliminate those supports from the model.
i have modeled it to include the frame..
i did a test run where i used very thick concrete
(definitely rigid) against a relatively small
load at the ends and still getting tension on 2 to 5..
looks like the closest i could to "actual"
is remove 2 to 5..
unless any other good advice here..
newbie,
staad is giving you the correct answer as you have it modeled.
my suggestion:
add additional nodes.  at least one in between each of the support nodes that you have.
copy all support nodes down 3".  these copied nodes will be your new supports.
remove all support parameters.
assigned "fixed" supports at the new support nodes.
put stub columns in from the new support nodes to the old support nodes.
the two stub columns on either end should be fixed at the base, and pinned at the beam.
assign "compression only" to the remaining stub columns.  this is under commands --> member specifications or the specification page icon.
re-run it.
let me know how it goes.
chip

chipb, there is a compression only support also in staad that i tried similar to what you are proposing but i dont it's the right approach.. it is just giving me different iteration and re routing load path if particular support is in tension..
what am trying to do is for staad to consider the "rigidity" of carrier beam similar to when
you do mat foundation as assumed rigid..
what is your "transporter"?  the top of a truck?  if you know the stiffness of the "transporter", you can model your supports as springs.  otherwise, mike is correct--unless you anchor the interior portions of the carrier beam down, all the load will go to supports 1 and 6.
daveatkins
dave, what if i am assuming transporter and carrier beam
as infinitely rigid?
if you want to assume the beam and transporter as perfectly rigid, then why do you need a model?  the load will be evenly divided between your supports.
naturally, we know this is not true.  
if the transporter is perfectly rigid and the beam is not, then supports 2 through 5 are in tension by inspection.
we know this is probably not true either.
what is true is that both the beam and the transporter are not perfectly rigid and the center of the beam may or may not lift off of the center of the transporter.  the only way to approach this accurately is to model both the beam and transporter's stiffness correctly.
two ways to get your arms around the problem:  1 - go ahead and assume that supports 1 and 6 support all of the load or 2 - make sure they put some wood blocks to support the beam on the transporter.  this will allow the beam to flex and help redistribute some of the load towards the center.
if you "heard" it on the internet, it's guilty until proven innocent. - dcs
approach 1 is what i am currently doing..
thanks all for your help..