Need a book? Engineering books recommendations...

Return to index: [Subject] [Thread] [Date] [Author]


[Subject Prev][Subject Next][Thread Prev][Thread Next]

As a said in my previous post, my experience with STAAD and master/slave
functions were with older versions (around version 21 or so...before it
became STAAD Pro).  I deliberately put in that disclaimer because REI
could have changed/fixed that function to work as you say it does now
because I just don't know since I no longer use STAAD (that is another
story).  I _DO_ that older versions of STAAD implemented the master/slave
function such that it ignored torsional effects if used to "model" a
diaphram.  I specifically tested this in STAAD at the request/suggestion
of another engineer at my company.  We also test this in RISA since RISA
came out with their diaphram feature about the same time.  He (the other
engineer) was aware that STAAD did not do it "correct" when when using the
master/slave function, so he believed that this would also be true of
RISA.  But it turned out that RISA did not use a master/slave method to
achieve the diaphram modeling.

If STAAD is using a true master/slave function, then it cannot model a
diaphram.  As Daryl pointed out, when a joint is slaved to a master joint
(in true master/slave fashion), it is forced to translate identically to
the master joint in that plan.  That is if the master joint moves 3" in
the x directions and 3" in the z direction, then so will the slave joint.
Thus, the slave joint cannot move a different distance than the master
joint meaning that the "diaphram" cannot rotate.

Now, I fully admit that maybe REI has changed the master/slave function to
no longer function as a true master/slave relationship but to truely
function as a diaphram.  But in order for this to be the case, then the
"slave" joints would have to move not identical to the "master" joint but
rather move in such a fashion that the relative distance between the
"slave" joint and the "master" joint stays constant (assuming a rigid
diaphram is being modeled).  This would mean that if the slave joint was
50 ft away from the master joint in the x direction and 60 ft away in the
z direction, then those distances would be maintained if eccentric loading
is applied but that the "slave" joint might only move 2 in in the x
direction and 2 in the z direction, while the "master" might move 3 in in
each direction.  If STAAD can now do this, then REI has in fact added a
true diaphram function to STAAD since the versions that I used since this
is NOT a master/slave type relationship/function.

FYI, I don't need to have something in front of me to comment on it.  If I
don't have it in front of me, then I make sure that I disclose the fact
that I am not in a position to truely verify the issue in question but I
am rather going by memory or supposition.  Thus, I will disclose that my
level of "surity" is lower.

I mainly point this out originally because many people don't realize the
potential flaw in using a master/slave function to model a diaphram...I
did not until that other engineer pointed it out to me (in my case, it was
not too much of a problem since I was not doing any 3-D models with or
without wanting to model diaphrams).  Thus, I thought that I would share
my learning experience.  It appears that you have already verified/tested
this so you did not need to "benefit" from my learning experience, but
others might.


Ypsilanti, MI

On Fri, 6 Sep 2002, Dave Langhone wrote:

> Scott
> First of all, I am amazed you are commenting on software without having them in front of you. Your memory must be reaaaaaly good, my friend.
> There is a very simple way for you verify the accuracy of the master slave aspect in STAAD. Open the example problem # 26 that comes with their program. It is a multi-storey building example with 2 load cases. Load case 2 consists of loads which cause the structure to undergo torsional deformation when seen from above. Run the analysis and view the displaced shape. You will observe that the deflected shape does indeed correctly reflect the behavior that a structure of this type will undergo when subject to such loading. If you wish to explore the matter further, you can work with example number 9 also. If you create a model similar to that, where the plates are substituted with a master-slave (DIAPHRAGM in ZX plane) command, and an eccentric load is applied at one of the top corners, you will see that the answers are remarkably close to what you will get if you apply such a lop-sided load on the example 9 model. I'll be happy to send it to you if you wish.
> I-cheeeee-ro rules! (I love my M's)
> Dave
> PS
> For Gerard, please don't generalize on what "American" engineers think or don't think. I worked for the USACE (who by the way standardize on STAAD.Pro) and I know comments like that gets everybody into trouble.
> Dave,
> I am curious...what do you use the master/slave funciton for?  Is it
> per chance to "model" the diaphram action in a 3-D model?  If so, you may
> want to be a little careful with STAAD.  While they may have updated the
> function, I do know that in older versions (not too old) the master/slave
> could not be used to get a true rigid diaphram effect.  The master/slave
> function makes ALL the slave joints translate the same as the master joint
> in plane of interest.  The result is that using the master/slave method
> will not allow for torsional rotaion of floor plane due to relative
> stiffness of lateral frames/walls.  In otherwords, your model (if using
> the master/slave function to model the diaphram) will be ignoring the
> torsional effects of the lateral loads.
> RISA does have a diaphram option that at least appeared (it has been a
> while and I don't have RISA to use at this time) to include the torsional
> effects.
> HTH,
> Scott
> Ypsilanti, MI
> ---------------------------------
> Do You Yahoo!?
> Yahoo! Finance - Get real-time stock quotes

******* ****** ******* ******** ******* ******* ******* ***
*   Read list FAQ at:
*   This email was sent to you via Structural Engineers 
*   Association of Southern California (SEAOSC) server. To 
*   subscribe (no fee) or UnSubscribe, please go to:
*   Questions to seaint-ad(--nospam--at) Remember, any email you 
*   send to the list is public domain and may be re-posted 
*   without your permission. Make sure you visit our web 
*   site at: 
******* ****** ****** ****** ******* ****** ****** ********