Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations KootK on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Distributed Load vs. Point Load 2

Status
Not open for further replies.

PatBethea

Mechanical
Nov 16, 2006
142
I had a SE size a beam for me that will be used to support three steel trusses (columns will sit under the two end trusses). In his calculations, he assumed that the load from the trusses was distributed over the length of the beam. The trusses are approximately 14' apart. I'm trying to understand this. Were the trusses 16" or 24" OC, a distributed load is easy for me to see. When sizing a beam or a header, is there a truss spacing at which you would start to consider the loading a point load as opposed to a distributed one?
 
Replies continue below

Recommended for you

I suspect the engineer was distributing loads the same way as cds72 (but with 3 trusses not 5). If this is the case then the centre truss load is wL/2 and peak moment is wL^2/8.
 
Ignore the bit about "3 trusses instead of 5". I misunderstood cds72's phrase.
 
Why are the trusses 14' apart? Are these holding up some kind of roof? SIP's, I hope?
 
depending on the load, 3" deck will span 14' continuously

Dik
 
Thanks for all the replies.

The reason I did not just go back to the SE for clarification was that I wanted to get it straight in my head first.

COEngineer was correct - getting the shear and bending moment diagrams out of Excel was alot more work than I anticipated.

Yes, Michfan, these trusses are holding up a roof.

PEinc - you got about the same result in 7 or 8 lines that it took me all afternoon to arrive at. Using point loading, my maximum moment was 142 kip-ft. The SE got a maximum moment of about 200 kip-ft using distributed loading. This result was counter-intuitive to my way of thinking. However, I am now convinced that his beam will work.
 
So, if my assumed-for example 10 kip load per truss gave a point load moment of 70 kip-ft, your 142 kip-ft moment must have been caused by a 20.28 kip truss reaction. Correct? Your span between columns was 28ft?
 
The truss reactions were roughly 16 kips. There was a small (relative to the truss load) distributed load that made up the difference (also, one of the columns is not centered directly under a truss - it is offset to the inside by 18"). I didn't mention it before because I didn't think it mattered in the overall scheme. I was trying to understand what was done.

I still don't quite understand why a distributed load was used for the three truss loads. True, it gave a higher bending moment and maximum shear which results in a larger beam being selected. So it was more conservative. But was it too conservative? I'm not going to ask that it be changed, but it seems to me like a point-loaded beam gives a more realistic analysis. I guess it was just a short cut to save time on the part of the SE.
 
PatBethea,

You are looking at it from a mechanical perspective where you may be doing a design for 100000 cars(say) and even 1% would make a lot of difference.

In Structural engineering we are always designing 1 off structures and the additional time it takes to cut things down to the bare bones usually outweighs the savings. The material cost is only a proportion of the total build cost.

knowing the shortcuts is a big part of our profession.
 
If we were talking about one or two beams, I might agree. But if many beams are 30% over sized, pretty soon the dollars add up!
 
If it's early in the design of the building, he might have used a conservative shortcut to leave room for error and revision down the line.
 
I interpret the problem as a beam with a truss at midspan of the beam and a truss at each end where the beam is supported. I would design the support beam using a discreet point load at midspan equal to the middle truss reaction, not a uniformly distributed load. The moment would be equal to PL/4, where L = the beam span.
 
What the SE did was perfectly acceptable. Provided that you follow the correct tributary areas for each element, the PL/4 moment from the truss rxn is equal to the wl^2/8 moment that the supporting beam sees. From the posts above, the most common mistake is dividing the point load by the length of the beam to get the equivalent distributed load.
This will not work.

Turn over an envelope:

Use easy numbers:

Draw a 10'x10' bay with columns at each corner.
Bisect the bay with one intermediate beam.
Use 50psf roof load
Analyze the south beam.

Using the point load:
Wtrib for intermediate beam = 5'
RXN from intermediate beam = wl/2 = 50x5x10/2=1250#
PL/4=1250*10/4= 3125#'

Dist. Load:
Wtrib of south beam=5'
w=50x5=250plf (NOT the 1250 from above divided by 10)
wl^2/8=250(10^2)/8=3125"'

Same moment. Size beam, check shear and deflection. done.

Take it further if you wish... try non-square bays, try multiple intermediate beams, draw the moment diagrams, and here's what you'll find:

Even number of spaces between intermediate beams means the udl moment will equal the point load moment. Odd spaces mean a slight disparity, but the UDL moment is slightly more conservative, but it is much easier and quicker. I always use UDL unless there are 3 spaces.

dik's above post said the same thing, only more succintly.
 
If you have a point load, why not analyze it as such? It's much easier to design the beam for a point load than to explain and prove that an assumed distributed load gives the same answer - sometimes.
 
I agree... but the point to my post is that it doesn't matter which way he analyzed it. Both methods yield the same answer. I would have done it the same way as the SE.
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor