Existing Steel Beam

[engr567]

I have an existing roof beam, and a new RTU is to be supported by this beam. I was told that an existing beam can be overstressed a maximum of 10% and still be within code allowance? Does anyone know about this? If so, where can I find this info in IBC 2006? Thanks in advance.

 

[JAE]

hokie66,
In the US codes, (IBC) you can reduce roof live loads based on tributary area similar to floor live load reductions. In the older codes (Uniform Building Code) there was a simple method:
0-200 sf - 20 psf roof live load
201-600 sf - 16 psf roof live load
>600 sf - 12 psf roof live load

The new codes are similar but provide a formula for the actual RLL to be used.
You don't reduce snow loads of course.

 

[woodman88]

Maybe this is why the 2009 IBC Section 3403.3 and 3404.3 has been changed to state "...an increase in design gravity load of more than 5 percent..."

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.

 

[CANPRO]

engr567...I think you need to analyze the beam in its existing condition...get your maximum factored moments, shear, reactions...then analyze your beam with the RTU. If the ratio of new/old (for moment, shear, or reactions) is greater than 1.05 then you need to start looking at the entire load path and reinforce the members as necessary along the way.

I think DST148 said pretty much the same thing in an earlier post.

 

[ash060]

You can't increase the force. They are referring to the internal forces in the member, shear, moment, and axial. It doesn't matter how the external forces are arranged on the member, the internal forces can't be increased by more than 5%.

If there is a 200 k-ft moment in the member due to some arrangement of loading, and a new point or uniform load is added to the member, the new moment can't exceed 210 k-ft. Same goes for shear and axial.

 

[engr567]

IBC 2009 and 2012 are clear. Makes sense now, 5% is based on gravity load. Thanks woodman88.

 

[hokie66]

Back to the live load thing, I suggest that the mechanical unit is a live load, so the portion of code required live load in the area of the unit is in fact the unit. Does that help? Not if snow load controls.

 

[wannabeSE]

Hokie,
The IBC and ASCE 7 definition for dead loads include fixed service equipment, such as . . . ventilating and air conditioning systems.

If the mechanical unit is tall, I would take a quick look at the wind load reactions (from overturning) from the mechanical unit to see if it adds a significant load to the beam.

 

[ekaj2000]

Be careful with the 5% provision. I've recently done some work in the telecom industry (adding antennas and equipment shelters to rooftops), and on a few occasions have come across sites where engineers have used the 5% rule of thumb on one retrofit after another. Sure, it may be within the 5% over the existing condition, but is it within 5% over the original design condition? Just make sure additional loads beyond the initial design condition weren't already added to that beam before you say it's within 5% and further analysis is unnecessary.

Jake

http://www.pelicensemanager.com

 

[hokie66]

My point about the loading is that you won't have both the mechanical unit and another load in the same place. Whether you call the mechanical unit live or dead, there is only one superimposed gravity loading on that area...except for snow load.

 

[shobroco]

I'll stick my neck out again (I forget who but someone disagreed with me about this in another post) & I'll say the snow load on a mechanical unit is insignificant. I'm guessing you don't have a lot of experience with it in Australia, hokie, but we get a bit here in Canada. If snow is building up on your RTU, you've got more immediate troubles than weight. Your building must be getting pretty cold with the heat shut down. Yes, I know not all RTUs are for heat, but the vast majority are producing or exhausting heat in some way & they don't accumulate snow on them.

 

[jdgengineer]

I like the idea shobroco, but what happens if the building is vacant and unrented for 6 months and all the mechanical equipment is turned off?

 

[shobroco]

It's unlikely that the heat is completely off unless the building is abandoned, in which case it may be a benefit that the roof comes down. It saves demolition costs (that's a joke, okay?). I've had discussions before over too many what-if scenarios & I figure at some point you have to decide if you're comfortable with the probabilities. That's all we're working with anyway, isn't it?

 

[DST148]

@jdgengineer - Ref ASCE 7-05 commentary - Brief power interruptions and loss of heat are acknowledged in the Ct = 1.0 category. Consequently, heated structures need not be designed for this unlikely event. Some dwellings are not used in the winter. Although their thermal factor may increase to 1.2 at that time, they are unoccupied, so their importance factor reduces to 0.8. The net effect is to require the same design as for a heated, occupied dwelling.

@shobroco: I acknowledge your experience and observations, however,.....
For snow loadings, the American code uses Canadian code as a point of reference for thermal, aerodynamics effects etc. As per ASCE commentary : The combined consideration of exposure and thermal conditions generates ground-to-roof factors that range from a low of 0.49 to a high of 1.01....., all regardless of their thermal condition.
I have worked in Ontario Canada on projects related to pharmaceutical industry. This is what we have done - for new projects full snow load on the units; for checking adequacy of existing members, we started with full snow load on the units, reduced on few occasions, but never less than 2/3 of the roof snow load.
I agree with you, these are subjective.

 

[jdgengineer]

Thanks for the insight DST. I don't use snow loads in my part of the country s a lot of this I am not familiar with. Seems to me not using any sow load on an area is different than using a Ct of 1.0 though.

 

[CANPRO]

If you get enough wind to get the worst case drift against the RTU, what are the chances that there will be snow left on top of the RTU?

I suppose you could have a 1/50 year snow storm with a large drift against the RTU and then get another significant snow storm with no wind which would load the top of the RTU (all while the RTU is not putting out enough heat to melt the snow)...but I think that is very unlikely.

Granted I am very new at this, but with that logic I have never considered snow loads on top of the RTU.

 

[waytsh]

Sorry to bring this thread alive again but what about the condition where you are within the allowable 5% increase but the beam is now overstressed? Can the beam be reinforced to the original loads and code or does it now need to be reinforced to the new loads and code?

 

[BAretired]

Drifted snow around the RTU is frequently more than the weight of the unit itself. If drifted snow contributes to the load on your beam, it must be considered.

BA