Yankee
Registered User
Where is it that you see the 2.5 safety factor? (and what exactly do you mean by 2.5? 2.5%?)GHRoberts said:
-
Welcome to the new and improved Building Code Forum. We appreciate you being here and hope that you are getting the information that you need concerning all codes of the building trades. This is a free forum to the public due to the generosity of the Sawhorses, Corporate Supporters and Supporters who have upgraded their accounts. If you would like to have improved access to the forum please upgrade to Sawhorse by first logging in then clicking here: Upgrades
Point Loads
- Thread starter Jobsaver
- Start date
I can't speak to the safety margin for sheet goods but most building materials have a healthy safety factor. Steel is homogenous and carries 1.67:1 ultimate/allowable if I remember right. Structural lumber is very variable and has a 2.1:1 safety factor with a 95% exclusion limit. 95% of the pieces in the pack should fail at a load at least 2.1 times allowable load with 5% of pieces allowed to fail between allowable load and the 2.1x safety margin. There will be pieces that are 5-7 times stronger than the allowable load in most packs, I try to put them where they will do the most good. Often at the sawhorses I can remove the limiting defect in a stick and bump the grade up. The gradestamp is segregating volumes of lumber into groups, I can select through the group and put the individual pieces in the best places.
Don't forget that each joist on a beam is really a point load even though we call the beam uniformly loaded... Magnitude does matter but picking the exact place where uniform becomes point gets a little like some folks up the road, where does the good book say we can plow with a mule but not with a tractor.
Don't forget that each joist on a beam is really a point load even though we call the beam uniformly loaded... Magnitude does matter but picking the exact place where uniform becomes point gets a little like some folks up the road, where does the good book say we can plow with a mule but not with a tractor.
Jobsaver
Registered User
Thanks for all of the posts. Usually, I try to participate more in the threads I start, but got busy after posting the OP.
The purpose of the thread is to try and establish better guidelines for sizing headers and girders using the IRC header and girder charts found in chapter 5, particularly R502.5(1) and R502.5(2). One desirable result from discussing this issue is to better determine what, if any, point loading specifically on headers, and girders, can be accounted for in these tables: Starting with headers, for examples:
A 32” door header supporting a point load from a ceiling beam that carries the ceiling only.
A 101” 2-ply #2 SYP 2x12 cased opening header supporting a point load from a ceiling beam that carries the ceiling only.
The same header supporting beams carrying ceiling and roof bracing loads.
And, are there specific rules of thumb that can be applied for like circumstances?
I find it curious that in the ’95 CABO, per table 602.6, MAXIMUM SPANS FOR HEADERS LOCATED OVER OPENINGS IN WALLS (feet), these spans are listed for a 2-ply 2x12:
(note reduction in spans with ’06 IRC):
1995 CABO (602.6): #2 grade lumber with 10’ tributary floor and roof loads (italics mine)
Supporting roof only: (presumably, this includes ceilings?) 12’
One story above: 10’
Two stories above: 8’
Headers in walls not supporting floors or roofs: 16’
(presumably, ceilings only?)
2006 IRC (R502.5(1)): #2 grade lumber assuming 20’ building width (italics mine)
Supporting roof only: (presumably, includes ceilings?) 9-9 (roof and ceiling)
One story above: 7-1 (w/clear span floor)
One story above: 8-1 (w/center bearing floor)
Two stories above: 5-6 (w/2 clear span floors)
Two stories above: 6-8 (w/2 ctr bearing floors)
Headers in walls not supporting floors or roofs: No listing
(presumably, ceilings only?)
What accounts for the reduced spans?
Why does the IRC fail to address sizing headers for “ceiling loads only?
Most homes built in my ahj are stick-built, and commonly feature multiple pan or tray ceiling systems comprised of built-up beams, laminated beams, LVL’s and dimensional lumber. Often, beams supporting portions of the ceiling (only) will rest above a window, door, or cased opening header. At other times, the beams will carry a portion of the ceiling and a partial roof load, predominately from roof bracing. Sometimes the additional roof load will be a minor load resulting from a purlin brace. At other times, the additional roof load will be more significant, say a 1500-2000 pound load resulting from supporting the bottom end of a half-valley.
I am clumsily working a few examples using the beam calc tables listed by DRP earlier in this thread. And, as previously mentioned, am interested in determining to what extent a building inspector that depends on the tables can account for any point loads accumulating on headers, and develop safe rules of thumb.
(Later in the thread, I will address the same questions as apply to girders).
Note: The example I am working on now is for a 101” 2-ply #2 2x12 header having a point load resulting from one end of a center ceiling beam resting above the header.
20 x 20 ceiling (uninhabitable w/o storage. 10 psf live load, 5 psf dead load – R802.4(1)).
So far, I have determined that the beam carries a tributary load of 3000 lbs, resulting in a 1500 lb. point load on the header.
The purpose of the thread is to try and establish better guidelines for sizing headers and girders using the IRC header and girder charts found in chapter 5, particularly R502.5(1) and R502.5(2). One desirable result from discussing this issue is to better determine what, if any, point loading specifically on headers, and girders, can be accounted for in these tables: Starting with headers, for examples:
A 32” door header supporting a point load from a ceiling beam that carries the ceiling only.
A 101” 2-ply #2 SYP 2x12 cased opening header supporting a point load from a ceiling beam that carries the ceiling only.
The same header supporting beams carrying ceiling and roof bracing loads.
And, are there specific rules of thumb that can be applied for like circumstances?
I find it curious that in the ’95 CABO, per table 602.6, MAXIMUM SPANS FOR HEADERS LOCATED OVER OPENINGS IN WALLS (feet), these spans are listed for a 2-ply 2x12:
(note reduction in spans with ’06 IRC):
1995 CABO (602.6): #2 grade lumber with 10’ tributary floor and roof loads (italics mine)
Supporting roof only: (presumably, this includes ceilings?) 12’
One story above: 10’
Two stories above: 8’
Headers in walls not supporting floors or roofs: 16’
(presumably, ceilings only?)
2006 IRC (R502.5(1)): #2 grade lumber assuming 20’ building width (italics mine)
Supporting roof only: (presumably, includes ceilings?) 9-9 (roof and ceiling)
One story above: 7-1 (w/clear span floor)
One story above: 8-1 (w/center bearing floor)
Two stories above: 5-6 (w/2 clear span floors)
Two stories above: 6-8 (w/2 ctr bearing floors)
Headers in walls not supporting floors or roofs: No listing
(presumably, ceilings only?)
What accounts for the reduced spans?
Why does the IRC fail to address sizing headers for “ceiling loads only?
Most homes built in my ahj are stick-built, and commonly feature multiple pan or tray ceiling systems comprised of built-up beams, laminated beams, LVL’s and dimensional lumber. Often, beams supporting portions of the ceiling (only) will rest above a window, door, or cased opening header. At other times, the beams will carry a portion of the ceiling and a partial roof load, predominately from roof bracing. Sometimes the additional roof load will be a minor load resulting from a purlin brace. At other times, the additional roof load will be more significant, say a 1500-2000 pound load resulting from supporting the bottom end of a half-valley.
I am clumsily working a few examples using the beam calc tables listed by DRP earlier in this thread. And, as previously mentioned, am interested in determining to what extent a building inspector that depends on the tables can account for any point loads accumulating on headers, and develop safe rules of thumb.
(Later in the thread, I will address the same questions as apply to girders).
Note: The example I am working on now is for a 101” 2-ply #2 2x12 header having a point load resulting from one end of a center ceiling beam resting above the header.
20 x 20 ceiling (uninhabitable w/o storage. 10 psf live load, 5 psf dead load – R802.4(1)).
So far, I have determined that the beam carries a tributary load of 3000 lbs, resulting in a 1500 lb. point load on the header.
Last edited by a moderator:
Jobsaver
Registered User
Some preliminary results are indicating no problems with point loads resulting from "ceiling only" loads. Comments?
My '93 CABO must be set up differently, sorry, I don't have a '95. Building widths in mine start at 24', headers are in 402.6- In our PM discussing reading and adjusting design values in the NDS Supplement we came up with an Fb for#2- 2x12 SYP of 975 psi. This kicks me into my table 402.6e. (This is labelled SPF...I think I'm seeing a bit of history. In the mid 90's the In Grade testing program derated 2x12's a bit and adjusted values so there may be several factors at work. Also notice the low Fv values, around '03 or so a math error at AF&PA was found from the '40's. Fv values are now about 190% of the old values...probably all more than you wanted to know right now!) Anyway, I'm not sure how much to make of the discrepencies between new and old tables without an old engineer to clarify all this....
I do believe you are on the right track mathwise though;
I do believe you are on the right track mathwise though;
It would depend on member, span and load but for your 2-2x12's @ 101" span with 1500 lbs yes it passes easily.
dhengr
Bronze Member
Jobsaver:
Point loads or concentrated loads are not included in these tables, and there is no simple rule of thumb to include them. They are more complicated to tabulate than uniform loads. You must know some real engineering and understand the development of the beams formulas and be able to manipulate them to include or add point loads to the tabulated info. You’re sorta asking for a Ph.D. in Structural Engineering in one forum post. But, I’m glad you are finally considering the possibility that there might be loads on headers and girders in addition to the simple uniform loading from some tributary floor widths, which became the primary design basis of your original question about the 3 or 4 ply built-up 2x6 center line fl. girder with about 6' spans, on a 20' wide bldg. I had this post pretty well along before I saw your post #27, so we’ll get to some of that later.
First lets talk a bit about the IRC and the two tables R502.5(1) & (2), and I’m looking at the 2006IRC also. The IRC is set up pretty conservatively to keep you BO’s and builders out of trouble, much of the time, without needing engineering help on many of the problems you encounter, as long as you follow the IRC and understand its limitations. It is based on the IBC and the old std. “design in accordance with acceptable engineering practice.” This does not mean every condition you encounter can be taken care of without some engineering involvement, or that every situation is covered by a table or code section, and you must know when you need that help. We would never set it up that way for fear that we would go hungry for lack of paying work.
The footnotes and sub headings in these types of tables are very important and every word must be read for its most restrictive meaning. I believe these two tables basically originated a number of code generations ago, long before the ICC, IBC or IRC, the general format looks familiar, but I haven’t really dug into their exact assumptions and variables or span values for a long time. However, let me lay down a few ground rules, or some of my thoughts about how they work and how the values were determined. Starting at the top of table R502.5(1):
1.) These are for girders and headers in exterior walls which will pick up half the roof load on a std. gable roof system, whether framed with rafters or metal plate connected trusses, plus some portion of the ceiling/attic load as indicated in the left column of the table. And, while you see 20, 28 & 36' for bldg. widths, a 2' overhang is probably included. Then they may also pick up floor loads, either center bearing (or near center bearing), which means that one half of each jst. span length goes to its exterior wall; or clear spanned which puts all the floor loading on the two exterior walls, as if using parallel chord fl. trusses. Draw some cross sections of these different bldgs. for all five of the vert. groupings in the left column of the table, with a view toward how these various loads get to the two exterior walls. Pick some reasonable DL’s & LL’s and tabulate them, at each level, on each wall or header, in lbs./lf of wall, and let’s talk about and check them, before you start using the beam formulas. Also note that the snow load increases as you move to the right in the table and that the bldg. width increases moving to the right under each snow loading sub-group, increasing the wall loadings. And, you should start to see heavier loadings and members and/or shorter spans for a given member as you move right and downward in the table. This should answer some of your questions in post #27.
2.) Now the second item down in the heading: they allow for four different groups of lumber, and then the all important footnote ‘b’, #2 grade lumber. This sets the allowable stresses for our design and tabulation, or the max. span length for a simple beam; this will be primarily a bending and deflection consideration problem, but may be a horiz. shear stress issue too, for the given pieces of lumber. But, if you look in the NDS, you’ll find that each of the groups of lumber have different allowable stresses and E’s (modulus of elasticity), so they will have used the lowest of the lumber species groups for the tabulation. They will also have used some set of average DL’s for roofs, ceilings and floor systems, and will likely have considered some typical amount of interior walls, which can be considered a uniform loading on the girder when they are parallel to the girder. Walls perpendicular to the girder cause concentrated loads on that girder and should usually have doubled joists under them, and are more than likely not included in these tabulations, that’s why I kept harping at you about them. You really must know when DL’s loads and various other load conditions exceed the assumptions used in developing these tables. And, that’s obviously not being understood, nor are those assumptions well spelled out in the tabulations, other than to say assume average DL’s.
3.) As you move down the page and to the right, for any given size of header, you finally need more than one or two jack studs. This is because the header reaction and thus the bearing stresses (and compression perpendicular to the grain) have gotten large enough so that 1.5" or 3" of bearing is not enough bearing length, and furthermore the jack studs acting as columns need to be larger for their assumed length and loading.
4.) I haven’t really studied all the variables and assumptions used in developing these tables for some time now, one would have to do some calcs. on random members at various loads and spans to start to hone in on the assumptions which were used in developing these tables. Finally, it might be easier to talk to the people who actually developed these tables for this code to really ascertain all of their assumptions and minor variations of variables. But, the basic engineering remains unchanged from when they were first developed years ago.
The table R502.5(2) involves essentially the same thinking as above, but obviously doesn’t include roof loads, but may include ceiling/attic loads, and must assume stacked bearing walls right over the girder. Table R502.3.1(2) for joists may be the simplest table to do some basic beam calcs. on, to get your feet wet, since it has the fewest hidden assumptions, in its development. The stress grade, member size and span length are obvious, the DL & LL are straight forward, except for the adjustment for the jst. spacing to convert the load from #/sf to #/lf. Then you apply the beam formulas to get moments, shears and deflections, and go to the NDS for actual stress calcs. and compare them to the allowable stresses and deflections, to determine if the member passes or fails.
DRP has given a couple bending moment equations in recent posts, M=WL/8 for a uniform load and M=PL/4 for a point load at center span and noted that if P=W, in magnitude, the bending moment and thus the stresses are doubled for the point load on a simple beam. He also showed the beam formulas for a simple beam, uniformly loaded, post #17, in DarrenE’s thread on ‘number of jack studs,’ and I believe he started to outline some of the stress calcs. for your original girder question. I believe his beam formula figure #1 came right out of the NDS code books, which show many different loading conditions and types of beams. There are other places to see these same formulas.
I believe changes in these tables from one generation of the code to the next will have most to do with the poorer quality lumber we are having to use these days, and thus lower allowable stresses and E values. There are also changes in the code requirements and various adjustment factors in the codes. The underlying engineering, strength of materials, stress analysis, etc. has not really changed. All of the above is what we keep calling “design in accordance with acceptable engineering practice,” picking a girder or header from these tables is not engineering, even though the tables are based on acceptable engineering practice. If you are going to do anything other than picking a header from the tables, after truly understanding the limitations of that table, you are starting to do some engineering and you better understand what you are doing and your own limitations. You really must be comfortable determining DL’s & LL’s, tributary areas, types of loads, concentrated or uniform, load paths from origin of the load to the bearing soil. And then you must be comfortable using the beam formulas and combining them as appropriate for the types of loads or spans involved, or you should probably be asking for engineering help. Since, just as for Architects and Engineers, you should really not be practicing engineering activities beyond your experience and training level. Thus, the admonition that if it isn’t specifically in the prescriptive code (the IRC) you should get engineering help, or have the builder get an Engineering involved.
Point loads or concentrated loads are not included in these tables, and there is no simple rule of thumb to include them. They are more complicated to tabulate than uniform loads. You must know some real engineering and understand the development of the beams formulas and be able to manipulate them to include or add point loads to the tabulated info. You’re sorta asking for a Ph.D. in Structural Engineering in one forum post. But, I’m glad you are finally considering the possibility that there might be loads on headers and girders in addition to the simple uniform loading from some tributary floor widths, which became the primary design basis of your original question about the 3 or 4 ply built-up 2x6 center line fl. girder with about 6' spans, on a 20' wide bldg. I had this post pretty well along before I saw your post #27, so we’ll get to some of that later.
First lets talk a bit about the IRC and the two tables R502.5(1) & (2), and I’m looking at the 2006IRC also. The IRC is set up pretty conservatively to keep you BO’s and builders out of trouble, much of the time, without needing engineering help on many of the problems you encounter, as long as you follow the IRC and understand its limitations. It is based on the IBC and the old std. “design in accordance with acceptable engineering practice.” This does not mean every condition you encounter can be taken care of without some engineering involvement, or that every situation is covered by a table or code section, and you must know when you need that help. We would never set it up that way for fear that we would go hungry for lack of paying work.
The footnotes and sub headings in these types of tables are very important and every word must be read for its most restrictive meaning. I believe these two tables basically originated a number of code generations ago, long before the ICC, IBC or IRC, the general format looks familiar, but I haven’t really dug into their exact assumptions and variables or span values for a long time. However, let me lay down a few ground rules, or some of my thoughts about how they work and how the values were determined. Starting at the top of table R502.5(1):1.) These are for girders and headers in exterior walls which will pick up half the roof load on a std. gable roof system, whether framed with rafters or metal plate connected trusses, plus some portion of the ceiling/attic load as indicated in the left column of the table. And, while you see 20, 28 & 36' for bldg. widths, a 2' overhang is probably included. Then they may also pick up floor loads, either center bearing (or near center bearing), which means that one half of each jst. span length goes to its exterior wall; or clear spanned which puts all the floor loading on the two exterior walls, as if using parallel chord fl. trusses. Draw some cross sections of these different bldgs. for all five of the vert. groupings in the left column of the table, with a view toward how these various loads get to the two exterior walls. Pick some reasonable DL’s & LL’s and tabulate them, at each level, on each wall or header, in lbs./lf of wall, and let’s talk about and check them, before you start using the beam formulas. Also note that the snow load increases as you move to the right in the table and that the bldg. width increases moving to the right under each snow loading sub-group, increasing the wall loadings. And, you should start to see heavier loadings and members and/or shorter spans for a given member as you move right and downward in the table. This should answer some of your questions in post #27.
2.) Now the second item down in the heading: they allow for four different groups of lumber, and then the all important footnote ‘b’, #2 grade lumber. This sets the allowable stresses for our design and tabulation, or the max. span length for a simple beam; this will be primarily a bending and deflection consideration problem, but may be a horiz. shear stress issue too, for the given pieces of lumber. But, if you look in the NDS, you’ll find that each of the groups of lumber have different allowable stresses and E’s (modulus of elasticity), so they will have used the lowest of the lumber species groups for the tabulation. They will also have used some set of average DL’s for roofs, ceilings and floor systems, and will likely have considered some typical amount of interior walls, which can be considered a uniform loading on the girder when they are parallel to the girder. Walls perpendicular to the girder cause concentrated loads on that girder and should usually have doubled joists under them, and are more than likely not included in these tabulations, that’s why I kept harping at you about them. You really must know when DL’s loads and various other load conditions exceed the assumptions used in developing these tables. And, that’s obviously not being understood, nor are those assumptions well spelled out in the tabulations, other than to say assume average DL’s.
3.) As you move down the page and to the right, for any given size of header, you finally need more than one or two jack studs. This is because the header reaction and thus the bearing stresses (and compression perpendicular to the grain) have gotten large enough so that 1.5" or 3" of bearing is not enough bearing length, and furthermore the jack studs acting as columns need to be larger for their assumed length and loading.
4.) I haven’t really studied all the variables and assumptions used in developing these tables for some time now, one would have to do some calcs. on random members at various loads and spans to start to hone in on the assumptions which were used in developing these tables. Finally, it might be easier to talk to the people who actually developed these tables for this code to really ascertain all of their assumptions and minor variations of variables. But, the basic engineering remains unchanged from when they were first developed years ago.
The table R502.5(2) involves essentially the same thinking as above, but obviously doesn’t include roof loads, but may include ceiling/attic loads, and must assume stacked bearing walls right over the girder. Table R502.3.1(2) for joists may be the simplest table to do some basic beam calcs. on, to get your feet wet, since it has the fewest hidden assumptions, in its development. The stress grade, member size and span length are obvious, the DL & LL are straight forward, except for the adjustment for the jst. spacing to convert the load from #/sf to #/lf. Then you apply the beam formulas to get moments, shears and deflections, and go to the NDS for actual stress calcs. and compare them to the allowable stresses and deflections, to determine if the member passes or fails.
DRP has given a couple bending moment equations in recent posts, M=WL/8 for a uniform load and M=PL/4 for a point load at center span and noted that if P=W, in magnitude, the bending moment and thus the stresses are doubled for the point load on a simple beam. He also showed the beam formulas for a simple beam, uniformly loaded, post #17, in DarrenE’s thread on ‘number of jack studs,’ and I believe he started to outline some of the stress calcs. for your original girder question. I believe his beam formula figure #1 came right out of the NDS code books, which show many different loading conditions and types of beams. There are other places to see these same formulas.
I believe changes in these tables from one generation of the code to the next will have most to do with the poorer quality lumber we are having to use these days, and thus lower allowable stresses and E values. There are also changes in the code requirements and various adjustment factors in the codes. The underlying engineering, strength of materials, stress analysis, etc. has not really changed. All of the above is what we keep calling “design in accordance with acceptable engineering practice,” picking a girder or header from these tables is not engineering, even though the tables are based on acceptable engineering practice. If you are going to do anything other than picking a header from the tables, after truly understanding the limitations of that table, you are starting to do some engineering and you better understand what you are doing and your own limitations. You really must be comfortable determining DL’s & LL’s, tributary areas, types of loads, concentrated or uniform, load paths from origin of the load to the bearing soil. And then you must be comfortable using the beam formulas and combining them as appropriate for the types of loads or spans involved, or you should probably be asking for engineering help. Since, just as for Architects and Engineers, you should really not be practicing engineering activities beyond your experience and training level. Thus, the admonition that if it isn’t specifically in the prescriptive code (the IRC) you should get engineering help, or have the builder get an Engineering involved.
Jobsaver
Registered User
DRP said:
This helps me to understand some of the differences between the older and newer codes. I had imagined that some of the difference also occured because during the compilation process to create the ICC codes, at every turn the most conservative information was chosen. It does not stand to reason that the panels of experts performing the compilations did their own math to re-establish span tables, including each span in relation to each footnote selected.dhengr said:Jobsaver
One would have to do some calcs. on random members at various loads and spans to start to hone in on the assumptions which were used in developing these tables. and go to the NDS for actual stress calcs. and compare them to the allowable stresses and deflections, to determine if the member passes or fails.
I believe changes in these tables from one generation of the code to the next will have most to do with the poorer quality lumber we are having to use these days, and thus lower allowable stresses and E values. There are also changes in the code requirements and various adjustment factors in the codes. The underlying engineering, strength of materials, stress analysis, etc. has not really changed.
Why is there no distinction for headers bearing "ceiling loads only"? In my eyes, it is a flaw in the compilation resulting in grossly conservative, even missing, information.
Of course "rules of thumb" can be established to determine to what extent the maximum spans for headers listed can accept "ceiling loads only", point or otherwise.
I submit this equation: think of it as a test question on a exam:
Prove the following:
For any interior header selected out of table R502.5(2), for a typical single story home, under what circumstance will the header fail given the following conditions?:
The interior header is supporting ceiling loads only for no more than two rooms having a combined gross ceiling area of 700 sq. ft.
There exists no more than one concentrated load on the header resulting from a ceiling beam.
I am exploring this question, which includes performing a number of calcualtions using the maximum spans given in table R502.5(2), to ascertain whether or not typical loading confiqurations found in my ahj are overloading table sized headers, despite certain point loads.
Remember:
Most homes built in my ahj are stick-built, and commonly feature multiple pan or tray ceiling systems comprised of built-up beams, laminated beams, LVL’s and dimensional lumber. Often, beams supporting portions of the ceiling (only) will rest above a window, door, or cased opening header.
The simple answer is, no. Or, you must hire an engineer whenever a supporting pan or tray ceiling beam rests over a header.
These are both inadequate answers that reflect an ideal that is not my reality.
Last edited by a moderator:
Daddy-0-
Moderator
I make them post the load all the way to the foundation with solid dimensional blocking beside the rim board. Otherwise, you have defeated the purpose.
At what point? Under every door jack or under significant loads? I think it would depend on the magnitude of the load and the compressive strength of the rim. Aside, most of the engineered rimboard has higher compressive strength than sawn lumber.