-
Welcome to The Building Code Forum
Your premier resource for building code knowledge.
This forum remains free to the public thanks to the generous support of our Sawhorse Members and Corporate Sponsors. Their contributions help keep this community thriving and accessible.
Want enhanced access to expert discussions and exclusive features? Learn more about the benefits here.
Ready to upgrade? Log in and upgrade now.
Floor Girder Problem?
- Thread starter Jobsaver
- Start date
dhengr
Bronze Member
Jar:
Given your twitter like brevity, what comments, by whom? That looks like just another way to skin the same cat. But, the engineers didn’t really give you this printout did they, they licensed the program to you, with some instructions about how to use it, and with some disclaimers too, right? Is there some mysterious difference btwn. pages 1 & 2 of your attachment, I couldn’t find any?
I keep screaming for a full accounting of the loads on that girder, so far we have the 1st fl. DL (10#/sf) and LL (40#/sf) on two 10' long simple span joists framing into that girder for a 10' tributary width, leading to a 500#/ft. uniform loading on the girder; the joists are not 20' long and continuous over the girder; and there are no partitions, 1st fl. opening jamb loads, no ceiling, attic, or roof loads being added to that beam; and I doubt that. Without seeing the plans and having some real involvement in the project, or sufficient experience working with that builder so I know the quality of his work and how he interprets what I tell him to do, I don’t know if I’m getting the full story, and I do know who’ll be hung out to dry if things go wrong. Because your very own statement, “This is what the “engineers” gave to me.” would me misinterpreted. I’ll usually give my opinion and general guidance on the problem, as best I understand the word picture that you have presented, but I just can’t (won’t) do real engineering over the internet.
As to your printout, I don’t particularly like the format either, but could probably get used to it after using the program for a while. I would like to see what you input, it looks like all the answers are there, but I want to study them a bit more before commenting. That is usually the case with any of the programs or spreadsheets out there, you have to use them often enough to get comfortable with the input and output format, and what you can expect from them. Also, their limitations, exceptions and the like must be kept constantly in mind when using them. Your’s or DRP’s might not be exactly the way I’d write the program, or the input or the output, but we should all get basically the same answer to a given problem. So, now we have the IRC tables for joists and girders, we have DRP’s spreadsheet, and we have your BeamChek v2010 program which purports to comply with NDS 2005, which I don’t have a copy of, but I think I can muddle through, if you guys give me a clue now and then. I suggest you use your program, just as I did Jobsaver, to work out a few of the tabulations in the IRC joist and girder tables, and lets see how they compare and if we can explain any differences.
Given your twitter like brevity, what comments, by whom? That looks like just another way to skin the same cat. But, the engineers didn’t really give you this printout did they, they licensed the program to you, with some instructions about how to use it, and with some disclaimers too, right? Is there some mysterious difference btwn. pages 1 & 2 of your attachment, I couldn’t find any?
I keep screaming for a full accounting of the loads on that girder, so far we have the 1st fl. DL (10#/sf) and LL (40#/sf) on two 10' long simple span joists framing into that girder for a 10' tributary width, leading to a 500#/ft. uniform loading on the girder; the joists are not 20' long and continuous over the girder; and there are no partitions, 1st fl. opening jamb loads, no ceiling, attic, or roof loads being added to that beam; and I doubt that. Without seeing the plans and having some real involvement in the project, or sufficient experience working with that builder so I know the quality of his work and how he interprets what I tell him to do, I don’t know if I’m getting the full story, and I do know who’ll be hung out to dry if things go wrong. Because your very own statement, “This is what the “engineers” gave to me.” would me misinterpreted. I’ll usually give my opinion and general guidance on the problem, as best I understand the word picture that you have presented, but I just can’t (won’t) do real engineering over the internet.
As to your printout, I don’t particularly like the format either, but could probably get used to it after using the program for a while. I would like to see what you input, it looks like all the answers are there, but I want to study them a bit more before commenting. That is usually the case with any of the programs or spreadsheets out there, you have to use them often enough to get comfortable with the input and output format, and what you can expect from them. Also, their limitations, exceptions and the like must be kept constantly in mind when using them. Your’s or DRP’s might not be exactly the way I’d write the program, or the input or the output, but we should all get basically the same answer to a given problem. So, now we have the IRC tables for joists and girders, we have DRP’s spreadsheet, and we have your BeamChek v2010 program which purports to comply with NDS 2005, which I don’t have a copy of, but I think I can muddle through, if you guys give me a clue now and then. I suggest you use your program, just as I did Jobsaver, to work out a few of the tabulations in the IRC joist and girder tables, and lets see how they compare and if we can explain any differences.
Jar, nice, I tried the demo a few years ago, nice. Beamchek is more elegant than my efforts but the formulas are the same. It checked allowable compression perp to grain and gave minimum bearing area to avoid crushing the side of the beam. It then adjusted the span length and load to account for half the required bearing length at each end to arrive at the "true" span as per NDS. Curious why you , or it, didn't take the repetitive member increase, it is allowed here.
I was needing something that allowed me to input timbers of many species and sizes. I have a sawmill and one of the most diverse forests on the planet right outside my door. That was the limiting part of the programs I tried and what I and a few others needed. Try figuring a heavy timber barn swapping out a variety of sizes and species at the kitchen table manually. I've done it at a friends kitchen table with 7 kids running in and out, the mill idling in the background and we were trying to resize because the logs were opening up differently than planned. I wrote a few scripts and put them online. If I can get on their computer we can play "what if" a whole lot quicker than punching numbers on a calculator. I've written versions with drop down species lists that then automatically input the design values for common timbers but this is the simplest, so most versatile, we just used it with dimensional lumber, it'll work with lvl's as well.
I believe my program will come up more conservative on joist or rafter span lengths that are limited by deflection due to the fact that I am running the "pass/fail" on total load rather than LL only. Beamcheck should click right in with the awc spancalc. I can manually change the input to LL only and run again if I want to push the line, I don't feel a need to tell everyone that. If you're using this for conventional joists and rafters you're backing up to begin with. I let the javascript run out to the limits on decimals for simplicity, it could round but personally I just ignore beyond thousandths and convert to fractions in my head. You can view the code by right clicking a blank area on the calc and then click "view source". I bang nails for a living, feel free to modify the code and post your improvements.
Getting back to the original problem. If you run some checks against the girder table you'll see that a triple 2x6 is running a good bit less conservative than their trend or than a solid sawn 6x6 with the WSDD grade specs. I suspect they are leaving some wiggle room for the unknown potential conditions dhengr described and for the possibility that the girders in the prescriptive table may be side loaded. The math can be shown to work at 3 ply but there's certainly nothing wrong with 4 ply, there is no need to pass by the skin of our teeth.
Dr's Woeste, Bender and Loferski at VA Tech taught a short course on the '05 NDS and mixed in some other short topics, one was on floor vibration frequency... annoying floor vibration. The combination of joists designed to the limits and exacerbated by girders near the limits can produce some really annoying floors. When you run those calculations a spongy girder can easily screw up an otherwise acceptable set of joists. They suggested designing girders for L/600. Assuming 2x8 joists this floor is ok with either 3 or 4 ply vibrationwise since the spans are short but it did pick up >3 Hz going from SPF and 3 ply to all SYP and a 4 ply girder, a noticeably "tighter" feeling floor.
I was needing something that allowed me to input timbers of many species and sizes. I have a sawmill and one of the most diverse forests on the planet right outside my door. That was the limiting part of the programs I tried and what I and a few others needed. Try figuring a heavy timber barn swapping out a variety of sizes and species at the kitchen table manually. I've done it at a friends kitchen table with 7 kids running in and out, the mill idling in the background and we were trying to resize because the logs were opening up differently than planned. I wrote a few scripts and put them online. If I can get on their computer we can play "what if" a whole lot quicker than punching numbers on a calculator. I've written versions with drop down species lists that then automatically input the design values for common timbers but this is the simplest, so most versatile, we just used it with dimensional lumber, it'll work with lvl's as well.
I believe my program will come up more conservative on joist or rafter span lengths that are limited by deflection due to the fact that I am running the "pass/fail" on total load rather than LL only. Beamcheck should click right in with the awc spancalc. I can manually change the input to LL only and run again if I want to push the line, I don't feel a need to tell everyone that. If you're using this for conventional joists and rafters you're backing up to begin with. I let the javascript run out to the limits on decimals for simplicity, it could round but personally I just ignore beyond thousandths and convert to fractions in my head. You can view the code by right clicking a blank area on the calc and then click "view source". I bang nails for a living, feel free to modify the code and post your improvements.
Getting back to the original problem. If you run some checks against the girder table you'll see that a triple 2x6 is running a good bit less conservative than their trend or than a solid sawn 6x6 with the WSDD grade specs. I suspect they are leaving some wiggle room for the unknown potential conditions dhengr described and for the possibility that the girders in the prescriptive table may be side loaded. The math can be shown to work at 3 ply but there's certainly nothing wrong with 4 ply, there is no need to pass by the skin of our teeth.
Dr's Woeste, Bender and Loferski at VA Tech taught a short course on the '05 NDS and mixed in some other short topics, one was on floor vibration frequency... annoying floor vibration. The combination of joists designed to the limits and exacerbated by girders near the limits can produce some really annoying floors. When you run those calculations a spongy girder can easily screw up an otherwise acceptable set of joists. They suggested designing girders for L/600. Assuming 2x8 joists this floor is ok with either 3 or 4 ply vibrationwise since the spans are short but it did pick up >3 Hz going from SPF and 3 ply to all SYP and a 4 ply girder, a noticeably "tighter" feeling floor.
Last edited by a moderator:
Jobsaver
REGISTERED
But . . . what?GHRoberts said:
I looked at the second page expecting to see calcs for a 4-ply beam?dhengr said:
Respectfully, why? The balance of prescriptive spans, combined with the other elements prescripted in the IRC are thought to be conservative enough not to worry about a full accounting of the loads on an individual girder. In other words, if the codebook formulas are derivative of conservative engineering, then it is not necessary to have a full accounting of the loads in order to use the tables. I always imagined that when engineers, architects, and other design professionals were designing framing elements for a home, something as basic as described in this thread, that even you guys turned to span charts here and again.dhengr said:
Note the following:
Jobsaver said:
Jobsaver said:
It seems to me that we are all saying the same thing. And, that every party is reaching the same conclusion: A 3-ply girder will probably work okay, but it is close, and depending on where the walls are . . . you might want to make it a 4-ply just to be sure.DRP said:
jar546
CBO
In all actuality and in real life, the poor planning of the masonry contractor would have to bear responsibility for the costs to be incurred for engineering services if they want to build outside of the prescriptive code requirements.
I would need a stamped spec sheet from a PE if they did not want to comply prescriptively. That is truly the bottom line.
I would need a stamped spec sheet from a PE if they did not want to comply prescriptively. That is truly the bottom line.
As a "PE" I do not work for the AHJ. I provide information to my clients. That information is never stamped. My clients have the option of passing the information on to the AHJ if they desire. (Unless I am doing the actual physical construction the engineering is for "informational" purposes only.)jar546 said:
I believe the code requires "engineering" not work by an engineer. Certainly not work containing an engineer's "stamp."
---
The nice part of engineering is that anyone can do it for themselves. Most people don't have a "stamp."
jar546
CBO
Little thread drift.
Why I like BeamChek V2010:
1) OK so it is relatively easy to use but relative is a relative word. I still find myself calling up the experts to verify I am inputing the correct data
2) I get updates about 3 times a year as the manufacturers specs change which make for better accuracy
3) Just about all manufacturers are on there. Have yet to not have an LVL make/model present
4) We can look at heavier solid lumber such as 4x6, 6x8, etc
5) It does steel beams
6) It does more than just simple spans
7) It can do concentrated loads
8) It is an all around versatile program
Now WHY I we use the program (this is where I expect to be flamed)
1) To check submissions that are not prescriptive yet come from an architect and not an engineer.
2) For my own personal enjoyment.
Here is an example of a recent use:
We had a recent submission for a 24' wide ranch home with clear span roof trusses. The main girder in the basement was specified by the architect as a (3) 2x10 spf with 1/2" ply in between. The vertical support posts were 9'5" apart. The bedrooms were split up by bathrooms, hallway and a laundry room so there was no particular area where it was all bedrooms.
Table R502.5(2) Prescriptively shows that this beam is inadequate for supporting 1 floor whether he attempts to reduce to the 20' width column, bump up to the 28' column or interpolate in between.
I cannot verify what the plywood will do, especially when it is not prescriptive and there will be obvious joints not over supports for this multi-span. Bottom line, I cannot take that into consideration.
To give the architect the benefit of the doubt, I inputed the girder he planned into BeamChek and it still failed.
When I sent the plan review back, he argued that the prescriptive table does not include 24' wide and he just put this in another town with no problems during plan review. I told him that his problem was beyond the prescriptive code and that I ran it on beam check to see if it would pass and it of course failed. That did not stop the argument from him so I said to either change the beam to meet the code prescriptively or hire and engineer and provide a stamp and calculations for our records.
Basically, I just wanted to see how close the prescriptive tables were and also use it to re-enforce my decision since I was dealing with a DP.
BTW, he later submitted the same plan to another town that used me to do the plan review and he never learned his lesson and failed again.
Why I like BeamChek V2010:
1) OK so it is relatively easy to use but relative is a relative word. I still find myself calling up the experts to verify I am inputing the correct data
2) I get updates about 3 times a year as the manufacturers specs change which make for better accuracy
3) Just about all manufacturers are on there. Have yet to not have an LVL make/model present
4) We can look at heavier solid lumber such as 4x6, 6x8, etc
5) It does steel beams
6) It does more than just simple spans
7) It can do concentrated loads
8) It is an all around versatile program
Now WHY I we use the program (this is where I expect to be flamed)
1) To check submissions that are not prescriptive yet come from an architect and not an engineer.
2) For my own personal enjoyment.
Here is an example of a recent use:
We had a recent submission for a 24' wide ranch home with clear span roof trusses. The main girder in the basement was specified by the architect as a (3) 2x10 spf with 1/2" ply in between. The vertical support posts were 9'5" apart. The bedrooms were split up by bathrooms, hallway and a laundry room so there was no particular area where it was all bedrooms.
Table R502.5(2) Prescriptively shows that this beam is inadequate for supporting 1 floor whether he attempts to reduce to the 20' width column, bump up to the 28' column or interpolate in between.
I cannot verify what the plywood will do, especially when it is not prescriptive and there will be obvious joints not over supports for this multi-span. Bottom line, I cannot take that into consideration.
To give the architect the benefit of the doubt, I inputed the girder he planned into BeamChek and it still failed.
When I sent the plan review back, he argued that the prescriptive table does not include 24' wide and he just put this in another town with no problems during plan review. I told him that his problem was beyond the prescriptive code and that I ran it on beam check to see if it would pass and it of course failed. That did not stop the argument from him so I said to either change the beam to meet the code prescriptively or hire and engineer and provide a stamp and calculations for our records.
Basically, I just wanted to see how close the prescriptive tables were and also use it to re-enforce my decision since I was dealing with a DP.
BTW, he later submitted the same plan to another town that used me to do the plan review and he never learned his lesson and failed again.
jar546
CBO
George, you are really working hard to find something to be a detractor about, huh?GHRoberts said:
BTW, I don't care who hires or pays the PE or who the PE is working for as it is the permit applicant who must submit it to the AHJ before work can continue or begin.
You said you require an engineer's stamp. Show me the code section that requires an applicant to provide a stamp.jar546 said:
----
I am amazed that you admit you don't know how to enter data into Beam Check, but you accept the results you get from doing so.
Jobsaver
REGISTERED
jar546 said:
Given a choice, should I be counting on someone, I'd put my confidence in the man that is not afraid to ask a question, and understands the meaning of a sentence.GHRoberts said:
We've shown that the engineering on this is pretty basic, we have general agreement on methods, results and interpretation of the results. Code interpretation is a bit more mixed. I don't feel a need to make the contractor get a stamped design for a simple girder. Jar shows that blindly trusting a RDP's stamp might not be wise. Knowing how to design a simple beam is much more useful for all parties. Out of the box software, or using my calc has it's benefits, it gets you up and running quickly. It has some drawbacks, you may or may not understand what it is doing. That has been my point in participating in these threads, either sharing what I know or learning from those who know more. I'd be more than happy to continue that part of the discussion. It is counterproductive to decide that we cannot discuss engineering topics because we might miss something, using that approach we will most certainly miss everything every time!
As for the code interp. Look at the stipulations on grading... pretty clear and focused ( I happen to disagree with them as well), now look at the call for engineering... much broader. I pointed out footnote 3 in your CABO for a reason. The intent is clear, simply show that you have considered non prescriptive methods in a rational way. Just as with those grading requirements you are slowly rewriting the law through commentary and interpretation. Jobsaver, you sound alot like our old BO. If I ran into trouble he was my first call. I know the answer now will be "get an engineer" so I don't bother him much. It is a two way street and unless the inspector has all the time in the world he simply isn't going to look in all the right places. When we were working together he already knew the places I was concerned with.
As for the code interp. Look at the stipulations on grading... pretty clear and focused ( I happen to disagree with them as well), now look at the call for engineering... much broader. I pointed out footnote 3 in your CABO for a reason. The intent is clear, simply show that you have considered non prescriptive methods in a rational way. Just as with those grading requirements you are slowly rewriting the law through commentary and interpretation. Jobsaver, you sound alot like our old BO. If I ran into trouble he was my first call. I know the answer now will be "get an engineer" so I don't bother him much. It is a two way street and unless the inspector has all the time in the world he simply isn't going to look in all the right places. When we were working together he already knew the places I was concerned with.
Jobsaver
REGISTERED
DRP: Your post sums up my thinking as well. I prefer to be a resource, not just another link in the chain of bureaucracy. And, to be a good resource, one needs good resources.
I am currently a mediocre resource aspiring to become a good one. To know all of the methods, each of the interpretations, is the goal. Until then, I will sustain my ambition by asking a lot of questions to those accomplished that know more than I.
I am currently a mediocre resource aspiring to become a good one. To know all of the methods, each of the interpretations, is the goal. Until then, I will sustain my ambition by asking a lot of questions to those accomplished that know more than I.
Did you notice that in post #24 Jar had the span wrong, the tributary "width" wrong, and the dead load wrong?Jobsaver said:
You might notice that in post #32, Jar indicated that he would accept his own unsealed unlicensed opinion if BeamChk (improperly used - wrong span, wrong tributaqry "width", and wrong dead load) gave "OK"s. But if a licensed professional submitted the same it would need a seal. (Read all that Jar writes about what he does. Not just those points that support your point of view.)
Perhaps you could be helpful and tell us why the numbers are wrong. (It is impossible to determine some of the correct numbers from the information given.)
Jobsaver
REGISTERED
GH: The span given in the OP is 68". Jeff's beamchek printout shows the beam span to be 5.67'. This is correct. I do not understand much of the balance of the printout. Where is the tributary width indicated? The dead load? At the bottom of the table portion of the document, I see printed, Uniform LL:400, and, Uniform TL: 500 = A, but these fiqures do not mean anything to me as presented.
I do not read post #32 the same as you. While anyone can make a mistake entering data, it appears to me that Jeff is particular to say that he has not mastered making the determination of which data is required in every circumstance, (I have experienced the same problem using beam sizing software, and will sometimes call someone more proficient to guide me through a particular equation.)
Also, he is particular to say in what two circumstances he (his company) uses the program:
"Now WHY I we use the program (this is where I expect to be flamed)
1) To check submissions that are not prescriptive yet come from an architect and not an engineer.
2) For my own personal enjoyment."
Then he proceeds to give a specific historical incident that includes checking a beam designed by an architect that seemed suspect, and that did not meet the prescriptive requirements of the code, rendering it further suspect, while dealing with an architect, who in effect, instead of doing and producing the math, said, "this is the way we always do it". Should that same architect have produced an appropriate solution in writing, sealed (professional verification), the solution would be accepted.
Making an issue of whether or not a Certified DP's document, or opinion, has to be stamped or not is a nonsensicle argument . . . of course it is in the BO's prerogative to require a stamped document before accepting a solution.
Lastly, concerning Jeff's post #32, he stated, "this is where I expect to be flamed". Maybe . . . from a registered DP that cannot admit the possibility of self-error.
I do not read post #32 the same as you. While anyone can make a mistake entering data, it appears to me that Jeff is particular to say that he has not mastered making the determination of which data is required in every circumstance, (I have experienced the same problem using beam sizing software, and will sometimes call someone more proficient to guide me through a particular equation.)
Also, he is particular to say in what two circumstances he (his company) uses the program:
"Now WHY I we use the program (this is where I expect to be flamed)
1) To check submissions that are not prescriptive yet come from an architect and not an engineer.
2) For my own personal enjoyment."
Then he proceeds to give a specific historical incident that includes checking a beam designed by an architect that seemed suspect, and that did not meet the prescriptive requirements of the code, rendering it further suspect, while dealing with an architect, who in effect, instead of doing and producing the math, said, "this is the way we always do it". Should that same architect have produced an appropriate solution in writing, sealed (professional verification), the solution would be accepted.
Making an issue of whether or not a Certified DP's document, or opinion, has to be stamped or not is a nonsensicle argument . . . of course it is in the BO's prerogative to require a stamped document before accepting a solution.
Lastly, concerning Jeff's post #32, he stated, "this is where I expect to be flamed". Maybe . . . from a registered DP that cannot admit the possibility of self-error.
Jar did it right, lets work it longhand;
10' trib width x #40LL= 400 plf LL
10' trib width x #10DL=100plf DL
LL+DL= 500PLF TL
Max bending Moment= WL/8
W= (5.67'x #500)+#34 = #2869
M= (#2869 x 5.67')/8= 2033 ft-lbs
2033 ft-lbs x 12 =24396 in-lbs
section modulus = bd2/6
3-2x6's... (4.5 x 5.52)/6 = 22.69"3
24396/22.69= 1075 psi extreme fiber stress in bending.
the extreme fiber is the one furthest from the neutral axis, the strap of wood along the bottom edge of the beam has 1075 psi of stress. Look at Jars adjusted allowable stress, 1138 psi... we're safe. I took issue here, since there were 3 plies he could have adjusted upwards another 15% giving an Fb of 1308.
Blueberry muffins just got out, class dismissed
10' trib width x #40LL= 400 plf LL
10' trib width x #10DL=100plf DL
LL+DL= 500PLF TL
Max bending Moment= WL/8
W= (5.67'x #500)+#34 = #2869
M= (#2869 x 5.67')/8= 2033 ft-lbs
2033 ft-lbs x 12 =24396 in-lbs
section modulus = bd2/6
3-2x6's... (4.5 x 5.52)/6 = 22.69"3
24396/22.69= 1075 psi extreme fiber stress in bending.
the extreme fiber is the one furthest from the neutral axis, the strap of wood along the bottom edge of the beam has 1075 psi of stress. Look at Jars adjusted allowable stress, 1138 psi... we're safe. I took issue here, since there were 3 plies he could have adjusted upwards another 15% giving an Fb of 1308.
Blueberry muffins just got out, class dismissed
Last edited by a moderator:
jar546
CBO
Nice work guys. Looks like we all came to the same conclusion using our own methods, mine of which is a licensed software program.
Jobsaver
REGISTERED
Jobsaver said:
Thanks again DRP. I owe you an apple.DRP said:
Now I see I missed the footnote at the bottom of the Beamchek document, "Uniform and partial uniform loads are lbs per lineal ft."
Last edited by a moderator:
jar546
CBO
True, unless you strategically place spliced where there is neither negative nor positive moment. The baseline of the curve so to speak.Richard said:
Joints would, of course, be over the supports but for such small spans only one of the 3 members would need a joint at each support which would crate a much stronger and more stable girder. I realize the builder will use continuous lengths anyway and there is nothing wrong with exceeding the code reequirement for strength and deflection but it demonstrates why the prescriptive tables can be a poor substitute for god engineering design.