Floor Girder Problem?

[Jobsaver]

Why is there no span chart for a built-up girder comprised of (3) 2x6's, or (4) 2x6's.

Single-story pier and beam having 20' building width. Piers are constructed of double 8x8x16 cinderblock, solid grouted, 7' oc. Longest span between pier edges is 68".

Table R502.5(2) allows for a minimum girder of (2) 2x8 to span 5-9.

The circumstance is that the builder set the foundation up for a 5.5" height girder.

Solution A: Lower pier height to accomodate 2x8's.

Solution B: Build-up sill plate 1.75" so top of sill will match 2x8 girder height.

Solution C: Build-up multiple ply 2x6 girder.

How many 2x6 plies?

 

[fatboy]

Not specified in the IRC = engineer's analysis for the 2 X 6 built-up girder. Do you have anchor bolts to accomodate an additional 1.75" sill build up? If not, again, engineering solution. About the only non-engineered solution would be to lower the pier, but solid grouted? Ugh, won't be fun.

 

[GHRoberts]

Perhaps you can notch the ends.

But it should be easy enough to use a thicker bottom plate. Nail 2 2x's together and drill the top one for the nut, bottom one for the bolt.

---

Is this an inspection or your construction? Inspectors should not be offering solutions.

 

[DRP]

AWC's (AF&PA) Wood Structural Design Data;

http://awc.org/technical/spantables/index.html

This section of the multi part download would include the tables for that span;

http://www.awc.org/pdf/wsdd/c2b.pdf

 

[KZQuixote]

Perhaps you can notch the ends. But it should be easy enough to use a thicker bottom plate. Nail 2 2x's together and drill the top one for the nut, bottom one for the bolt.

---

Is this an inspection or your construction? Inspectors should not be offering solutions.

Sill plate is PT. Doubler sill plate is a regular 2X6. Put a coupling nut on the AB and run a 1/2" bolt down into the coupling nut through the doubler sill.

Bill

 

[KZQuixote]

Originally Posted by GHRoberts. " Inspectors should not be offering solutions".

This is the real world right? Where folks try to get along and find solutions that allow all the parties to save face.Bill

 

[GHRoberts]

This is the real world right? Where folks try to get along and find solutions that allow all the parties to save face.Bill

I agree with you. But tell me "Who is responsible when the inspector gives 'wrong" advice?"

---

I did s search using "wood girder span tables" A responsible source gives the following allowable loads for girders spanning 8':

4x6 223 pounds/ft

6x6 237

2x8 412

It would appear that 6" of depth is not going to do it.

 

[Jobsaver]

Thanks for the ideas. Good ideas for anchoring the double plate GH and Bill.

DRP: I looked up both of these charts before posting. I did not find them helpful. How does one go about applying these particular documents to the problem? To be clear, I am asking this question earnestly.

Also, puzzle me this everyone: 2006 IRC TableR502.5(2): One floor only:

How is it that a 2-ply 2x8 girder will span 5-9, a 3-ply will span 7-2, but a 4-ply will only span 5-10?

Next, technically, the code allows us, for widths between those shown, to interpolate spans. Now, I don't interpolate, I decipher (think Jethro Bodine - Beverly Hillbillies, it is what we do in Arkansas). My deciphering is as follows:

One can increase the span by 24.5% by adding one additional ply to a 2-ply 2x8 girder (5-9 to 7-2). One can increase the span by 25% by adding one additional ply to a 2-ply 2x10 girder 7-0 to 8-9.

It stands to reason that one can increase the span by about 25% if one adds one additional ply to a 2-ply 2x6 girder, and, for damn sure, if one adds two additional plys to a 2-ply 2x6 girder.

In my example in the OP, I need to get from 4-6 (2-ply 2x6 girder) to 5-8, a 25.7% increase in span.

Comments?

 

[DRP]

George, It would depend on load, span and material. Notice the species in George's post is unknown, white oak or balsa? Span you just quoted is 8' where the OP is spanning <6'... quite a difference.

your idea of notching would be a prescriptive solution as would the WSDD if the load is within the table's range.

Checking section properties;

section modulus of a double 2x8-26.28"^3

of a quad 2x6- 30.25"^3... stronger in bending if identical grade

Moment of inertia of a double 2x8-95.27"^4

of a quad 2x6- 83.19"^4... greater deflection if identical grade is used

The codebook table would be using the weakest species combination listed, AR is SYP country.

I just posted on top of you jobsaver, I'll get back with what I can. Is this a center girder under the floor only?

 

[Jobsaver]

I just posted on top of you jobsaver, I'll get back with what I can. Is this a center girder under the floor only?

Center girders. Assume a uniform load. (Aren't these the assumptions of Table R502.5 (2)?).

I can do the math using the tables provided, but I am unclear how a multiply 2x6 built-up beam relates to, say, a 6x6 on the span chart?

Interestingly, I am looking at a 1989 CABO One and Two Family Dwelling Code, Table R-602.2.1a, Allowable Span for Girders Supporting One Floor Only, and Table R-602.2.1b, Allowable Span For Girders And Required Size Of Columns And Footings To Support Roofs, Interior Bearing Partitions and Floors.

Note: One problem with blogging about engineering principles is that it seems that every other sentence one begins to type as a statement ends up being another question.

 

[DRP]

In some surroundings it never hurts to clarify that one is speaking earnestly

I've been typing and have just checked for your response so some of this duplicates what you have confirmed.

My '06 reads 2 ply 2x8- 5-9, 3 ply 7-2, 4ply 9-0... either a misprint in yours or you jumped a line while reading?

Assuming this is a center girder supporting one floor only... the description of the table.

The girder is supporting halfway to the exterior wall on the left, 5', and halfway to the exterior wall on the right, 5'. The total width of floor being supported is 10'. The codebook lists the load as 40 psf plus the dead load of 10 psf...50pounds per square foot total load.

50 psf x 10' of tributary width= 500 plf, pounds per lineal foot, bearing on the beam.

The girder is spanning roughly 6'... 6x 500 lbs=3000 lbs total load

Go to the table I posted, run down to 6x6... 4 plies (6" w x 5.5" tall) is greater than a 6x6 (5.5"x5.5") and we are getting some increases for multiple members and dimensional lumber, in other words this is quite conservative.

Run across the table until w>500 and W>3000... second column. Look up , you need a species and grade with Fb >1000. Look down in that cell E needs to be >.981, Shear >76psi

Go to Lowes and pick up some #2 SPF... from the awc- Fb-1308, E-1.4, Fv-135.

I'd slide over to the SYP pile #2 Fb-1437, E1.6, Fv-175

Now go back to the WSDD 6' span table, look at the 4x6 (3.5" x 5.5") Look at the Fb 1600 column, a 3 ply is 4.5"x5.5". That would work in SYP, you can ask for #1 if you want to clearly pass without cyphering further.

This is actually the way the old CABO 1&2 family table was set up. I've tried to keep this prescriptive, we can discuss how to check it using accepted engineering practice if you'd like. Make sense?

 

[DRP]

LOL, just saw your edit. in the CABO look at S=10', 40 psf...6-6"

Do you have footnote 3 in your edition?

 

[GHRoberts]

George, It would depend on load, span and material. Notice the species in George's post is unknown, white oak or balsa? Span you just quoted is 8' where the OP is spanning <6'... quite a difference.

My post was to indicate that according to some sources the benefit of multiple plys seems to diminish quite quickly. And that 8" depth seems to be beyond the reach of multiple plys of 6".

I am very happy to accept that the obvious engineering that says 3 plys will support 50% more load (plus the 15% property increase) than 2 plys. But there appear to be engineers who use the same engineering references that do not rely on the obvious engineering and get a much different answer.

 

[dhengr]

Jobsaver:

Part of the reason that there isn’t a table for various built up girders made up of multiple 2x’s is that fastening them together to truly act as a unit becomes a little tricky. And, we also start to worry that there might be unanticipated loads on them, something other than a nice uniform load in lbs./ft. from the joists, or that the joist loading on one member is not being properly transferred to the other members of the girder. It just really becomes difficult to tabulate everything, to fit every situation, under every circumstance, as more possible variables enter the picture and should be considered for a proper design. I haven’t had time to study every detail of DRP’s posts, but they certainly look about like the advice I would give. If you know what you are doing with section properties, stress grades of the lumber, accumulation of loads on the girder, any special loading considerations, deflection considerations, you can do just what he is suggesting. The trick is to watch out for any exceptions that the tables don’t cover, or don’t allow for. Watch out for notching any members to fit, that’s a very special case and generally not allowed, and follow directions for nailing the girder together.

What are your loads on that girder? Certainly, they will be at least 10' of floor area (width), unless the fl. joists are continuous over the girder, then the girder loading will be greater. Is their a wall line over or near the girder, what’s its loading per ft.; and walls crossing the girder are concentrated loads on the girder, do these walls carry any ceiling loads, attic loads, or roof loads? Any jamb loads from 1st fl. openings are concentrated loads on or near the girder, which really can’t be accounted for in a simple tabulation. Once you have a firm handle on these types of questions, and there may be others which a plan review would reveal, you can take a first shot at the problem (generally a conservative approx.) by looking at the joist tables. Table R502.3.1 (2) shows a #2 SPF 2x6 will span 10'-3" and carry 50#/sf or 50#/lf at 12" spacing. And, as a rule of thumb 3 or 4 of them will carry a proportionally greater load. Actually, you see that a #3 SPF 2x6 spaced at 24" is carrying 100#/lf (that’s 2 x 50#/sf) and that’s closer to your span length. You should be careful what controls your design, bending, shear or deflection, and duration of loading should be considered too. So, while this is a reasonable first approx. as a way to look at your problem, you should be careful that there is nothing which prevents this from being reasonable. How do you tie these girders together over the piers, or are they continuous over the piers, and that’s a whole different problem, although not impossible either.

Also, what is the condition of the bearing surface on the conc. blk. piers and how do you tie the girder down to the pier? You say double 8x8x16 blk. pier; is every other course laid perpendicular to the previous course so you have a bonded pier, or do you really have two piers side by side?

 

[jar546]

Adding layers of nominal lumber to increased width does not help as much with deflection as adding depth which is probably why there are limitations on layers in the prescriptive charts

 

[DRP]

It actually has to do more with what dhengr said, getting the girder to act as a unit... which really only matters if the joists are side hung. If the joists are resting on top I could go wider than the table, with engineering, not efficient, but efficiency isn't the controlling requirement here. If the joists are on top the girder simply needs to be attached together but not neccessarily "knitted" into a unit.

My post was to indicate that according to some sources

"or in accordance with AF&PA/NDS"

I've been referring to that source for my info as per code.

The rest of the concerns are addressed by reading the tables, code and NDS.

You may notch a girder, if you do I would follow the NDS and traditional practice, slope the notch rather than making a sharp 90* corner to avoid splitting at that re-entrant corner.

If you are worried about acting as a unit it can be a solid sawn 6x6, just be aware that design values change to heavy timber above a 5x5... a #2 6x6 carries lower design values per square inch than a dimensional 2x6.

 

[Jobsaver]

LOL, just saw your edit. in the CABO look at S=10', 40 psf...6-6"Do you have footnote 3 in your edition?

3. Spans and girder sizes may be computed independently of the above table when designed in accordance with accepted engineering practice and substantiating data are provided and submitted as required.

Yes. I read footnote 3 to say; Do the math to determine all imposed loads on each individual girder, determine deflection per species, etc. . . . as dhengr outlines in his post:

. . . becomes a little tricky . . . start to worry . . . It just really becomes difficult to tabulate everything, to fit every situation, under every circumstance, as more possible variables enter the picture and should be considered for a proper design. . . . , you can do just what he is suggesting. . . . The trick is to watch out for any exceptions that the tables don’t cover, or don’t allow for. If you know what you are doing with section properties, stress grades of the lumber, accumulation of loads on the girder, any special loading considerations, deflection considerationsWhat are your loads on that girder? Certainly, they will be at least 10' of floor area (width), unless the fl. joists are continuous over the girder, then the girder loading will be greater. Is their a wall line over or near the girder, what’s its loading per ft.; and walls crossing the girder are concentrated loads on the girder, do these walls carry any ceiling loads, attic loads, or roof loads? Any jamb loads from 1st fl. openings are concentrated loads on or near the girder, which really can’t be accounted for in a simple tabulation.

If you know what you are doing with section properties, stress grades of the lumber, accumulation of loads on the girder, any special loading considerations, deflection considerations . . . I don't.

My post was to indicate that according to some sources the benefit of multiple plys seems to diminish quite quickly. And that 8" depth seems to be beyond the reach of multiple plys of 6".I am very happy to accept that the obvious engineering that says 3 plys will support 50% more load (plus the 15% property increase) than 2 plys. But there appear to be engineers who use the same engineering references that do not rely on the obvious engineering and get a much different answer.

Are you happy to accept that 4 plys will span the extra inches in the equation at hand?

Adding layers of nominal lumber to increased width does not help as much with deflection as adding depth which is probably why there are limitations on layers in the prescriptive charts

.Still, it would be convenient for me right now if the tables did include the same calculations for muti-ply (x3 & x4) 2x6's, as they do for multi-ply 2x8, 2x10, and 2x12.

Somewhere between the engineer, who knows what reference material and what math to use to solve equations, and the builder, who says, "this historically works", lies this building inspector, who has to try to explain in plain language why doubling the size of a thing doesn't necessarily add a bit more span to it.

Thanks all for the perspectives. I need to go back to school so I can better help fix the residential code in places that matter.

 

[GHRoberts]

"Are you happy to accept that 4 plys will span the extra inches in the equation at hand?"

(yes) I am happy to accept the engineering that DRP outlined. It is what I would do. (I did not read his post with any more care than to determine that his approach is the same as mine.)

 

[jar546]

If there is no roof load and just a 40 lb live load + 10 lb dead load, If your tributary load is only 10' and your span is only 5.67 feet, (3) 2x6 will work fine.

 

[DRP]

I agree,

This is my "spreadsheet" for simple beams and might help explain things one step deeper jobsaver. It's just the AWC beam design formulas written into a simple script;

http://www.windyhilllogworks.com/Calcs/beamcalc.htm

Enter:

Load 3000 lb

Span 68"

Width 4.5"

Depth 5.5"

Fb 1308 - #2 SPF

E 1.4

Fv 135

Click "show result"

Works just fine, deflects 1/8" + a blond one under the entire load.

Now for grins, back out the 4.5" width and insert 1.5", back out the 5.5 depth and insert 9.25"... we're trying a single 2x10. Watch the deflection number as you click "show result". Less timber and the deflection dropped about in half. Remember Jars comment above? Old carpenters say "deeper is cheaper"

Oh, if they want to use a solid 6x6 timber it'll need to be a #1 in SYP

 

[Jobsaver]

a blond one

LOL!

Thanks again for all of the comments. Also, DRP, for the beam calc link (cool!).

 

[Jobsaver]

Is this an inspection or your construction? Inspectors should not be offering solutions.

The reality is that I have lived in this community as a working adult in trade-oriented work for thirty years. People ask my advice, and I believe it is my civic duty and job to give it, carefully.

This is the real world right? Where folks try to get along and find solutions that allow all the parties to save face.

And, the building community is happy to call when they discover a problem, in order to correct the problem, before the problem becomes a bigger problem. Virtually always, it is a given at this point, that a mistake has been made in the "design" of the construction. So, Bill is right. It is about helping people with whom one has a good working relationship save face in light of an error they made, without compromising the integrity and intent of the code, permitting, and inspection processes.

I agree with you. But tell me "Who is responsible when the inspector gives 'wrong" advice?"

Not having a design professional to assume responsibility, the permit holder is responsible; It is no different when an inspector misses something entirely when performing an inspection. The inspector loses credibility, but the permit holder is still responsible for his project. This is one of the reasons we strongly encourage individuals "building thier own home" or project to cause the trades they are hiring to pull the permit, unless, they are truly willing to accept full responsibility as thier own contractor.

The circumstances that caused the establishment of this thread:

One of our very reputable builders made a mistake, confusing what used to work under the SBCCI codes he worked under for many, many years, (see the CABO code referenced earlier in this thread), but that is not in any obvious way prescriptive since the regional codes were consolidated into the IRC. He set up a foundation for 2x6 girders.

Upon working with his framer, another very reputable tradesman in this community, to put together the framing package delivery, the "mistake" of not being able to meet the prescriptive requirements of the IRC girder table was discovered. An hour later, the builder called me and explained the situation looking for possible solutions. Of his own accord, he ruled out sideloading the girders instead of toploading them as planned, (hence I did not even mention this option in the OP).

As is my custom, I put forth some effort to offer solutions that I will accept as a building inspector, to remedy a specific one-time problem, with the caveats that this particular solution will not become a new construction practice in my ahj, and that I am offering my best advice only as a Building Inspector, not as a Design Professional. I try to do enough research to the best of my ability and means, to only offer consevative solutions that will fulfill the intent of the code.

Also, my Department Head and direct supervisor is a P.E. She directs me to carry on in my usual manner in which she has placed her confidence.

The folks that hire and use the inspectors that respond in this forum have placed a confidence. And, as inspectors, our reputations and credibility are at stake. This is a powerful incentive to carefully proceed to offer good advice where it counts, in the field, onsite, talking to the men and women that build things for a living.

This forum and the people responding here comprise another great resource for my town. Thanks.

 

[dhengr]

Jobsaver:

I applaud you for trying to be helpful and for trying to solve a builder’s problem rather than just red flagging it and walking away with a ‘when you get it fixed, call for another inspection, and we’ll see.’ But, he shot the bull, you didn’t, and unless you know what you are doing you probably shouldn’t stick your neck out a mile to solve his problem, you might get accused of committing Engineering, and we all know what a crime that can be. Alternatively, it seems like a great working relationship within your JHA, when buildings will come to you having discovered their own error, rather than trying to hide it and hoping you’ll overlook it, or it’ll go away.

Your problem isn’t rocket science, it’s a fairly clean and simple engineering problem, and we about beat “designed in accordance with acceptable engineering practice” to death in another thread, and unless you know what that is and how to do it, you might not be doing it acceptably, and thus probably shouldn’t be doing it. The loads aren’t tough math, they are just a more complete accumulation and understanding of the loads and load paths, concentrated loads, etc., a thorough mind set during that process. Bending moments, shears, bending and shear stresses and deflections are real engineering, but still fairly elementary engineering, for your problem. And, one way to start to approach this problem is as DRP, George and I have suggested, by using tables that were not exactly done to meet your problem. But, here too, the trick is that you have to know what you are doing, you have to know about how the tables were developed; in particular, any limitations or exceptions or special considerations. Most of these tables are for uniform loads only, because any concentrated loads can be located anyplace on the beam and can have a significant affect on the beam’s design. DRP has done a very nice job of laying out the basic steps for approaching your problem, but again, the disclaimer at the bottom of his spreadsheet is pretty important to understand. It isn’t usually the well laid out calculation steps in the spreadsheet that go wrong, it is the input and assumptions made in compiling the input or the interpretation of the results. And, having the experience and intuition to know when an output answer just doesn’t look right. To pull DPR’s leg a bit, I usually don’t worry about more than 4 or 5 decimal places for stresses and defections, since I’m not that sure of my load assumptions. You never did comment on any unusual loadings, concentrated jamb loads, wall loads and the like, and that’s important because they are deviations from what the tables or DRP’s spreadsheet are calculating or are based on. And, then the assembly and installation of the girder become an important issue too. An interesting exercise on your part, now that you have DRP’s spreadsheet, might be to go back to the joist or girder span tables and see if you can agree with their results for loads and span lengths, etc. and see which stress or deflection controls the design, and report back to us.

DRP..... Several of your posts here are really a nice presentation of this problem, from a builder’s perspective. You are making me a convert, less of a doubting Thomas; but we’ll still have to talk a bit more about long shed dormers, without much roof sheathing and without rafter ties. I think we could work together quite well, and understand each other in the process. It’s usually a pleasure working with builders who have a bit more than the normal builder’s intuition about why something works or what size a beam has to be. I said before that I thought you got screwed on your oversized cedar deck beam, and now I really believe that if you applied the same finesse to that problem as you have here. If you showed me that, as a BO, I would not have made you get an engineers stamp on it. The BO probably didn’t know as much about the problem as you do, so then he didn’t know what to do if it didn’t exactly follow the code tables. Your post #20, shows exactly the process the engineer would normally go through with the 2x10 and 6x6 as possible options, to try to hone in on a best solution.

 

[jar546]

This is what the "engineers" gave to me. Comment?

http://inspectpa.com/download/GirderSample1.pdf