IRC 2000 - Unreinforced Foundation Walls Table R404.1.1(1) 1

[pcronin]

I am a structural engineer in the lovely state of New Jersey and occasionally my firm deals with residential projects. My question is limited to unreinforced CMU basement walls and the wall thicknesses provided in this chart.

If I calculate the soil pressure on the wall based on Rankine theory with good soil properties (friction angle of 34 degree, unit weight of 115 pcf) the walls shown in this chart NEVER work. Even if I assume active pressures (assuming flexibility at the top connection) with a Ka=0.283 and precommpression due to the self weight of the house (even though 50% of the walls are non bearing and won't have axial load), the calculation tensile stress (M/S - P/A) always exceeds the allowable tension of 25 psi (ACI 530, Ungrouted Type M/S mortar).

How does the IRC justify the thicknesses in these tables? If my firm is involved in a basement wall, it is usually due to cracking and it is difficult to explain that the IRC “Guidelines” do not meet the material design codes. This is very difficult to explain to Joe Homeowner because he just thinks I am a typical over-conservative engineer.

 

[concretemasonry]

You will never be able to justify the thicknesses because they are empirical values and not engineered. I wrestled with trying to justify the tables for about 25 years until the proven performance in basements became obvious. They are proven to work if applied within the limitations imposed.

If you try to "engineer" a basement wall instead of using the accepable tables, you may be doing a disservice to the home owner by choosing the wrong method.

One of the reasons the tables have stood the test of time is that they are based on actual performance. When you try to do detailed calculations, the assumptions made for ease of calculations compound the "gray area". A 12" thick wall 8' high really in flexure? How about actual end conditions? Aspect ratios of panels? Actual strengths rather than ASTM determined strengths for different purposes? Then compound these assumptions with the assumptions made for the behavior of soils.

I have been involved in masonry design and construction for 40 years and initially tied to resolved what appeared to be contradictory to my structural engineering training. Since I started designing nasonry in California, I thought I could engineer any masonry.

I am at my lake home and do not have access to all of my library, but do have a copy of the ACI 530-02 that I participated in. ACI 530 and several earlier documents are the general basis for the masonry portion of the current codes. The references I am currently using are to show the concepts. Obviously, some portions were modified slightly when the original ACT 530 was written into International codes.

The concept of empirical design is contrary to many engineers, but it is necessary. A parallel to empirical design is the old claims by aeronautical engineers that the "bumble bee" cannot/should not fly, but it does.

There are several general design methods that may be used for basement walls, but not arbitrarily intermingled. In 530-02, they are Allowable Stress, Strength Design, Prestressed Design and Empirical Design. Obviously, you do not pick and choose portions of one design method to use in another method (i.e., tensile stress from the Allowable Stress section in the Empirical Design section where ther is no requirement regarding tensile stress).

When you use the Empirical Design method, you must use it within the scope and limitations. The 530-02 specifically eliminates Sections 1.7 (Loading), Section 1.8 (Material Properties), and Section 1.9 (Section Properties) except as noted. This was done to eliminate confusion of the different sections.

The behavior of a masonry wall is far more complex than a concrete wall because it is composed of materials (masonry unit, mortar and grout) with differing properties. I have seen 5000 psi(f'm) hollow prisms built and tested at 7 days using mortar that was less than 2500 psi.

In Brazil, where good structural masonry design is critical, empirical design is used for low rise buildings. Other methods are used in the design of 20 story intermitantly reinforced 6" concrete masonry loadbearing masonry buildings.

Empirical design is a method that recognizes the proven performance of masonry and permits a simple design method that is applicable for certain applications.

Dick

 

[pcronin]

OK. I can understand that empirical methods have worked in the past and I am not concerned with the wall failing because I have seen taller walls work. I am concerned with taking on liability when the only thing I can use in my defense are comments similar to, "Well, all the other walls are built like this and they work." I do not mean to be argumentative, just open to ideas related to this discussion. As calculated, the wall I am looking will experience 60 psi in tension with only 25 psi allowable. But then again, the soil properties are conservative, I accounted for a surcharge due to a patio adjacent to the wall, but I assumed it was free draining soil and didn’t account for any hydrostatic pressure.

If a structural engineer gets involved in residential projects it is because the town is requesting a signed and sealed letter and/or calculations. The town just wants someone on the hook if something happens to the wall and it is not worth the liability for the smaller fees in residential construction.

If all of these walls work all the time, then maybe the code allowable tensile stresses are too low. I would rather see the design codes revised to reflect reasonable values than just a blanket statement that even though the numbers don't work, it will be fine.

If two engineers specify a wall type with one actually designing it for the code required loads and the second designing it empirically, which one is the "better" engineer? The one who follows the proper design methods will engineer a stronger and more expensive wall, but the empirical method would be a wall that might just work without the code required safety factors. Which engineer is providing a service with his client’s best interest in mind? Safety, cost, durability, etc.?

 

[civilperson]

Tensile strength of concrete is in the neighborhood of 6 x Square root of compressive strength. The 25 psi given is very low, a FS of 10-15. Use of empirical methods is defensible with good design practices, i.e. granular backfill and drains to eliminate hydrostatic pressure, no surcharge, limited compressive effort on backfill adjacent the wall and good surface drainage or downspouts extended away from building.

 

[concretemasonry]

pcronin -

In reference to your last paragraph, Empirical design is in the codes and is a PROPER, acceptable design method. Just because it is proven, fast, economical and realistic does not mean it is not performing a professional service for the client. I don't recall seeing the "factor of safety" mentioned anywhere in the concept, especially if you are not required to calculate some stresses. There is no "best" design method. Each method is accepable if used properly and each has different types of structures it is most applicable for. Perhaps the choice of design methods leads to the idea that the fees are too low in relationship to the time involved.

Using the empirical design method for a basement wall does not require you to calculate or investigate the tensile stresses and no allowables are given for that design method for that reason! Design methods and allowables go hand-in-hand.

If you really feel the allowable tensile design stresses (which are not a part of Empirical design) are too low low, spend 10 years to get on as a voting member of the code body that writes the codes. Then you can correlate information, documents, sources, new materials, and put everything into enforceable code language that works in the office and field. After that, you will have to wait for it to be accepted by your critics and peers.

You just have to use the appropriate design method and implement it to the best of your ability. If you chose a different method and end up with a very unusual design design because of the realistic limitations of that procedure, that is your choice. There is no "cook book" method, but experience teaches what makes the most sense for the conditions and type of structure.

If you really want to have fun and impress the client with sophistication and calculations, design the basement with Strength Design or Prestressed masonry concepts. They will also provide a design meeting the code requirements.

Granted, a basement is probably one of the most difficult items for an engineer to grasp. In some ways, is it much more difficult to design properly than a 6" load bearing high rise.

P.S. there are many structures that do not comply with the current empirical design and they are over 1000 years old.

Dick

 

[PSlem]

ACI just released Residential Concrete Code and increased allowable tensile from 5*f'c^.5 to 7.5f'c^.5 since no trouble had been noted in res. walls.

 

[concretemasonry]

The walls in question are masonry (CMUs) and not concrete.

Dick

 

[cap4000]

Also you should check UBC 97 just for a comparsion of the tables. There is quite a difference. Nice tread. Good Luck