Eccentricity Check For Spread Footers
Eccentricity Check For Spread Footers
(OP)
I need a sanity check.
I've designed many basic shallow foundations for my structures over the years with, in my opinion, a pretty straight forward and well established procedure (I think). But over the years, I've seen designs from other engineers come through that make me wonder HOW in the world can they can justify the numbers, so to speak. Not only does this now make people ask the question why my footer is typically larger, but it introduces doubt in their minds, not to mention my own.
Let me explain.
The first thing I check with a spread footer is eccentricity. If the resultant of the loads (e = M /P) does not fall within the the middle third of the base, then uplift, on some part of the base, will occur. Also, the traditional elastic theory for finding bearing pressure (q = P/A + MC/I) no longers applies, at least in the sense of using the full base width. From what I understand, meeting this eccentricity criteria is a must, and sound practice. (NOTE - I suppose if you allowed less-than-full base contact that you could use an iterative process to find the actual effective base width, but all my references state that it's not recommended since it leads to high bearing pressures and unnecessary settlement.)
Next, I'll check both sliding and overturn. These are pretty straight forward. Just to note, overturn should NEVER govern if the eccentricity criteria is met.
Lastly it's bearing pressure. Provided that the eccentricity criteria was met, I use q = P/A + MC/I. Obviously q-max must be less than the allowable. q-min should be equal to or greater than zero since eccentricity criteria was met; a negative value (uplift) can't occur.
Lately I've seen some designs come through (from outside sources) that just look unusually small. Most of the time the cases involved are rigid frames where there is a large outward thrust at the base. Typically, there's 2 ways to deal with this outward thrust:
1) Design the footer independantly; the thrust load is resisted simply by the footer's geometry/weight, or
2) Use hairpins (or similar) into a floor slab to tie each side together and eliminate the thrust load on the footer. The slab acts as a tie.
In the recent cases I'm talking about, the engineer used method #1 from above; they're independantly stable....supposedly. However, I'm fairly certain that if I had the column reactions that there's NO WAY that the footers proposed meet the required eccentricity requirement. Granted, I don't have any official involvement in the project, but voicing my opinion makes me look like I'm just being overly critical of someone else's work.
Has anybody ever run into a similar situation? Does anybody else feel that the eccentricity check for spread footers is overlooked frequently?
Or.....am I missing something?
I've designed many basic shallow foundations for my structures over the years with, in my opinion, a pretty straight forward and well established procedure (I think). But over the years, I've seen designs from other engineers come through that make me wonder HOW in the world can they can justify the numbers, so to speak. Not only does this now make people ask the question why my footer is typically larger, but it introduces doubt in their minds, not to mention my own.
Let me explain.
The first thing I check with a spread footer is eccentricity. If the resultant of the loads (e = M /P) does not fall within the the middle third of the base, then uplift, on some part of the base, will occur. Also, the traditional elastic theory for finding bearing pressure (q = P/A + MC/I) no longers applies, at least in the sense of using the full base width. From what I understand, meeting this eccentricity criteria is a must, and sound practice. (NOTE - I suppose if you allowed less-than-full base contact that you could use an iterative process to find the actual effective base width, but all my references state that it's not recommended since it leads to high bearing pressures and unnecessary settlement.)
Next, I'll check both sliding and overturn. These are pretty straight forward. Just to note, overturn should NEVER govern if the eccentricity criteria is met.
Lastly it's bearing pressure. Provided that the eccentricity criteria was met, I use q = P/A + MC/I. Obviously q-max must be less than the allowable. q-min should be equal to or greater than zero since eccentricity criteria was met; a negative value (uplift) can't occur.
Lately I've seen some designs come through (from outside sources) that just look unusually small. Most of the time the cases involved are rigid frames where there is a large outward thrust at the base. Typically, there's 2 ways to deal with this outward thrust:
1) Design the footer independantly; the thrust load is resisted simply by the footer's geometry/weight, or
2) Use hairpins (or similar) into a floor slab to tie each side together and eliminate the thrust load on the footer. The slab acts as a tie.
In the recent cases I'm talking about, the engineer used method #1 from above; they're independantly stable....supposedly. However, I'm fairly certain that if I had the column reactions that there's NO WAY that the footers proposed meet the required eccentricity requirement. Granted, I don't have any official involvement in the project, but voicing my opinion makes me look like I'm just being overly critical of someone else's work.
Has anybody ever run into a similar situation? Does anybody else feel that the eccentricity check for spread footers is overlooked frequently?
Or.....am I missing something?





RE: Eccentricity Check For Spread Footers
RE: Eccentricity Check For Spread Footers
RE: Eccentricity Check For Spread Footers
Michael.
Timing has a lot to do with the outcome of a rain dance.
RE: Eccentricity Check For Spread Footers
However, I take exception with the assumption that the eccentricity must be within the middle 1/3 of the footing. It is not a difficult calculation to calculate the soil pressures for a footing in partial uplift. If you're designing to a Load Combination that uses 0.6DL + 1.0WL then it makes more sense to design this way as the overturning safety factor is already accounted for by the drammatic reduction of your dead load. If you still require full bearing for these types of LC's then you are (IMHO) being unnecessarily conservative.
RE: Eccentricity Check For Spread Footers
As far as the latter part....
I see what you're saying in regards to that load combination, mainly due to the relatively predictable dead load of the footer.
However, solving for the soil pressures on a footer base that has partial uplift, if I'm not mistaken, has to be an iterative solution:
- Guess a lesser-than-full value for the footer length, B
- Recalculate A & I
- Refigure qmax and qmin
- Repeat unitl qmin equals zero (or close to it)
- Assumed value of reduced B is now correct
Every reference I've ever seen recommneds staying away from this situation though. I've never seen where keeping the eccentricity within the kern of the base was not recommended.
RE: Eccentricity Check For Spread Footers
RE: Eccentricity Check For Spread Footers
So....
Is the procedure I listed above for findng the reduced footing area reasonable? Anybody have a better suggestion?
RE: Eccentricity Check For Spread Footers
Let 'q' be the maximum pressure (height of pressure triangle),
Let 'd' be the length (less than B) of the pressure triangle.
Let 'L' be the width of the footing.
(1) Sum of vertical forces equal to 0. i.e., q/2*d*L-downward forces=0
(2) Sum of moments about center of footing equal to zero, i.e.,
Any applied moments/thrustxdistance-q/2*d*L*(x-d/3)=0.
solve for q and d.
RE: Eccentricity Check For Spread Footers
RE: Eccentricity Check For Spread Footers
You know that the centroid of the triangular load is at H/3 corresponds to the location of the load eccentricity. Therefore, you know the length of the soil bearing triangle H. The total axial force is also known. And, it must equal q_max*H/2 * width, right? Therefore, you can easily solve for q_max.
For Seismic and Wind loads, I believe it is common to use footings in partial bearing... more so now that we've got these 0.6DL load combinations. Though, it would not be advisable to do so for a gravity only load combination.
RE: Eccentricity Check For Spread Footers
RE: Eccentricity Check For Spread Footers
I see what you're saying.....that's much easier than what I had in mind. Obviously this only works when "e" falls within the base length.
So.....in regards to the original post.....I guess keeping the resultant within the middle third is't always needed. You learn something everyday! Thanks everybody.....
RE: Eccentricity Check For Spread Footers