1967 foundation design 1

[penpe]

A heater structure installed in 1967 at a refinery needs to be updated, which will increase the dead load slightly and increase the height somewhat. It's around 150 feet tall including the 70 foot stack, so wind loading is significant. They'd like to verify that the foundation is adequate. I have the records indicating what wind loads were imposed then according to UBC, but I don't have footing design notes. My initial analysis indicates that the eccentricity M/P puts the resultant force well outside of the middle third of the foundation, (and our foundation design software says the same thing). Is there a load factor that should be applied to wind loads before computing eccentricity?
The dead load is 422 kips, live load is 36 kips, total shear from wind 60 kips, moment at foundation base from wind is 3600 ft-k. Foundation is 20' x 36' x 4.5' with no soil on it. The footing dimension responsible for resisting the moment is the 20'. The soil net bearing capacity is 2100 psf.

I'd like to be able to prove that the foundation WAS adequate according to wind loading (verified UBC 1967) and foundation design standards of that time - before I try to update to today's loadings and design standards, but everything I try ends up failing due to wind loads. What am I missing?

Thanks!

 

[SlideRuleEra]

I'd like to be able to prove that the foundation WAS adequate according to wind loading (verified UBC 1967) and foundation design standards of that time...

UBC 1967, wind load requirements are attached. Calcs were straight forward at that time. More or less just wind pressure (psf) at various heights above ground on horizontal projection of a structure.

 

[penpe]

Thank you. Actually the wind loads per height were verified as being appropriate. The problem is that the resultant moment appears to overwhelm the dead load so that the eccentricity is too great for stability as the resultant pressure is outside of the middle third of the foundation.
The UBC you sent refers to Chapter 28 for footing design. I did a quick search for that and didn't find it. Can you provide that document? Thanks so much, SlideRuleEra!

 

[penpe]

Thanks again. Don't bother sending Chapter 28. I have it now. Don't know yet whether it will do any good...

 

[CarlB]

Does code specify the resultant must be in the middle third? I'd think for overturning analysis, just compare overturning to resisting loads, and show a FS=1.5 or more. The resultant location is important for foundation structural design, and checking maximum bearing pressure, but I understood the middle-third was a "desirable", not a must have.

 

[SlideRuleEra]

The dead load is 422 kips, live load is 36 kips, total shear from wind 60 kips, moment at foundation base from wind is 3600 ft-k. Foundation is 20' x 36' x 4.5' with no soil on it.

The footing dimension responsible for resisting the moment is the 20'.

If you are considering overturning of the heater + foundation, wouldn't:
Dead Load = 422 kips + 470 kips (20' x 36' x 4.5' x 145 lb/ft[sup]3[/sup]) = 892 kips

If so, is anchorage of the heater to the foundation adequate for loads of that magnitude?

Seems 422 kips dead load would be for overturning of just the heater, not the foundation.

 

[penpe]

Thanks for your input. I have looked at the Factor of Safety for overturning, and feel certain that the wind can't tip the structure/foundation over. Structure dead weight 422k, foundation weight 486k, overturning moment at foundation base 3592 ft-k. For simplicity, loads centered on 20 foot dimension of footing, resisting moment is 9080 ft-k. So F.S. is 9080/3592 = 2.5. So this makes me comfortable that the wind loads of 1966 can't tip it over. (SlideRuleEra: The actual heater structure is supported on (12) 18" x 18" concrete columns that are 7 feet tall, each with 3/4" base plates, (4) 1 1/8" diameter anchor bolts - so I think anchorage to foundation should be adequate).
My Civil Engineering Reference Manual the method to compute service soil pressure qs = Ps/Af + soil dens x footing thickness + soil on top +/- (Ms x B/2)/I. result should be less than allowable soil pressure.
The result I find is that using the moment from wind takes soil pressure beyond what is allowable - by about 30%. This formula is using unfactored loads, and is intended to determine the footing area.
I found elsewhere that the service load Ps is DL + LL. Would service moment Ms include moment from wind loading?

 

[1503-44]

You have described everything but how you calculated wind load, so I cannot tell if factors should be used or not. It would be helpful to see exactly how you have calculated the wind load and its overturning moment. You have apparently taken flat wind load on a flat projection. Are the heater and stack composed of flat surfaces? Is the stack not round? Have you applied appropriate shape factors?

Usually there is a dead load, a service, or operating load, and a test load. Sometimes there is an additional live load, usually only applied to service platforms.

3/4 PL for 1.125" bolts sounds thin and like there is no uplift. What's the distance between column rows?

I believe that formula is for soil bearing acting across the entire footing (no uplift).

Einstein gave the same test to students every year. When asked why he would do something like that, "Because the answers had changed."

 

[SlideRuleEra]

Would service moment Ms include moment from wind loading?

Yes, if you are checking wind loading effects.

You are having an inordinate amount of trouble with these calcs because you are trying to combine modern methods with the way these type calcs were performed in 1967. In my recollection, there were no "service" loads or "factored" loads in 1967, just "loads"... which were applied using a suitable safety factor. Look through the 1967 UBC, I don't believe you will find any required "load combinations"... that's more modern.

 

[penpe]

SlideRuleEra, thanks. My colleague and I have debated about what load factors to apply in the analysis, as it pertains to 1967 and today. I've concluded that the service soil pressure doesn't exceed allowable when using the 1967 existing unfactored loads. Applying a formula for actual pressure beneath footing (pmax, pmin = (P/BL) x (1 +/- (6e/B)) gives the result that actual pressure doesn't exceed allowable soil bearing pressure. (e = eccentricity = M/P) The formula yields result pressures 1972 psf and -799 psf. My reference book says if eccentricity is large, a negative soil pressure will result, but since soil cannot carry tensile stress, such stresses are neglected. Along with the overturning factor of safety result of 2.5, I'm more comfortable with concluding that the foundation design is adequate, (that and the fact that it's still standing after 56 years!).

1503-44: Wind loads for existing structure were calculated using 1967 UBC parameters at the different elevations, applying factor 1.0 for square or rectangular faces, and factor 0.6 for round or elliptical. Wind loads for updated structure are according to ASCE 7-16 using many more divisions of elevations, and applying F = qz x G x Cf x Af where qz = .00256 x Kz x Kzt x Kd x Ke x V^2. I guess this now begs the question: Is it appropriate to apply the same formula for actual pressure beneath footing, and be satisfied that the foundation is adequate if those pressure results don't exceed allowable soil bearing pressure? Our foundation analysis software applies all the various load factors and combinations of loading per ASCE, but the analysis stalls because "base reaction acts outside combined middle third". I get this "calculation invalid" result using existing structure's moment due to wind, as well as updated heater structure current ASCE wind loads, so I pretty much have to rely on calculations done by hand.

 

[WinelandV]

penpe,

It appears that your eccentricity is outside of the kern of the foundation. If you're getting a negative bearing pressure, the P/A +/- M/S model is no longer valid. Essentially, you're going to have an increased positive pressure, because static equilibrium needs to be maintained. See the image below. If you're doing analysis on a "per unit length" basis, omit the L value in the q[sub]max[/sub] equation.

Please note that is a "v" (as in Violin) not a "y".

 

[1503-44]

I don't see anything wrong with the wind load calcs.
Yes, apply the same factors for soil bearing calculations.
It would appear that your soil bearing calculation program forgot to include the case where eccentricity uplifts the upwind side of the foundation, as wineladv mentions, however the calculation is simple enough to do by hand. You can follow the calculation he shows there. That is also probably going to be your critical case for control of overturning. Righting moment / overturning moment must be greater than 1.5

Einstein gave the same test to students every year. When asked why he would do something like that, "Because the answers had changed."

 

[SlideRuleEra]

Agree with windlandv's illustration. In this your case, it's not a question of calculating q[sub]max[/sub]... based on what you have told us, q[sub]max[/sub] is known and equals to 2100 PSF (soil net bearing capacity). Using that value, back figure eccentricity. Then using that eccentricity back figure resisting moment. In 1967, safety factor was whatever the Engineer considered appropriate, not necessarily 1.5.

 

[1503-44]

I did a number of these types of vessel, heater and reactor foundations at a number of large refinery specialist engineering companies in the mid 70s and all used 1.5 stability criteria.

Einstein gave the same test to students every year. When asked why he would do something like that, "Because the answers had changed."

 

[penpe]

Thanks again. Further investigation reveals that the soils report says the allowable pressure is 2900 psf when wind load is present. If I use your approach and determine eccentricity by using the allowable pressure, the eccentricity falls within the kern. I'm not understanding how that helps. If we know the moment and the axial load it seems eccentricity of forces determines the pressure on soil, not the other way around.

Answering 1503-44: righting moment / overturning moment = 2.5 for our loading, so tipping over isn't a concern. Unless the soil fails, uneven settlement, moves center of gravity causing a momment from dead load, higher pressure on soil, etc.

 

[SlideRuleEra]

If I use your approach and determine eccentricity by using the allowable pressure, the eccentricity falls within the kern. I'm not understanding how that helps.

It helps because the bearing pressure, magnitude of load, and eccentricity are related variables. The equation that winelandv presented describes how these variables are mathematically tied together. Change one, and one or more of the other variables must change to balance the equation. If you know allowable bearing pressure... now we are told it is 2900 PSF, and you know load magnitude, then eccentricity for those values can be accurately calculated.
Using that eccentricity, magnitude of the resisting moment (that does not cause soil overload) can be accurately found.

And another issue, to me, the calcs are NOT about "helping" or trying to get the structure to "pass". The calcs are about accurately finding out what is going on with the structure... then you can reliably "deal" with whatever the numbers reveal.

1503--44 - I'll defer to your experience. I know that many safety factors were historically used for decades before being formally incorporated into codes. Since the OP is referring to UBC 1967, I wanted to point out (to the best of knowledge) that UBC 1967 does not require a specific safety factor.

 

[1503-44]

When eccentricity falls within the "Kern region", q_min is >= 0 and is at the upwind edge of the footing. There is no uplift of the footing; the soil below the footing is under compression at all points.

When outside of the Kern, q=0 at some point on the footing, the pressure diagram appears as shown in wineladv's diagram. The footing experiences uplift in the region everywhere q =0.

SlideRuleEra, I also do not recall seeing any stability requirement in any of the common building codes of the times. We rarely (basically never) designed any refinery structure to any common building code IBC, UBC, SBC, etc. Refinery and chem companies had their own design specifications, as did the engineering companies. All used SR_min of 1.5 I'm not sure what they do today; I basically switched to compressor, pump Station and pipeline design in 1980. Except for the odd dehydrator, slug catcher, pump foundation, or light oil field structure, and heading the structural engineering dept of Enron's primary engineering contractor in the early 90's, I have not done anything related to structures, ex for a couple of pipe supports, since then.

Einstein gave the same test to students every year. When asked why he would do something like that, "Because the answers had changed."

 

[penpe]

Again, I appreciate all of your perspectives. Sliderule, it is useful information to reverse engineer and determine the magnitude of resisting moment that doesn't cause soil overload, but we can't control the wind! We will need to inform the client that if we were designing the foundation by today's standards we'd make it bigger; so that there's no uplift, reaction falls within the middle third, doesn't overload the soil, etc. Given that the structure SEEMS stable and updating the heater only pushes the soil overload another 5% over the reported allowable soil pressure, maybe they'll decide to live with that risk. I think that catastrophic failure is VERY unlikely. I assume failure would be slow like the leaning tower of Pisa, and action would be taken when/if a problem is observed.

1503-44. it helps to know that your experience is that refinery structures weren't typically designed to the standards we would use today. I've been repeatedly questioning the validity of my results, (double-checking every aspect because my assumption was that the foundation was probably OVER-designed). I didn't expect to find that by analysis the foundation isn't adequate using today's design standards. Turns out the answers DO change over time.

 

[1503-44]

Offshore platform design standards were revised due to the effects of hurricane Katrina and others that hit the Gulf of Mexico in the mid 2000. Wave heights and wind load increased. Surely the wind loads along the onshore regions of GOM have also increased since the 70's, but I don't know for a fact.

You're right. The answers are always changing. It's a very dynamic universe.

Einstein gave the same test to students every year. When asked why he would do something like that, "Because the answers had changed."

 

[penpe]

winelandv, thanks for your input. Would you consider the weight of the footing as part of P (=R) in determining eccentricity?