Angle of Soil Uplift 1

[abarker]

Can someone out there point me in the direction of information on how to estimate the angle of uplift for deadman type concrete anchors (in this particular case, it's for analyzing the pullout resistance of anchors for a guyed communication tower). This angle is used to define the trapezoidal "wedge" of soil above a foundation which can be used to resist uplift forces. I've been scouring my books (and the internet) for information, but it seems there's very little information a can scare up. The only information I've seen says assume a 30 degree angle. I just can't believe that's universally true! My intuition tells me it should be related to the friction angle of the soil. Would "45-Phi/2" be a reasonable estimate? Any words of wisdom would be appreciated.

 

[Focht3]

I assume that you are looking at a purely vertical lift.

There are many schools of thought on this; for dug spread footings, the conservative answer is to include only the weight of soil directly above the footing. Why? Because of the disturbance of the soil during the installation process.

Underreamed piers can be treated differently; if you are dealing with that footing type, it depends on the depth to diameter ratio. Let me know if you need more info on this -

 

[QSE]

Focht3

What you suggested is very conservative. Anyway, we like to pick your brain, thanks for your input.

Cone of uplift varing from 20 to 30 degrees have been used for years by utilities for many of their transmission line towers. Full scale tests have been done, although these are not readily available data. The ones that we have seen, do suggest better results than 10 degree. Normally, each utility have their own criteria and foundations basically follow the prescribed limits.

We have used USBR # 10 recommendation for many years and have no problem so far. USBR 10 is the design standard N0 10 for Transmission structures. We have an old copy and do not know if copies can be downloaded from the net.

Focht3, do you have a copy of USBR?

qse

 

[Focht3]

No, I don't - if you find an online copy - please post the link.

I've looked at a lot of the work supported by EPRI (Electric Power Research Institute) - with the exception of the work done at Cornell under Fred "the Troll" Kulhawy, it's mostly junk. (Fred's a good friend - and a very smart man.) Unfortunately, Cornell's work is too difficult for most geotechnical engineers to use with confidence. And watch out for CAISSON - based on a paper by Brinch Hansen (in the 1960's?) It gives very unconservative results for lateral loading of short drilled piers -

I'm not familiar with uplift tests by EPRI. Do you know when they were done? By whom? What controls were used?

My "conservative" answer was based on a particular construction method and direction of loading. It is the correct assumption for purely vertical loading when the footing excavation was done with a backhoe, the excavation spoils are dumped over the completed footing and the loading occurs shortly after the foundation was completed. If you are talking about combination vertical and lateral loading, a different type of foundation, and/or different construction details, my answer may not be correct. You can arrive at your own assumptions about the footing construction that are different than mine, which will affect your answer. Just be aware that some field crews get in a hurry and take short cuts, which can affect the validity of your assumptions.

I've spent a lot of time on construction sites...

 

[abarker]

Focht3,
Yes, the component of interest is the vertical lift. Typically, I also go the traditional route and consider only the weight of soil above the foundation, but in this case it's an existing guy tower that is being analyzed to see if the foundations are suitable. The designer tells me he needs an angle of uplift of 35 degrees to have the required safety factor. These anchors appear to be "bank poured" with the bottom of footing about 8' below grade. I'd also agree that disturbance of the adjacent soils is a concern.

QSE,
We also provide geotechnical explorations for transmission line structures. I'd be interested in tracking down a copy of USBR #10, to help me understand my clients design needs more thoroughly. I sometimes get conflicting information concerning what's important. As near as I can tell, there seems to be two schools of design, emperical and load/deformation (i.e., p-y curves). I understand what's important for analyzing for allowable deformation, but the emperical method is a little more fuzzy in my mind. Does the USBR #10 use an emperical format?

 

[Focht3]

In general, I would answer your question:
Would "45-Phi/2" be a reasonable estimate?
as 'Yes.' But only on the side being pulled...

Where is the structure? How old? Soil profile? Do you have any information about the original construction?

A crucial consideration: What is the consequence of a failure of the anchor? How likely are the critical loads to ever occur? Is reliability of the structure very important, or is the risk of a failure of the anchor - and the tower being out of service for 3 to 9 months - acceptable given the cost of augmenting the existing anchors?

Most owners will accept the failure of the guys and tower, but are unwilling to accept the failure of the anchors. Put the monkey on the designer's back: "The existing anchor has a factor of safety of only ___ with respect to the maximum loading provided by ___. The selection of acceptable risk rightly belongs to the Owner; we judge that the risk of failure has changed from 1% from the previous loading to ___ (5%? 10%? 25%? 99%?) under the new loading. The likely failure mechanism would consist of a pullout of the anchor itself until structure collapse occurred. The anchor will have to be augmented as outlined in 'Anchor Enhancement' in order to restore the perceived risk of failure to about 1%."

You can also fudge a bit by describing your typical assumptions for uplift resistance - and why you are deviating from it in this case. Describe the risks and rewards of the modified approach, then leave it to the designer by also providing a design for an augmented anchor. This way, you gave the "right" answers and described the risks/rewards involved without angering the designer or your client. And you don't have to be the bad guy. It won't keep you from getting sued, but it will help your defense immensely.

 

[abarker]

The partial set of plans have a 1990 date, so I'd say the tower is at least 10 years old. Of course, the foundation page of the plan was missing. But we did probes to determine approximate foundation sizes (and we cored the pads to determine thickness). The tower is in the Central part of Wisconsin. Soil profile is generally a clayey sand to 6' then sand and gravel below it (glacial till origin). It's my understanding the critical loading situation is a combination of ice on the tower and the design wind load (90 mph, I believe). So the likelihood of both occuring at the same time is relatively low.

I like your risk analysis, my company already follows a similar philosophy! Fortunately, for now they haven't asked me "is a lower FS alright," only "can you give me a 35 degree uplift angle." From what I'm seeing, my answer currently is just "no". I think they'll just go out and pile somemore soil over the anchors to increase the resistance. But if the next phase involves a supplemental letter, you can bet I'll be qualifying my recommendations! Thank-you for your thoughts.

 

[QSE]

Balla presented an article to the "soil mechanics and foundation Engineering" in 1961 on uplift resistance for transmission line towers. Tests were made in dense sand and "failure surface for shallow footings was approximated circular in elevation and that the tangent to the surface of ground contact was at an angle approximately 45-phi/2 to the horizontal" How about this, I have to read this several times to get it.

Balla's article is in French, and I copied this "---" from a reference made to Balla's article by Meyerhof and Adams. The trick of course if the defination of "shallow" footings. If d=depth and b= width, what would be the ratio d/b for different values of phi?


For an overview of foundation practice in transmision line, see if you can get a copy of cigre paper 22-99(WG08/07)36. This article made references to other researches on the subject matter.

Another cigre article 22-99(WG08/TF4)is on probalistic design of transmission line structures foundations.
All on loose sand and nothing on dense sand as far as test data are concerned.

qse

 

[VoyageofDiscovery]

Hi abarker,

Typically when these foundations are constructed they excavate enough material in order to place the deadman.
Thus when I calculate the dead load above the deadman, I use the backfill phi from vertical to determine the extents of the spread.

Piling soil on top of the anchor is called berming and I usually use 45+phi/2 and 45-phi/2 from horizontal at the base of the deadman on the windward and leeward side respectively for the extents of the berm as it also contributes to the slip failure mechanism. However, berming always raises a red flag for me so I always check my lateral capacity for both ultimate "wedge" failure and passive local resistance against deformations at the face as serviceability is an issue for communications equipment on the tower.

Remember to use submerged unit weights where water table is an issue.

A past edition of Bowles has an explanation with regard to shallow and deep as well as long and short.


Regards

VOD

 

[abarker]

VOD,
Thanks for the information. Concerning berming, on some existing tower sites we've also noticed that they tend to place the majority of the soil across the front half of the anchor, the remainder wraps around the guy wire support (with very little of the berm being on the side of the anchor away from the tower). I can now see that the difference between active and passive failure plains is likely a portion of the reason they do this. But is it also just for the convenience of being able to access the cable support (is this what you mean by "serviceability&quot? I would have thought distributing additional berm weight over the entire anchor area would be more effective.

 

[VoyageofDiscovery]

Hi abarker,

I can't quite grasp what you mean by berming "across the front half" and "wraps around the guy wire support".

I do not bury the above-ground anchor shaft and fish plate as these need to be accessed for guy wire retensioning, so I slope the ground in this local area for access.

With respect to "serviceability", this is a design criteria for the performance of the antennas on the tower.

There are two types of soil failure with regards to a deadman anchor for communication towers, ultimate and expected performance also called service(allowable). The "wedge" failure is how we define the ultimate resistance, while passive pressure resistance defines local yielding of the soil directly in front of the face of the deadman. In the service case I use a factor of safety of 2 for dry and 1.5 for submerged.

At ultimate capacity of the anchor, antenna performance is not an issue, however in order that the tower perform throughout its life, a "serviceability" check using passive resistance should be done for the allowable loads with the appropriate factor of safety.

Here in Canada, structural engineers use Limit State Design, which requires us to check the ultimate load case and the service load case (allowable). This may be why I sense some confusion by my use of the word "serviceability".

The US EIA Standard for the design of Antenna Supporting Structures does not incorporate Ultimate Design philosophy,
so your structural engineer may be at a loss if you mention anything about ultimate loads. That said, I would design for the performance based (allowable) loads typical in the US.

Regards

VOD