Rock anchors

[engleprechaun]

I am working on a design for a comm. tower for remote cell sites in alaska. Most of these sites make use of 150ksi all thread drilled and embeded into the bedrock. It is the only option to keep these towers down. The problem i am having is with the pretension part. Our factored desgin load to one rock anchor is about 120 kip. We are currently tensioning to 105 kips, however, we are starting to second guess the need to place so much pre-tension on the rock anchors and if so how much do we put on them. The loads have a factor of two due to TIA criteria. Dose anyone know if this large of a tension is needed and where i could go to get documention? Any input would be greatly appreciated. Thank you in advance!!!

 

[PEinc]

If the tendons are pretensioned, them much, if not all, of their elastic stretch will have occured. If you lock off the tendons at a lower load and if the tendons become loaded to the higher load, then the tendons will stretch more. How much additional elastic stretch can the towers accept wihout presenting a problem?

Tiedown anchors for dams are often locked off at 100% to 110% of the tendon design load. This prevents the dam from lifting up when the tiedowns become loaded to their design load. You don't want a dam to lift up off its rock foundation. For your towers, this uplift movement might not be a big problem. Also, deflections (elongation) should not be calculated for factored loads.

 

[ishvaaag]

Except some mandatory code indicates otherwise, I would prestress this prestressed connection for only a bit more any service load it is likely to see. Shear should be taken by a key or then the additional effect of delivering enough normal force to avoid relative slippage at the anchor's interfaces should be in place. I think this is the usual practice for presstressed bolts and the like, and the reason that comes to my mind supporting this is better fatigue behaviour due to lower standing stresses. With this force in place the interfaces neither should see any aperture; load and security factors are there (and figure in the selection of the rod or anchor) but they are NOT guesses that the structure will see these forces at such level; just that we want the elements as big just to cover enough adversarial circumstances the code wants us to care for.

 

[PEinc]

The footing would need to be designed to support the vertical load of the tower and the lock-off load from the tiedown anchor. Therefore, you may need to increase the size of the footing. This is why some engineers try to minimize the lock-off load. If you use a threadbar tendon, you can lock of at whatever load you need. If you use strand tendons, you need some minimum lock-off (seating) load to assure that the strand wedges hold tight if and when the tendons see the full design load.

As far as shear, you didn't mention it so I assume it is not a problem. Also, tiedown anchors are not prestressed bolts.

 

[ATSE]

Using 150 ksi threaded bar is not my first choice. You may come to the same conclusion after you see one of these rock snap during routine handling in low temperatures.

Except for high-strength-low alloys, this 150 ksi steel is almost always brittle (low elongation before tensile fracture).
Williams Form offers a 75 ksi threaded rebar (conforms to A615) which works well and is very ductile. I'm sure Dywidag offers something similar.

Keep in mind that the higher your tensile pre-stressing load, the more your creep via your grout (and also creep in your pilecap if you're using one). Over the long term, your prestress will likely dissipate to less than half unless your grouting is extremely robust and your concrete is f'c>5ksi.

Better to stick with mild steel (slightly larger diameter) and not prestress.

 

[ishvaaag]

Of course rock anchors are not prestressed bolts. You can have them not presstressed if you want, or may be the convenient solution. Some even prestress without allowing some free length. The point here is that we are not here (I think) to show who is smarter, but to provide some help.

As I have said, I live and practice in Spain. Here since 2005 geotechnical studies are mandatory. 90% percent of them are entirely useless, showing 3 common defects:

1. Not providing probabilistic correct estimates of settlement, but worst case 10 times higher.

2. Repeating as parrots "bulb of pressure" influence blah blah where the compounded integrated effects of loads at their position, whichever the class of foundation shows that except for superestructure rigidity effect, settlements scarcely differ. Not meaningful advertences of local risks, just generalized omission of the technical truth.

3. Consistent with this, recommending k30 values producing astonishing settlements the building never will see. Plus, some seeming to think a ballast modulus it is the sacred approach to designing foundations.

I can live with this, and when I point this to them, even if reluctantly, they admit it. So I admit, yes, prestressed bolts are NOT rock anchors. Yet I would love anyone seeing any of my posts look for the information within, and not if it seems I have outstretched some analogy.

Stay friends.

 

[BigH]

ATSE hinted at a question I would ask - why do you need to prestress unless there is a need for "no (read that barely any) movement" permitted under what is most probably transient loading? Why not passive anchors? (e.g., grouted rock bolts - or dowels) I was on the DEW line years ago and we anchored with Dywidag thread bars (2 inch) - we did not prestress that I can remember (years and many beers ago).

 

[PEinc]

I agree, BigH. Dams are locked-off tightly because they shouldn't have movement. A tower may be different, as you mentioned.

ishvaag, it's not a question of who is smarter. It sounds to me like engleprechaun is describing tiedown anchors rather than prestressed bolts. Maybe I'm right; maybe I'm wrong. However, he does mention these connections as rock anchors. Therefore, my response is appropriatly related to tiedown anchors. BigH's response is also appropriate as it relates to structure movement and elongation of the tension connection member. If the tower footing sits directly on rock, the tiedown connections might be rock bolts, prestressed or not. If the tower sits on soil overlying rock, the tension connections should probably be tiedown anchors with unbonded and bonded lengths. In that case, elastic elongation and lock-off loads are considerations. As for your last response, it sounds to me like you are venting your general dissatisfaction with geotechnical engineers. I'm not sure how your last response relates to this thread.

 

[engleprechaun]

I thank you all for the valuable input. To clear up some things, they are "rock bolts" as the rock is the main foundation system. We use the bolts to tie 4' to 6' tall piers to the ground. Also "no movement" is and can be very important for these particular types of towers. They have microwave dishes on them which, depending on the distance of the shot, can make "catching" the signal difficult. So keeping the dishes in the same spot becomes more crucial the further away form each they become. Thanks again.

 

[born2drill]

ATSE, why so hesitant about the 150ksi threadbar? For ground anchors I've never used anything other than 150ksi bar or 270 ksi strand.
I'm not sure how cold you mean when you say "low temperatures," but I've been crashing and banging those bars around for years in some pretty harsh upstate NY winters and never seen the dramatic failure you described.

 

[hokie66]

High strength bars (or strands) make sense for stressed anchors, but not for passive ones. Deflection is just relative to bar size.

I doubt that it makes much difference whether active or passive anchors are used for these towers. There will be a lot more deflection in the tower itself than in the anchors.

 

[ishvaaag]

PEinc,

It seems yo still are not wanting to see if what I said in my post holds something of value to the question, and that is applicable to both presstressed bolts and rock anchors, so maybe it was not as an outstretched analogy as to merit be pinpointed as a separate comment.

Respect venting against, heavens, you may take for sure geotechnical experts here or anywhere are very low in my list; I am old enough to see that people earning their bread working in anything stay in other level that so many (interjective) one meets in his life.

I simply thought useful to pinpoint something that happens here, that, by the way, it is a fault still not committed by some geotechnical engineer, simply because maybe there's still no one with such title here(the career has just been created and there may be still no graduates); these things are made here usually by geologists maybe signed as well by an ICCP, that is, a Civil Engineer of here, Ingeniero de Caminos, Canales y Puertos. I must say that the ICCP's are for sure the more competent of the pair without doubt. They come from responsibiliy, not theory.

And remember, I am an architect. Here we were trained to do our structures, and those that so want, so do. This may come to surprise to some engineers, because the practice elsewhere is or may be different. Yet no one should be surprised, architects have been in charge of firmitas since Hammurabi through Vitrubius, and even today we could, on the statements of standing laws from the nineteen thirties, if we were ready to fight for, project a main bridge as long it is in an urban environment. We would be not only the architect of the bridge, but the engineer of the bridge. Maybe that Calatrava is both an architect an an engineer is not casual. Surprising? A publication of Dragados y Construcciones, then the main firm doing civil works in Spain expounds over a hundred of bridges of the nineteenth and early twentieth century in Spain. Almost all were the design and sign of an architect.

I have never had been in fear of liberty. So I don't think illustrative anecdotary of personal and local experiences may damage to anyone, the information appearing where it appears. Very contrarily, I think anecdotary is what makes stories interesting, you note through it that what told is true.

Take my comment above. Negatively, you may think is what you have said is. Positively, anyone reading such thing may reflect and think, hey, this guy is saying something important: the users of geotechnical studies want information meaningful to them. And so in his mind rests a note on that indicating overestimated settlements in an order of magnitude and going home to sleep well, because the worst that can happen has been covered in such excess, is unethical, rendering the information useless. I am just asking more precision; exactly the same you demanded from me.

 

[PEinc]

I am sorry but I cannot understand what you are saying. I will make no further comment.

 

[kslee1000]

ishvaaag:

I don't know what was the trigger, but as an architect practicing in structural, I commend your sophistication in engineering as a whole. However, you may need to spend some time on geotechnical theories as well to make yourself a complete all-around, if you lack of confidence on the fact telling practitioners.

 

[BigH]

I am having a bit of difficulty in some of the above response, too, however, I think I understand the gist. Apologise to the original post as we are diverging from the post - but having an interesting discussion.
I think, to paraphrase a quoate that Focht3 (when he was active on the forums) like to put forth, is that soil is a different animal than steel and concrete which has fairly consistent properties and homogeneously applied within the material (unless, like concrete it is/was a major construction problem.
ishvaaag did indicate that geotechnical engineers did not provide probabilistic correct estimates of settlement - sometimes 10x different. He has a point - 10x is too obviously too high (for settlement but not for coefficient of permeability). Still, one must remember, statistically, is that the coefficient of variation of soil properties (std-dev/average) typically ranges from 30 to 50 percent and can be as high as 100%. The average of a soil property can be bracketed by in the order of 2 to 2.5 standard deviations - and yet these may be exceeded. (see Baecher and Christian "Reliability and Statistics in Geotechnical Engineering", page 140) The inherent variation in measuring soil properties is high - depends on the type of test, the bias of the operators, the disturbance of the soils and the like. This is all taken into account when a geotech writes a report and provides reccomendations (and, don't forget, variability due to poor to good construction practices). I remember my mentors telling me - if you do a consolidation settlement estimate and are within 30 percent of the real value - "hell, you've had a good day." So, while ishvaaag rightly criticizes geotechnical estimates when viewed on the macro-level and with 100% hindsight, it appears that we all must step back and look, then, on the micro-level. It is a tough business - once the foundations are in there is very little you can do about reinforcing them. With structures, you can brace, remove and replace, etc.
It is interesting that Spain permits architects to design bridges (in urban settings). Ishvaag - your view are appreciated - that is why this site, with engineers from so many different countries, cultures and the like, being able to provide their views.

 

[miningman]

This is all good stuff and I have use Wiliams Bolts as well as Dywidags etc extensively. The original poster referred to ROCK anchors and in my experience it is the quality of the installation techniques and the type of grout that determines the ultimate effectiveness of these systems. Are we talking epoxy resin, cementious grout, or some other speciality grout.?? The best stuff I ever used was called Rockloc... havent seen it on the market for years... messy as hell to install but the steel bars broke in tension when I performed a pull test... most others failed (at lowwer loads) in shear at either the rock/grout interface or the bar/grout contact.

 

[kslee1000]

The OP has clearily define the material except grout, but uncertain on the level of pretension required. I think there was a similar discussion on "welding, bonding & fastener engineering" forum with a lot of good feed backs. Why not try a search on that forum see if there is clear answer to the question.

 

[ishvaaag]

Thanks ALL for your comments above. I understand the -BigH- referred statistical variability of soils being high, and that those signing the geotechnical studies want it recognized. The same happens (in its degree) with calculation of RC deflections, but the results have been retouched enough to show a resemblance of reality. I also accept, know, kslee100, I need more knowledge in everything, soil mechanics and geotechnical studies as well. Everyone knows more than me on something; that's to be someone! So how not seasoned professionals in their field. Yet one needs to point, in "Modern" law, say since the early nineteenth century, the Código Civil (Civil Code), on the old standing tradition, singled out the architect as personalyy liable of the "vices" of the soil (a kind of explicitly named singularization still to be seen for almost all other professions). So in general our profession has been reasonably attentive to the issue of soils and good foundations, proportionally to the times. Till recently, we have had to make it quite unassisted (except of course by the enlightening guide of everyone that said something of value in the field and was at our reach). When assisted, the realities of building transactions and level of knowledge, were not enough sometimes to warn off some problems; I remember a 1980 project for hundreds of flats (with tens of seasoned professionals having something to say about) where having a geotechnical study and building on piles could not avoid evident dishing settlement action that fortunately was moderate enough to not cause legal action; the piles, better, the buildings, seemed to float over some substantial clay layer at some depth, that, if detected, didn't lead to a better solution.

And yes, I understand all this if anywhere should be in another post, yet it shouldn't harm here, neither; as Sy Barry's The Phantom said, "Looks do not kill".

To end, we can't manage the use of settlements 10 times those reasonably expected. All doing structural analysis know that imposing vertical displacements in the 1 foot range to framed structures is going to make something ugly to the structural design. Our clients can't afford that; we can't afford to say so to our clients if not warranted. And almost all know these values are not usually seen even with the most basic foundations. Imposed settlements has been to my knowledge traditionally adscribed to the probabilistic hypotheses, that is, just like earthquakes, as something one needs to consider as something happening but to deal with at the service level as an eventuality (this distinction surely was made at the time of first factored load codes, since before all was service level). Of course, settlements are to occur and earthquakes may not; I even could agree in changing the traditional view; yet I think we should keep pace with reality.