Structural Design of Soldier Piles

[MHasch]

I work for a small subcontractor that installs temporary soldier pile and lag walls. We out source all of our engineering and I believe that the methods our structural engineer is using to size our beams may be overly conservative and preventing us from being competitive.

Item 1: For bending our structural is using 0.6Fy based on ASD. I see FHWA referencing 0.66Fy. I had another engineer tell me that based on the new codes we should be using the Plastic Modulus (Zx).

Item 2: Arching – Our structural engineer is currently neglecting this for the sizing of the beams. I see referenced from Peck suggesting using 2/3 the moment and NAVFAC suggests using 80% of the moment for braced/tieback walls. It is also be neglecting when estimating deflections.

Item 3: When we install tiebacks at steeper angles and apply axial loads our structural engineering is considering the soldier beams unbraced in both directions. It is my thought that the lagging and soil provide bracing along the minor axis.

Item 4: When determining the unbraced length for evaluating axial loads our structural engineer is taking the distance from the lowest tieback to a point below the bottom of the excavation that the bending moment has been dissipated. This seems conservative to me.

Item 5: In several instances we have notched the flange and run our tieback just off center of the soldier pile. Our structural engineer is requiring us to put straps on the soldier piles to prevent torsion and account for the eccentric loading. It is my viewpoint that a lot of this torsion is taken out through friction associated with the flowable fill and soil around the soldier pile. I believe it is more appropriate to only install the straps on an as need basis. When we tension the anchors, if it starts to rotate we will install the straps.

I am interested in how other people are handling these issues and references that I can use to present my case to our structural engineer.

 

[FixedEarth]

Do you have a typical job details.

For example, 16 ft cut, cantilevered, 8 ft spacing, 3 by lagging and selecred beam of W 14x61 with 19 ft embedment.

Also mention the type of soil, ground water level and any surcharges within 1.5H & if it is private job. This way we can compare site parameters with what we would recommend.

 

[DRC1]

Item 1 there really is no differene between .66 Fy and .67 Fy. Really should look at the section you are using. Although almost all W sections are compact, Most H-sections are not compact and thus do not qualify for .66Fy. Also since soil loading is only estimated and may vary, I usually do not design up to allowable. As for plastic modulus, that is for LRFD design,where the limits you mention are for ASD design. You can not simply switch S for Z. It is a different proceedure, that will get you to approximately the same answer.
Item 2: I am not familar with this, although I have seen it used. It is not widely used. If you do not know the basis for it, you should not apply it. That said it is recognized that back walls are stiffer than braced or walls with deadmen. This is because the load is locked into the anchor prior to excavation below the anchor and the tieback load forces the pile into the retained soil, densifying and strengthing it. However the actual effect is not readily quantifiable.
Item 3: Generally slope is 45 degrees or less, I assume this provides sufficent lateral bracing. If axial loads are due only to the anchor, I think you are okay. If loads are axial loads due to some other condition, you would need to assess the risk and consequense of buckling.

Item 4: Lagging, properly installed should provide lateral bracing of the Y axis, unless again,if you are applying large axial loads, you may want to look more closely at the bracing.

Item 5: This one I am definately siding with your structural. Again I have seen this one rarely and do not like it. there are rotation issues which induce shears, flange and web bending issues. More common is to install a wale and put the tie midway between the piles.

In general, the cost oof a heavier pile is not a sinificant part of the price. Generally it is the labor that drives the price. Too much conservatism is not a good thing, but since we really do not have a good idea of what loads are actually applied to the piles, some caution is in order. I would talk about items 3&4, but the others, I do not disagree with the structural, except possibly for 1, where Fy may be too high.
Good luck hope this helps

 

[jdonville]

MHasch: Good questions. They've forced me to take a hard re-look at some of *my* typical shoring design practice.

DRC1: Excellent point about HP sections not being compact for flexure (typically the web is NC, and the flanges are *just* S, depending on section).

For HP sections, my quick calculations indicate that 0.6Fy (NOT 0.67Fy) for ASD, and 0.9Fy for LRFD are the appropriate factors to use for strength, based on AISC guidance.

J

 

[MHasch]

Everyone,

Thanks for the responses and sorry for the delay in getting back. My computer went down.

FixedEarth: We have been looking at a lot of different projects with varying soil conditions so I can’t really give you a typically situation. I am trying to get a better idea what the industry standards are and how others are dealing with the issues I listed above.

DRC1: I do not understand what you a saying about the 0.66Fy not be applicable to HP. I think I need to get a copy of the ASIC codes so I can understand this portion of the design.
Does anyone know a reference indicated that the lagging “properly installed” would brace the piles. This would be helpful in dealing with our consultants.
Item #5 – I understand what you are saying. My challenge is our company has a relationship with a company that does this on all there project so our field guys push really hard for this so I know I’m going to have to continue to deal with this.

Jdonville: Are you saying that based on ASD W-shapes - 0.66Fy is appropriate and for HP – 0.6Fy is appropriate?

 

[DRC1]

Most H-piles have flange widths equal to the depths. Most W-shapes do not. The .67 is applicable to shapes that meet a ratio of flange width to thickness ratio. We are renovating our office so I have my books packed away right now, but it is part of the AISC Specifications. Since the W shape flanges are shorter for a given thickness most (But not all) meet the criteria for .67Fy. H-piles on the other hand, have longer flanges for a given thickness. These flanges, under compression, are more likely to buckle, thus have lower allowed compressive stress. Thus almost all (but there are a couple of exceptions),H-Piles are not compact and thus can only be loaded to .6 Fy.

As for offsetting the hole I don't know what to tell you. As I said, I have seen it done, but I really do not like it. Some day a flange will collapse and some one will have a nasty failure.

Hope that helps.

 

[PEinc]

Cutting holes through the front and back flanges, and sometimes the web, of a soldier beam drastically reduces the strength of a soldier beam. However, tiebacks are often installed this way. I have seen soldier beams bend laterally away from the side of the beam where the flanges have been cut. When I have done this, I have always used a larger than normal soldier beam.

When installing tieback anchors, it is not uncommon to drill 4 to 6 inch (or more) diameter holes. That means the holes in the beam need to be big enough to pass the drill bit or casing pipe. Cutting 4 to 6 inch holes in the flanges is a significant reduction in the bending strength of a beam. Cutting a hole through the web can affect the ability of the beam to resist shear. It is easy to replace the missing web, but harder to symemetrically replace the flanges.

The next question is when do you make the thru-beam connections in the soldier beams. Before or after the beams are driven? If you make them before installing the beams, you need to be sure you can get the beam installled with the connections at the correct elevations. If you make the connection after driving the beam, you need to make sure the beams do not fail during the connection fabrication (especially at the upper tieback connection during a cantilevered condition).

http://www.PeirceEngineering.com

 

[PEinc]

If you are drilling-in the soldier beams, you are better off using soldier beams made from paired channels or wide flange beams with a space between them for the tieback anchors to pass.

http://www.PeirceEngineering.com

 

[mudman54]

#1-agree 0.6 seems conservative for temporary
#2-the only arching we consider is for the earth pressures on the soldier pile socket below the dredge line; usually 3 x diameter but we divide the passive by FS=2
#3-we try not to go steeper than 45 degrees
#4-we take unbraced length from the tieback to the point of fixity
#5-mostly we use wales; now that you mention it, i have seen tied-back soldier piles without wales; like PEinc says you would have to use some sort of back-to-back system and not drill through the soldier pile.

can anyone share with us the single pile tie-back design?

 

[MHasch]

There are 2 types of through pile connection that we are looking at. Both of these are constructed prior to installing the beams. If they are being driven the stiffeners may be installed after driving.

The one that creates the heartburn for everyone and has the eccentric loading on the tiebacks consist of notches cut our of the flanges on one side of the beam with the notch being cut in such a manor that there is a hook to keep the tieback bar in place. 2 steel plates are welded to the inside of the opposite flanges to increase the structural capacity of the beam in this area and stiffeners are welded between the flange above and below the tieback. The connection plate is also welded in place. This is what our operations guys want to do.

The other method that I push for is cutting a hole in the center of the beam at an angle welding a pipe in place and then welding a plate on the front and back to increase the structural capacity in this area. Obviously the flange width has to be large enough to do this. The plates are typically wider than the beam to get enough steel to reinforce this area. There is the issue of making sure you have sufficiently reinforced the area, but you are not inducing a lateral load on the tiebacks. If there is not enough flange width the plates could be welded on either side prior to cutting the hole in the center. I like this a lot better, but I'm getting some resistance from operations.

 

[PEinc]

My position on arching: Arching does not apply to design of the soldier beams. It applies to design of lagging and other wall facings between soldier beam and to shotcrete facings between nails. There are even some references that apply an arching effect to the design of continuous braced wales where more of the wale load is considered to be concentrated at the braces rather than being uniform throughout the spans bewtween braces. When soldier beams or soil nails are loaded, the soil between them bridges or arches. Therefore, there is less load on the lagging or shotcrete as you near mid-span.

If you ever saw a soldier beam wall where lagging has fallen out, you can often see that the soil between the soldier beams falls out leaving a stable, backward curving surface or arch - as long a there is good contact between the soldier beams and the soil at the soldier beams. That is why you should not remove soil behind soldier beams, why lagging should have good contact (no voids) with the soil behind, and why you should fill soldier beam drill holes completely with lean concrete or low strength flowable fill. These drill holes should not be filled with cohesionless sand, gravel, or crushed stone because these materials can run out when lagging is being installed. When it runs out, any arching is lost and the soldier beams can move backward from the tieback loads or the soil can push forward to overload the lagging or another facing.

There are other reasons (not including arching, to use higher allowable bending stresses for soldier beam design. These reasons may include: an allowable overstress for temporary applications, continuity over several support points (brace or tieback levels), and the nature of the empirical earth pressure envelopes which were originally derived to calculate the maximum, anticipated, random(?),bracing load (not the design load to the soldier beams or sheet piling). Different text books focus on the different applcation of these reduction methods. I've seen reductions for soldier beam loads of 10% up to 33%. For temporary wall, apply one of the reductions. For permanent walls, it is not uncommon to design the soldier beams or sheet piling for 80% of the load from an empirical earth pressure envelope. Bracing or tieback anchors are designed for the full load. The key to staying out of trouble in wall design is to not double-dip or compound reduction factors. It is easy to justify use of one reduction factor. In my opinion, it is folly to use more than one.

http://www.PeirceEngineering.com

 

[DRC1]

MHash, As for the first method, could you attach a sketch? Having trouble visuallizing the connection. Is this strictly for bar anchors or also for strand? Are the 2 plates that are added to the oppositte side of the beam from the anchors set parrallel or perpendicular to the flanges? Just trying to get the right picture in my head.
I have done the second method many years ago. Made a bunch of piles with a pipe set at a down angle of 15 degrees, as that was wat I typically used and was using on this project. Cut out the web and flages and welded it in. Put two cover plates on and welded those on also. I belive I extended the plates back about 18 inches or so from the hole to eliminate shear flow problems. Cover plates were somewhat thicker than flanges and a little narrower. In the end it cost me about the same as if I had done a wale. Piles were reusable. Set hole down 8 feet from top.

 

[VAD]

MHasch

Interesting comments. Keep on plugging away. Keep on working for the Contractor. The experience is invaluable and has allowed you to cultivate your thought process. We often need to ask contractors to do the design as in many instances their experience goes beyond some of the theories that we are using. Risk however is an important aspect of decision making and this should be recognized and incorporated in our design approach. I would suggest that you provide some of the practical experience through discussion with your structural engineer. My experience is that the structural engineer and geotechnical engineer need to collaborate unless your structural enginer is well versed in geotechnical engineering and can be considered a Foundation Engineer.

With respect to some of the questions raised PEinc has provided some good insights. Your heartburn issue seems to be quite acceptable to me. Here is where as typical desk engineers we fail miserably. Keep on pushing and plugging away for waht you have concieved. Unfortunately textbooks do not provide these ideas unless they are written by an experienced field person. I have seen welders come up with brilliant solutions. Anyhow.

Regards

 

[MHasch]

DRC1

There is a detail in "Earth Retention Systems Handbook" by Alan Macnab that is similiar to what I am describing. It shows installing strapping to take the rotation. We put in our first project with type of connection. However, even though the borings did not show rock, there were several 2 ft layers of limestone that we encountered so I doubt there was much load that ever made it the wall. We did tension the anchors without any rotation problems so we skipped placing the strapping. I have also seen several project with this done in sands without any strapping.

 

[wmta]

This discussion is interesting for my point of view. There are many variables that should be considered in the total design.

1. Soil type and design values of both active and passive values.( critical )

2. Surcharge loading - can be very critical

3. Allowable over-stress - duration of loading a factor and quality of steel.

4. I believe it takes using good engineering judgement to not have the design overly conservation.

5. Side pockets can and are used on many majors projects (see photo). Details may effect design stresses in the pile.

Not sure all the factors in a great design can be covered in a short discussion.

William Tucker