Are cap plates required on H-piles? 4

[cliff234]

We are designing a building supported on 75-ton (service) H-piles. We showed cap plates on the piles and were told that cap plates are not needed – and were given a table produced by Bethlehem steel in 1965 (copy attached) to justify the plates are not needed. The table shows cap plate sizes for various piles – and then says that the plates are not needed if the pile cap is “properly designed and reinforced”. I have been scratching my head trying to figure out what this means. It seems that our entire profession (asking around it seems that no one in our area (Northeastern US) uses cap plates) is hanging their hats on this statement with no information to justify it. Why did Bethlehem Steel provide the cap plate table if a note at the bottom of the table essentially says, “Never mind – you don’t need the plates.” What is a properly “properly designed and reinforced" pile cap? (All of our pile caps are properly designed. Does this mean that if we improperly design the pile caps, then we need the plates? That makes no sense. I am thinking that “properly reinforced” might mean that we need bars crossing over the pile (confinement steel) – so our detail will specify that 2 bottom bars pass over the piles in each direction.) I checked the bearing stress on the concrete between the pile and the concrete and it’s not even close. We need 150k (service) bearing capacity and we have about 46k (service) bearing capacity with 5 ksi concrete. Does anyone have information to justify eliminating the cap plates? Can anyone shed light on this for me? How can I justify not using bearing plates on 75-ton H-piles? (I asked this question previously about a year ago. That was a smaller project. It's built. No cap plates. This current project is bigger. We also have an adjacent building that will be using 150-ton H-piles.)

 

[bookowski]

I posted a very similar question on here not too long ago (link). I had a different pile type but same issue. I did a lot of digging and found that there isn't much consensus on the the pile to cap transfer mechanism/design. As I looked into it and some of the suggested/recommended detailing I'd find that I had local bearing stresses > 15ksi in the concrete. I asked some geotechs about it and the most common answer that I received was that this is how it's always been done and it's worked fine. In my area I see piles from geotechs with 500+ ton capacity, no way to detail that transfer within code limits without crazy plates and bars etc. but people use only pile embedment all the time here.

Not sure if you're using strut and tie but I found issues with trying to approach it this way as well, s&t limitations at nodes seem to cap your capacity way below what the pile design allows (about 1/2 of just a straight bearing check since it doesn't allow increase in effective area). I found some papers and aashto recommendations about raising that limit in s&t but nothing codified.

Can't offer any help other than saying that you're not crazy, it doesn't make much sense. It seems like it works in practice and over a very small confined area you can probably reach huge bearing capacities compared to what we design with, but it's not in a code that I could find.

 

[SlideRuleEra]

cliff234 - What size HP are being used?
Also, what length of each HP is embedded in the cap?

http://www.SlideRuleEra.net

http://www.VacuumTubeEra.net

 

[cliff234]

SlideRuleEra, HP10x42 with 6" embed into bottom of pile cap.

 

[SlideRuleEra]

cliff234 - I'll get back with you this afternoon.

http://www.SlideRuleEra.net

http://www.VacuumTubeEra.net

 

[cliff234]

SlideRuleEra, Thanks!

We were able to find the 1947 research study upon which the "no cap plates are required" statement was based. We are comfortable not using cap plates with the 75-ton piles and are now looking into confirming that we don't need plates on the 150-ton piles (HP14x73 piles).

 

[Ingenuity]

Can you call Bethlehem Steel and ask them?

Sorry, extreme sarcasm

 

[KootK]

Is the research something that you can ethically share Cliff? I've been interested in this since the threads that Boolowski referenced.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[KootK]

The clip below shows what CRSI has to say in their latest pile cap guide. Some of this stuff really does strain credulity in my opinion. Supposedly, I could install a 400 ton pile with 4" side cover and sleep easy. Makes me wonder what it means to "properly design" the cap.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[SlideRuleEra]

Neither the HP14x73 nor the HP10x42 are going to need cap plates for those loads (300 kips and 150 kips, respectively).

Unlike virtually any other structural load, this one is applied inside the "properly designed and reinforced" pile cap.

First, ignore friction of concrete on the sides of the HP. Allowable is in the 20 to 40 psi range and is handy for uplift calcs, but usually ignored for compression.

An HP14x73 (without a plate) has a crossectional area of 21.4 in². For a load of 300 kips, that's a compression stress of 14 ksi applied to the concrete (inside the rebar cage). Note: For an HP10x42 with 150 kip load the compression stress is 12 ksi.

Of course that's unacceptable compared to 5 ksi concrete... but that is not a valid comparison.

The volume or concrete inside the rebar cage is large compared to the volume of concrete being stressed by the compression load at the top of each pile. The concrete mass itself will dissipate the stress in many directions.

Additionally, if the pile's compression was going to cause the cap to fail, either by splitting or crushing, the volume of the cap (inside the rebar cage) will have to increase. That is not going to happen at a spot with 14 ksi - the rebar cage will maintain containment. Even if the concrete "crushed" in a small volume at the top of the pile, crushed concrete will have the same volume as intact concrete... an observer would not know that anything happened inside the cap.

In fact, putting a plate on top of an HP will have an unintended negative consequence. During concrete placement, the Contractor will have to get consolidated concrete under the plate and between the flanges of the HP. This is virtually impossible to do with any assurance of success, especially since the plate is only 6" above the bottom of the cap. IMHO, this alone (having plates with questionable concrete below it) is enough to compromise the design.

http://www.SlideRuleEra.net

http://www.VacuumTubeEra.net

 

[cliff234]

KootK, I'll post the URL where we found the report on Monday.

SlideRuleEra, Thank you for the feedback. We have bottom bars in the pile caps and we specified that we want 2 bars spaced 6" apart over the piles in both directions. we are looking for those bars to confine the concrete sufficiently to prevent the H-piles from punching through. The research report had test specimens with no reinforcing steel! (It also looked at specimens with spiral ties and with cap plates.) When you referred to a "rebar cage" were you assuming spiral tie confinement steel? We were considering the bottom bars running over the piles to provide confinement.

 

[Agent666]

I have not read each and every reply, but to me it's an upside down column effectively, knowing this ask yourself if you would be happy embedding a column say 8" into some concrete with no baseplate for the loads being considered....

My answer is no dependant on the magnitude of the loads. Put the plate on and demonstrate how the load transfer to the concrete occurs and that measures are taken to ensure placement of concrete won't be an issue. I often also see and use Shear studs for transferring the loads as it's better for concrete placement, you just need to take the pile a lot further into the pile cap.

 

[SlideRuleEra]

cliff234 - Confinement does not have to be sophisticated or strategically located, but bars only above and below the pile but are not what I would consider confinement - especially for the piles at the cap's perimeter. Compression could "blow out" the side of cap.

What is important is to have plenty of steel and have it well distributed. As long as the top, bottom and all sides have have reasonable reinforcement, there is no way the volume inside this cage can increase under foreseeable compression stress applied inside the cage. In other words, don't cut corners on cap size and reinforcement. Having HP embedded more than the traditional 6" (say 12" instead) is also a benefit... of course that means a thicker pile cap, too.

IMHO, the reasonable options to not having proper confinement is either more piles or larger (more crossectional area) piles. Loads of 300 kips on HP14x73 and 150 kips on HP10x42 is really pushing to the limit.

http://www.SlideRuleEra.net

http://www.VacuumTubeEra.net

 

[thaidavid40]

I can easily visualize and concur with SRE's concerns regarding the ability to effectively consolidate the concrete mix under a cap plate. If the localized bearing stress on the concrete above the H-pile section still causes you concern, then there is an alternative answer. Weld some thick doubler plates onto the H-pile flanges to thicken them and increase their bearing area, and thus reduce the local concrete bearing stresses to a level that you are comfortable with.
Dave

Thaidavid

 

[KootK]

All of the structurally active concrete and rebar in one of these "properly designed" pile caps would be located above the top of the pile. Would that not ameliorate a good deal of our concern for quality concrete beneath a potential cap plate?

knowing this ask yourself if you would be happy embedding a column say 8" into some concrete with no baseplate for the loads being considered....

This is pretty persuasive in my opinion. Conversely, I've known many an engineer to not sweat axial only base plate, even though they distribute stress very poorly, because no one ever seems to have heard of one failing. Maybe this is why.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[cliff234]

Here is the link to the elusive report upon which our entire profession appears to justify not using cap plates on H-piles.

http://www.ohiomemory.org/cdm/ref/collection/p15005coll2/id/702

 

[KootK]

Thanks for that cliff. Also pretty persuasive. Apparently you can get upwards of 10 x f'c bearing stresses out of the contact surface between the piles and the cap based on the series A results. None too shabby...

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[canwesteng]

The research doesn't really shed any light on how we are meant to design the pile caps, other than to use a similar size as the tests, does it?

 

[KootK]

The research doesn't really shed any light on how we are meant to design the pile caps, other than to use a similar size as the tests, does it?

I'd consider that to be an accurate statement with one obvious exception: direct bearing stress at the pile/cap interface. In terms of modern design methodologies, I imagine that one would:

1) Use sectional methods and check one way shear, two way shear, and flexure OR;

2) Use strut and tie with the following modifications/relaxations:

2a) Don't sweat the nodal stress check at the interface between pile and node AND;

2b) Consider your tie development length to start at higher shear side of the steel pile.

As bookowski mentioned, there are a few failure modes that can go south on you when using code strut and tie to evaluate.

One feature of the testing that would differ from real life is that, in the testing, the axial load would be resisted by a concrete compression field/strut parallel with the pile. In a real life, multi-pile scenario, that strut would be coming in at an angle. That would exacerbate the compression stress in the struts and some of the node faces.



I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.