Plate Loading Tests

[SUGARCANE]

As a test house in the UK we frequently conduct plate loading tests. The method essentially being based on the requirements of BS 1377: Part 9, ensuring that the plate is loaded such that the settlement recorded is in excess of 1.25mm. We then evaluate the data in accordance with Dpt. Transport Design Manual HD25/94 which gives us the following data:

1. Modulus of Subgrade Reaction (K762) in kN/m2/mm
2. An equivalent CBR Value
3. Elastic Modulus E

This is not always sufficient for our clients (generally earthworks contractors). We are continually being asked (particularly for building platforms as appose to road construction) if we can tell from the results if they have complied with their specification requirements and acheived a specified bearing capacity (generally 75-150kN/m2 dependant on site use).

THIS IS WHERE WE GET STUCK.

The loads that we apply to the plate vary dependant on how competant the soil is that we are testing. In general, we would be putting loads upto 250kN/m2. What we generally tell our clients is that we can interpolate from the load/ settlement graph a settlement at any given bearing pressure applied to the plate and that somebody somewhere (generally the design consultant) should know if this is acceptable.

Can anybody please help. Is there anyway we can determine if our Clients have acheived their contractual requirements.

Keep it simple. We are not all geotechnical engineers!. Nonetheless, we get a hell of a lot of queries from you guys so its not only us that struggles to find the answers.Remember, its supposed to be us that does the testing, you to do the clever stuff.

Anyway, a worked example would be good.

Heres hoping

 

[Focht3]

SUGARCANE:

I'm afraid you will remain stuck. At least in my opinion...


Your clients are asking a question that you cannot really answer. Plate load tests (PLTs) have a limited depth of influence generally taken as twice the plate width; so your PLTs only really evaluate about the first 0.6 meter or so of soil. A footing that is 4 meters wide and 7 meters long will have a depth of influence of about 8 meters. Your PLT, if applied to this footing, would miss the influence of 7.4 meters of soil...

And a PLT deflection of 1.25 mm is too small to model the effects of footings that will deflect 25 to 100 mm while in service. It's just not representative.

My advice? Avoid these requests - tell your clients that, "its supposed to be us that does the testing, [let the geotechnical engineers] do the clever stuff."

 

[BigH]

Focht3:

 

[SUGARCANE]

Focht3

Thanks for the speedy reply.

I hear what you say regards how representative the test is for the example given. However the fact remains the designers are specifying plate tests a means of testing a materials acceptability. Somebody somewhere must be evaluating our data and deciding if the earthworks fill has been placed to acceptable standard.

In some instances, areas of fill are shallow, may be 2 or 3 meters which means, going back to you example, that a foundation may influence existing ground that our earthworks contactors have no control over.

Does this mean that the designers are specifying the wrong tests! Surely if a designer wants to ensure that a design is safe for such a zone of influence, wouldn't zone tests or SPT/DCP's be more appropriate, preferably before works commence to prove existing ground is of an acceptable bearing capacity to full depth of zone of influence.

If we wanted to ensure that material placed by an earthworks contractor is acceptable, wouldn't insitu tests such as DCP's (to depth of fill placed) or reprentative bulk samples tested for shear strength be more appropriate.

OR, back to plate tests. If we know the limitations on the depth of influence is equivalent to approx twice plate size, is there a means of determining the bearing capacity of the material tested and for us to qualify this in our reports. At least this is actually determinimg the bearing capacity of the material placed as appose to the whole depth of influence under the foundation. We can ammend plate size to suit depth of fill placed. If we did this would we need to try and penetrate plate signifcantly more than 1.25mm

This is where it gets all contractual i know but everyone still asks the testing house. At the end of the day we are working for earthworks contractors who want to know if their earthworks have been done in accordance with the specification (not if 5 meters below their fill is suitable ground)


Thanks again

 

[jheidt2543]

Well, open mouth and insert foot.

I usually look forward to comments by Focht3 and BigH. Focht3's discussion was really thoughtful, but his advice is H.S., in my humble opinion. Contractors are paying for analysis and advice and I think the engineering profession should make every effort give them what they are paying for.

As a testing firm, Sugarcane should be providing an answer to the question, "does this compacted fill material meet the required bearing capacity?" If the tests that are requested can't answer that question, why perform them? Why not tell the client that, to get the information requested, a different test is required? Or, maybe, based on Focht3's discussion, a series of plate tests at various elevations as the fill is placed is what is required.

There is an answer to the question and someone has to step up and provide it, within known tolerances. We can't require a contractor to perform to certain tolerances and then say we can't tell him if he met them.

 

[Focht3]

jheidt2543's comments are worth discussing - and appropriate to the circumstances. (No "foot in mouth disease" that I observed...)

I think the main problem has to do with the acceptance criteria: SUGARCANE initially said,

We are continually being asked (particularly for building platforms as appose to road construction) if we can tell from the results if they have complied with their specification requirements and acheived a specified bearing capacity (generally 75-150kN/m2 dependant on site use). (I added the italics and bold to this quote.)

Can you evaluate compaction quality using a series of plate load tests with depth? Absolutely. Can you evaluate the allowable bearing pressure beneath a "large" footing using "small" PLTs? Not in my opinion.

But I'm willing to listen to other opinions! What are your thoughts, jheidt2543 ?

 

[SUGARCANE]

Wow. My first time on eng-tips and it seems I have set the cat amongst the pigeons. I'm suitably impressed by the quick responses too.

I thought an extract from a recent specification might help (in case I have caused any confusion). Here goes:

"The surface plate bearing tests shall exhibit a minimum ground bearing capacity of 80kN/m2. Test plate shall be 450mm diameter and should be loaded to twice the required ground bearing capacity". And thats it, nothing else within a hefty document other than a B of Q saying how many.

Yes, any lab can do the a plate test to this spec, but can we tell them if its passed. Well no we can't!. So how does the person who wrote the spec know if the tests that we do meet his specification? Again, we don't know!

Yes we could, as jheidt2543 rightly points out, tell our clients a plate test will not give them this answer. But remember, contractors have a requirement to do the plate tests as per the specification. In this game of construction, we don't always get a specification prior to commencing the tests.

 

[BigH]

A bit of a if you will permit me.

First - you are a laboratory. You carry out a test as specified by an engineer or client. You carry out that test in accordance with either particular specifications (ASTM, BS, IS, etc) or "good" engineering practice (especially in those cases where the tests are not so well defined). You report the test results. You've done your job, in my opinion. This assumes you are a testing facility and not a lab tied in to an engineering facility. If I sent you a 32mm dia deformed steel rebar for physical and chemical testing, you would provide me (1) yield strength (2) ultimate strength (3) elongation at ultimate (4) bend test results, i.e. , cracks or not at bends (5) you might provide modulus of elasticity although most specs don't require it (except India's MORTH) (6) % of certain chemicals e.g., phosphorus, sulphur, manganese, etc. When test results are finished, you send me a test report saying that the material meets the requirements of AASHTO M31 or IS1786, etc. You do not tell me that the steel can't be used on my project. If the strength is 1% low, I can always add 2% extra steel by closing the spacing 20mm or so. You test, you report the facts and that, my friend, is your job. Of course, clients (especially those that do not know) will ask you 'Hey, is it okay??", etc. Answer at your own risk.

For the plate load tests - I have posted this elsewhere but it might be worth repeating. Current: Job in India. Design report borings show "decent" soils (N>5, Su> 50kPa) - not bad for a 7m embankment. Can stand - maybe with a small berm or 2:1 side slopes. But, whoa - change in design. Need retaining wall as the ROW is insufficient for an embankment slope. Designer wishes to put down a plate load test to determine bearing capacity. Fallacy?? Yes! Even at Su about 50, this is insufficient for a retaining wall of 8m high! Plate load test??? He wants the test "at surface" - there is a bit of a desiccated crust. If crust is 2m thick, what good is plate load test for the underlying non-crustal clay? I insist on a boring. Result: after 3m of crust (1.5m of clayey fill and 1.5m of desiccated crust with N = 5 to 7, we get the following N values at 1m intervals: 1, 1, 2, 2, 3, 5. Compare to design report of N>5 at these levels. What? Yes, the boring for design was through the existing 5m high embankment that had consolidated the clay for nearly 25 years. Bottom line: Your plate load test would never had shown the very soft organic clayey soils that exist and I envisage would never had foretold a failed retwall. This is why seasoned/experience professionals MUST be involved. They must determine if such tests are okay or not. Focht3 talks of footings/foundations as did I. Your PLT would be scant good unless you are certain that the soils "increase" in strength with depth below which you are doing the test.

PLT may be used for checking out compaction of fills - but kind of expensive, eh? Why not just use pentests (dynamic cones using 140lb hammer at 30inch drop) to get a penetration response. Why not just know that soils compacted to 95% MDD modified (heavy tamping) just don't settle (unless it is the foundation settling!) Okay - contactor is required to do the test. There must be some sort of basis for acceptance - if value is thus, then compaction passes; if not, it fails. Again, I would do the test as the testing lab and let the Engineer on the project determine if the test meets his acceptance criteria. Contractor needs to work with the Engineer in setting up the ground rules before work commences or very soon afterwards.

Interesting thread - thanks for your kind time in letting me put forth the above - cheers to all.

 

[Focht3]

Wow, I sure am glad that I didn't get a chance to finish my post earlier today and had to wait until now to complete it! Although BigH "stole" my soapbox!


jheidt2543:
I hope that "H.S." stands for hindsight, not "equine excrement."

This thread involve two different issues that are intertwined in the example posed by SUGARCANE. The first question is, "How far should we venture from our immediate responsibility to perform PLTs and report the results?" And the second question - not directly asked by SUGARCANE but implicit in my posts and BigH's, is "Is the plate load test (a "small footing&quot appropriate for evaluating the allowable bearing pressure of a "large" footing?"

It should come as no surprise that I concur completely with BigH on both questions.

On Question #1
The acceptance or rejection of the PLT results rightly belong to the specifying authority, not the contractor or his laboratory. I understand the pressure that can come from a contractor; I have worked in a number of areas where the contractors call the shots with respect to construction inspection and testing (including my present area, unfortunately.) If the specification is clear and unambiguous, then you can state whether the test was performed in accordance with the specification, and whether the test results meet the specification. You should stop there - or expose yourself to potential aggravation, if not liability. In my view, it is foolish to venture beyond one's area of expertise and responsibility. And you will almost certainly aggravate at least some of the engineers of record for "stepping out of bounds."

On Question #2
Please review my first post in this thread to review my reasoning for rejecting the use of PLTs to evaluate allowable bearing pressure. And BigH did a superb job of providing a case history to document the validity of my reasoning. The misapplication of the plate load test is a real, and all too common, problem. At least where PLTs are commonly done. (Thankfully, they aren't used in my areas of practice.)

A good thread - worth the time and effort.

 

[Focht3]

P.S. SUGARCANE:

You generated a pretty good discussion on your first "try." Congrats! (And keep your sense of humor!)

 

[VAD]

Well, I am going to try and duck Focht3 hammer. By the way how do you incorporate the smileys. The interpretation of HS has me still laughing. Excellent responses Focht 3 and BigH. I am not schooled on the plate load test in practice except knowing that it was used in the early days by Mc Leod in Canada in airport runway design and in early designs. I think we may have attempted it once but when something is not fashionable in a jurisdiction then it is generally not given much thought.

My interest in this thread is to understand the relationship between the testing lab and the contractor. It seems that the Contractor has to undertake tests on the contract and provide same to the Client's Engineer. I worked on one such project years ago in the Netherland Antilles where the test was done by the Contractor and we presnted the results to the Engineer for acceptance. I was at the time working for the Contractor.

With the End Product Specifications or End Result Specs, the contractor is responsible for quality control i.e doing his tests on a mix that he wants to be accepted for use say in a paving project. The test is done by a testing firm generally and the results are passed to the Client's Engineer for acceptance.The mix design that is passed is generally what is expected to be approved as in such cases the requirements are pretty clear cut and the contractor generally knows about this from the lab.

I am not sure of the details of the practice related to sugarcanes jurisdiction, but ther may be some obligation to assure the contractor that the information meets the specs. Well maybe not. Again I rather understand first the contract before making a definitive statement.

Nice discussion. I still like the hammer/mallet.

Cheers

 

[SUGARCANE]

OK OK OK. I feel like I have been suitably battered with the weighty hammer of the geotechnical engineer.

I think if I go back to my original thread, I answered my own question. That is "..let us do the testing and you guys do the clever stuff".

That will also be my answer to all designers, engineers and consultants who in future ask if their plate tests have passed. Either that or tell them to visit eng-tips and find out for themselves! From now on we'll stick to what we know best and I promise never to tread on the big toes of the geotechical engineer ever again.

Seriously, this question is asked at least twice per month and its one that we have never been able to answer. Its a shame that BigH and Focht3 aren't specifying the testing. At least then we would ALL know who should be doing what.

Thanks to all contributors.

 

[Focht3]

VAD:

I'm glad the H.S. line amused you. I think it is generally better to handle things with humor when possible. And I was laughing as I typed it...

Regarding smileys: they work about the same as text formatting. Once you know the name of the smiley, enclose it in square brackets [] without extra spaces. Try it - they help add "meaning" to your messages, and they're fun! You can find them by clicking on the 'Emoticons/Smileys' link in the 'Step 2 Option' part of the 'Your Reply' box (where you type in your post.) Be sure that, as a minimum, the second and third boxes are checked ('Emoticons/Smileys' and 'Process TGML').

Some smileys are static; others are animated. Try them out -

 

[BigH]

Sugarcane - being from England - you should tread - it's us Yanks who'll say "Don't tread on me!". Seriously, though, your post has provided for a fantastically fun and enlightening interaction between some pretty good engineers - such posts are most welcome! We'll come to you for testing!



ps Where is dirtdoc anyway? Miss his threads on testing!

 

[TJWATKINSPG]

Great discussion! The only thing that I think got dropped is that as Professional we may need to direct our clients (or educate them) as to which test to use for their needs. In this case, of course, the test was prescribed before the testing or for a different purpose. Therefore the answer must be that that information can not be gained from this test.

The truth will set you free. Best of luck. Geodude

 

[Focht3]

BigH:

I haven't "heard" from DirtDoc in awhile, either. If he's 'out there' and not posting, we miss you! Please come back!

I'm trying to find out if he's still around - another way...

 

[jheidt2543]

Having just finished my "equine excrement" sandwitch I'll get back in this discussion. My, how much has transpired in one day!

Sugarcane's posting of the specification clarifies what is asked for - a test. The AVERAGE testing lab will perform the test in the field and the soil either withstands (passes) "twice the required ground bearing capacity" or it doesn't (fails). The fact that the test maybe meaningless in this application is beside the point for the time being, that is what the specification calls for. This could be the end of the discussion.

Focht3 and BigH rightly raise the red flag, this is the wrong test to answer the contractor's main concern, "am I meeting the required bearing capacity?" The specification writer either pulled the wrong spec. off the shelf or didn't understand what test or series of tests should be called for.

Following up on Focht3's Question #1, I agree that acceptance or rejection of the test results lay with the specifing authority. However, the contractor certainly has a vested interest too. As his work progresses, he has to know whether he is meeting the specifications or whether he has to stop and rework an area. This has to be decided in "REAL TIME" and wrong decisions cost real money. I think we can all agree that halting work to wait for the formal report is unrealistic, hence the contractor's question.

My original post was to raise the point that we can't leave the contractor hanging in mid air. If the answer to the question can't be had because the wrong test is specified, I think the testing lab has a professional responsibilty to say so and recommend to whom ever hired him, the contractor or the specifying authority, the correct testing proceedure.

An aside: this has been a fun discussion, it ruined my taste buds, but I thought about a lot of engineering. Thanks!

 

[SUGARCANE]

To clarify for jheidt2543, we will always inform our clients where (and why) we have any reservations with any testing requirements, whether it be for plates or any other part of our service. We are a proactive lab and frequently get involved with problem solving, suggesting alternatives for consideration as we fell appropriate.

For the example specification quoted above, it was agreed with our Client and their Consultantants prior to commencing any testing that we load the plate to 160kN/m2 in 6 approx equal increments. We just reported the settlement values and the data is all yours and our jobs done (sorry, i should say ours and yours, were all in this together, yeah right!).

I thought I'd had enough of this subject but someone just dumped another plate test specification on my desk. This goes as follows:

Load 600mm plate in increments upto 300kN/m2
at 100kN/m2 settlement should be less than 3mm
at 300kN/m2 settlement should be less than 10mm

Now at least with this spec we have clealy defined acceptablity limits which we can all understand and more importantly for our contractor, we can tell him in 'Real Time' if his fill is up to scratch.

For this site, again the fill area is shallow, only 1 or 2m is some places. I'm not sure on final use of site, likely to be light industrial.I know it note a lot to go on but can anyone fill me in on what is the designer trying to model here! He's come up with acceptability limits, How!

It appears to me that he is trying to model a foundation which will exert somewhere between 100-300kN/m2 presumably. If my assumtion is right, surely such foundations will influence more than the 1.2m that the plate will. We are then back to why we are doing them. Are plate tests anygood at all for confirming foundation designs.

I NO LONGER UNDERSTAND WHY PEOPLE WANT PLATE TESTS FOR EARTHWORKS FILL OPERATIONS. AAAAARGGG! Still, brings in good money, ha ha.

Given that the plate tests are coupled with insitu density tests to confirm compaction less that 5% AV and lab tests to confirm MC is with acceptable limits based on PL and OMC, isn't it just a little bit overkill to ask for plate tests as well.

Hell, your gonna wish I'd never started this. [YAWN]

 

[Focht3]

SUGARCANE posted the following questions & comments (in italics) and I responded in plain text:

Are plate tests anygood at all for confirming foundation designs.

I don't think so - unless the plate and footing are about the same size and the sustained footing loads will be quite low.

I NO LONGER UNDERSTAND WHY PEOPLE WANT PLATE TESTS FOR EARTHWORKS FILL OPERATIONS.

It's a "feel good" test. It doesn't really help evaluate the problem, but they feel like they have done something meaningful. And the person that specifies the PLT doesn't understand geotechnical engineering.

Given that the plate tests are coupled with insitu density tests to confirm compaction less that 5% AV and lab tests to confirm MC is with acceptable limits based on PL and OMC, isn't it just a little bit overkill to ask for plate tests as well.

Absolutely. PLTs are a waste of time and money when you have monitored the fill placement, and the placement guidelines are appropriate to the intended use.

Hell, your gonna wish I'd never started this.

Nah!

 

[BigH]

Finally - and I see it in India - you may just have real "old" guys specifying the works and as "they have always done PLTs, they continue to put them in." These are guys who wouldn't really know the nuances about tests - just that they ARE done as a matter of course. Just when you think the the "cancer's" relapsed, it rages again! FUN ! ! !