Open question: Are we doing less lab testing? 2

[DarthSoilsGuy]

I've been reading posts lately and i feel like i'm seeing a reoccurring theme. i think back to my old Geot/CMT company.

When i would work on CMT work where either we or other companies did the Geotech report, i would hardly ever see any lab testing besides natural moisture, Proctor, maybe a wash #200 sieve, maybe Atterberg limits, and very rarely a hydrometer to supplement boring logs for the geotech report.

On DOT work we would take undistrubed samples for the state to run tests on. I'm assuming the state performed triaxial or unconfined testing on the samples.

My question is: Did there use to be more lab testing, specifically triaxial, consolidation, or __________ testing, performed by Geotech companies for commerical and industrial construction?

In my head, i imagine that:
There use to be and as the old Geotech Egrs picked ran the tests over and over again, they got a good feel for the expected info to come out of the tests. Once this happened, less tests were run and info was based on experience for commercial efficiency (less expensive lab costs). Then the oldies retired and now you have an industry guessing about soil properties without having good experience to back it up.

Am i way off base here?

-DSG

 

[dgillette]

No, you aren't off base.

Years ago, I had the pleasure of working with an old guy named Carroll T. Coffey, a lab tech with ~40 years experience. He'd worked on a lot of the research USBR did in the 1960s and 1970s. He could rub the dirt between his fingers, spit on it, rub it around some more, and tell you a reasonably accurate PI. He had good rules of thumb for drained friction angles (which also turned out to be accurate). Carroll could look at any material and tell you the Proctor max density pretty well also. He got all that from a lot of years of lab and field testing. If he was still alive, we would need hardly any test capability for those things. (He also loved what he was doing and teaching it to the young engineers, most of whom now have some gray hair. He retired one year about Christmas time, and his coworkers bought him a big screen TV. He watched the Christmas TV specials and the NFL playoffs, then switched off the TV and came back to work until he had a stroke many years later.)

The thing that he did not have experienced-based intuition for was undrained strength, which we now have more appreciation for in earthquake engineering. I think it's also a harder thing to deal with by rules of thumb and squeezing of dirt, since it is so sensitive to void ratio, stress history, etc., in comparison with phi' for example. (How much does phi' vary between samples tested at Proctor max density and at 92%? Not a whole lot.)

Anyway, where was I going with this? Oh, yeah, we used to do more lab testing, and in many areas it's OK that we don't do as much anymore (triaxial for effective-stress parameters, direct shear) because we, as a profession, have a reasonably good handle on certain parameters. I don't think that can go on forever, however, for things like undrained strength, cyclic resistance, etc. that are sensitive to small variations in density.

 

[iandig]

I would concur with the previous post that we don't do as much testing as we used to. This is in the UK and on that basis it starts to look like a general trend. I agree with the previous comments that we now have more data to look at than we used to, bearing in mind that soils testing is still a very 'young' profession. We have been making concrete for thousands of years, but testing soils for less than a couple of hundred.
A couple of points I would add though is that:
1. Some engineers only test sufficient to prove their assumptions - OK with many years experience [comments please]
2. Some engineers don't know why they are testing and as such rely almost soley on historical/reference values [comments please]
I am an ex CMT who saw the volume of testing done reduce along with the rate for each test. At the moment we are charged about the same for a test as we we used to charge in 1990! It all seems that the importance of geotech testing is not what it was, however we are about to go through a change with the introduction of BS EN 1997 [Eurocode 7] which will replace BS 5930, and this may well have a short-term impact on the testing done in the UK. The new standard emphasises the use of effective strength parameters in determining soil properties such as the allowable bearing pressure, whereas prior to this quite often only undrained parameters would have been used. Only time will tell and how we 'interpret' the standard - implied [rest of Europe, 'if this is what you do this is how you do it, if its not then don't'] or implicit [UK - 'you will do it this way']

 

[DarthSoilsGuy]

i always forget to consider international aspect of the forum.

My experience is in the Southeast and New England areas of the USA.

to add. i've seen more and more field testing and inspection required, and especially so since the entire USA is more or less using the building code developed in California now (a recent change). There hasn't been any increase in lab testing that i can tell however. maybe more requirements for the material manufacturers but not so much for geotech/cmt.

 

[jdmm]

As the world becomes more experienced they become more confident and require less testing be as the world becomes more litigious the engineers become more cautious and require more testing to confirm their opinions. It is a shame but this has been brought on by a legal sensitive system and lobbyist system in the US. Sorry for all of us.

 

[fattdad]

Not sure I agree with jdmm. Developers/owners are looking for the cheapest "bid" for their geotechnical and field testing services. As a result, pavements are designed without CBR values, earthworks are specified based on a geotechnical report that only has visual classifications (i.e., no Atterberg limits or percent sand data) and projects are run with "a" Proctor - heck it all looks the same, eh?

Too many geotechnical testing firms are falling into the trap that if a fill "proofrolls" it must be compacted - well it isn't true, especially when the contractor is placing 2-ft thick lifts!

It frustrates me no end and non-engineered "structural fills" are causing big problems. All over a few bucks.

Off my high horse now. . . .

f-d

¡papá gordo ain’t no madre flaca!

 

[BigH]

I believe in lab testing - but not overkill. For instance, I have recently scheduled lab testing on a 35 m boring - basically a soft clay all the way. I scheduled some 3 consolidation tests, one undrained triaxial with porewater pressure measurements, 6 Atterberg limits, 18 water contents (not every water content was scheduled with an Atterberg), no sieve and hydrometers on the silty clay. Yet the drilling/lab testing company wanted to add sieve and hydrometers on every undisturbed sample, add Atterbergs to every scheduled water content. Definite overkill - makes them money. Point I am making - schedule the lab tests but make them relevant and not just to fill up a page. I agree that when I worked steady in Canada in the 70s to 90s, we didn't do nearly the number of tests done that were done in the 60s and early 70s. Couldn't afford to do it with the cost of the lab testing and a 1965 $3000 investigation going for $1400 in 1993.

 

[jdonville]

I agree with fattdad that too often testing is skimped because the geotech firm wants to win the bid.

As a consultant, I always fought to perform moisture testing and at least a hand penetrometer over the depth of at least one boring, as this information can be extremely helpful in characterizing the subsurface conditions present at a site.

I would characterize the Atterberg limits as being more useful information that particle size distribution, as the latter can usually be well-estimated based on visual examination or less frequent sieve/hydrometer analyses.

Now that I am out of consulting and into the contracting side of things, I get to have the last word over what tests get performed after discussing with the geotech for design-build projects.

I am still disappointed with the lack of test results for projects where we are asked to bid a shoring or foundation system based on an existing geotechnical report. Often the report does not address important design issues for what we are asked to bid and we have to include additional geotech in our bid.

Jeff

 

[Touma08]

I hear you Jeff. I too was a consultant for a while and now I am more on the contracting side. Indeed, quite often, we ask the client's permission to access the site and conduct our own soil investigation, and ususally it is an out of pocket cost, just to feel confident enough to price (and be competitive) a turn key project. I have read geotech reports perhaps > 50 pages long and came up with nothing relevant to the actual design: not enough relevant testing and alot of pictures and irrelevant mumbo.

I believe, based on the discussion so far, the problem seem to be international: part economics & competition, part arrogance (we know it all), and part ignorance.

An example: here we have researchers still trying to figure out the best way to model the bearing capacity problem for a footing resting on a ultilayered soil medium and are utilizing F.E. models to come up with a solution for two layers, yet I had someone the other day say "Who does bearing capacity calculations anymore, for the last ten years I've used nothing but charts".

Are we, as geotechs, overdoing it ? sometimes I do have my doubts, especially when you see consultants accepting work based on a person jumping over the finished section (I bet that beats the 2' problem of fatdad) and considering that compacted to specifications.

 

[msucog]

i would like to see the geotechnical industry as a whole make improvements. we've got lowball firms out there that prostitute themselves and in turn create unfounded headaches for the rest of the industry. by trying to be competitively priced, we foresake lab testing, additional exploration, or the much needed time to actually evaluate the data. by doing all of these things, we open ourselves up to litigation.
take for instance mse walls. we have actually been sued because a wall fell down. we were specifically hired to only perform density testing at a specified frequency because the wall designer didn't bother to reitterate all the exploration and testing that was needed per their own report. we ran those tests like we were supposed to but because the wall designer dumped so much blame on the owner (and tried to dump on the the testing firm) everyone involved was drug into court. the wall designer tried to defer all points regarding the design, construction of the wall, and acceptance/rejection of the material on to us. and now, when we deal with mse walls, instead of being "reasonable" in our approach to the project, we get very specific because we're not going down that road again. the lawsuit didn't stick to us but we still wasted lots of money and manhours putting all our ducks in a row to defend ourselves. and now, i see that owners really appreciate it when they realize the wall designer has specified $50-100k worth of exploration and testing but sold them a "cheaper" solution to a cast in place wall. i've actually seen the wall designer tell the owner that the wall is more expensive because the testing firm is too expensive. and then the owner simply goes and hires the lowball firm that does $3k worth of testing. we say let the lowball guys have it and take all the risk associated with it. in the past two years, i've probably passed on a half-million dollars worth of exploration and testing for mse walls because we could not get questions answered (questions such as the required tests needed, frequency, testing parameters, etc). however, we still do the testing on the rest of the project and exclude the areas around those walls from our workscope. we lose no work in that sense and shield ourselves from the liability.
i've taken the attitude that we'll do the exploration and lab testing necessary for us to feel comfortable with the project or we won't take the work at all. enough of the lowballing and shopping around...the industry as a whole needs to reform itself and i'm afraid litigation is going to be the primary catalyst to push that change.

 

[MRM]

In the last decade or so, we've made a lot of money by entering projects midway already in trouble due to a lack of lab testing and field testing during construction, i.e. engineered fills that are anything but, wrong MD values, incorrect compaction methods, soft soils down deep that were not accounted for, etc.

There seems to be little risk for us since we're only investigating what went wrong and how it should've been done. Our fix to the problem isn't the cheap way out either. The client goes for it too because they can finally see what happens when they cut corners. Plus, because of that, the money's much better too.

There are a couple of companies that we've contemplated sending Christmas chocolates their way as a thank you. But seriously, every time they low ball a bid, we get ready to go...

 

[Touma08]

Funny but very sad. It is good that you have the opportunity to benefit. For us it has been a big problem. Over the last 30 years, our organization have worked hard at establishing a reputation of quality, which naturally comes at a cost, and because the clients' learning curve has been too slow, and because some are too proud to admit they were wrong, they, more often than none, end up calling upon someone else to continue, or even pay to correct the problem and continue with the same contractor. Consequently, we lose a lot of business due to "intruders".

This particular thread has been very interesting to me because I was under the impression that the problem is somewhat regional/restricted, only to find out that it is more widespread than I had originally thought. 20 years ago, I started preaching "client's education" within our firm; interesting to find out that after all this time, the concept is still just as valid, if not more, and Inetrnationally.

 

[DRC1]

My favorite story on how the owner saved a few bucks on testing:

We did a small bridge for our DOT. The bridge was designed by a cosultant who did both struct. & geo. The bridge was probably 30 ft long and 25 feet wide to be done in 6 (yes, six) stages. The bridge was designed as a double barrel cast in place culvert bearing on native soil(no piles). Each stage was to be built inside a sheetpile cofferdam. The bid price was about $300,000 and was to take 9 months. After we built the first stage, we pulled the sheets to start the second stage. Durring the course of pulling, the culvert sank 9 inches and the wing wal sank 11. The job was switched to pile foundations, work became etremely more complicated. The final cost $3,500,000 and duration was over 3 years.

So during the course of the ensuing litigation, it turned out that the designer had requested additional borings, and laboratory testing for strength and index. The state, decided it did not need to spend the estimated 5 to 10,000 dollars, and deleted the borings and testing from the design scope. They felt the borings would fully characterized the soil. We had samples tested after the inital failure, and for $25, found out from an atterburg that the silt had a natural water content at or nearly at the liquid limit. These soils are inherently unstable during vibratory sheetpile driving, and generally have poor bearing capacity. so, by saving $25, the DOT spent over $3,000,000

 

[Touma08]

Way to go.

 

[Patgeotech]

I agree with fattdad above, developers/clients are looking for the cheapest bids and sometimes these bids are far inadequate for the site.

Working in South Africa, the above is a big problem as I think clients base their decision on the cheapest geotech bid, which often neglects important geotech testing (yet due to cleverly worded reports, they get away with it when things go wrong). I agree that one does certainly get a 'feel' for the soils/rocks in your neighbourhood, but sometimes that problematic soil pops up and this is when there should be no getting away from the testing.

Due to the testing expense (expense perceived by developers and I will have to add structural engineers), I have seen them use their own tests to assess the ground conditions and more than often getting it totally wrong. The question or thought arises "why did they not spend that 0.05% of the budget on the geotech testing" and it would ultimately have saved them the 25% expense to fix the problem. Yet they still do it - what are everyone elses thoughts on the matter?

 

[Touma08]

Education of the Clients and the Engineers. Not easy nor cheap. This is required from us before anyone else because we appreciate the gravity of the situation. It is a long and frustrating road, because it is a continuous effort, but we have been able to achieve some success. We have organized seminars with local chapters of engineering societies and presented something at every conference we attended etc...

I fully realize that it is a lengthy process and we have lost many clients but we got them back several projects later, and in almost all of the cases it was because, for every project we lost or won, we sat down with the project manager and gave him/her the basics on whatever subject we're dealing with. Some appreciate that right away and some take longer; it is a painful and costly road to both: us and the client, but someone has to do it, otherwise sooner or later we will find ourselves compelled to either: do what the others are doing which is unethical, or look for another profession.


Just to comment on the "structural" engineer issue, it is really interesting to note some of the theories we encounter from "them" (no offense to anyone guys):

1. At one of the places I worked at, we fought tooth and nail to convince the structural engineers that "micropiles" may actually carry load and may be classified as structural elements rather than "toothpicks" as someone put it. Even after performing load tests, somehow this tiny section could not be easily conceived as capable of carrying the intended load and the vote would be to go for the bulkier section.

2. When to use a Mat vs a Spread Fotoing. The tendency is to go for a Mat foundation regardless what the ground conditions are and definitely regardless of what the geotech fellow says. In the building I live in, we have a 4ft thick mat foundation and the bearing layer is limestone rock (RQD > 60% and >90% recovery). Simple 5' x 5' footings would have been ample.

and so on....

 

[SmokeyBear]

A very late reply; but...I do not understand how you can:- calculate settlements without some knowledge of stress history
assess overburden pressure without densities,
assess c' and phi' without strength testing,
classify soils (to the extent that different people logging the soils, call them the same thing) without Atterbergs.
We congratulate ourelves on the advances in numerical modelling, mathematical precision and coloured outputs, but use vague empirical relationships (modulus multiplier ranges from 150 to 500 times UCS, etc)
I don't believe in overtesting, but I suspect we should be able to improve on the rubbish we often get. Regular SPT's of between zero and 2 for many metres of "very soft" , "soft", "soft to firm" marine clays, often with such helpful comments as "increasing shell content". Personally, I find it difficult to design a basement retention system for 4 levels of basements with such data.
Similar information is also not uncommon for settlement assessment of large (40m dia. + storage tanks.
The comment we often hear from owners/clients is "why do you want more information? We gave you the geotech report"
I think the amount and quality of testing has really diminished over recent (15 years) to the detriment of design and overall cost of projects.

 

[dgillette]

In many or maybe most cases, phi' can be estimated with sufficient accuracy from collective experience with materials of a given type (in the form of tables in various manuals and text books). (Rarely does it matter much whether it is 36 degrees or 38 degrees.) c' is harder to nail down, and in many cases, it is reasonable to neglect it (provided you use a secant friction angle rather than a tangent friction angle). Undrained shear strength sometimes called "c" (NOT c') and c for partially saturated materials cannot be estimated so simply with any credibility.

Regards,
DRG

 

[SmokeyBear]

We do a lot of work in soft clays and over the past 25 years a large testing data base has been collected. It suggests Su ranges from around 10 to 30kPa, phi' between 20 to 28 deg and c' between 1 and 5 kPa.
While I agree with dgilllete et. al. that for "stronger" soils, the difference between phi=36 or 38 deg is unlikely to be critical, a couple of degrees or a couple of kPa does have a very significant in soft clays. Focussed testing helps decision making and reduces the multiple "what-if" scenario I often see - running upper and lower bound analyses, without any real feel for what is likely/credible.
In fairness, I neglected to mention in-situ testing in my previous post. We see a lot of this now, CPTu, Dilatometer and shear wave. BUT; the interpretations for realistic parameters are numerous and some are contentious, to say the least. The real benefit seems to be the ability to profile areas at lower cost than drilling many boreholes. However, some bores and some laboratory tests should/must be done to aid in correlating the in-situ tests.
I guess what I am really trying to say is that soil mechanics is still a bit of a black art and rather than look up a data base of some form, prod it and poke it a bit. Your dirt may not be the same as that in the data base.

Cheers

 

[doolittle2]

doolittle2

In the UK certainly on contracts I have been involved with there has been not only a reduction in the volume of earthworks testing carried out but a reduction in the range of tests.
Add to this the fact that most testing is carried out by labs employed by and for the contractor a lot of them quite small and dependent on contractors for their cash flow and you end up with a strong feeling (shared by many of my colleagues) that expediancy is likely to take precedence over integrity.
I have no axe to grind with the testers, we all are to a degree prisoners of cash flow and as I see it the answer is simple and is in the clients hands, instead of costing testing through the contractor cost it directly and make the testers answer to the client base.