Forensic Evaluation of Engineered Fills 1

[Ihatelawyers]

Reading ASTM E141 it seems to me that when evaluating existing fills the standard practices of our industry are falling short. E141 indicates that location of and evaluation of the units being examined (dry density of the engineered fill) should be performed in exactly the same way as done for the sample (i.e. nuc's or sand cones during construction). That is ... using SPT unit weights (or shelby, or whatever) to evaluate the relative compaction of an engineered fill is bunk. Am I reading E141 right? Seems like the degree of precision and accuracy needed for unit weight determination for geotechnical analysis would fall short of the accuracy and precision bar necessary for verification of an engineered fill.

 

[dgillette]

Not familiar with E141, but if there are those whom you hate involved, there is probably enough money involved* that it is reasonable to insist on careful test pits with high-quality sand cones and/or nukes (with trench correction and sand cones to verify them!), and maybe some hand-trimmed block samples. If it's too deep for test pits, then dry-cored samples may be your best bet for density, if it's the right kind of material. It might even make sense to use the procedures that Castro and Poulos used to use to track changes in density during drilling and sampling of liquefiable sand for their steady-state method.

SPT is nothing but an index test, like liquid limit or consumer price index. Plain old Shelby samples in engineered fill probably would probably be too disturbed for good density numbers.

*Inexperienced lawyer generally costs more than a top-shelf engineer with gray beard and a long list of projects and publications. How many lawyer hours does it take to equal the cost of two days of a small back hoe and operator, plus an engineer and a technician to do the testing, and a little bit of lab testing like density, gradations, water content, and such? Not many.

Q. What do you call it when a lawyer falls into the test pit?
A. Free backfill.

 

[Ihatelawyers]

Unfortunately, it is the 'top-shelf' engineer with the 'gray beard' and the long list of projects and publications that is using unit weights obtained from SPT's and shelbys to characterize the site as 70% compacted. Really, 70%???

I've looked exhaustively for studies that support or refute 'gray beards' methods but haven't had any luck. You have any ideas there?

(E141 is ASTM's Standard for "Acceptance of Evidence Based on the Results of Probability Sampling" ... Seems like our forensic/expert engineers should be held to some standard.)

Thanks for you input.

 

[dgillette]

Is he saying 70% of maximum, or 70% RD?

You would have to try pretty hard to place any kind of nonhydraulic fill at only 70% of maximum with motorized equipment!

 

[Ihatelawyers]

70% of ASTM D 1557 to be exact!!!

I'm with you. I've been watching earthwork for over 30 years. If any of my tech's came in and said 70% I'd tell them they were using the nuc upside down. You and I both know 70 is rediculously low, but finding a study or reference that states that... good luck.

 

[BigH]

You get that by dumping, light spreading and an elephant pass! 70% sounds more like relative density - he is implying that the SPT values of the fill are such that he can correlate the values to a chart of SPT vs Relative Density. Hunt (Geotechnical Engineering Investigation Manual) has discussions. Chart and a correlation by marcuson and Bieganoousky are attached. In the correlation, sigma is the effective overburden pressure.

An SPT N value would be in the order of 70% according to these sources.

I once saw a lab manager fired when the client asked him what the specific gravity of the project sand was - and he said 2.65, approximately. The client jumped up and down and screamed "How could it be - it is only about 1.56 g/cc (about 15.5 kN/m3). Of course, the client was mixing up specific gravity and maximum dry density. Perhaps your grey beard is having a "senior moment".

 

[Ron]

Hey BigH...I've had those!

As for correlation between SPT "N" value and the modified Proctor...not good. As for the Shelby tube...depends on the material whether the densities will be significantly disturbed. It would be similar to a drive sleeve method which is certainly valid in materials passing a No. 8 sieve or so.

Opining that compaction in "engineered" fill is at 70% of the modified Proctor is patently absurd. As BigH and others have noted, you can get 70% from just dumping and wetting.

As for determination of density, any direct method should prevail...sand cone, shelby tube or drive sleeve. The nuke would be a second tier choice, but not a bad one.

As for the gray beard...age doesn't necessarily beget knowledge or wisdom. Does he have 30 years of experience or 1 year of experience 30 times?

 

[Ihatelawyers]

I wish that were the case Big H (thanks for taking the time to post the reference, though, I appreciate that!). He doesn't correlate SPT to RD... he uses the SPT to obtain samples upon which he measures the unit weight (strike 1, right?) ... he then takes a bulk sample he's obtained from his boring (failing to consider that a 5' sample interval in the boring might get him a big enough sample, but that 5' interval reflects over 5 days of grading with potentially 3 (or more)different soil types... there's Strike 2). And then he compares his mod proctor determined from his bulk sample to the density he measured from the SPT (STRIKE 3!!!! BACK TO THE BENCH!). And then he makes the magnanimous statement ... "The fills were inadequately compacted." (Here's where he should be ejected from the game!)

My handle would have been, "I hate idiot engineers who should never have had a stamp and ought to be tossing pizzas for a living." But it wouldn't fit (and everyone hates lawyers - so I remain anonymous), and unfortunately there are times in my professional career that I haven't wondered if I just tossed a pizza!

 

[dgillette]

Can you tell us the back story, or would that put you too far out in the open?

 

[Ihatelawyers]

I don't think I'll 'expose' myself ... what I'm talking about is a pervasive practice in my geographic location and my profession. Someone calls into question the compaction quality of a deep fill(i.e. greater than 5' let's say)... with or without distress to overlying improvements. The 'expert' goes out, does borings, obtains unit weights from SPT's or shelbys (we're not a very refined community ), and proctor samples from the cuttings, and declares an inadequately compacted fill. I'm just trying to find references that would allow me to evaluate the accuracy of their methods. SPT/shelby unit weights are fine for ball-park and dialing in an estimate for geotechnical design ... but to use as a basis for acceptance of an engineered fill... In my opinion, they're too disturbed, too variable. Shelby maybe okay ... but you'd have to evaluate the degree of precision or discuss the limitations of your methods. I'd just love a reference that could tell me I'm right or wrong, or provide some indication as to level of precision associated with these methods.

 

[BigH]

I presume there are no records of the fill placement - so you end up either with an engineered fill - or a variably placed fill of "who knows what". I'd look at the investigation like I would any investigation - based on the facts as I discover. If I have fairly uniform materials and uniform SPTs, I would evaluate as per normal procedures and never back-track to % of MDD. Similarly, if I ran into fills of variable materials and variable SPT results, I'd look on it as a "crappy" site and design accordingly. You should be able to determine if the "fill in place" is adequate for your intended use.

Now if the investigation is for assessing the fill as placed against the project specification - that is different and I agree that his method is highly suspect - especially if one is in gravelly fills or fills with gravel sizes - the split spoon is inadequate for this due to the gravel size to opening aspect.

 

[Ihatelawyers]

Unfortunately, the investigation was for assessing the fill as placed against the project spec. Without a reference or a research project showing the variability in precision and accuracy is excessive ... it's just my humble opinion against his. Which just means we all lose.

Thanks for your input everyone. If any of you run across any references in your travels that discuss the variability of UW test data from boring sampling methods, please keep me in mind. Happy Holidays!

 

[theCorkster]

Back in the mid-80s there was some work performed by the Bureau of Reclamation in Denver in the Division of Research that as I recall involved placing and compacting sand at a target density into a large mold (say 5' by 5' by 10' high). I believe that SPTs were taken as well as sand cone tests. Unfortunately, I don't recall who was the principal investigator. However, I don't believe sufficient testing was performed to develop precision and bias between the methods to measure density.

I'd stand by the case that selecting material from cuttings and then comparing it to a small "undisturbed" sample from a specific interval is quite a leap, even from a geotechnical perspective. Unless the fill is a processed material without much gravel or uniform fine-grained soil, the comparison is based on a guess. If I was dealing with that testimony, I would definitely want test pits to visually confirm the viability of this approach. Photographs with physical property tests showing a high degree of variability might sway a judge/jury.

If the fill contains sand and gravel, it is possible the coarse material can break down as a result of drilling and laboratory processing, including compaction, resulting in an erroneously high laboratory maximum density.

I noted that you did not respond to BigH's comment regarding compaction records during construction. If that is the case then I warrant a higher degree of evaluation as dgilette described is appropriate since the fill's content is really not known.

 

[Ihatelawyers]

Thanks 'Corkster'! Sorry ... I forgot that part about the QC testing during mass grading. Yes, density tests and construction observation services were provided by an engineering firm during mass grading. But that has all been undone by an engineer with a drill rig, a scale, and an inability to evaluate or consider the limitations of his/her methods. (Our firm was not the one that provided the construction testing services. But we could be/would be in the same pickle if someone decided to hire this engineer to evaluate one of our fills, right?).

Thanks for your input. It's kind of funny, seems like we're all pretty much of the same opinion but undoing what has been done is difficult without the benefit of a publication. It just looks like an engineering stand-off to those in position to decide if there is or isn't an issue.

 

[GTeng]

I read the posts pretty quickly, so maybe I'm missing the technical details, but here is a slightly different take.
In this case it seems like a good lawyer is exactly what is needed. A contractor signs a contract to place and compact fill to the requirements of the contract, which include compaction tests by nuclear gauge or otherwise, but not usually SPTs. So it seems like a good lawyer could stop any legal action by others right there. Similarly, a testing agency is contracted to test by those same methods. The results of any other test are irrelevant.
I'm probably just stating the obvious. And, I suppose that if law were actually practiced to the requirements that it meet the rules of common sense, we wouldn't need any lawyers.

 

[Ihatelawyers]

"if law were actually practiced to the requirements that it meet the rules of common sense, we wouldn't need any lawyers."

Can I get that tatoo'd without a copyright infringement?

Your position that results of any other tests are irrelevant is in step with ASTM E 141 ... but unfortunately in my neck of the woods ... we haven't reached that level of refinement. It's whoever tests last, tests best (i.e. they have the most credibility) and validity of the methods don't enter into the equation.

Thanks for the input.

 

[oldestguy]

In my humble experience everything in the file of an expert is open for the opposition to look at. I'd insist that all of his field notes, the logs and samples be available for inspection and review. Ask for the names of the drillers and indicate they may be part of the witnesses to be called up. Any computer programs used have to be furnished.

This can be a pain in the butt for the expert, but why not go that route and look for holes in it.

Ask for his reference material. After all, this can't be just something he made up from thin air. What is the "standard of the industry" that he meets with his criteria for data to be used for such a conclusion? Then, show any calculations he makes. A good lawyer needs a little help on this, since they can make black lok white if they are smart.

All this stuff usually can be found with interrogatories if they have not already limited the listing of such witnesses.

On this subject, I have OK'd many a random dumped fill for use as a building site using SPT, but never tried to tie that into percent compaction. The words "loose, firm, dense, very dense, etc." as used by drillers don't have numbers attached as I recall.

On the other hand, I have been set back by a contractor or two when our guys on the job were asked to test "percent of Modified Proctor density" of nearby natural ground where we have OK'd the placement of foundations and the number was far below what we specked for the fill, like 82 percent for a clay as an example. Then we scrambled with testing unconfined compression, etc. on the fill as well. That brings up the question I sometimes ask "Why 95%?". Not many can explain that other than to say "Then we know we got it".

 

[BigH]

Oldestguy - why 95% indeed! Several years ago I posted a reference to a paper from Ground Engineering that talked about fills and the 95% Fixation.

 

[Ihatelawyers]

Wow ... I thought I was the only one. I've performed consol tests - several times - on soils remolded to 90 or 95 - and have seen very little or no significant difference (sometimes the 95 moved more - that either shows the limitations of the test or the limitations on my ability to remold a sample - eh?).

That's part of the overall issue I have with our industry's approach to forensic engineering. Brand x comes through post-construction and gets 89% compaction with their testing program in lieu of the 90% specified - and they mandate to remove it all and do it over again. I have not seen a forensic engineer (locally) temper their findings with documented precision and bias between samples or variations in post construction testing methods - For instance a sand cone in the bottom of a test pit is at best equal to a sand cone on a flat pad being compacted and more likely the sand cone is compromised to at least to a limited degree ... and depending on the care of the one performing the test ... possibly to a great degree. And then ... I have never seen locally where a forensic engineer evaluates the potential strain associated with their 89%.

Maybe I need to move??? Anyway ... thanks everyone for your input. Maybe my 'counter part' is a reader of posts and will begin to rethink his approach.

 

[moe333]

One of my professors stated that in areas of high seismicity some granular saturated fills at 90% could be subject to liquefaction wheras granular saturated fills at 95% would not. I think there is some merit to this statement although I have not thought about it too much.