Field Density Testing 9

[Mazzman]

Is it standard practice in your area (where are you?) to conduct periodic "backup" tests using sand cone (ASTM D1556), drive cylinder (ASTM D2937), or other methods to check the accuracy of tests conducted using a nuclear density device (ASTM D2922)?

Do you know of any authoritative published sources that recommend or specify that such "backup" tests be conducted?

 

[jdonville]

Mazzman,

I can honestly say that I have never ever had to perform backup density tests, but that I have taken grab samples from the compacted material to perform moisture content checks (usually for clayey soils).

Jeff


Jeffrey T. Donville, PE
TTL Associates, Inc.

http://www.ttlassoc.com

 

[RCEJD]

If I am understanding the question, there is extensive literature including the US Army Corp and many Cities and Counties that specifies custom and practice is that you should perform 1 sand cone for every 10 nuclear density tests. In So Cal this is standard in most areas, and it is a good practice.

When the sand cones are performed, we usually burn the samples off in the field (you could bring the samples back to test the moisture but this would delay results). We perform projects regularly moving 100,000 to over a million cubic yards and this is standard.

Sand cones with Nuc gages is a good practice, and although the technicians and contractors resist performing the checks because of time, it should be done.

Hope this helps.

 

[BigH]

I agree with RECJD that this is good practice - besides, at the beginning of a large job, you might even want to calibrate the nuke to the sand cone and laboratory moisture content. Too many years ago, we would use the nuke but if the contractor questioned it we would use the sand cone or Washington Densometer - then he would "shut up and accept." I also used to find that the nuke's moisture content (on our material) was too wet - that is the lab moisture was only 70% of the nuke's moisture. Now, this was the time of reading off charts and I am sure that things have improved tremendously, but, the feel of the sand cone vs the "black box" of the nuke - does make good sense to correlate from time to time. Good post RECJD!

 

[RCEJD]

BigH ... you too. We constantly run into issues of certain minerals in rocks throwing off our nuke gages. I should have mentioned at the beginning of the job testing because it puts any of the typical contractor arguements to bed early.

 

[Mazzman]

Thanks for replies. My experience has been that it is good practice to conduct correlation tests with the nuclear gauge. I'm familiar with the USACE requirements for such, but rarely see it specified in commercial projects. I'm told that the nuclear method ASTM doesn't suggest or state that backup tests be performed (I'm going to check it and other ASTM requirements).

Regarding certain minerals in soil: when I worked in North Carolina (I'm in KY, now) the nuclear gauge was never used in the residual granite micaceous soils in parts of the state. The high mica content in the soils made for some very strange gauge readings.

 

[GeoPaveTraffic]

Backup tests are almost never used in the Missouri and Illinois areas where I work.

 

[techmaximus]

Mazzman writes, “The high mica content in the soils made for some very strange gauge readings.”

Strange readings? How so?


Techmaximus

 

[Mazzman]

To: techmaximus

It's been quite a while since I worked there, but as I recall the gauge would give erroneous moisture content values due to something in the mineralogy of the mica. I know that it was common practice to conduct field density tests using a method other than a nuclear density gauge.

 

[Rjeffery]

Mica has a low molecular weight and therefore absorbs the neutrons much like water does. This causes a false (high) moisture content reading...

 

[techmaximus]

Mica is not a neutron absorber. And even if mica were a neutron absorber it would cause the gauge to give a false LOW moisture reading.

Elements like boron, chlorine, or cadmium absorb, or capture, neutrons because their nuclei have large cross sections. The absorption of neutrons causes LOW moisture readings because the gauge measures thermalized neutrons and if neutrons are absorbed, or captured, they don’t get measured.

Mica gives a false HIGH moisture reading because like gypsum, coal, fly ash, organic material, and phosphates it is hydrogen rich. The gauge measures neutrons thermalized, or slowed, by collisions with hydrogen to determine moisture content. So the higher the hydrogen content in a material the higher the moisture reading.


Techmaximus

 

[Rjeffery]

techmaximus,

It is easier to explain that low density atoms absorb neutrons. Note that Chlorine, Bromine, Cadmium are all low density atoms. Water is used as shielding in most (If not all) nuclear reactors in the US because if the 2 Hydrogen atoms bonded to the Oxygen atom. Yes your explanation of the thermalization is correct, essentially slowing down the neutron and therefore imparting less energy to the collector (sensor).

I have to explain this almost on a daily basis to "the public". It is easier to explain that, "The neutrons go out and bounce around. Some are lost and some are detected. The difference between the expected return and a return with 0=zero water tells the gage how much water is present." Therefore loss of neutrons = high(er) moisture. Start talking about high order calculus, fast and slow neutrons and well...

I know it is dumbed down but then "They" stay away from me and my gage because they don't want any of my neutrons to get lost in them, possably because I remind them that they are over 90% water and could effect my gage reading… He he he

 

[techmaximus]

Troxler: “The gauge measures moisture by determining the amount of hydrogen present in the material. If the material contains compounds that are high in hydrogen forms other than water an inaccurate reading may result.”

A licensed PE nuclear engineer imparted the following information to me: It’s true enough that low density atoms absorb neutrons; however, water, specifically light highly purified water, is used in US nuclear reactors to transfer energy from the fuel to the turbines. Water is also used in US nuclear reactors as a shield but not because “2 Hydrogen atoms bonded to the Oxygen atom” or because water absorbs neutrons. While water does absorb neutrons and is thus a good shield most neutrons are absorbed by boric acid that is added to the water specifically for that purpose. Water with boric acid makes a better shield and a better conductor of energy than say a curtain of lead plus water is a lot cheaper and easier to maintain.

Your contention that, “Therefore loss of neutrons = high(er) moisture.” is totally incorrect. The gauge takes into account the tiny fraction of neutrons absorbed by H2O. This is the reason that the absorption of neutrons by other elements like boron, chlorine, or cadmium cause erroneous LOW readings: not enough neutrons are being thermalized and thus counted by the detectors.

The gauge determines the H2O content of the material being tested NOT by calculating released neutrons vs returning neutrons but by calculating released neutrons vs returning thermalized (slowed by collisions with hydrogen) neutrons. The gauge takes into account the tiny fraction of neutrons lost by absorption to H2O. The detectors in the gauge are insensitive to returning neutrons that are not thermalized (been slowed by colliding with hydrogen).

Elements that act as neutron absorbers cause the gauge to give false LOW moisture readings.

Elements like mica give false HIGH moisture readings because they are hydrogen rich causing too many thermalized neutrons to reach the detectors. The moisture content of a material is directly proportional to the number of thermalized neutrons reaching the detectors.


Techmaximus

 

[techmaximus]

RCEJD writes, “We constantly run into issues of certain minerals in rocks throwing off our nuke gages.”

Elements in the material being tested may cause false high or low moisture content readings but these same elements do not affect the wet density readings. If the moisture content reading is high or low then, of course, the dry density and percent compaction readings will be low or high; however, a moisture correction factor can be used to make the moisture reading correct and thus the dry density, and percent compaction readings accurate. I have yet to meet the fill material that I couldn’t get an accurate test out of.

Correlation tests? That depends on the following:
Is the material being placed in lifts of required thickness?
Is the material being placed and compacted with appropriate equipment?
Is the material at or near optimum moisture content?
Do we have an accurate proctor and have we run check points to verify we are using the correct proctor?
Is the fill material free of deleterious material (organics, oversized rocks, etc)?
Are we probing the material with a small diameter steel rod to aid in the detection of upper low consistency material as it is being placed and compacted and are we finding the material to be hard and stiff?
Have we observed the hauling vehicles moving over the fill material and have we observed that no deflection, pumping, of the material is occurring?

If the answer to these questions is YES, then no way am I running any silly correlation tests. Pass or fail, my running a test with a nuke gauge, a sand cone, or a drive cylinder is only going to confirm something I already know.

As far as “contractor arguments”, these can be stopped with a tactful reminder of who controls the pencil that is writing up the test results.


Techmaximus

 

[Mazzman]

Well said, Techmaximus. I have always believed that there is no substitute for the eyes of a trained soils technician inspecting fill placement and compaction. The density test (regardless of method) becomes merely confirmation of what the eyes see. But, our files need to have documentation of the test results, because that is what attorneys will focus on if there is ever an issue. That, too, is the source of my origianl question about backup tests for the nuclear gauge. Maybe I'm old school, but I feel better seeing direct measurement of weight and volume (ie: sand cone or drive tube methods) to document density as confirmation of the less direct measurement of density and MC with a nuclear gauge. There is a mixed response in the posts here, but after reading them, I am beginning to believe that no backup tests with the nuclear gauge meets the "standard of care" for geotechnical professionals.

 

[CIVGEOT]

With all due respect to some of the discussions, the postings on this thread that appear to discount a checks and baalnce system that are generally followed by prudent engineers is of bigger concern than the belief that the standard of care (better stated custom and practice) is no backup tests.

After 20 plus years of geotechnical engineering I understand that in some areas of the country they may not experience the issues that are experienced in So Cal. However, I will tell you that if you perform in the private sector in this part of the country and do not perform the necessary (silly ?) checks and balance, your projects will not be approved, and/or an attorney will hang you on the stand ... and I say that from a position of experience as an experienced expert witness who practices in this jurisdiction on small to large residential tract developments.

I have no doubt, and I agree with the punch list descrived above should ALWAYS be in the mind of any experienced tech in the field, and should be a common sense check list. But to some how infer that the gauge is always right ... or that you can always rely on the gauge is flawed in many respects.

The eyes of a trained tech are valuable, but how do you quantify and qualify their observations ... I had a grandma once that drove for 60 plus years, experience MAY OR MAY NOT always be the correct experience.

This post intends no disrespect but only to point out a circular logic that sometimes occurs in the engineering field when certain procedures are not followed.

Good luck, this has been an interesting post.

 

[Mazzman]

I did not intend to discount the value and necessity of documented test results. My point was that the FIRST indication that structural fill is being adequately compacted is to watch the equipment work on it - that helps determine when it is time to take tests. I want my project file to be full of test results that indicate the structural fill was compacted in accordance with project specifications, retests were conducted in areas showing less-than-specified compaction, and preferably, checks conducted on the nuclear gauge.

My original post was made because we have been debating this in our office. My 20+ years of experience at Firm A suggests "thou shalt conduct backup tests". Now, at Firm B, they tell me they don't think such backup is necessary. Because contractual "Standard of Care" language is typically locality-specific, I need to gauge the usual practice locally with regard to this topic.

 

[BigH]

Just one point on "Standard of Care" - there is the sort of "mandated" SOC but you as an engineer also should have your own SOC - if you think that you need more than the minimum, you should do it. I started on nukes back in 1975. They were primative; I haven't really used one personally since the 1980s and I am sure that they have come a long way - but, . . . just as you would do hand-checks on computer software, etc., you should be aware that the "black box" does have limitations and you should be able to satisfy yourself that you are comfortable as a PE with the results and implications thereof. Many jobs - it doesn't really matter if you get 90% or 95% relative compaction from an engineering point of view - but legally you might have a problem on a contractual basis. But the bigger more important jobs - it seems like I would do some checks using other equipment. Just like I would use field vanes to check SPTs or even piezocone data.
- hope you enjoyed Spurs win!

 

[techmaximus]

Mazzman wrote, "I need to gauge the usual practice locally with regard to this topic."

Good one, a nice play on words. It’s a good thing you are “checking” this topic out.

If you want checks in the file, have the techs do checks. Running a drive tube in the footprint of a nuke gauge test is not only reassuring but educational. Cooking out a few moisture contents is also beneficial to get an idea if offsets are needed. However, I have not found any ASTM requirement for backup tests for nuke gauge tests. Each of the in-place density tests (sand cone, drive tube, nuke gauge, etc) is believed to be a stand alone test. The most important and most widely neglected check is the proctor check point. I’d like to see one proctor check point for every 3 density tests or at least one a day.

BigH writes, “you should be aware that the "black box" does have limitations and you should be able to satisfy yourself that you are comfortable as a PE with the results and implications thereof.”

Every piece of equipment and each test method has its own set of limitations. In my book the satisfaction with regards to results comes before the gauge (or sand cone or drive tube or whatever) leaves the shop. If the tech has been properly trained as to the limitations of the “black box” (or whatever piece of equipment he has been issued) and is sufficiently experienced, off he goes because I am positive I will get valid results. If I wasn’t positive I’d get valid results, well some other course of action would be taken.




Techmaximus

 

[RCEJD]

Good comments from CIVGEOT and BIGH ...