Moisture content Clay 3

[DOUGDOUGHNUT]

I am an Supervisor Engineer where the Contractor has placed the road sub-base onto the sub-grade, it was proof rolled and passed. Then down came the rain the since then the road has started moving under a load, wittnessed when the construction traffic drove on it.

Would it be sensible to expose the subgrade to the sun to dry out and then re-compact the clay. I am trying to avoid the additional cost of stabilisation.

The contractor has now asked me what the maximum and minimum % that would be acceptable for the moisture to vary from the OMC.

My understanding of the optimum moisture content in clays was that it is the moisture content that will produce the best compaction.

Could there be any problems with using the OMC this highly reactive clay?

Thanks

 

[BigH]

First, please provide some details on the clayey subgrade - was it an expansive clay (you did say reactive) - what were the PI values? What were the specs on compaction? What type of roller did you use? Was the layer crowned so as to shed off the water? What was the subgrade design CBR - soaked or unsoaked?
When you have a subbase that gets saturated (water isn't draining) it creates many problems should heavy trucks travel on it - you are experiencing these now. Normally, for embankment construction you would compact 1 to 2% dry of optimum (although the specs will permit up to 2% wet). If this was an expansive clay, I would compact on the wet side - so that increase in water content would not create a severe swell. - but for normal clay (low activity) under normal circumstances, I would use -2% to +1% and err on dry side.

 

[Rjeffery]

In Virginia, the DOT's specification is 2% either side of Optimum Moisture content. That is OM * 1.02 or OM * 0.98 (not OM + 2 or OM - 2)

 

[BigH]

Rjeffery - isn't that a bit restrictive - say OMC is 6% as for crushed stone base; then you only allow range of 5.94 to 6.06%??? - or is this just for subgrade? Still, if you have a sand subgrade with OMC = 11%, then you are in a zone of 10.9 to 11.1% - don't see how a contractor can achieve such accuracy. I've never seen anything but using +2% or -2% or so.

 

[BigH]

sorry about math - that would be 5.88 to 6.12 if at 6% or 10.78 to 11.12% if at 11% - and who reports moisture contents to 1/100? Damn, take my math as 5x!!

 

[dmoler]

Just letting drying out in sun might not work too well. Should be scarified first, then dried, then recompacted. I suppose the density tests were performed with the nuke gauge, and that would mean the top 2 inche may be dried but not the full 12 inches (nuke takes moisture from top 2 inches or so).

The questions from post 2 need to be answered. However, if the water table is at or above the interface of the subgrade and base, compact all you want with whatever moisture you want and the same problems will happen again.

 

[techmaximus]

"(nuke takes moisture from top 2 inches or so)."

When using a nuke gauge, the moisture content is determined down to 6”. The moisture content of soil (or whatever material being tested) is determined by a source within the gauge emitting neutrons which are scattered or reflected back to a detector within the gauge. Neutrons penetrating the material are slowed (thermalized) due to their reaction with hydrogen (or other material) within the soil. The detector within the gauge is not sensitive to the fast neutrons, only the slowed (thermalized) neutrons. This is why the higher the moisture content of the material being tested, the shallower the depth of measurement. The higher the moisture content of the material, the less deep the neutrons have to travel to get thermalized and detected. If fill material were wet enough to only be tested down 2”, I suspect that testing would be unnecessary as visual observation would suffice. I mean that sub-grade would be as shaky as cafeteria jell-o.

SCET - Techmaximus

 

[iandig]

I disagree with the previous comment on the Nuke measuring the moisture to a full 6". I would strongly suggest that the moisture is only 'assessed' in the upper 50mm, not to full depth. This is based on guidance AND experience, and highlights the importance of understanding how the equipment works. Density is 'assessed' over full depth, moisture only within the upper 50mm or so. If in doubt, its quite easy to check. Prepare a calibration box, let the surface dry out, then take moisture samples through the layers at 25mm intervals. We did this to 'enlighten' an earthworks contractor that allowing just the surface to dry was not giving a true reflection on the condition of the material.

 

[BigH]

From past experience, I don't trust the moisture contents of nukes. I used to take samples at the probe locations and do lab tests to determine the moisture content - then for each type of soil that we were compacting (say Ontario MTC-Granular B or Granular A), I developed correlations between the nuke moisture and the lab moisture. I found out that the nuke was on the high side. Of course, it is machine and material dependent - but if you do a lot of testing of the same soil, I would establish such relationships - it might help when you know (or pretty sure) you have the compaction but the tests just don't measure quite up to it.

 

[techmaximus]

Only 2”, hmmm. I don’t know if I’d stick on 2”.

Troxler 3430 Operators Manual: “The depth of measurement, or depth at which 98% of the counted neutron pass before reaching the detector, is a function of moisture content.”

Depth (inches) = 11 – (0.17 X M), where: M – moisture in pcf

I rounded off the following figures from the Troxler 3430 Operators Manual. Material that has a moisture content of 6 pcf is being tested to a depth of 10”. Material that has a moisture content of 21 pcf is being tested to a depth of 8”. Material that has a moisture content of 32 pcf is being tested to a depth of 5”. According to Troxler, to get only a 2” test on moisture the material would have to have an off the chart moisture content of 55 pcf!

The higher the moisture content of the material being tested the shallower the depth of measurement. For instance, crusher-run that is 6% moisture is being tested way closer to 10” than 2”. Also, if this were not true why would they even make the gauge with a 12” probe on it? It would be useless to run compaction tests with a moisture assessment only to a 2” depth averaged with a density assessment to a 12” depth.


SCET - Techmaximus

 

[dmoler]

Sounds like some have not done field work lately. The first nuclear source is in the tip of the rod. That only measures the total (wet) density. The second is the source (different type of nuclear source that responds differently to water) located in the testing box a few inches from the rear. This reads moisture by deflecting and such as somewhat described in the manual. Be careful of manuals, use your eyes and common sense.

The machine picks up moisture from around the top area. Dont believe me? Run the test normal, then put your foot at the rear and run it again. The moisture increases dramatically, and then the dry density decreases (old trick I learned as a contractor in Texas against unsuspecting field techs, hehe). So, does that correlate well with your theory, techmaximus? BihH is right, but a bit more work to get a k factor, oven drying and all.

Testing is one thing, but sometimes insticts and observations are more important for quality control (I tell all my techs under me "did they roll it? was a compactor there? what type? did you think it was hard and compacted? are the contractors trying to pull a fast one? fail it when you think it fails, and I will always back you up, regardless of test results)

 

[techmaximus]

The nuke gauge picks up moisture from around the top area. OK, I never wrote that the gauge did not pick up moisture from around the top area. I quoted what Troxler wrote, that the higher the moisture content of the material being tested the shallower the depth of moisture measurement. Conversely, the lower the moisture content of the material being tested the deeper the depth of the moisture measurement. The depth to which the gauge measures MC is determined by the MC of the material.

BTW, the gauge detects hydrogen, any hydrogen, not just water. So materials like PVC or HDPE (materials used to make some types of boots), or mica (that’s the shiny silica minerals that occur in igneous and metamorphic rocks) produce false high MC readings. So the gauge is sensitive to material containing high concentrations of hydrogen that are placed close to the source and this will affect the depth of the moisture measurement during a test. But the moisture measurement during such a test will still be much deeper than 2”. Until you hit an MC of 50+pcf you are not going to touch anything as shallow as 2”.

Also, you could place a depleted uranium plate behind the gauge and reflect an endless steam of neutrons back to the helium-3 neutron detector within the gauge all day long and this would NOT get you a higher MC. The helium-3 neutron detector only detects neutrons that have been thermalized, slowed by collisions with hydrogen. The detector is not sensitive to the fast neutrons off of the americium-241:beryllium. Deflecting neutrons that have not been slowed by collisions with hydrogen will not get you a higher MC.


SCET - Techmaximus

 

[BigH]

dmoler - I agree with the "observational" method. I once had problems getting 98% MDD Modified with a crusher run and we pounded the hell out of it. One reason I went to checking out the real moisture contents. I also knew one very experienced geotech who putatively used to take a "broken" machine to site - take it out his car for all to see - then he spent an hour walking about the site - kicking the dirt (old "heel test"), watching the contractor work, etc. Pick up his nuke - back in the car and drove off.
and

 

[dmoler]

place a depleated uranium plate and reflect....(gasp)...

...Anyway, the old manual (10 years ago, last time I read it?) said 2 inches. The new manual I downloaded somewhat expands on the moisture depth reading. I would forget the equation, which seems inaccurate, and look at the curves on the next page and where that equation 'line' was developed and is located at within the graph.

For the model 3440, a 20.8 pcf moisture has total moisture depth to about 5 inches. But, the curves indicate that 85% of the moisture is read in the top 2 inches. The remaining 15% come from the next 3 inches, as the curve flattens out (to infinity?). The exact way that the curves were developed I am not sure about, but the curves seem to be in line with my overall field observations.

 

[dmoler]

Actually, about 62% in the top 2 inches, the other 40% in the next two. (typo, my bad)

 

[techmaximus]

Be careful of manuals and forget formulas when they don’t support your claims…(gasp)

Haven’t read manuals and kept up with technological advancements for 10 years…(gasp)

Maybe as part of your PDHs this year you should take Troxler’s (or another nuke gauge manufacturer’s) training class in order to refresh/update your knowledge base.

The graph in the 3440 Operators Manual clearly shows (to me at any rate) that the 20.8 pcf moisture has a total moisture depth to 7.5”. The manual states that a neutron from the americium-241:beryllium source must hit a hydrogen nuclei an average of 19 times in order to be thermalized, slowed enough, to be detected by the helium-3 neutron detector. The graph shows that if the material being tested (assessed) has an MC of 20.8 pcf that no more than 2% of the neutrons pass 7.5” of depth in order to be thermalized and subsequently detected. If 62% of the neutrons are thermalized in the top 2” and 38% of the neutrons are thermalized in the remainder of the material then it still takes 7.5” of material to get a 100% assessment. Also, the drier the material is the deeper that first 62% of the MC test is. Look at the next graph down for the MC of 5.95 pcf where 62% is at a depth of 3.8” and total depth of MC assessment is at 10”.

So just for fun I called the source, no pun intended, Troxler, and the product technical support person told me that I was reading and interpreting the information correctly. She confirmed that Troxler stands behind the equation, the graph, and that the gauge reads MC to an average depth of 8” based on an MC of 15 pcf and deeper if the material is drier. If the material in the top couple of inches is wetter or drier than the underlying material (like 4” or 6” of crusher-run of 6% OMC over a clay soil sub-grade of 23% OMC) then the gauge will assess some of the clay with the stone as far as MC goes. This why offsets are sometimes needed.

You wrote that 62% of the MC is determined in the top 2” and that the other 40% is in the next two inches. This adds up to 100% of the MC being assessed in the top 4”. First it was 2”, and then it was 5”, now it’s back to 4”. It turns out that to get an </=2” assessment the material would have to have an MC over 50 pcf. That's just not right and no amount of shaving the numbers will get you to 2". Why can’t you just be big enough to say oops, I got that wrong, it’s not just to 2”, I learned something new today?


SCET - Techmaximus

 

[Rjeffery]

TechMaximus, Not good sportsmanship to rub people's noses in their mistakes. Still ASTM (I beleive) requires that a sandcone test method be run to confirm every 'n' test done by the nuclear method.

Big H, 2% either side of optimum is tight, I admit, for crushed stone. But for silts and clays as subgrade (not sub-base) where optimum can be up to 50%...

It is the contractor's job to place the material correctly at optimum conditions for the expected use and service life!

 

[dmoler]

No, I dont usually sit around and read the latest versions of the troxler manual. If I knew it was such a hot research topic, I would have paid more attention.

I did tests similar that iandig posted above for some extensive evaluation of landfill clay liners once to evaluate nuclear testing effectivness. Again, the top 2" dominate the moisture readings for the nuke gauge. I wish you the best with your method, Techmaximus, if it works well for you. I suppose I have my own curves that I use, and my buildings are still standing, so Im happy.

 

[BigH]

Rjeffery - nice point on mistakes; we all make them. Your point on high OMCs - still, if OMC is 50%, then 0.98 or 1.02 x would be +-1% of 50%. Damn hard to get a clayey soil in that narrow range. Personally, I don't worry so much about the range of OMC - because OMC is a function of compactive effort - OMC is less for modified than for standard; OMC is less for 15 ton vibratory roller than for modified;, etc.

 

[Ron]

Doesn't take much effort to verify the moisture content by oven drying. Nuke gages measure moisture in "backscatter" mode.

No substitute for observation and "reflection". If the material is pumping, it's too wet. As for range relative to OMC, it varies with the material. For clayey sands and clays, the closer to optimum, but slightly on the "wet" side, produces compaction with less difficulty. Keep moisture content within about 2 percent of optimum for this type of material. Sands and graded aggregate base materials are usually much less moisture sensitive and can be compacted at broader range of moisture contents.