Density Tests 1

[belzer]

I want to verify dry density and wet density numbers on a Troxler 3440.

I know how to verify dry density---proctor x % of compaction=
dry density number.

How do I verify the wet density number?

Thanks

 

[Drumchaser]

Hey woofar

I will not dispute the extensive research program. Since we all know that research programs are what make the world spin.
I would think Troxler has done some research as well.

Having gone to the Troxler web site (you made me do it)looking up the Troxler 3440, going to the Advanced Operations tab, is the language relating to "oven dried" moisture correction procedures, and pertinent formulae.



As I have stated it should be routine to correlate if moistures vary (%O.D. vs. %M reading from gauge) then corrections are allowed to be dialed in (this model appears to do it for you if the O.D. is input correctly)
Corrections are very rare/uncommon on soils, for us.

As the "manufacturer" states there are materials that will cause false %M in the gauge readings. For us it is usually recycled ACP mixed in with base materials (your term bitumen?)

Rarely have I had to use the two knobs for “moisture correction” on the gauge. But our spec.s do allow it under Test-Method xyz.

 

[fattdad]

Thought #1: If you want to calibrate your nuke, do a side-by-side sand cone test, as referenced above. That'll show you the wet/dry density of the in-place soil.

Thought #2: If you want to determine the "percent compaction" get a bulk sample at the location of your density test. (Wow! That'd be a lot of proctor tests, eh?)

Thought #3: Many field technicians use their nuke gauge like a $5,000.00 rebar.

There is a HUGE problem with folks assigning a Proctor value that confirms their "gut" instinct, i.e., that the soil is dense enough it just must have 95 percent relative compaction. This is particularly a problem when the soil is below optimum (i.e., with dry strength).

I'm not saying that there should be a proctor for each denstiy test, but for the case of construction failures, the first thing I look for is whether the quality control program included confirmation proctors or one-point confirmations that the proctor that was used was appropriate.

End of rant.

f-d

¡papá gordo ain’t no madre flaca!

 

[woofar]

I can't help myself I just have to keep responding even though we're probably way of track from the original thrust of this thread.

To drumchaser,

We are basically coming from the same line, a few technical differences here and there, but the most important part of testing, as you stated, is the quality of the person doing the assessment.

Your mention of recycled ACP brings up an interesting situation that we encountered last year. We were testing some recycled pavement that was a mixture of bitumen (sprayed bitumen not ACP) and varying ridge gravels. Initially we were testing it with one for one sand replacement testing but the results came out so consistent that we ended up calibrating the nuke to the material and using an assigned proctor value. According to the rules that isn't possible (supposedly).

Assigned values in our part of the world require six initial proctor tests and then updating the value with a new proctor test every third lot(5-10 tests per lot). We were able to do this despite the fact that there were visual(color) variations in the material, although, the material was sourced from the same quarry.


To Fattdad,

We have a similar problem here, although not with lab staff but with construction staff. In Australia, pavements have to pass a density test as well as a proof roll. A proof roll consists of running a fully loaded water truck slowly over each part of the pavement while an inspector walks alongside and looks for any little bit of deflection under the trucks wheels.

A poorly compacted DRY pavement will pass the proof roll but fail the density testing. A well compacted WET pavement will pass the density testing but fail the proof roll. Trying to explain the difference between hardness (dry strength) and density is extremely hard. Most people in the construction side of things just can't grasp the concept.

Also your single point proctor's take me back. That practice got banned here back in the early 90's, unjustly so in my opinion, these days it is either 1 proctor per density test or the aforementioned assigned value. The proctor samples are taken by expanding the field density hole. (Yep that is a lot of proctors, eh!)

Cheers
Michael

 

[Drumchaser]

woofar – you make some good points

It is apparent your field and mine are very different. Evidently you are on a different continent. Quite sure your soils have different properties than the ones we have here.

We sample the (soils) source at varying depths and locations (and colors), and have a full blown multi-point proctor, L.L., P.I. (linear bar if appl.) etc. Our spec.s require a minimum of 95% - 98% of the Da depending on the P.I. of the materials. In this area the native soils are mongrel (that is to say they may change 20 FT away.
There is no way to run a proctor and series of lab tests for all the differing soils we will encounter. (Projects with 400K to 1 million CY)

“A well compacted WET pavement will pass the density testing but fail the proof roll.”

Sounds like you were/are working with sand (low P.I. matl.s)



----------------------------------------------------------------------------------------------------------------------fatdad

1)We have a test method for and do occasionally use the sand cone.
2)I would agree there are technicians that have no business operating a gauge.

“confirmation proctors or one-point confirmations that the proctor that was used was appropriate”

3)Your confirmations / expectations far exceed that of our normal industry standards.

 

[darthsoilsguy2]

woven geotextile is crap! geogrid works better!

Yarrr!!!!

Threadjackin Pirate

 

[woofar]

OK, I'll bite!

You're not only off topic and out of line, your contention is fundamentally flawed. You are
comparing two great products that are used for totally different applications. Your statement
is the equivalent of saying, concrete is better than asphalt.

The stupidity of such comparisons is mind boggling.

Go Away!!!

 

[mb27]

I guess I'm not fully versed in the area of one-point proctors. Are they only performed in the area of a nuke test? IE, is it the same as doing a sand cone to verify a nuke gauge, only simpler and faster? And how close or far away do the numbers have to be to verify or negate your gauge's numbers?

 

[woofar]

Hi mb27,

The proctor test determines the Maximum Lab Dry Density(MLDD).

On a project job you can develop a library of proctor curves for each different material type. Then, when you do a Field Density test in a certain material, you can knock out a single proctor point at estimated optimum and then match the single point with the correct curve from your library. This method also requires a visual comparison, thus, the jars of material that Drumchaser mentioned earlier in this thread.

How you go about the determination of the MLDD has no bearing on the comparison between a sand cone and a nuke.

Reading back on what I have just written, I realise that my point may not be totally clear, if that is the case then please post back.

Cheers
Michael

 

[mb27]

I get comparing it to the proctor library, I just wasn't sure if it was intended to be another way to verify the nuke. From what you're saying, it's more of a guide to which proctor to use and not an indicator of compaction.

 

[dirtsqueezer]

Woofar,

Perhaps some direction on where to find your info? But I bet I'd still think you're crazy. There is a procedure for testing trench backfill that involves taking a standard in the ditch, but besides that I know no 'adjustment' for site-specific testing. The test gives the same numbers as a sand cone; how do adjust that? I'm just speechless. Trust your gauge. That's all I can say.


-d

 

[woofar]

Hi Dirtsqueezer, I will address your post point by point



.....Perhaps some direction on where to find your info?

Did the link I provided in my earlier post not work for you? If that is the case, then please let me Know.



.....But I bet I'd still think you're crazy.

Yes, my wife and my friends would tend to agree with you there, but only as a general observation, never when discussing technical matters.



.....There is a procedure for testing trench backfill that involves taking a standard in the ditch, but besides that I know no 'adjustment' for site-specific testing.

Yes, you do need to do a new standard count when testing in a trench. You have to take a new standard count whenever the conditions for a changed background radiation exist.



.....The test gives the same numbers as a sand cone how do adjust that?

I have done many thousands of such correlations and the only place that the numbers match is in the wet density determination. In the moisture determination there is always variability. The reasons for this variability are many and varied and you can find a full explanation of this at the aforementioned link.


.....I'm just speechless.

Can't help you there.



.....Trust your gauge. That's all I can say.

I agree totally, the gauge is very trustworthy, however, like any tool, it does have it's limitations and to use it correctly you must fully understand those limitations.


Cheers
Michael

 

[WHEngineer]

Just a note pertaining to the question of moisture correction. Most nuclear density gauges measure water content by counting "slow neutrons" which lose energy through interaction with water molecules. The density testing is only concerned with free moisture (which can be burned off in oven drying). In areas where bound water is present in the matrix, the nuclear desity gauge will give incorrect moisture values. We have a large percentage of these materials in the SE USA, which include minerals such as mica and montmorillinite clays. In our area, the wet density is generally measured with a nuke gauge and we burn off moisture samples separately. Troxler has a good application bulletin on its website that describes this problem and methods to correct moisture readings.

 

[Ron]

I would fire a technician or engineer (and have done so) for "adjusting" numbers, for any reason other than a valid calibration factor. That is the type of activity that makes it difficult for legitimate testing laboratories to compete against such practices and demeans the entire profession.

Properly obtained and validated data are extremely important to the evaluation process. We can deal with correct data, whatever its course. We cannot evaluate incorrect data and the consequences of such can be dire.