Nuclear Gauge Testing on Tailings Material

[littlegrego]

Hey,

Im doing some compaction control testing on bauxite tailings material. I keep getting weird inconsistent results when i use the nuclear gauge, but when i do a sand replacement it is fine.
Can anyone tell me whats happening? they are compacting the hell out of it.

** Nuclear gauge IS calibrated externally & Internally **


Any help would be appreciated. I cant get my head around this one..

Thankyou

Andrew

 

[Ron]

Bauxite is the ore for aluminum. Any metalic material will cause variation in the results of a nuclear density gauge because the metalic material will either randomly bounce or absorb the radiation. Keep in mind the way a nuclear density gauge works....it emits radiation at a set distance from the counter. The number of ions counted compared to those emitted is related to the density of the soil. Any metalic material in the path of the radiation will cause unpredictable variation in the counts, thus variation in the results.

Sand cone density is more appropriate and more accurate for this type of testing.

 

[littlegrego]

Ron,

I Have tested the bauxite before for use on haul roads and etc..the densitys came up then around 90% of the time. Would you think because this material i am testing now ( Tailings ) would have a higher metalic property? or should the bauxite have higher Metalic Property?

Now i'll have to go out and do a bazillion sandies..

Cheers,

Andrew

 

[Ron]

The tailings would have more unpredictable properties than the ore. I don't know what types of contaminants might be in the tailings that are in smaller quantities than in the ore, but there is a good chance of concentration of some constituents within the tailings. Have it tested chemically to see all the stuff that's in there as compared to the ore and their relative quantities. That will probably help you figure out the variability.

 

[Hoaokapohaku]

How do your nuc tests on average compare with the sand cone results on average? The strength of nuclear gauge testing is that even though an individual test may vary from a theoretical "correct" value more than a sand cone test, the nuclear density results can more accurately represent the whole of the material represented because it is practical to take many more tests using the nuclear method rather than using the sand cone method.

The phenomenon that Ron mentions regarding deflection of radiation occurs with non-metallic materials as well. In fact, though we measure the overall density of a sample, a sample contains materials of varying densities and the shape and orientation of the various particles can affect an individual test result. I find that when the overall average density is marginal when compared with a modified proctor or Gmm (for asphalt), the variability of the nuclear gauge results as measured by the standard deviation is greater, often much greater, than material compacted on average a few points above the minimum. However, as long as you have statistically valid sample sizes, the average difference between nuclear results and sand cone results remains very close to the same regardless of whether the material is marginally or well compacted and regardless of the increased standard deviation of the nuc results in marginally compacted materials.

I would be very surprised if the above did not hold true in materials containing metallic metals as well. Even if metallic particles are more likely to deflect radiation, that would result in a greater amount of variation in nuclear test results, but should not invalidate them as long as you have adequate sample size. In addition, we think of "metallic" materials as reflective based on our experience, but that is a property of a material exposed to visible light. Gamma emissions from a cesium 137 source is another matter.

I suppose all of the above goes out the window if the material being tested contains a gamma-emitting element that will be measured by the densometer.