GAMMA RAY BASICS
Many elements are naturally radioactive as a result of basic particle physics. Gamma ray logs measures the number of natural gamma rays emitted by the rocks surrounding the tool. This is often proportional to the amount of shale in the rocks, but there are other causes of gamma radiation. The spectral gamma ray log breaks up the total gamma ray response into three components, namely those due to potassium, thorium, and uranium. These measurements are used to distinguish the mineralogy in a shale or other radioactive minerals.
The log can be run in air or mud filled open holes, and also in cased holes, although the response is attenuated by the cement and pipe thickness.
In the early days of the logging industry, gamma ray flux was recorded in micrograms Radium equivalent per ton (ug-Ra equiv / ton) prior to about 1960. After that time, logs were calibrated in API units based on known radiation levels of artificial formations in test pits located in Houston. The usual scale for old style logs was 0 to 10 ug Ra and 0 to 100, 0 to 120, or 0 to 150 API units for newer logs. There is an exact conversion between ug-Ra and API units, but since the old logging tools were rarely calibrated, this conversion is seldom useful. The pragmatic solution is to multiply ug-Ra by 10 to obtain an approximate API units scale.
Radiation is naturally erratic. A stationary detector facing a given gamma ray flux will not see a constant stream of gamma rays. To obtain a reliable count rate, measuring instruments record the total number of emissions over a period of time, known as the time constant. For most gamma ray tools, the time constant is 1 or 2 seconds to obtain a smooth log curve. The differences in count rates between one time constant and another are called statistical variations.
An empirical relationship between potassium content and gamma ray API units is reproduced below for the standard gamma ray logging conditions of 8" borehole, 10 lb/gal mud and 3 5/8" scintillation NaI detector type tool. This relationship was originally developed by the author while calibrating gamma ray log response to potash content of potash (sylvite and carnallite) beds in 1963. For other borehole environments refer to appropriate borehole correction charts.
The flattening effect at high count rates is due to the dead time of the detector system. Dead time is the time it takes to transmit the recorded pulse to the surface. For other tool types, with different detectors and dead times, the relationship must be found by calibration. Newer tools (post 1980) have a linear response up to 1000 API units.
Special purpose gamma ray tools, such as those used by USGS
in mineral investigations, are not calibrated to oil field
standards. Conversion to oil field or mineral values will require
calibration on a project-by-project basis.
But detectors age, and tool sensitivity varies, so A is not constant over time and we need a secondary calibrator, namely a jig with a near-constant GR source. Still at the test pit site and immediately after finding the sensitivity constant A, we place the jig a fixed distance from the tool and note how many cps it adds to the local background radiation. Since we know A for this tool at this moment, we can determine the number of API units that the jig represents at that distance from the tool. Suppose this jig adds 200 API Units to the background while at the test pit. The equation is GRapi = 200 + BKGapi. BKGapi is the GR background in API units. However BKGapi is unknown, but could be estimate from BKGapi = A * BKGcps, or any other arbitrary value.
This is not a great method because we don't know the
background radiation level in API units (only in CPS). So the
process in the field is iterative and imprecise.
For a two point calibration, we determine the
difference in count rates caused by placing the jig at two known
distances from the tool.
SPECTRAL GAMMA RAY LOGS
Log scales may vary but uranium and thorium are usually
scalle in parts per million (ppm) and potassium in percent.
Curve names may also vary but POTA, URAN, and THOR are common.
If uranium is known in ppm, total gamma ray can be corrected for
Gamma rays emitted by the rocks rarely reach the detector directly. Instead, they are scattered and lose energy through three possible interactions with the formation; the photoelectric effect, Compton scattering, and pair production. Because of these interactions and the response of the sodium iodide scintillation detector, the spectra are degraded to the rather “smeared” spectra shown above.
The high-energy part of the detected spectrum is divided into three energy windows, W1, W2, and W3; each covering a characteristic peak of the three radioactivity series. Knowing the response of the tool and the number of counts in each window, it is possible to determine the amounts of thorium 232, uranium 238, and potassium 40 in the formation. There are relatively few counts in the high-energy range where peak discrimination is best; therefore, measurements are subject to large statistical variations, even at low logging speeds.
By including a contribution from the high-count rate, low-energy part of the spectrum (Windows W4 and W5), these high statistical variations in the high-energy windows can be reduced by a factor of 1.5 to 2. The statistics are further reduced by another factor of 1.5 to 2 by using a filtering technique that compares the counts at a particular depth with the previous values in such a way that spurious changes are eliminated while the effects of formation changes are retained.
Gamma Ray Log (GR)
Spectral Gamma Ray Log (NGT)
Page Views ---- Since 01 Jan 2015
Copyright 1978 - 2018 E. R. Crain, P.Eng. All Rights Reserved