Page 349 - Standard Handbook Petroleum Natural Gas Engineering VOLUME2

916 Reservoir Engineering

Pulsed neutron capture
Depth of investigation -6 in.
Vertical resolution -1y2 ft
Resistivity
Depth of investigation -1 in. to 5 ft
Vertical resolution -1 in. to 5 ft

This large range reflects the large number of resistivity devices available.

Carbon-oxygen
Depth of investigation -2 in.
Vertical resolution -l'/* ft
Nuclear magnetism
Depth of investigation -1 '/4 in.
Vertical resolution -3 ft
Dielectric constant
Depth of investigation -1 l/z in.
Vertical resolution -1 1/2 in.

These measurements are bulk measurements of the formation sampled. Thus,
if there are heterogeneities due to fractures, permeability variations, etc., these
will result in errors for elemental volumes. However, since for example, a low
permeability zone may not contribute to a tertiary recovery process, this feature
of the measurements may be an asset rather than a liability.

Accuracy of Logging Methods
The estimated uncertainty in residual oil saturations from electric logs is 5%
to 10% under optimum conditions and could easily exceed 10% under less
favorable conditions [349]. Accuracy of saturations derived from electric logs
is generally in the range of f15 saturation percent [324], which is clearly
inadequate for residual oil determinations.
From the laboratory measurements of the C/O ratio, it is known that the
accuracy of this technique is poor when porosity is low [218]. Oil saturation
probably cannot be reliably determined at porosities less than 15%. However,
even in a 30% porosity sand, the accuracy of oil saturation measurements would
only be f12.5%. While this uncertainty could be reduced by taking repeated
measurements at the same depth, stability of resolution may still be a problem.
The accuracy requirement for the determination of residual oil is so great
that the use of the log-inject-log technique is often required. Although this
technique is applicable to various logging methods, practical application has been
limited to the PNC and NML. The term log-inject-log (LI-L) refers basically to
the fact that the measurements are made with control over the properties of
the fluid saturating the rock volume under' study. Assumptions with this tech-
nique are that the formation is at residual oil saturation; no oil is displaced
from the formation during injection the true total cross-section is measured;
the brmation water is completely displaced within the radius of investigation by
the tool; and that shrinkage of oil due to gas stripping during injection is negligible.
With the pulsed-neutron-capture LI-L technique, accuracy of i5 $6 oil satura-
tion is obtainable [349,352]. For best results, the contrast in salinities of the
two waters should be as great as possible. While this technique has been