INSTRUCTION/ COMMENTS

STEP
1 Setting up the test - ensure your measuring method and position are consistent for every measurement. You will be measuring very small volumes and small changes in height
at which the measurement is taken can change the result of a single measurement (ie. change in "hydrostatic head"). A small (1L) measuring beaker with gradations for
measuring 1/10ths of litres (or smaller) is preferable. In deep wells of large volume with a large density and/or temperature difference between the original fluid and the new
underbalanced fluid, early measurements may be large (10+L over 10 minutes), so be prepared.

2 Collecting the data - the test in deep wells is likely to take 2-4 hours before the large wellbore and outer-annuli effects settle down and the data begins to form a trend. Expect
to gather data for an additional 2-4 hours (12-24 points) to ensure the trend line is representative and defendable, especially if the conclusion is a failed test that will require time
and expenditure to repair - take the time to be sure. Therefore, if a crew change or relief is likely, ensure the data collection method is consistent no matter who is taking the
measurement and be prepared for a 4-8 hour duration before a result is clearly identifiable. Having said this, if the trend is consistent, declining, and meets the criteria for a good
test, it is possible to conclude the test earlier.
TIP: Early data can be misleading - again, with large drawdowns and large volumes, exposure of the new, cold fluid to multiple annuli (7" x 8 1/2", 9 5/8" x 12 1/4", surface
casing, conductors, risers), may show effects related to expansion/ contraction of steel, fluids in outer annuli, etc. What you are looking for is a consistent trend of data that is
representative of the annulus you are testing e.g. the 7" liner lap (tbg or dp x 'A' annulus). Over time, the temperature expansion effect of the 'A'-annulus will exhibit flow which
follows a logarithmic decline. The Horner Plot is attempting to show the trend of this decreasing flow to indicate if, over time, it will diminish to zero (when the fluid temperature
reaches steady state), or whether there is indeed flow from the annulus being fed from somewhere else (failed inflow test where flow does not go to zero at infinite time).

3 Plot the data - on the table in the tab, "Horner Plot", fill in the information in the Yellow cells. The table has the calculation for Horner Time (HT), where HT = ((DT+T)/DT), in the
appropriate column (Green cells). This is the data that will plotted in the lower right graph. Note that the x-axis is on a logarithmic scale. This is the most common representation
of this data, and should be used in preference to the alternative chart as shown in the lower left (if Excel is not used at the wellsite, the alternative graph may be easier to
physically plot on graph paper if semi-log paper is not available). This second graph plots the natural logarithm of Horner Time, ln((DT+T)/DT), calculated in the final column of the
table. You can see the similar trends, but the x-axis is not logarithmic and the ln-caculation is done first, then plotted on a normal scale. A trend line on both graphs can be
represented by the same equation.

4 Interpret the data - on the 1st graph, when DT goes to infinity (infinite time), Horner Time calculated will trend to DT/DT (the finite T becomes insignificant), or 1.0. Therefore, if
flow is diminishing at a rate that will go to zero (wellbore steady-state conditions) before infinite time, then the trend line should cross the x-axis prior to HT=1.0. On the second
graph, plotting ln(Horner Time), the trend should cross the x-axis before ln(HT) goes to 0.

5 Apply a trend line - one can use the automatic trend line function in Excel, but be warned that this will calculate a trend line that honours all the data, even the erroneous early-
time data which is not necessarily representative of the annulus you are concerned with. It is preferable to apply a manual trend line through the later data once a consistent
trend line becomes apparent. If you do use Excel to illustrate a trend line, you can modify what data is used for the trendline by "hiding" rows of data in the table (select the entire
row, and choose Window > Hide). Once hidden, the data point will not appear in the plotted chart and therefore the data will not be included in the trend calculation.
Unfortunately, when this is done, the data also disappears from the chart altogether and for historical purposes, this is not preferred as it masks potential data points that form
the whole picture of what is going on, making the data look better than perhaps it was.

6 Adjust the graph formats - once the graph is plotting the correct data, the scales of the axes can be modified to be illustrated correctly within the graph area using the right-
click function, and editing the scales etc. It is advisable to leave the axes crossing at Horner Time = 1.0, as any data to the left of this intercept is imaginary (beyond infinite time).
If data is plotted to the left of the y-axis on this graph, it is possible that the incorrect data is being plotted (ln-double-dip). Check the Source data for the graph to make sure it is
correct. The most common settings for the scales will be y-axis set to include all the data, and x-axis at one logarithmic interval (1.0 to 10.0), but use your judgement to make the
graph readable and representative.

7 Prepare the graph for presentation - finally, add titles to the graphs, and annotate if required. This graph should be published to Livelink, Discovery and copies should go to
the Asset for the historical wellfile and records. The graph will be critical should an annular leak ever occur.

NOTES
Auto trendline (Excel), close to intercept? - if the trend intercept is very close to the y-axis intercept at 1.0, one can right-click on the trendline and go to the Trendline
options. On the third tab, there is a box to tick for "show equation". If this is selected, the equation of the trend line is calculated and shown, and should show the actual intercept
in the format y = mx + b where m is the slope of the line, and b is the intercept. A negative intercept intersects the x-axis (good).

Source material - it should be noted that the two primary sources for information on inflow testing and Horner Plots at time of writing are the EP 1500-2002 document, and the
Wells Distance Learning Package for Round 1 and 2. Both of these documents currently contain errors in the sections related to this analysis (see below).

Floater Operations - beware of wave/surge activity and tidal influences when performing inflow tests on Floating Drilling Operations. Primarily, ensure the test is set up to
measure fluid coming from the well that is being compensated for wave action, if at all possible (measure annular volumes through kill/choke lines in preference to riser/flowlines
if possible, or up drillstring where applicable). Secondly, ensure any corrections for tidal influences are taken into account. Over the period of time of a test (4-8+ hours), this could
be significant. Finally, use all data supplied, measure as carefully as possible, and ensure any variables that can be taken out of the measurement have been compensated for.
Adequate planning for the test will be the crucial.

FINAL GUIDANCE - remember that the Horner Plot is only a tool. It is meant to provide a meaningful trend analysis to data collected over a shorter period of time to allow
identification (and elimination) of temperature effects acting on an annular flow such that one might be able to infer whether the annulus is exhibiting a leak to inflow forces, or
whether this flow will eventually trend to zero. Be prepared to discard the data if it proves inconclusive. Circulate the well back to heavy fluid, and monitor what comes to surface
on bottoms-up. Is there a significant gas peak (significantly greater than seen during previous circulations)? Once the test packer is unset, and the well circulated from below the
packer, is there a similar gas peak? Was the packer leaking? (this should be positively tested on the annulus prior to the beginning of the test). Use all the data available to assist
the Team in coming to a conclusion. It will always be a judgement call, and it should be understood that the Horner Plot is not always as definitive as one might want it to be.
 INFLOW TEST HORNER PLOT TABLE Only fill in the yellow cells
WELL: En08-P DATE: 14-Jan-15

Time since last circulation (T) = 60 MIN. Inflow Test 9 5/8 csg
(Note: DT = elapsed time since beginning test)

DT Volume Volume FLOWRATE {(DT+T)/DT} LN((DT+T/DT)

5.00
mins gal litres l/min (Horner plot)
2 5 0.5 1.89 0.38 13.000 2.565
3 10 0.25 0.95 0.19 7.000 1.946
4 15 0.5 1.89 0.38 5.000 1.609
5 20 0.25 0.95 0.19 4.000 1.386
6 25 0.3 1.14 0.23 3.400 1.224 4.00

Flowrate [lpm]
7 30 0.35 1.32 0.26 3.000 1.099
8 35 10 37.85 7.57 2.714 0.999
9 40 0.25 0.95 0.19 2.500 0.916
10 45 0.5 1.89 0.38 2.333 0.847
11 50 0.4 1.51 0.30 2.200 0.788 3.00
12 55 0 0.00 0.00 2.091 0.738
13 60 0 0.00 0.00 2.000 0.693
14 0 0.00 #DIV/0! #DIV/0!
15 0 #DIV/0! #DIV/0! #DIV/0!
16 0 #DIV/0! #DIV/0! #DIV/0!
17 0 #DIV/0! #DIV/0! #DIV/0! 2.00
18 0 #DIV/0! #DIV/0! #DIV/0!
19 0 #DIV/0! #DIV/0! #DIV/0!
0 #DIV/0! #DIV/0! #DIV/0!
0 #DIV/0! #DIV/0! #DIV/0!
0 #DIV/0! #DIV/0! #DIV/0! 1.00
0 #DIV/0! #DIV/0! #DIV/0!
0 #DIV/0! #DIV/0! #DIV/0!
0 #DIV/0! #DIV/0! #DIV/0!
0 #DIV/0! #DIV/0! #DIV/0!
#DIV/0! #DIV/0!
0.00
#DIV/0! #DIV/0! 0 10 20 30 40 50 60 70

#DIV/0! #DIV/0! Elapsed time [min.]

#DIV/0! #DIV/0!
#DIV/0! #DIV/0!

Inflow 9-5/8" casing

6.00

5.00

5.00 4.50
Flowrate [lpm]

4.00

Flowrate [lpm]
Manual trendline
4.00
Manual trendline 3.50

3.00

3.00
2.50

2.00
FAIL
2.00
1.50

1.00
1.00

0.50
PASS
0.00
0.00
1.000 10.000
0.000 1.000 2.000 ((DT+T)/DT)
ln((DT+T)/DT)