SAMPLING and QAQC

For Gold (Au) and Copper (Cu)

Harman Setyadi
Database Manager, J. Resources Nusantara
Mine Engineering Study Program 
Faculty of Mine and Petroleum Engineering,
Institute of Technology Bandung
 Course Outline
1. Process Control
2. Quality Control
3. Exploration Stages
4. Exploration Drilling 
5. Sampling Error
6. Assay Error and Control
7. Statistic Process Control (SPC).

2
1. PROCESS CONTROL

3
PMBOOK
Processes Group

PMBOOK

5
PMBOOK
2. QUALITY CONTROL

7
 QUALITY ASSURANCE and
QUALITY CONTROL (QAQC)

Why do we need Quality Control (QC) ?

• Required by most securities regulators and financial institutions 
when reporting Exploration results & Resources or Reserves, thus 
providing an indication of the reliability and level of confidence in 
the data being reported.

• Provides a check on the quality of all information loaded into the 
Geological Data Base.
 Traditional Quality vs. Mistake Proofing

Traditional Inspection
Result
Sort
Worker or Don’t Do Defective At Other
Machine Error Anything Step

Discover Take Action/ No Next

Error Feedback Defect Step

Source Inspection
“KEEP ERRORS FROM
TURNING INTO DEFECTS”
SBM, ITB

9
 DEFINITIONS of QA and QC
QUALITY CONTROL (QC):
Achieved by following good WORK PRACTICES and set PROCEDURES
in sampling, sample preparation and assaying.

QUALITY ASSURANCE (QA):

Achieved through collection of supporting evidence which can
DEMONSTRATE the reliability of the data.

Requires assessment of the QC data to determine if results meet

required precision (reproducibility) and accuracy (bias).
3. EXPLORATION STAGES

11
 Exploration Business
• Different the Common Business,
• High Risk, High uncertainty, long term investment,
• Non renewable  Resource Depletion
• Currently – high social & environmental Issues
 EXPLORATION STAGES
• Step‐by‐step to reduce risk,
– To find  to prove  to measure
• Different method and/or approach in every step depend on 
the exploration strategy,
• Different type data collection,
• Different method in data visualization for exploration 
evaluation,
• Data growing and data continuity along the exploration 
work
 General Exploration Information Development
Asset Discovery: How to Find Asset Development: How to Prove Asset Inventory: How to Measure

Large Ares, small sample Density, Selected Area, Higher sample Density, Small/ Selected Area, Higher sample
Simpler database structure & System Moderate database structure & System Density, prepare for mining information
HIGH COST per Sample Collection LOWER COST Higher sample Density HIGH COST Higher sample Density

Detailed Geophysical, Geochemical Survey Detailed drilling, advance geological study,

Regional Geophysical, Geochemical
and Geological Mapping, scot drilling metalurgycal, economic valuation
Survey and Geological Mapping
Target Selection – GIS Analysis, 3D Resource Modelling & Calculation
Prospect Generation – GIS Analysis
Resource Indication

START External FINISH External

Constrains Constrains
FINISH

Reconnaissance, Semi Detail Detailed

Public Geo Prospecting Evaluation Exploration Evaluation Exploration Geological
Database programs Program Program Model

Geotechnical
Model

Geo-hydrology
Explora tio n Exploration Geo Exploration Model
Geo Database
Database Geo Database Geometry
Model/
Constrain
Resource Ore Reserve
Modifying
Calculation/ Calculation/
Factors
Model Model

Relational Object-Oriented External Factors: Economics,

Flat File Legal, marketing, Social,
Database Database
Environmental FINISH

Geographic Information System

(Setyadi, H., dan

Mining Spesific Packages (Mine Planning, Resource Calculation)
Anggayana, K., 2012;
Setyadi, H., 2013 (a))
4. EXPLORATION DRILLING
 JORC TABLE1
Section 1: Sampling Technique and Data
• Sampling Technique: Method of sampling, should be 
representative. Sample determination
• Drilling Technique: drill type, DD, RC, RAB, Bangka
• Sub Sampling Technique: core – non core, sample preparation
• Drill sampling Recovery
• Quality Assay of Data and Laboratory: CRM, Dup, Check Assay.
• Verification of Sampling and Assaying: Sample duplicate storage
• Location of Data Point
• Relation of data to geological structures,
• Sample security

17
 Exploration Drilling Data Control Elements
DOWN HOLE CORE CORE BULK Point
COLLAR SURVEY TRANSPORT MARKING DENSITY Load

Pick up data error Survey Method Access to Drill Site Driller Marking Balance Problem Human Error
Proposed
Collar Core Packing Core Packing Training RQD
Water Level Water Impurity
Data Transfer error Transportation
Magnetic Field Core Transport Water Oscillation Inspection Human Error
Core Production
Surveyor data error Instrument
Actual Surveyor errorChange of Custody Core Production Human Error Training
Collar
Transfer error Tracking Error Core Re-marking Transcription error Human Skill
Data Transfer error Calibration

Transcription
Sorting Balancing CalibrationGas Impurity
Washing InstrumentData Transfer Data
error Geology,
Security
Analytical Labeling Sample Resource
Drifting ContaminationData Reading Observation Technical
Method Mix-up
Support Model
Drill Core Mine Design
Production Flexibility
ANALYSIS
Interval Error Accessibility
Drying Sample Mix-up Pulverizing ContaminationLabeling

Sorting Crushing Homogenizing Splitting Geologist Connectivity

Skill
SAMPLE Geology Integrity
PREPARATION Contact
Sample Size Sample RecoverySample Mix-up Compatibility
Mineralization
Labeling Splitting Contamination
Out Put Proposed and Compiled
Alteration Format
SAMPLING by: Harman Setyadi, PMP
ASSAY
Supt. Geology QA - September 2006
GEOLOGY DATABASE
SAMPLE LOGGING SYSTEM
 COLLAR
• UTM‐Geographic‐Local Grid coordinate system,
• Distinguished between plan and Actual Survey,
• Survey Method
• Transciption Error.

Validation
• Comparing Plan with Actual,
• Plot on the Base Map
 Down Hole Survey
• Survey Requirement Standard, at 25m then 
repeated every 50m.
• Compass versus down hole Survey (at Zero depth)
• Different/change of the azimuth and dip (<5˚),
• Must not exceed to the end of hole (EOH)
• Typo errors.
 Core Transportation
• How the transport Core,
• When should be transported,
• How to handle core
• How long to take time for transportaion.

Quick Logging Reporting,
• Quality and its comparison with the detailed logging.
 Sample and Assay
• Sampling Procedure and control,
• Sample Preparation Protocol and control,
– Preparation protocol: Recived‐dry‐crush‐secondary crushing –
splitting – milling – splitting.
– Sub sampling protocol, 
– Sample size criteria out put for spliting,
– How the reduce and to get representative sample for 0.25 Kg 
from 2‐5 Kg.
• Laboratory Assay Protocol and control,
– Method, digestion, sample charge: 30/50 gram for FA, 10, 100, 
1000 gram for CN, 0.5 or 1.0 gram for BM element.
– Instrument Reading used
– Laboratory internal Check, calibration and maintanance 
procedure. 
 Drill Sampling interval
• Reffer to the geological control: Lithology, 
mineralogy, alteration.
• One meter for regurlar sampling, min 20cm max 
120 cm.
• Consider the sample recovery > 25% or 1 kg 
sample volume.
• RC sample, just by one or two meter sampling 
interval
 Check Sample
• BLANK sample, Baren rock 1 in 40 sample (2.5%), 
after the high grade (if possible).
• Field Duplicate 1 in 20 sample,
• Coarse Reject Duplicate 2.5% and Pulp Duplicate 
5%.
• CRM 1 in 20 sample (5% sample)
• External Laboratory (IT vs SGS; SDO vs IT), 5% 
pulp duplicate sample.

Check Duplicate selection systematic vs Random.
5. SAMPLING ERROR

25
 SAMPLING

AIM OF SAMPLING:

• To collect a representative Sub‐Sample from the 
Original Sample.

• Minimise errors.
SAMPLING DEFINITIONS

1. LOT
2. SAMPLE
3. SUB-SAMPLE

SPECIMEN?
 ERROR
Can be broadly classified as:

• SYSTEMATIC:
A consistent error in the system.
Shows up as a BIAS.
Worse type of error to have.

• RANDOM:
Occurs occasionally.
Shows up as poor PRECISION.
 SAMPLING ERRORS
•Sampling error is the variance of the grade of the Sub-
Sample from the true grade of the Original Sample.

•Sources of sampling errors:

Nugget effect – Fundamental Sampling Error.

Segregation – e.g. layering in calico bag.

Extraction – e.g. dust loss; plucking of core.

Contamination – e.g. water; open hole.

Preparation – e.g. operator; contamination.

FUNDAMENTAL SAMPLING ERROR (FSE)
•FSE states that small samples, particularly from
a low grade lot, will generally not return the
actual grade of the lot due to differences
(heterogeneity) at a particle scale.

•The FSE is a fundamental characteristic of the

material type and CANNOT BE ELIMINATED.
FUNDAMENTAL SAMPLING ERROR (FSE)
•“Gy’s formula” is basis of modern sampling theory,
developed by Pierre Gy in the 1950’s.
•Updated by Francois-Bongarcon in the 1990’s.
•Predicts the RELATIVE VARIANCE of a sample (i.e.
characterizes its reproducibility). It is a function of:
Macroscopic properties of the ore;
PARTICLE SIZE d (P95 nominal size); and
Sample and lot MASSES (MS and ML).
It is easier to use a NOMOGRAPH to derive a
sampling protocol.
 ACCEPTABLE PRECISION

SAMPLE PREP STAGE PRECISION

Pulps 10%

Preparation Duplicates 15%

Field Duplicates 20%

 WHERE ERRORS OCCUR
Classification of Errors (after Bongarcon)

Estimation Errors

Taking the sample

Sampling Error Geostatistical Geological
Estimation Error Interpretation Error

Sample Collection Analytical Error Modelling Kriging Error

Error Error

Localisation Error Materialisation Error Fundamental Sampling

Error

Geological
Delimitation and Grouping and
Extraction Error Segregation Error Errors
Geostats Errors

Sampling Theory
CORRECT SAMPLING

Three critical areas:

 WEIGHT of Sub-Sample.

 PARTICLE SIZE at each Sub-Sample stage.

 METHOD used to take Sub-Sample.

 SAMPLE TYPES
•Solid – Diamond
core.

•Chips – Generated
by conventional &
RC drilling.

•Rocks – Face,
stockpile grab
samples.
 DETERMINING A
SAMPLE PREPARATION PROTOCOL

•Conduct orientation test work – produce a Sampling Nomo

graph.

•Document sample preparation protocol:

Produce a Flow Chart.

Communicate to all Geologists and the Lab.

•Conduct Check Sampling using Duplicate Samples.

•Modify sample preparation protocol if Check Sampling

results indicate a problem.

•Audit Lab regularly (every 3 months) for compliance.

DO NOT ALLOW A SAMPLING PROBLEM TO CONTINUE UNCHECKED.

SAMPLE PREPARATION FLOWCHART
NQ FULL CORE (5.5kg) BULK DENSITY
NQ HALF CORE (2.75kg) SG_RPTCHK
(Newmont will nominate
DRY samples)
(1100C +/- 50C)
INSERT FLUSHES
1. After designated samples.
2. Before each batch (6 flushes).
JAW CRUSHER
(Coarse Crush)

LM5 PULVERISER
(85% Passing 75m) 1 to 3 PASSES HOMOGENISE

ALL SAMPLES GRIND CHECKS

SCOOP PULP RESIDUE 75m (1 in 40)

ALL
1) PAPER PACKET - 400g
SAMPLES
(MASTER SPLIT)

ALL 2) PAPER PACKET - 500g

SAMPLES (ASSAY)

1 in 50
3) PAPER PACKET - 150g
SAMPLES
PULP_DUP
(PD:aaxxxxxx); (Au)
6. ASSAY ERROR AND CONTROL

38
 ASSAYING
•Digest – getting the metal into solution.
•Finish – reading the grade of the metal.
•Choosing a suitable assaying technique for an ore type is
critical:
An experienced Chemist must ALWAYS be consulted.

Adequate testwork must be performed on the ore type.

•4 techniques can be used for gold – Fire Assay, Screen Fire

Assay, Acid Digest and Cyanide Leach.
•Choice of technique depends on:
 Ore type
 Level of accuracy required
 Lower detection level required
 Turnaround time for results
 Cost
 FIRE ASSAY
• Preparation
 The sample is mixed with a flux which includes lead and silver.

• Fusion
 Conducted in fusion furnace at about 1,100°C.

 Gold is collected by lead (from flux) to produce a lead button.

 The lighter silica slag collects all impurities.

• Cupellation
 After solidifying, the lead button is placed in a cupel.

 The cupel is placed in a muffle furnace.

 The lead is absorbed into the cupel, leaving a prill of Au/Ag.

• Parting
 Process where the silver is separated from gold.

 Use Nitric acid (dissolves silver) and Hydrochloric acid (to produce
aqua regia).
• Finish
 The solution containing gold is read on an AAS machine.
QAQC - TERMS
•Primary sampling and assaying

•Standards

•Blanks

•Grind Checks

•Check Assaying

•Check Sampling
 QUALITY CONTROL PROCESSES
DEFINITIONS
Primary Assay:
•Performed by Principal Lab.
•Performed on Primary Assay Sub-Sample.
Check Assaying:
•Performed by Umpire Lab.
•Use same pulp packet as Principal Lab.
•Checks for bias.
Check Sampling:
•Performed by Principal Lab.
•Performed on Duplicate Samples.
•Checks whether sampling method is providing a
representative Sub-Sample.
 Check Sampling (cont.)

•The Check Sub-Sample must be taken using exactly

the same sampling technique as the Assay Sub-Sample.
•Assaying must be conducted at the Principal Lab.
• The Check Sub-Sample must be prepared and assayed
using exactly the same methods used for the Assay
Sub-Sample.

Any variances or errors should be due to the sampling

protocol. Modify if incorrect.
 REVERSE
DRILL SAMPLE CIRCULATION
GENERATION

DIAMOND
CORE

ASSAY
SUBSAMPLES

FLOWCHART
INSERT
CHECK BLANKS RIG
SAMPLES & DUPLICATES
FLUSHES

SAMPLING COARSE CRUSH COARSE

and
GRIND 90% <3mm DUPLICATES
CHECKS
GRIND DUPLICATE
CHECKS ASSAYS

ASSAYING
PULP PULVERISE PULP
GRIND 90% <75u DUPLICATES
CHECKS

PULP REPEATS REPEAT

RESIDUE PULP ASSAYS

INSERT
STANDARDS

PRIMARY
PRINCIPAL LAB ASSAYS

INSERT
CHECK STANDARDS
ASSAYS
&
GRIND UMPIRE
CHECKS LAB
 PROTOCOLS for STANDARDS (CRMs)
Standards are used to determine ACCURACY in Lab.
Properties:
• The Expected Value must be accurate.
•The Standard Deviation must be low.
• Must be homogeneous, cannot contain coarse gold.
•Must be stable over time:
Sulphides can oxidise with time in humid conditions.
Particles with extreme density differences can segregate during
transport.
•Should be made from a similar rock type to Primary Samples:
This can be a problem with commercial standards.
•Must have a certificate with details of the validation process.
SELECTION of STANDARDS
At least 5 standards should be selected for insertion on a
rotating basis:
Low gold grade (e.g. < 1.0g/t)
Gold grade near the Mill Feed cutoff (e.g. 3g/t)
Medium to high gold grade (e.g. 3 to 20g/t)
High gold grade (e.g. 20 to 50g/t)
 The above 4 standards must be available as both oxide and
primary material.
Where possible, all standards should be visually similar.
Always attempt to match the rock type of the standard with the
rock type of the samples being assayed (matrix matching).
INSERTION OF STANDARDS DURING SAMPLING

•Generally insert 1 per 25 samples.

•Three standards must be included in each assay
batch (74 samples for ALS).
•Need to identify with lab the batch size & number
of internal checks they use in each batch.
•Must know the position of the Standard in sample
sequence.
•Maintain a good numbering system at Tanami!
 PROTOCOLS FOR BLANK MATERIAL
Blanks are used to monitor contamination in sample prep.
Selection & Properties of Blanks:
Similar rock type to primary samples.
Made from coarse material (e.g. drill core or rock chips).
Made from hard material to remove smeared Au.
Barren of all assayed elements.
Place within a mineralised zone where possible.
 Assign sample number in sample sequence.
INSERTION OF BLANKS DURING SAMPLING

• Insert 1 per 25 samples.

• Insert more if problems occur at lab.

• Place within a mineralised zone where possible.

• Assign sample number in sample sequence.

 Population

Sample
Sample
Sample

7. STASTICS PROCESS CONTROL

50
 Parameters vs. Statistics
Population: All the items that have the “property of interest” under study.

Frame: An identifiable subset of the population.

Sample: A significantly smaller subset of the population used to make an inference.

Population

Sample
Sample
Sample

Population Parameters: Sample Statistics:

– Arithmetic descriptions of a
– Arithmetic descriptions of a sample
population – X-bar , s, p, s2, n

– µ,  , P, 2, N
OSSS LSS Black Belt v9.0 - Measure Phase 51
 Accuracy and Precision
Accurate but not precise - On Precise but not accurate - The
average, the shots are in the center average is not on the center, but
of the target but there is a lot of the variability is small
variability

OSSS LSS Black Belt v9.0 - Measure Phase 52

 Accuracy vs. Precision
ACCURATE PRECISE BOTH

+ =

Accuracy relates to how close

the average of the shots are to
the Master or bull's-eye.

Precision relates to the spread

of the shots or Variance.
NEITHER

OSSS LSS Black Belt v9.0 - Measure Phase 53

 CALCULATION OF BIAS
(ACCURACY)
• For a Standard:
BIAS = (ASSAY – CERTIFIED VALUE) ÷ CERTIFIED VALUE

• For an umpire check assay:

BIAS = (PRINCIPAL LAB – UMPIRE LAB) ÷ MEAN

SAMPLE TYPE ACCEPTABLE BIAS

Reserves, Grade Control 5%
Exploration 10%
CALCULATION OF PRECISION (RSD)

ASSAY PAIR:
1. Calculate MEAN
2. Calculate STANDARD DEVIATION (SD)
3. Calculate RELATIVE STANDARD DEVIATION (RSD)

RSD = SD ÷ MEAN (expressed as %)

e.g. For a Repeat: Au1 = 8.7 g/t Au
Au2 = 10.0 g/t Au
RSD = 10%
 GENERAL REQUIREMENT FOR QAQC
QAQC should be done for all sampling and assaying.
EVERYTHING!!!

•Standards, Blanks and Grind Checks:

Assess daily.

Act Immediately – the primary samples in the batch are

suspect.
The batch may have to be re-assayed.

•Check Assays, Check Samples:

Assess monthly.

Correct emerging trends.

•Report Results and Actions:

Action Register.

Identify issue and the corrective action taken.

 Elements of Control Charts
Developed by Dr Walter A. Shewhart of Bell Laboratories from 1924
Graphical and visual plot of changes in the data over time
– This is necessary for visual management of your process.
Control Charts were designed as a methodology for indicating change in
performance, either variation or Mean/Median.
Charts have a Central Line and Control Limits to detect Special Cause variation.

Control Chart of Recycle

60 1

UCL=55.24
Special Cause 50

Variation Detected
40
Individual Value

30
_
X=29.06 Process Center
(usually the Mean)
20
Control Limits
10

LCL=2.87
0
1 4 7 10 13 16 19 22 25 28
Observation

OSSS LSS Black Belt v9.0 -

57 Control Phase
 STANDARDS/ CRM
Standards check for bias during assaying.
Part of Primary Sampling &Assaying and Check Assaying process

Acceptance of the results is g/t Au Action Required

Action Line
determined by using “warning lines” 3
Warning Line
and “action lines”, which are 2
calculated using the standard x Recommended Value
deviation () given by the supplier. 2 Warning Line
3 Action Line
Action Required
The “warning lines” are equal to 2
standard deviations (2). Sample Number / Time

The “action lines” are equal to 3

standard deviations (3).
 Au_ppm (AA)

01

8.0
8.5
9.0
9.5
10.0
10.5
11.0
03‐Jan‐
04‐Jan 11
07‐Jan‐‐11
10‐Jan 11
11‐Jan‐‐11
13‐Jan 11
14‐‐Jan‐‐11
17 Jan 11
19‐Jan‐‐11
21‐Jan 11
22‐‐Jan‐‐11
25 Jan 11
28‐Jan‐‐11
3 ‐Jan 11
040‐Jan‐11
07‐Feb‐11
09‐Feb‐11
11‐Feb‐11
11‐Feb‐11
15‐Feb‐11
16‐Feb‐11
21‐Feb‐11
25‐F eb‐11
01 ‐Feb‐11
03‐M a ‐11
04‐M arr‐11
08‐M a ‐11
10‐M arr‐11
11‐M a ‐11
14‐M arr‐11
16‐M a ‐11
17‐M arr‐11
20‐M ar‐11
21‐M a ‐11
22‐M arr‐11
25‐M a ‐11
26‐M arr‐11
28‐M a ‐11
30‐M arr‐11
‐
03 M ar‐11
06‐Apr‐‐11
07‐Apr 11
STD5

13‐Apr‐‐ 11
18‐Apr‐ 11
20‐Apr 11
23‐Apr‐‐ 11
26‐Apr 11
28‐Apr‐‐ 11
01 ‐Ap 11
03‐M ayr‐ 11
06‐M a ‐11
07‐‐M ayy‐11
08 M a ‐11
12‐M ayy‐11
16‐M a ‐11
17‐M ayy‐11
19‐M a ‐11
21‐M ayy‐11
23‐M a ‐11
24‐M ayy‐11
26‐M a ‐11
28‐M ayy‐11
29‐M ay‐11
‐M ‐1
ay‐ 1
11
STANDARDS – MODIFIED APPROACH

Lab ‐2SD
Lab +2SD
Lab Mean

Certified Value
Rejected Values

5 per. Mov. Avg.
Accepted Values

Moving Avg +Lim

Moving Avg ‐Limi

(Accepted Values
BLANKS
Part of Primary Sampling and Assaying process.
Check for contamination during sample preparation.
Check for swaps.

g/t Au
Action Required

0.2g/t Action Line

0.1g/t Warning Line

Sample Number
 INTERPRETATION of BLANKS
•Look for trends.
•Are there any unacceptable results.
•What corrective actions need to be taken.
BLANK No ... at ... LAB
Au g/t 10.00
(log
scale) 2.34
1.00 0.83
0.36 Action Line (0.2 g/t)
0.13
0.10 Warning Line (0.1g/t)

0.01
31-May 10-Jun 20-Jun 30-Jun
DATE
 CHECK SAMPLING
•Check Sampling is the process of taking
Duplicate Samples to check whether the
sampling method is providing a representative
Sub-Sample.

•The level of precision required is:

 ± 20 % for Field Duplicates
 ± 15 % for Preparation Duplicates
 ± 10 % for Pulp Duplicates
 CHECK ASSAYS
Check Assaying determines
whether there is an assay bias at
PRINCIPAL LAB. Warning Line (+10%)
Action may
ASSAY B be required
We don’t know which Lab is g/t Au Reference Line(x=y)
correct.
Standards determine accuracy. Warning Line
(-10%)
Use scatter plot.
The “warning lines” are set to Action may
precision. be required
ASSAY A g/t Au
The Precision is calculated using
RSD. 10x Detection
The Bias is calculated as
(Difference – Mean) / Mean
INTERPRETATION of CHECK SAMPLING
•Look for trends.
•Are there any unacceptable results.
Example 1 – Field Dups •What corrective actions need to be taken.

SC ATT ER PL OT
CHE CK S AM P LE S at … LA B
ALL G RADE RA NGE S +20%
25.00

20.00

-20%
15.00

10.00

5.00

0.00
0.00 5.00 10.00 15.00 20.00 25.00
AS S AY A A u g/t

•Good result. The scatter plot shows good precision.

•1 duplicate occurs below the 20% warning line. Investigate this sample.
Thank You / Terima Kasih
Harman Setyadi
Equity Tower, 48th Floor
Sudirman Central Business District, Lot.9
Jl. Jendral Sudirman, Kav 52-53
Jakarta Selatan
e-mail: Harman [email protected], [email protected]

T +6221-515 3335
F +6221-515 3579

www.jresources.com