A New Large Pump For Oil Sands

A New Large Pump for Oil Sands
Michael Bootle
Principle Engineer Weir Minerals
Calgary Pump Symposium 2015
Presenters
Michael Bootle
1984-1987: Application Engineer and Regional Sales

Manager for Lawrence Pumps Inc.
1987-1990: Carpenter
1990-2000: Field Service and Senior Field Service Engineer
for Warman International
2000- 2014: Senior Design Engineer for Weir Minerals North
America
2014: Principal Engineer for WMNA
Why the Need? HI Slurry Classification

Based on
Specific Gravity
& d50 Particle
Size
Typical Oil Sands
Hydrotransport and
Tailings Slurries:
100 400 micron
1.4 1.6 sg
Why the Need?

HI Head Recommendations
Based on Slurry
Classification:
Class 1:
Class 2:
Class 3:
Class 4:
105 m
73 m
55 m
40 m
Wear history has

shown these
recommendations
are not
conservative
Variety of Pump Types

for Different Applications
N s and ns =
Ns =
ns =
N Q
3/4
rpm USgpm
ft
3/ 4
rpm m 3 / sec
3/4
Ns (US) = ns (Metric) x 51.64
Specific Speed, Efficiency and Pump Size
Early Oil Sands History:

Warman 450 STL Pumps on Plant 5 PSV Uflow
(4) 450 STL
450 mm discharge
500 mm suction
950 mm impeller
41.7 ns
D2/D1 = 1.93
Installed mid 80s

Neoprene Lined
6000 - 8000 hr life in
late 80s, early 90s
Early Oil Sands History:

Plant 5 Warman 450 STL Specific Speed
Horizontal Pum p
Calculation of
Specific Speed:
450 L
CURVE SHOWS APPROXIM ATE PERF ORMANCE FOR CLEAR WATER (In ternation al Test Stan dard ISO 9906:199 9 - Grad e 2 u nless other wise sp ecified). For me dia other than water ,
corr ections m ust be made for density, visc osity and/or othe r ef fects of solids. WEIR M INERALS reser ves the right to ch ange p ump perf orma nce an d/or delete im pelle rs without notice. Fram e
su ita bility must be che cked f or each d uty and drive arr angem ent. Not all fr ame alternatives are necessarily a vaila ble fr om each m anuf acturing centr e.
60
6 00 rpm
Estimated 2015 System Curve
50 % 6 0%
70%
50
8 0%
40
Hr
NP S
12m
4 00 rpm
20
8m
3 50 rpm
Max imum Flow
Gland Sealed Pump

(No rm Max r /min)
545
700
Mi n P assag e Si ze
103m m
C urve
R evision
2
R evision Notes
MI NIMUM P AS SAGE SIZ E ADDE D
R eference
MAD ISON TESTS 840303, 04, 05;
A4-0-1037; N PSH : MAD ISON
TESTS 841011, 12
Las t Iss ued
Jan 93
Is sued
Mar 06
6m
5m
4m
10
- KW )
560
560
600
900
S eal
Polymer
Metal
3m
Multiply by 51.67 to obtain Specific

Speed in terms of US gpm & feet:
41.7x 51.67 = 2155
5
950mm
Closed
Material
Metal
Liner
3 00 rpm
n s = 41.7
Impeller
Vanes
Vane
Type
P art No
STL45147DP2
80 %
30 4 50 rpm
600rpm (4640 m 3hr ) /(3600 sec/ hr )
(41 .4m)3 / 4
83%
EFF IC
IE
5 00 rpm
NCY
3/ 4
450mm
500mm
Frame ( Rat ing

ST
SX
G
GG
85 %
10m
ns =
83 %
5 50 rpm
Head, H (m)
ns =
N Q
P ump
D isc harge
Suc tion
Average Flow
Min imum Flow

2000
4000
F lo w Rate, Q (m/hr)
6000
8/20 15 Weir Miner als Au str alia ( PT C)

All Rig hts Re serv ed
TYPICAL PU MP PE RFORM ANC E CUR VE
WPA450A22/2
1997 Plant 5 Process Change:

North Mine Hydrotransport Feed
Draglines, bucket
wheels, conveyors
and tumblers
changed to truck
and shovel, in pit
crushing and
hydrotransport

Increased max solid size:
Tumblers:
9/16x 2
North Mine screens: 2 x 3
Increased flow rates & head:

Flow:
13,000 to 20,000 gpm

(3000 to 4500 m3/hr)
Head: 46 to 70 ft
(14 to 21 m)
10

Decreased 450 STL Life:
Cutting of Neoprene liners
Increased impeller vane and
back shroud wear
Material change required to

restore adequate 450 STL life:
High chrome liners
Hyperchrome impellers
On two pumps, larger 500 STL

wet-ends installed:
Greater wear life with
standard high chrome iron
11
Syncrude Aurora PSV Uflow:

Warman 650 ULP Installation
Part of Aurora SWQR:
Larger 5 x 5 Screens
(2) Warman 650 ULPs

650 mm Discharge
800 mm Intake
1455 mm Impeller
39.8 ns
D2/D1 = 1.83
Installed 2006
12

Warman 650 ULP Operation
240-1-G1 Average:
240-2-G1 Average:
7881 m3/hr
26.3 m
1.45 sg
306 rpm
175-225 micron
7397 m3/hr
20.2 m average
1.47 sg
282 rpm
175-225 micron
Barely Class 4 Slurry:

40 m limit
13

Warman 650 ULP Wear Life
Impeller:
2500 - 3500 hours
Throatbush (suction liner):

3100 - 3800 hours
Volute Liner:
3500 - 5500 hours
Frame Plate Liner Insert (back liner):

4000+ hours
14

650 ULP Impeller Wear: 2500 - 3500 hrs
Mainly vane leading edge wear
Despite operation near BEP, back shroud intact
15

650 ULP Throatbush Wear: 3100 - 3800 hrs
Relatively uniform wear pattern occasional gouging near ID
OD wear at edge of tungsten carbide cladding since
addressed with OD cladding and elastomer gasketing
16

650 ULP Volute Wear: 3500 - 5500 hrs
Smooth wear pattern associated with BEP operation
Little wear at ID junction with side liners
17

Desired Wear Life -2014
Syncrude Requisition MR 240-G-5000 issued
May 2014 with requirement for:
Casing/volute liner life of at least 8500 hours
Impeller and throatbush/suction liner life of at
least 4250 hours
Substantial increase in wear life will require

larger pump design to increase life
18
Effect of Specific Speed on Pump Wear
With permission, from Prediction of Slurry Pump Wear and Cost

- Sellgren and Addie, WEDA XXV, New Orleans, 2005
19
Effect of Suction Diameter on

Suction Liner Wear
With permission, from Prediction of Slurry Pump Wear and Cost

- Sellgren and Addie, WEDA XXV, New Orleans, 2005
20
Impeller Geometry and Its Effect on

Specific Speed
Values of Specific Speed, ns, using m3/sec & m
Courtesy of Hydraulic Institute
21
Syncrude Leading Edge Wear Model

Dominant mode of impeller wear is impingement
damage on the vane leading edge
Rear shroud wear at the vane leading edge can also be
significant
Vane thinning and wear at the trailing edge is minimal
22

Dan Wolfes wear model showed impeller vane wear can be dramatically
reduced by lowering leading edge peripheral and impact speed
23

Key assumption is that vane wear is related to
leading edge velocity to the 2nd or 3rd power
Benefits to lower specific speed (bigger diameter):
Increase OD/ID ratio
Decrease rotational speed, inlet peripheral speed and
impact speed for a given head
Decrease leading edge wear rate
Increase vane length
Increase impeller wear life
24

Desired Future Operating Conditions
Max Duty:
10,500 m3/hr
45 m
1.6 sg
420 micron
400 rpm
Normal Duty:
7900 m3/hr
24 m
1.46 sg
232 micron
300 rpm
Normal Duty
comparable to
recent operation
25
Syncrude Aurora PSV Underflow:

Existing Weir Pump Options
650 ULP: 57 OD
Current life
indicates 8000
hours unlikely
with this design
650 MCH, MCR, &

MCU: 69-70 OD
Lined and unlined
low pressure mill
circuit designs
600 HTP: 62 OD
Hydraulically too
small for required
volume flow at
low head
26

Warman 650 HT Installation
27

New 650 HT (74 Impeller) Design
Max Duty:
Normal Duty:
10,500 m3/hr
45 m
1.6 sg
420 micron
300 rpm
7900 m3/hr
24 m
1.46 sg
232 micron
225 rpm
Approximately
75 to 100 rpm
lower than
existing 650 LP
28

650 HT Duty Summary 3500 BHP Motor
Normal duty at approximately 224 rpm and 1269 HP power draw

Max duty requires 312 rpm and 3653 HP at 1.6 sg and 420 microns
Max duty requires 306 rpm and 3125 HP at 1.46 sg and 232 microns
Project
Slurry Case
3
Flow Rate (m /hr)
Total Head per Pump (m)
Solids d50 (micron)
Occurrence Time (%)
Stream Specific Gravity (-)
Weight Percent Solids (%)
NPSHa (m)
Pump Selection
Quantity
Head Ratio (Roudnev-Wilson)
Efficiency Ratio (Roudnev-Wilson)
Required Water Head (m)
Required Speed (rpm)
Water Efficiency (%)
Expected Slurry Efficiency (%)
Expected Power Draw (kW)
Expected Power Draw (BHP)
NPSHr (m)
3
Approximate BEP Flow (m /hr)
Percent of BEP flow (%)
Normal
7900
24.0
232
average
1.46
50.60
> 15 m
0.96
0.96
25.0
224
83.0
79.7
947
1269
4.6
7800
101.3
Syncrude PSV Underflow Retrofit

Max. with 1.6 sg & 420 d50
10500
45.0
420
1.60
60.23
> 15 m
650 HTP
2
0.92
0.90
48.9
312
84.0
75.6
2725
3653
7.2
10800
97.2
Max. with average sg & d50

10500
45.0
232
1.46
50.60
> 15 m
0.96
0.96
46.9
306
84.0
80.6
2331
3125
7.1
10600
99.1
29
Alternative Applications for 650 HT(P)

Prior to settling
on final design,
application
opportunities
were reviewed:
Showed need
for high
pressure and
low pressure
designs
Showed need
for high and
low flow
designs
30
650 LP vs. 650 HT Wet-end Comparison

650 HT substantially larger
with respect to:
Volute area
Volute thickness
Impeller diameter
Vane length
Throatbush thickness
Lower ns:
29.4 vs. 39.8
Larger OD/ID ratio:

2.47 vs. 1.83
Expect significant increase in

wear life:
650 LP
Details to follow
650 HT
31
650 LP vs. 650 HT Impeller and Volute

Comparison
IMP. EYE DIA. 760mm
IMP. O.D. 1880mm
VANE LENGTH 1726 mm
MID VANE DIA. 550mm
IMP. EYE DIA. 794mm
IMP. O.D. 1455mm
VANE LENGTH 980mm
MID VANE DIA. 573mm
Thickest in
High Wear
Areas
1.5 to 2
times 650 LP
Volute
Thickness
Constant Thickness Wall
650 LP
650 HT / HTP
32
650 LP vs. 650 HT Installation:

Pump House 1
Pump House 1 with low 12.5 tonne overhead crane had
little room to move spares particularly volute liner
650 ULP
650 UHT
33
650 LP vs. 650 HT Installation:

Pump House 2
Pump House 2 with high overhead crane has greater
room to move spares still 12.5 tonne crane
650 ULP
650 UHT
34
Casing Types, Radial Load

& Preferred Operating Range
True/Near Volute
80 to 120% BEP
Semi Volute
60 to 110% BEP
Circular/Annular Volute
40 to 100% BEP
35
650 LP Velocity Plots at 302 rpm
36
650 HT Velocity Plots at 224 rpm

VAR IATION OF VELOCITY AROUND THE C ASING FOR VARIOUS FLOW R ATES
AVERAGE INTERNAL VELOC ITIES, m/sec
18
PERCENT
B.E.P.
FLOWRATE
16
14
30%
12
40%
10
50%
60%
70%
T A
2
0
A
90%
100%
C
F
80%
110%
120%
E
F
130%
CASING WALL SECTIONS
37
650 LP vs. 650 HT Relative Velocity &

Velocity2 at Normal Duty
Velocity2 is a good general indicator of relative
wear for pumps of different size
650 LP vs. 650 HTP - Aurora PSC U'flow: 7900 m3/hr @ 24 m
650 ULP @ 302 rpm
650 HTP @ 224 rpm
Section
Vel (m/s)
Velocity ^ 2
Vel (m/s)
Velocity ^ 2
T
9.60
92
6.26
39
A
10.87
118
6.84
47
B
11.74
138
7.30
53
C
12.31
152
7.64
58
D
11.98
144
7.91
63
E
11.75
138
8.13
66
F
11.69
137
8.30
69
G
11.64
135
8.45
71
H
9.45
89
6.51
42
38
650 LP vs. 650 HT
2
Velocity
Plots
650 LP shows characteristic peak associated with two arc volute

650 HTP has better, flatter, lower profile
39
8000 Hour Volute Goal:

Predicted 650 HT Volute Wear Life
Existing 650 ULP wear life: 3500 5500 hours: Use 3500 hours
Relative velocity2 factor: (12.31 m/s/8.45 m/s)2 = 2.12
Relative thickness at predicted peak wear areas = 2.0
Predicted Life based on Geometry change alone:
3500 hours x 2.12 x 2.0 = 14,840 hours
3
650 LP vs. 650 HTP - Aurora PSC U'flow: 7900 m /hr @ 24 m

650 ULP @ 302 rpm
650 HTP @ 224 rpm
Section
Ve l (m/s)
Velocity ^ 2
Ve l (m/s)
Velocity ^ 2
T
9.60
92
6.26
39
A
10.87
118
6.84
47
B
11.74
138
7.30
53
C
12.31
152
7.64
58
D
11.98
144
7.91
63
E
11.75
138
8.13
66
F
11.69
137
8.30
69
G
11.64
135
8.45
71
H
9.45
89
6.51
42
40
4000 hr Impeller Goal:

Predicted Life: Syncrude Wear Model
Pump:
650 LP
650 HT
Speed:
302 rpm
226 rpm
Inlet mid-vane diameter:
573 mm (22.56)
550 mm (21.65)
Inlet peripheral speed:
9.06 m/sec
6.45 m/sec
Outside diameter:
1455 mm (57.28)
1880 mm (74.02)
Vane length:
980 mm (38.6)
1726 mm (68)
Existing 650 LP Impeller wear life: 2500 3500 hours: Use 2500 hours
Using Dan Wolfes program and the above imputs:
Power
650 LP Baseline
650 HT Baseline
Predicted 650 HT
Life
3rd
251 hr/in
491 hr/in
9362 hr
2.5
229 hr/in
400 hr/in
8858 hr
2nd
207 hr/in
325 hr/in
7861 hr
41
4000 hr Throatbush Goal:

CFD Wear Analysis
8100 m3/hr @ 24 m:
650 LP @ 294 rpm
650 HT @ 225 rpm
650 HT shows lower velocity at midpoint and seal dam:

16 m/sec vs. 18 m/sec at midpoint
11.5 m/sec vs. 14.2 m/sec at seal dam
11.5 m/s
14.2 m/s
650 LP
18 m/s
650 HT
16 m/s
42

CFD Wear Prediction
Existing 650 LP wear life: 3100 3800 hr:
Use 3100 hour
Wear Calculation at Eye:

Relative part thickness at eye = 1.60
Velocity2 ratio at seal dam: (14.2 m/sec/11.5 m/sec)2 = 1.52
3100 hr x 1.60 x 1.52 = 7539 hours
Wear Calculation at Mid-point:

Relative part thickness at mid-point = 2.49
Velocity2 ratio at mid-point: (18 m/sec/ 16 m/sec)2 = 1.27
3100 x 2.49 x 1.27 = 9803 hours
43

Rough OD/ID Ratio Prediction
650 LP OD/ID: 1.83
650 HT OD/ID: 2.47
Relative wear:
Assume similar trend:
34.3m
= 2.31
14.8m
34.3
14.8
Predicted life:
3100 hr x 2.31 x 1.6 = 10,000+ hours
Relative wear
Relative thickness
1.83
2.47
*Chart used with permission, from

Prediction of Slurry Pump Wear and
Cost - Sellgren and Addie, WEDA XXV,
New Orleans, 2005
44
650 HT Adjustable Throatbush

The 650 HT features a full
face adjustable throatbush
(suction side liner)
Adjustment:
Increases head output
Increases efficiency
Reduces required speed
Reduces throatbush wear
Throatbush
(Suction Liner)
Adjustment Bolt
Intake Flange
Seal
Retaining Stud
45
Effect of Throatbush Adjustment

Site A
Site B
Site C
Not Adjusted: Local Wear
Adjusted: Even wear pattern
46
Effect of Throatbush Adjustment

Gouging Wear Reduced by 33 to 42%
47
Additional 650 HT Photos
48
Additional 650 HT Photos
49
650 HT Wear Photos (3331 hours):

Throatbush
Generally smooth wear pattern:
18 24 mm wear adjacent to expelling vanes
50 mm divot a 6:00 oclock near eye (175 mm thick new)
50

Throatbush
50 mm depression near eye at
6:00 oclock position:
175 mm thick new this location
Throatbush adjustments:
1 mm at 200 hours
8 mm at 1200 hours
10 mm at 2660 hours
19 mm
Total
Gouging wear can be reduced

with more frequent adjustments
Regardless, part would easily
achieve 4000 hours run time and
likely 8000 hours with rotation
and/or more frequent adjustment
51

Volute and Frame Plate Liner Insert
Very smooth overall wear pattern.

Relatively minor 6 mm x 100 mm
( x 4) groove on fpli at 4:00
oclock.
Wear rates in the volute proper
range from 5.4 - 10.7 mm/1000 hr.
Wear life based on linear
progression at worst case location
predicted to be 10,000 hours till
hole through
Highest wear shifted to
approximately 120 degrees beyond
cutwater
52

Volute Cutwater
Very smooth wear pattern:
No gouging or twin spiral,
serpents back wear pattern
58 mm wear (measured from

discharge flange):
17.4 mm/1000 hr wear rate
Greater than 12,000 hours till
hole through based on linear
projection of current wear rate
53

Impeller
Front shroud in excellent
shape:
no obvious rounding over of
expelling vanes
smooth seal dam with
circular ID:
797 to 800 mm (760 mm new)
Trailing edge of main

pumping vanes in good
shape:
very minor gouging at OD
54

Impeller
Main wear to
leading edge of
impeller vanes:
60 to 61.25
(72.4 new)*
7054 to 8598 hr
life projected to
20 vane length
* Both measured on longer pressure side of vane
55
650 HT Wear: Operating Point Revisited

Aurora Tailings
system now has
two modes of
operation:
PSV to tails
distributor
PSV to composite
tails
Approximately 6 m
more head to tails
distributor
56
Two distinct
system curves:
To CT on BEP
To TD 90% of
BEP at higher
head
57

Operation at slightly lower flow rate results in
flatter velocity and velocity2 profiles:
58

Operation at lower flow rates results in a shift in wear
from the design section to cutwater and beyond.
59

Volute two dimensional wear model limited in the
presence of large particles:
Does not account for distance from impeller to volute
Does not account for gravity effects
For a top horizontal discharge orientation with a

uniform velocity distribution around the volute and
the presence of large particles, one can expect a
shifting of wear from the uppermost, furthest away
design section to the closer protruding cutwater
and lower area beyond the cutwater.
60
650 HT Wear Summary

Based on observed wear at 3331 hours we are able to
provide revised, more conservative predictions:
Component
650 ULP Life

Observed
Requested Life
650 HTP Projected

Based on Observed
Impeller
2500 3500 hr
4250 hr
6635 8598 hr
Throatbush
3100 3800 hr
4250 hr
4250+ hr*
Volute
3500 5500 hr
8500 hr
8397 10264 hr
FPLI
4000+ hour
8500 hr
Same as volute
* Extremely conservative estimate. Double this value likely with more frequent adjustment
Obviously, the next goal is to provide further refinements

to achieve 8500 hours on all parts.
61
Conclusion
The desire to maximize
bitumen recovery has
resulted in larger screen
size to minimize rejects.
Along with this increase
in screen size has come
step wise increases in
wear rates.
To provide adequate wear
life one can expect to see:
Larger pumps of lower
specific speed
Innovative design
features
Advanced materials
450 L
650 LP
650 HT
62

A New Large Pump For Oil Sands

Uploaded by

Copyright:

Available Formats

A New Large Pump For Oil Sands

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

A New Large Pump For Oil Sands

Uploaded by

Copyright:

Available Formats

A New Large Pump for Oil Sands

Calgary Pump Symposium 2015

1984-1987: Application Engineer and Regional Sales

Calgary Pump Symposium 2015

Why the Need? HI Slurry Classification

Calgary Pump Symposium 2015

Why the Need?

Wear history has

Calgary Pump Symposium 2015

Variety of Pump Types

Ns (US) = ns (Metric) x 51.64

Calgary Pump Symposium 2015

Specific Speed, Efficiency and Pump Size

Calgary Pump Symposium 2015

Early Oil Sands History:

Installed mid 80s

Calgary Pump Symposium 2015

Early Oil Sands History:

Estimated 2015 System Curve

Max imum Flow

Gland Sealed Pump

Multiply by 51.67 to obtain Specific

600rpm (4640 m 3hr ) /(3600 sec/ hr )

Frame ( Rat ing

Min imum Flow

Calgary Pump Symposium 2015

8/20 15 Weir Miner als Au str alia ( PT C)

1997 Plant 5 Process Change:

Calgary Pump Symposium 2015

1997 Plant 5 Process Change:

Increased flow rates & head:

13,000 to 20,000 gpm

Calgary Pump Symposium 2015

1997 Plant 5 Process Change:

Material change required to

On two pumps, larger 500 STL

Calgary Pump Symposium 2015

Syncrude Aurora PSV Uflow:

(2) Warman 650 ULPs

Syncrude Aurora PSV Uflow:

Barely Class 4 Slurry:

Calgary Pump Symposium 2015

Syncrude Aurora PSV Uflow:

Throatbush (suction liner):

Frame Plate Liner Insert (back liner):

Calgary Pump Symposium 2015

Syncrude Aurora PSV Uflow:

Calgary Pump Symposium 2015

Syncrude Aurora PSV Uflow:

Calgary Pump Symposium 2015

Syncrude Aurora PSV Uflow:

Calgary Pump Symposium 2015

Syncrude Aurora PSV Uflow:

Substantial increase in wear life will require

Calgary Pump Symposium 2015

Effect of Specific Speed on Pump Wear

With permission, from Prediction of Slurry Pump Wear and Cost

Calgary Pump Symposium 2015

Effect of Suction Diameter on