ASME

PTC 8.2-1 990

Centrifugal
Pumps

- .- -
."Y
.
ASME PTC*KB.2 90 I0759670 0083930 b I
 Date of Issuance:February 15, 1991

, . . - .

This document will be revised when the Society approves the

issuance of the
ne.xt edition,scheduled for 1995. There will benoAddendaissued to
PTC 8.2-1990. .. .. ,-

Please Note: ASME. issues written replies to inquiries concerning interpretation

of technical aspects of this document. PTC 8.2:1990 is being 'issu.ed with an
automatic subscription service.to the interpretations that wilj be issued to it
up to the publication of the 1995 edition..

ASME is the registered trademark of The American Society of Mechanical Engineers.

This code or.standard was developed under procedures accredited as meeting the criteria for
American National Standards. The Consensus Committee that approved the code or standard
was balanced to assure that individuals from competent and concerned interests have had an
opportunity to participate. The proposed code or standard was made available for public review
and comment which provides an opportunity for additional public inputfrom industry, academia,
regulatory agencies, and the public-at-large;
ASME does not "approve," ''rate,'' or "endorse" any item, construction, proprietary device, or
activity.
ASME does not take any position with respect to the validity of any patent rights asserted in
connection with any items mentioned in this document, and does not undertake to insure anyone
utilizing a standard against liability for infringement of any applicable Letters Patent, nor assume
any such liability. Úsers of a code or sta,ndard are expressly advised that determination-of the
validity of any such patent rights, and the risk of infcingement of such rights, is entirely their own
responsibility, . .

Pa<icipation by federal agency representative(s) or person(s) affiliated with industry is not to

be interpreted as government or industry endorsement of this code or standard. . .

ASME accepts responsibilityfor only those interpretations issued in accordance with governing
ASME procedures and.po1icie.s which preclude the issuance of interpretations by individual vol-
unteers.

No part of this document may be-reproduced in any form,

i . in .an electronic retrieval system or otherwise,
- .
without the prior written permission of the publisher.

. . . , .. .
- ,
. ..
Copyright O 1991, by
. THE
AMERICAN
SOCIÉTY OF MECHANICALENGINEERS
All Rights Reserved
Printed in the U.S.A.
., .
:
- _ . .. PERFOPMANCE
.BOARD,ON - .
TES," CODES PERSONNEL
. .. - _ . . ~

8.. *i
.. ,.a.. . . .

. ,
.
.
.J. S. Davis,Jr,; VicePresident
-.
'R.%rgensen,. Vice Chairman
W. O.Hays, Secretary

. . ...
. .

A. F. Armor W, L. Garvin R, P. Perkins

R. L. Bannister G. J. Gerber R. W. Perry
J. A. Booth R. Jorgensen A. L. Plumley
B. Bornstein . D. R. Keyser J. A. Reynolds ' -

W. A, Crandall J. E. Kirkland C. B, Scharp - .

H. G. Crim W. G. McLean J, W: Siegmund
J. S. Davis J. W. Murdock R. E: Sommerlad
N. R. Deming S. P. Nuspl J. C. Westcott
. .

. . . .
.. .
. t

,.

vi
 . . . .
CONTENTS

5,7.1 .................................................................................. 38
5.7.2 ................................ ................................................ 38
5.7.3 .................................................................................. 38
5.7.4 ................................................................................... 38
5,7.5 ................................................................................... 38
Complete Listing of ASME- Performance Test Codes
. .. ,-.
:*. ...........,.'........, 75

..

. .

. -

vjii
 satisfy TypeA, Type B, or mixed-type tests are given
1.3 This Code applies to the testing. of pumps utiliz-
in Section 4 of thisCodé. Descriptions of instruments
ing liquids or mixtures of miscibleliquids which have
and apparatus beyond those specified; but necessary
Newtonian viscosity characteristics.
to the conduct of tests under this Code; mav be found
in the ASME Performance Test Code Supplkments on
Instruments and Apparatus-(PTC 19 Series). If specific
1.4 If specific directions in this Code for any partic- directions in this Code for any. particular measure-
Uhr I-tIeaSUrement differ from lhose given in a refer- ment differ f r o h those given inthe PTC 19 Series,
ence code for similar measurements, the instructions the instructions in this Code shall prevail.
of this Code shall prevail. It is the intent of thisCode
that the meaning of all terms be understood and ap-
plied as defined in Section 2 of this Code. 1.7 The tests specified in
this Code may be con-
ducted in the manufacturer's shops, onthe user's
premises or elsewhere as agreed~úpon,provided such
1.5 This Code mandates testing prodeduresand ac-testsmeet. the requirements of~this-Code.
ceptable instrumentation for tests designated TypeA
and Type B. Subsequent reference in this Code will
be made to these as Type A andType B (see Section 1.8 Results of tests conducted in accordance with
3). Prior to the test, the parties to the test must agree this Code apply solely to the specific pump actually
1
 ASME.PTC8.2-1990' CENTRIFUGAL PUMPS

tested, The measuredtest data referonly to the pump

test system, The record of the test and the final test
report may include informationon the composite
unit, including driving and auxiliary equipment.
..-. 1.12 Measurement Uncertainty. By satisfying the
r
- .
,
:
,:
5
i .
.
instrument *accuracy'criteria.in Table I.I 1 and the ,
1.9 Tests of model pumps and/or substitute. liquids, balance of the procedural requirements of this Code,
if in conformance with all mandatory sectionsof this Types A and B testing will generally provide 95% or
Code, may be reported as tests conducted in accord- .greater confidence .that the measurement of the
ancewith this Code but, .as specified in parq. 1;6 of' testedparameters identified in.Table 1.I 1 will yield -
this ,Code,- pertain solely to that-.liquid and pump. results for which the bounds of the differences be-
Results extrapolated from tests of model pumps.and/- tween the reported test. results änd the true.values
or substitute liquids and presentedas another pump's are closely approximated as*follows: '

performance in -meetingguarantees, are outside ttie

scope of this Code. Although the method of exträp- Type A E
olation of the results to predict another pump's per- Discharge head -C 0.44% -C I .89%
Suction head +.0.91% 23.99%
formance may be agreed upon-by the-partiesto the
Total head [Note (I)] +.0.45%. r1;91%
test, such extrapolated results shall not be designated Efficiency +1.33% +2.45%
results-of a'test conducted in accordance with this Capacity -. -+0.81% . , :-+2,120/0 -
Code [see para. 3,4(b) and (gr andAppendix El. - .. .

.. Using PTC l!301)the foregoing values are based on

i . . . ' .. -
significant qpical--experience and precision indices
1.10 Reference by this Code to other codes or stan- for that experience. They are given in this Code-ás
dards shall. be. Enterpreted.to,impose the referenced generic valueswhich closely approximate-the bound-
code or standard in effect at-the timetheagreement ing values for type A and B tests. If specific test values
to conduct thetest is made by the parties to the test. äre required for the .overall.test,PTC 19;l shall be
~.
' . utilized to determine the specific test uncertainty.
. 1 .. ~.
This shall be done only by agreement of.the parties
1,I 1 Test. results adjusted for speeds outside- the to the test. Appendix C of this Code provides an ex-
range.noted in Table l.lsl- are outside the Scope of ample of-,the use of PTC 19.1 for pump test uncer-
this Code, Tests shall.be conducted at the specified tainty analys.is. .. . . . . L

2
CENTRIFUGAL PUMPS ASME PTC 8.24990

SECTION 2 -
 ASME P T C * 8 * 2 90 0 7 5 7 b 7 0 0083742 2 E

ASME PTC 8.2-1990 PUMPS CENTRIFUGAL

Abbrev.
Term Units Symbol Definitions
. .
. - 1 i . . -
'Totalpump head [ N o t e (2)] 'The algebraic.differencebe- H ft
tween the.total discharge
head and the total suction
head
discharge
Total
algebraic
The
head
sum the of Hd ft
[Note (211 pressure head,
elevation
the
.. . . head,and
velocity
at
the
head
the pump discharge referred
to datum (see Fig. 2.2)
Total
suction
head
The
algebraic
sum
of the. : H, ft
[Note (211 pressure head, the elevation
. .
head and the,velocity head at
the pump suction referred to
. ,
datum (see Fig. 2.2)
Pressure Head
head of liquid
equivalent
to H, ft
the gage pressure. This term
-. . . - . maybepositive(called
. ..
. . "head") or negative(called
~

.
.
~

. "lift")
when referredto
the
- .. . . pump datum (see Fig. 2.2) ..
Elevation
Vertical
head
distance
datum.
from H, ft
Can be either positive (above
datum). or negative (be1o.w
. - .datumj (see Fig. 2.2 and Figs.
5.7.1 through 5.7.5)
Velocity head Kinetic energy of fluid per H, ft
unit-mass'at the point of
measurement (see para, 2.7)
Vapor pre.ssure head.. . Vaporpressure ofthëfluidat HYP . . ft
the-pumping temper.atu_re ex- ,

pressed ás head ofliquid ..

being pumped during the test

Netpositivesuction headTotalsuctionhead-plusbaro- NPSH ft
metric head,-less 'the vapor . '

pressure head of the liquid

Net positive suction head NPSH available to the pump NPSHA ft
available referenced t o -datum (see Fig.
2.2)
Netpositivesuctionhead . TheNPSH of a pump at any NPSHR ft
associ-
'and
capacity
discrete
required
ated speed at which the per-
formance begins to
deteriorate .. .
Barometric head Atmospheric pressure ex- .. Ha, ft
pressed in feet of liquid
Total bowl assembly head Algebraic sum of the pressure Hb ft
. head, theelevation
headand
the velocity head determined
at the gage-connection, plus
the losses ,io equipment- be-
tween the bowl assembly and
the point at whichr the pres-
sure is. measured
torque--- -_I-_.--.- --.The-turning-moment-about ~~ - -T . ~-Ib-ftor-lbsin. ~ ~-
.the axis of a power transmit-
ting shaft
6
CENTRIFUGAL PUMPS ASME PTC 8-2-1990

2.3 Datum. The datum is the horizontal' plane 2.4 Pump. The pump is the compositemachinery
which serves as a reference for fluid energy termsfor which transfersmechanicalenergy to a liquidbe-
the pumptest. This plane usually passesthrough the Weensectionscalled suction- anddischarge.The
pump first stage impeller (refer to Fig. 2.2). pump includes all the components- required for the
. - . Horizontal
i Axial Flow
 ASME- PTC*K8-2 90 I0759670 OOBL94b T M .

CENTRIFUGAL PUMPS ASIWE PTC 8.2-1990

CENTRIFUGAL PUMPS ASME~PTC8.2-1990

SECTION 4 - .INSTRUMENTS.AND METHODOF

. .

MEASUREMENT

15
I
 ~ . .._
. ASPIE PTC*8.2 90 SlBl 0759670 O081951 3 IB1

lines, vent valvesshall be-providedatany high point .

4.7.1 The number of pressure taps will depend on
or loop crest to assure that instrument lines do not
the type of test. For Type A, four static pressure taps
contain any air or gas. Wh'ere~conditíonsprevent the i

shallbe provided spaced about the periphery

use of wet-tuBe type lines, dry-tube linesmust be;
Of the Pipe at the
suction and dischargemeasuringused. Oneor more dry-type pressuretaps .may be-
sections.Thepressure at the section i s to be takenused. For dry-type taps, lines must be
as the average of these four separate readings. .
used to provide visual assurance that no liquid exists
If any of the four readings differ by more than 1% in the line.
from the average, the cause should be found and the . i

"

discrepancy cohected, if possible. The four pressure

4.7.2 For Type B, a single suction and a single dis- 4.12 In applications with corrosive or hazardous
charge tap may be used(seeFigs.4.7.2,4.7.3, and fluids, complex mixtures, flocculent or dirty fluids, a
4.7.8through 4.7.10). limited or .continuous fluid purge (of inert gas or
17

"-=.-" - " -
ASME PTC 8.2-1990 CENTRIFUGAL PUMPS

Discharge

Undeiground discharge

III 7:
'.!
III

1 2 diameters

. -

FIG. 4.7.2 TYPICAL PRESSURE TAP ARRANGEMENT FOR

AN OPEN-PIT VERTICAL P U M P
CENTRIFUGAL PUMPS ASME PTC 8.2-1990
ASME PTC 8.2i1990 CENTRIFUGAL PUMPS

= .

" . .
. .
.. . .

.- .

Vent

FIG, 4.7.4 TYPICALPIEZOMETERRING MANIFOLD ARRANGEMENTS FOR

MEASUREMENTS OF HEAD USING CACES OR MANOMETERS

2
 r-

ASME

ASM E PTC 8.2-1990 CENTRIFUGAL PUMPS

Vent
(typical)

"

FIG. 4.7.7 TYPICALPIEZOMÉTER RING MANIFOLDARRANGEMENT

FOR MEASUREMENT OF HEAD USING USING
A'DIFFERENTIAL MANOMETER

. .
CENTRIFUGAL PUMPS ASME
PTC 8.2-1990 -
 ASME PTC 8.2-1990 CENTRIFUGAL PUMPS

Vent I

I .

Drain-, Y+
I
I"" I L.

(1 1 See oara. 4.7 fnr nine dimensinne

rience positive or negativë pressure, the manometer lable 1.I 1 and throughout.this Co&. - .

24.
ASME PTC 8,2-1990 CENTRlFlJGAL

Gage

Purge connection

Shut'offvalve

or instrument as

Slope gage line away from

gage 1 in. per foot

Air or water supply

(pressure must be twice that of
the maximum pressure to be
measured)

* 7
sure to notless than
twice or more than three times that of
maximum pressure to be measured)

Gate (shut-off) valve of material corrosion

resistant to pressure medium
IMPORTANT Always open this valve after purge is on.
Always close valve before purge is turned off.

GENERA'L NOTE:
For water purge, substitute a small-rotometer for purge
flow regulator. ,

FIG. 4.12.1 A TYPICALAIR OR WATER PURGE SYSTEM

26
.FIC..4.12.2 A TYPICALSEAL WITH INTERMITTENTPURGE .-

L/
 ' ASNE P T C * P - Z 90 W 075qb70 008Lqf72 8

ASME PTC 8.2-1990 .. CENTRIFUGAL PUMPS

 ~.

: ASME P T C g p . 2 - 98 m 0759670 00813b3 T I

CENTRIFUGAL PUMPS ASME PTC 8.2-1990

tachometer generators, optical or electricalrevolu- Not al¡ arrangements will -be suitable for all types of
tion pickup and frequency counters. In the case of a pumps. Modifications of these arrangments will be
pump driven by an ac motor, the pump speed may acceptable as longas the desired resultsare achieved.
be dètermined. from observations of the mean fre-
quency and motor slip.
4.41 In.Fig. 4.40.1, theflow.is suppliedfrom a pit
having a free liquid surface.The NPSHÀ can-bevaried '

4.36 A stroboscope may be used to determine ro- by changing the level in the-suction pit; The~maxi-
tative speed provided: mum NPSHA~attainableis with the liquid at its high-
(a) thé stroboscope shall.be synchronized to line est pit level; Losses in the suction pipe will reduce
frequency and the slip counted; or the net reading at .~ all levels,
(h) the stroboscope shall have been demonstrated
. .
to b e adequate to resolve the speed of rotation to
0.01% if direct readings are taken. 4.42 Figure 4.40.2 utilizes-a constant level suction
pit, with a throttling valve to change th&NPSHA. Cav-
itation at the valve should be avoided.- Flow into the
ASME PTC 8.2-1990 CENTRIFUGAL-PUMPS

- "

. . - ~.

.. . . . .
- FIG. 4.40.2 NPSH TEST ARRANGEMENT . " .
_ . . .- . ~.

. . ... - .
30
CENTRIFUGAL PUM'PS ASME PTC.8.2-1990
. ASME,PTG 8.2-1990 CENTRIFUGAL PUMPS

~.

NPSH constant

.T
. .

TestCapacity +

FIG, 4.44.1 CONSTANT NPSH

..
,-- Test range for each NPSH (see Fig. 4.44.1)

H
P
P
0.8H

. . FIG. 4.44.2 CONSTANT NPSH .

CENTRIFUGAL PUMPS - ASME PTC-8.2-1990

4.51,l After an initial calibration, verification of crit- 4.51.4 It must be recognized that some flow meters
ical dimensions, general condition, cleanliness, and are too large to calibrate using standard volumetric
pressure tap conditions is required at intervals not or gravimetric methods. In these cases, a calculation
greater than 12 months for. meters without moving may be usedbased on-the theory as given in ASME -

parts such as venturis, nozzles, and orifice plates. “Fluid Meters, Their Theory and Application.” Every
More frequent intervals of verification may be nec- attempt shall be made to verify this calculation by
essary for abrasive or dirty liquids, Additional verifi- use of other flow-measuring devices such as a pitot
cations ór calibrations may be performed by prior traverse, comparison to another calculated meter; or
agreement of the parties to the test. dye injection. . ~

A recalibration is required when dimensional ..

changesare found which would affect the original.
calibration. These metersshall be calibrated. with 4.52 Electric Motors - Drivers
their associated piping unlesS.thepiping conforms to
4.52.1 Type A Test.~Calibrated electric motors
the straight run distances recommended in the ASME
“Fluid Meters, Their Theory and Application.” When shall havean injtial calibration byactual load testing,
or ifthis is impractical, calculation bythe segregation--
flange or pipe taps are utilized, they are considered
of-losses method may~~be used. Thesë shall be done
as part of the meter, and the actual pipe or flanges
in accordance with the ANSI/IEEE Standard. (see para,
must be included in the calibration. - -
4.33) and performed on the actual motor. Calibra-
tions based on tests of a similar motor or one of the
4.51.2 Meters with mechanical movements and same manufacturer, design, or frame sizeare not-ac-
bearingssuch %sturbine metersand paddle meters ceptable.Recalibration or recalculation shall be done
are more susceptible to errors due to wear, A I year in the event of rewinding or-major overhaul.
recalibration interval is required. Since metersof this
SECTrON 5 - COMPUTATION. OF RESULTS
ASME .PTC 8.2-1990 CENTRIFUGAL. PUMPS

Datum -,
5.8 On a ve&ical pump, similar to those indicated in 5.11 Net Positive .Suction Head Characteris-
Figs, 5.7,4 and 5.7.5, the entrance losses to the pump^ tics. NPSHR is determined by utilizing dafa from
are charged to the pump. The average velocity head NPSH tests conducted in accordance with directions
of the flows.in the sump are small enough to be ne- provided i n paras. 3.12, and. 4,38through 4.44 of
glected. Therefore the total suction head is equal to this Code. ;.
, -

H,
5.1 i .I ln the case of a constant capacify test, as test
~

~NPSHAis reduced, the head generatedby the pump

will begin to change and break away from the linear
portion of the plot of' head versus test NPSHA. As
H,=O+H,+O=H,
NPSHA is further reduced, pump performance will be
impaired. The method to be used i n interpreting the
test results, and in establishing the breakaway point
5.9 Total Bowl Head. Total bowl head is defined in (critical point), or a magnitude of deviation from non-
para. 2.2. As indicated in Fig, 5.7.5, the measuring cavitating conditions, shall be mutually agreed upon
section is located immediately downstream of the in advance by the parties ti, the test.
pump diffuser section. Therefore, total bowl head is
CENTRIFUGAL PUMPS ASME PTC 8.2-1990
CENTRIFUGAL PUMPS -ASME PTC 8.2-1990
. .
. . . .

APPENDIX A
SUMMARY OF AGREEMENTS BY.. TH-EPARTIES-
TO THE TEST '. ~ -.

ThisAppendixcoversal¡ the -items noted in this code to bedeterminehand

agreedupon by the parties to the test, prior' to the test, Compliancewith the
format of this listing is nonmandatory.
ASME PTC 8.2-1990 -
CENTRIFUGAI PUMPS

Topic Code Paragraph Agreement

3.8 Number of readings at each test point (for Type
A tests only)

3.9 Testing at shutoff and at capacities below 25%

of BEP
3.1 1 Extent of pitot
tube traverse, if applicable

3.12 Are NPSH tests required?

3.1 3 If complete pump cannot be tested, indicate

altered condition

3.14 Indicate how -omissions 'in para.

3.13
are
accounted for

4.32 Option of using' non-cakbrated motor for Type

B tests with motor efficiencies as published by
manufacturer

4.40 Describe NPSH test

4.42 Arrangement ór indicate appropriate figure,

number of this Code

The undersigned parties herewith approve the above listed agreements.

Representing - . Signature Date

 ASME P T C * d - 2 90 W 0 7 5 9 b 7 0 O O B L 9 7 9 3

CENTRIFUGAL PUMPS ASME.PTG 8.2"1990

. I _

. .

Efficiency (q) -- 3960 (P) 1.341

For Discharge Head:

A, -= IT/^) D,Z (area of discharge pipe)

- 47
 ASME PTC*K8-2 90 0 7 5 9 b 7 0 0083780 T Is

ASME PTC 8.2-1990 CENTRIFUGAL PUMPS

The following equation is used to calculate the absolute bias limit:

B+ = [(e,, x B,,Y + (e, X B,), + (e, X B,), + (e, X ~,)211/2

In similar fashion, the absolute precision index is calculated as follows:

~.
- $fi..=[(e,, x S,,)' + (e, x Sp)2+ ( 8 , x 'SQl2+ (OD,, x (2)

Uncertainty is calculated by the following:

. .

U,,, = [BHd + ( L ,x SHJl (3)

. -

. .
Also,

For Discharge Head:

1
es = -
59.4
= 0.0168 ft3/lb

=
(5~605)
= 0.108 s/fi2
(1.27)2 (32.2)
aH. =, - 2Q' eDd =
2 (5.605)2 = -0.957
dDd ( I T / ~ )-~ D ~ ~ R ,( ~ / 4 ) ~ ( 1 . 2 7 ) ~ ( 3 2 . 2 )
\
. .

The following list of bias limits and precision indices is based on experience and is made for the independent
parameter of the preceding.equations for head and efficiency. Degrees of freedom is greater than 30. ~-

Parameter Bias Limit Precision Index Sensitivity (O)

-. P2
k1.75 psi = k252-psf +2,0 psi = +288 psf 0.0168 f t y b . .

P k0.004 Ib/ft3 . . ro.002 Ib/ft3 -26.01 W/lb

Qt "20.0 gpm = k0.0446 ds r 2 . 0 gpm = k0.00446 cfs . 0.108 s/ft2

Dd +0.001 in. = +0.0000833 ft O . . -0.957

These sets of values ace now substituted in Eqs. (1) and (2) to yield the discharge head bias and precisionlimits.
 . .

ASME P T C * 8 . Z 90 8759b70 0081982 3 Sll

ASME PTC 8.2-1990 PUMPS CENTRIFUGAL

For Suction Head: ~.

A, .= . (.rr/4)D,Z (area of suctionpipe)

_. -
”
UPl = - -- U.Ult30 IL-/Iu
JPl P 59.4

Bias Parameter Limit Precision Index Sensitivity (O)

pi 20.125 psi = C 18 psf 20.10 psi = 214.4 psf 0.0168 ft’/lb
P 50.004 Ib/ft’ 0.002 1wft3 -0.664 ftYb
Qt 220.0 gpm = 20.0446 cfs k2.0 gpm = k0.00446 cfs 0.0661 s/ft2
D, -1-0.001 in. = +0.0000833 ft O - 0.51 1

B&,, = [(0.0168 X 48)’ + (0.664 X 0.004)2+ (0.0661 X 0.0446)’ + (0.115 .X 0.0000833)2]1R ..

B., = 20.302 ft
SHS = [(0.0168 X 14.4)’ + (0.664 X O.ÖO2)’ + (0.0661 X 0.00446)2 + O’]‘”
SHs = 20.242 ft . .

For averaged result of 3 readings, N = 3. So,

SH, = 0,242/33/2= k0,140 ft

U,,, = [0.302 + (2 x 0,140)l = 50.582 ft at - 99% coverage

..or
U,,dH, = 0.582139.57 = 21.47% at _ 99%
. .
-
_ coverage .

U,, = [(0.302)* $. (2 x 0.140)211h = 20.412 ft at - 95% coverage

or .
- 95% coverage
.

U,JH, = 0,412/39.57 = +- 1.04%.at .

For Total Head:

All sensitivities are 1 by inspection.

B, = [(I X 4,23)2 + (1 X 0,302)2]1/2=. 24.24 ft

S.,,= [(I X 2.79)’ + (1 X ò.140)2]1n= 22.79 ft

50
CENTRIFUGAL PUMPS ASME PTC 8.2-1990

U,,, t- (2 x 2.79)] = k9.82 ft at -99% coverage

= [4.24 .
or
u , ~ H=, 9.8211 505.93 = +0.65% at -99% coverage
- _ . .

U, = [(4.24)2 + (2 x 2.79)2]1R= i-7.01fi at -95% coverage

- I ~ . ,.

or . -

UJHt = 7.0111505.93 = -C0.47% at -94% coverage . r

. .

. .

For Test Efficiency:

or
0.0100/0.7503 = i:1;33% at -95% coverage
51
ASME PTC 8.2-1990 CENTRIFUGAL PUMPS

Uncertainty Calculations for Flow, Head, and Power

Incorporating Corrections for Speed (Type A.Test)

The.following list of bias limits and pre.cision indices is based on experience, and is made for the independent
parameters of the succeeding equations, Degreesof.freedom is greater than 30. The correction for speed must
be used to.calculate its effect.on còmparisons of test parameters to design parameters. Theaffinity laws provide
the equations needed for this analysis.

@ForN,(tested rpm)
X = 3600 rpm and instrument is 20.1%
Absolute Bias Limit = k3.6 rpm
Absolute Precision Index =. + 3 rpm

*For P,(tested power)

X = -905.17 kW and instrument is -1-1.O%
Absolute Bias Limit = -C9kW
Absolute.Precision Index = “ 0 . 5 kW

. .
.For Corrected Flow .
Q = (NINt)-Qt

where
Q= corrected flow
N = design speedl(3560 rpm)
Nt = test speed ..
Q, = test flow. . - ~

-aQ= - Qt . O N = - - 251 5.70 - 0.699 gpmhpm

aN Nt 3600.0

. -a Q=- - N eQt = 3560 - 0,989

-
3600

Parameter Bias Limit Precision Index Sensitivity (e)

‘N O O 0.699 gprnhprn
N -1-3.6 rprn 23 rprn -0.691 gpm/rpm
Q “20 gpm -1-2 gpm 0.989

B, = 10’ + (0.691 x +
3.6)’ (0.989 x 20)’I’R = 19,94gpm *
+ +
So = [O2 (0.691 X 3)2 (0.989 X 2)2]‘R= k2.87 gpm
N = 3 for.three. readings, so S, = 2.8713’R = +I .65gpm

52
 ..

ASME PTC 8.2-1.990 CENTRIFUGAL. PUMPS

-BP= - N3 (3560)’
ept = - = 0.967
apt (3600Y .

on Limit Bias Parameter (O)

N O O 0.738 kW/rpm
Nt k3.6 rpm “3 rpm -0.729 kW/rpm ,,
Pt k9.00 kW +0.50 kW 0.967
. .
. .

Bp = [O2 + (0.729 X 3.6)‘ t. (0.967 x 9.00)’]ln = k9.09 kW

S, = [O2 + (0.729 X +
3)2 (0.967 X 0.50)21”2 = k2.24 kW
N = 3 for three readings,’so S, = 2.24/3’R = k 1.29 kW

UNCERTAINTIES
Substituting the above bias and precision limits in the uncertainty equation provides the total test uncertainty,
including the correction forspeed. (Note: Underlined uncertainties are the values used in para. 1.12.)

.For Corrected Flow

U, = [19.94 + (2 x 1.65)] = 223.24 gpm at -99% coverage

or
U,,dQ = 23.2412407.70 = k0.93% at -99% coverage

U, =‘ [(19.94)2 + (i -x-1 . 6 5 ) ~ p= 220.21 gpm at -95% coverage

or
U,JQ = 20.21f2487.70 = 50.81% at -95% coverage

.For Corrected Total Head

.. U,, = [5,09 + (2 X 2-12)] = k9.33 ft at -99% coverage

or
UAJH = 9.3311472.7 = kÖ.63% at -99% coverage

U, = [(5.09)z + (2 X 2.12)2]’” = 26.62 ft at -95% coverage

or-
U,JH = 6.6211472.7 = 50.45% at -95% coverage

*For Corrected Discharge Head: .: ..

U,,, = L5.12 + (2 X 2.14)l = 29.40 ft at -99% coverage

or
= 9.401151 1.3 = -+0.62% at -99% coverage
UADJHdc

URSS = [(5.1212 + (2 X 2.14)211n = 2.6.67 ft at -95% coverage - .I

or
URSJHdc= 6.6711511.3- =
.
kO.44% at -95% coverage

54
CENTRIFUGAL PUMPS- ASME PTC 8.2-1990

.For Corrected Suction Head:

or

or
= 0.351/38.4 = *0.91% at -95% coverage
URsJHsc

.For Corrected Power

U
,, = L9.09 + (2 X 1.2913 = * I l .67kW at -99% coverage
or
U,,dP = 11.67/875,33 = ?1.33% at -99% coverage

U,, = +
[(9.C19)~ (2 x 1 .29)2IlR = a9.45 kW at -95% coverage
CAVITATION
 ASME PTC 8.2-1990 CENTRIFUGAL PUMPS

inlet with stroboscopic illumination. As a practical blade inlets; frequently, model test pumps experience
matter,suchobservation is limited to cold water a leading edge laminar separation whereas the full
pumps and on a sizescaleagreed to between test scale may not. The physical forms and scaling laws
parties. Therethe location, cause and observed form of cavitation development of these two flows are.
of cavitation are readily visible from inception to de- known to be different. From this it would seem very
veloped states. desirable to carry out model cavitation tests at the
Because almost all pump operation occurs with highest possible Reynolds number.
some extent of cavitation leading eventually to ma- The concentration of free-air microbubbles has
terial erosion, an alternative type of test has been been shown to have a major effect on cavitation in-
used in which soft coatings are applied to blade inlet ception and even on developed cavitation for special
regions. These coatings may be paint, dye, electro- cavitation test bodies, shear flows and propellers. The
platings, or in one case frangible microcapsules of normal amount or fraction of volume occupied by
dye which may then be eroded by cavitation within these microbubbles is still very small, typically being
only a few minutes. The locations and ektent of cav- less than 0;OI %, and is thus far removed from' being
itation are then "witnessed" by the coatings but it is a %vo-phase" flow inwhich significant performance
problematic whether such tests can be used to pre- changes can occur.-Pump cavitation is of course sub-
dict actual machine life. ject to the same physical phenomena, but perhaps
because of the focus on operation with a significant
amount of cavitation, the "scaling" effects discussed
D4 Forms of Observed Cavitation. The physical ap- äbove have not been as controlling as for these ofher
pearance and forms of cavitation at inception and at flows.
more developed conditions of operation may differ
widely; cavitation may occur on the bladesurface
either asan attached cavity or as a disconnected D5 lmplicatidns for Pump Testing. The accepted
stream of travelling, growing and collapsing bubbles, scaling rules for cavitation in general, includin'g the
Cavitation also occurs in the blade tip clearance (if thermodynamic effect, make useof the difference be-
any) and wearing ring leakage flows, and usually then tween upstream total head or pressure and the head
.as clouds of bubbles, The "attached" cavities vary or pressure within the cavity. The cavity pressure dif-
widely in form; the cavity itself sometimes appears fers from the vapor pressureof the surrounding liquid
asan envelope o f small recirculating cavitating bub- by the-presence of the gas partial pressure, and the
bles, as. a "frostytr surface of more or less. wedge vapor pressure itself there may not be the sameas
shape seen from above with apex- upstream, and in that of the upstream bulk liquid if substantial cooling
some cases as a two-dimensional nearly glassy sheet of the growing region of cavitation occurs (this is the
springing from the leading edge. thermodynamic effect). The number of the growing
Several phenomena are believed to govern these cavitating regions/bubbles is related principally to the
. forms, chief of which is the Reynolds number, and free microbubbles concentration, which is itself in-
second is the air content of the liquid. The total air directly related to total air content, The normal mea-
content of liquid can be measured with a Van Slyke sure of suction pressure or head is-the suction total
apparatus. For example:the tofal air content-contains head minus the vapor pressure of the incoming pure
both free air existing in the form-of microbubbles 5 fluid or NPSH. For the reasons mentioned, this vapor
to 1O0 micrometers diameter typically and dissolved pressure may differ from the'one actually governing
gas. Although both are important for cavitation, the the development of cavitation.
free- microbubbles are believed to be the more ím- All these complex physical phenomena suggest
portant, A standard test-apparatus for microbubbles that a single physical scaling law is unlikely to cover.
is not yet available; for this reason, the total gas con- all cases of interest and this has turned out to be so
tent is often recorded so that relative degree of gas in a wide variety of situations for cavitation inception
saturation of the liquid is at-least known. as well as for cavitation erosion. There are some usë-
~ -Both the-Reynoldsnumber' and surfacmrorrgh'n-ess" - - ful -guidelin-es;--höw~~ëi: thãt"cãn be in'fèrt%?d
-from -
determine if a laminar .or turbulent flow prevails at these phenomena applicable to cavitation or NPSH
testing of pumps, as follows.
(a) Cavitation testing should be dqne at the highest
"

'Here ant consisfent measure of Reynolds number is.satisfactör'y, Reynolds number possible with available speed and
e.g.,-fnlet tip speed times blade thicknesskinematic viscosity. power limitations.

58
 ASME PTC 8.2-1990 CENTRIFUGAL PUMPS

~ - .. ~
~~ -
. TypicalExamplë Öf CãÏculations Relating a
Model Pump to a Prototype
A single-stage model, when tested, delivers 3,920
gpm against a head of 3 2 0 f t at 1,825 rpm and, with
a total suction head of 1.45 ft, hasan impeller di-
ámeter of 1.5 ft.
The largerprototype pumphas been
. - designed with a 6.8 fi diameter impeller and will pro-
~.
(b) The prototype pump capacity .-
. duce a head of 400 ft. .
..The intent of the example is to check the speed,
: capacity,and suction head for the above prototype,
. Applying the above relationships:
~

I fi
"
- I .~ .

(c) The prototype pump power

.= 1,825 E
(5)
NOTE: The above equations assume that the efficiency is the same
for prototype pump and model. . -
, " n = 450.1 rpm.
As noted above, the efficiency of the model may
not be the sameas the prototype pump. Exact hy-
draulic similarity will not be realized unless the rel-
ative
surface
roughness of the ,impeller
and pump
casing surfaces are 'the same. However, if the abso-
lute surfaceroughness in the model and prototype
Q = Q, (:)'G
pump are the same, the efficiency of the model will
be lower than the larger prototype pump, Generally,
it is not practical to model running clearances; there-
fore, the model effidiency can be further reduced; The
= 3,920 (:)E
degree to which the efficiency is reduced must be .
mutually agreed to by the parties to the test.
The efficiency of the pump model can then bè es-
timated by using the Moody Formula Q = 90,070 gpm .

' -
1-11
Ill-= (;)"(;y -.
The prototype will run at a speed of 450 rpm de-
livering 90,000 gpm against a headzof400 ft.
To checktheseresults, it will be noted that the
The
exponents n and should be
developed from Teed Of the model is
test data for a given type of pump on the basis of an
. adequate number of model and prototype tests. The fi"
N,,= n, - -
- 1,825 v?$E
___ = 1,510
value of the exponent n has been found to vary be- 32O3I4
tween O and 0.26, depending on the relative~rough- I

ness of the model to prototype surfaces and other and the of the P r o W P e Pump Will be
factors. In utilizingthe foregoing, it is recommended
that the parties to the test carefully review the data
' which
from the exponentsbeen
have empirically de- VG = 450.1
N, = NF vzpö '=. 1,510
veloped. 42O3I4 .

62

. ~ . . .
 .. . ..

CENTRIFUGAL PUMPS

The suction specific speed S for the model which

SA
should be the same as that of the prototype pump,
assuming the usual water femperature of 80°F as a
maximum probable value, is

NPSHA. =
n
nm
/-(v y ‘’?
r) .

\ I

where

NPSHA-= 42.78 ft . .

144
= -(14.7 - 0.5)
62.4 -
+ 1.45 = 34.22 ft and therefore

144
H, = 42-78 ”
(14.7 - 0.51.
62.4

1,825-
SA, =
(34,22)3r4

The prototype unit shouldperformsatisfactorily

with a positive suction head of 10.01 ft.
Further reference to model testing may be found
._
in PTC-~1
9.23 (GuidanceManuai-on Model Testing).. .

Assuming SA = SA,, compute thevalue of the pro-

totype pump NPSHA
. .

. -
 ASME P T C * d . Z 90 I0759b70 008199q T I

CENTRIFUGAL PUMPS ASME PTC 8.2-1990

ever, is affected.when ;he vorti'cesare of such a

magnitude that the entrained air blocks a portion of
the water passage. In such cases, unit shutdown will
be mandated bythe amount of vibration being ex- F5 As indicated, notall flow disturbancesare ob- .

perienced long before performance is affected. How- vious.Consequently, attention to any~and all signs
ever, good practice dictates that intermittent and that would signal the presence of such disturbances,
sustained air entraining vortices be eliminated before and the immediate correction of same, is the only
the test isway
done. in satisfactory
which.
be can
results obtained.
 CENTRIFUGAL PUMPS ASME PTC 8.2-1990

CONVERSDNS TO SI (METRIC) UNITS

. -

Multiplication Conversion Quantity Factor

- .
Acceleration ft/sec2 to m/s2 3.048* E-O1
- Linear standard gravity to m/s2 -65*
9.806 E+OO

in.2 Area to mz 6 6.451 E-O4

ft2 to m2 304*
9.290 E-O2

Density Ibm/ft3 to kglm3 846 1.601 E+Ol

Energy, work, Btu (IT) to J E+03 056 1.055

heat fi-lbf to 1 818 1.355 E+OO

Flow rate, Ibm/sec to kg/s 924

4.535 E-O1
mass Ibm/min t o kg/s 873
7.559 E-O3
Ibm/hr to kg/s 979
1.259 E-O4

797
6.242 E-O2

Specific weight Ibf/ft3 to N/m3

(force)

Temperature "F to "C

interval

67
 ASME PTC 8-2-1990 - CENTRIFUGAL PUMPS

ultiplication Conversion Quantity

Temperature, "F to "C t, = (t, - 32)/1.a
measured '. .
- .

Torque . - -Ibf-in. to N (m) 1.129 848 - . E-01

I - .
. . Ib%ft.to N .(m) . ~ - 1 . 3 5 5 818 i - ' Ë+O0

ft/sec Velocity, t o m/s 3.048* E "01

Viscosity,
centipoise to Pa (s) 1' E-O3
dynamlc - poise fo Pa (s) .. 1' E-O1
Ibm/ft-sec to Pa (s) 116.44 8 8 E+OO
Ibf-sec/ft2 t o Pa (s) 4.788
oz6 . E+OI

Viscosity, ' . - 1 .centistoke to mZ/s I* E-O6

stoke kinematic to m2/s I* E-O4
ft2/sec t o m2/s 9.290 304* E-O2

gallon
Volume (UStoliquid) m3- 4123.785 E-O3
ft3 to m3 2.831 685 E-O2
in.? t o m3 1 . 6 3 8 706 E-O5
to liter m3 I* E-- O 3
GENERALNOTE: The factors are written as a number greater than one and less than ten
with six decimal places. The number is followed by the letter E (for exponent), a plus or
minus symbol, and two digits which indicate the power of 10 by which the number must
be multiplied to obtain the correct value.

Example:

3.785412 E-o3 i s 3.785412 x 1 0 - 3 or 0.003785412

NOTES:
*Exact relationships in terms of the base units;
**gpm = US gallonsperminute.
CENTRIFUGAL PUMPS ASME PTC 8.2-1990

APPEN-DIX H . .

ADDITIONALGUIDAN'CE ON NPSHTESTING . .
 \

ASME PTC 8-2-1990 CENTRIFUGAL PUMPS

Noncavitating Cavitating

. .
NPSHA 2 NPSHA

FIG. H2.1 GENERAL NPSH VS. HEAD PERFORMANCE (CONSTANTCAPACITY)

ues of test NPSH are large enough, decreasingNPSH Determination of breakaway tangency points re-
incrementally will provide straight line results over a quires that small increments of capacity be recorded
range of NPSH values. for both the noncavitating head capacity curve and
In some pumps, especially those of high specific. the constant NPSH test condition in the anticipated
speed (high capacity, low head), it may be advanta- zone of breakaway. This is required in order to es-
geous to determine both head and horsepower char- tablish the performance characteristics well. enough
-".
" - acteristics during-constant-capacity NPSM-tests.-The-.--
"
to-preeisely-determine~the-breakawaypoint.---------
points of tangency for both head and horsepower are Figure H 3 depicts a common method of reporting
obtained as described for head in the preceding par- NPSH requirements based on tests suchas these. The
agraphs, In some cases, the point of tangency for NPSHR is identified as the NPSH at the point of tan-
horsepower may be observedto occur at conditions gency. Identifying this point depends on the same
different from the point oftangency for head as NPSH basic criteria as described in para. HZ.
is reduced, In such cases, characteristic curves'as From a practical point of view, it may be appro-
shown in Fig.-H2.3 can be derived on both head and priate to report NPSHR values at incrementally lower
horsepower bases. or higher capacities than those at the points of tan-
Some pumps, again especially those of-high sp.e- gency establishedby test. As stated in para. HI, PTC
cific speed (very low head per stage) are very sensi- 8.2 requires that the basis for reporting be clear.
tive to flow disturbances on the suction side of the
pump, .Repeated test runs in both increasing and de-
creasing NPSH increments may be required. H4 Analytical Techniques. Modern instrumenta-
tion, used directly, or coupled to electronic measur-
ingand- recording systems, provides atleast two
H3 Constant NPSH. If NPSH is maintained at a con- characteristics. which may be used to enhance the
stant value as flow is incrementally increased, a fam- reporting and interpretation of .NPSH test results.
l y of performance curves will result as depicted by First, test paraméters.may be varied in small incre-
Figs. 4.44.1 andH3.Forselectedvalues of NPSH, ments with accurate results, Secondly, data may be
there will be a limiting capacity. Attempts to increase taken rapidly with direct recording.which results in
capacity beyond this limit willresult in deterioration greater volumes of reliable d,ata. Such data may be
of head with little or no capacityincrease.This is reduced to 'empirically based higher order equations
- - - shown by the dashedhead capacitycurvesatvaried representing pump performance. While both manual
NPSH values on Fig. 4.44.1 and the constant NPSH and computer based techniques can be used to de-
performance curies on Fig. H3. The point of tan- rive the equations, reliable computer based cutie fit
gency, or break-off, is the point at which .the char- . codes are available to perform this function,
acteristic for a specific NPSH value initially separates Selection of a curve fit methodólogy must b e done
from the "noncavitating" headcapaci.9curve (see carefully, and correlation to test results must be con-
Fig. H3). firmed. Merely choosing a higher-order curve fit will

70
ASME PTG.8.2-1990 CENTRlFUGAL.PUMPS

P
f I _

-. -.

. .

Operating range

'I
0

. .

NPSHR

NPSHR 1%

NPSHR 2%

NPSHR 3%

. .

Capacity

FIG. H2.3 NP-SH-RAT BREAK-OFF AND ALTERNATE HEAD REDUCTiONS

.. . . .r. .
.. . . . , " ...
i
:
;,.
.-
 ~ ~~

ASME P T C r 8 . i i o mi 0759670 0 0 8 2 0 0 ~L l a i

ASME CENTRIFUGAL PUMPS 8.2-1990

76
PTC 21
PTC 22
PTC 23
PTC 23.1
PTC 24

PTC 25.3
PTC 26

PTC 28

PTC 29

PTC 31

PTC 32,l

PTC 32.2

PTC 33

PTC 33a

PTC 36
PTC 38

PTC 39,l

PTC 42
 A complete list of all PerformanceTest Codes appears
at the end of this book.

While providing for exhaustive

tests, theseCodesare so drawn
that selectedpartsmay be used
for tests of limited scope.

This document is printed

on 50% recycled paper.

50% RECOVERED PAPER MATERIAL

means paper waste generated after the
completion of the papermaking
process, such as postconsumer
materials, text books, envelopes,
bindery waste, printing waste, cutting
and converting waste, butt rolls,
obsolete inventories, and rejected
unused stock.

ISBN #0-7918-2085:8 C00590

. .