Inte rnatio nal Jo urnal o f Sc ie ntific & Eng inee ring Re se arc h Vo lume 3, Issue 3 , Marc h -2012 1
ISSN 2229-5518
Decoding the NGTE/Calvert Axial Flow
Compressor Computer Program
Tonye K. Jack and Robin L. Elder
(
Abs tract— Several models are used f or the design and perf ormance evaluation of axial compressors – thickness, clearance, eff iciency and others. How ever, tw o principal models f or the overall perf ormance evaluation are used – in North America, the US National Aeronautics and Space Administration (NASA) or Steinke Model, and in Europe, the UK National Gas Turbine Establishment Model (NGTE) or How ell/Calvert Model. In all, there are essential relationships and connects in some of these models. A n earlier investigator had applied the NASA model to a major UK gas turbine manuf acturer’s equipment w ith usef ul results. The aim of this research eff ort w as to evaluate some of the major models, decode the NGTE Calvert computer program since some of these relationships are experimentally developed and not available in the open literature, f ind the connecting relationships, through new models development .
Inde x Terms— Axial Compressor, Axial Perf ormance, Compressor, Compressor Perf ormance, Compressor Tip Clearance,, Multistage
Co mpressor, Off-Design Perf ormance, Stage Stacking, Diff usion f actors, loss coeff icient, blade aspect ratio .
—————————— ——————————
n the original unpublished paper, the title was given as “ A
Yet to be Concluded Summary Evaluation of the Calvert
At Stall:
Program”. The current title depicts more accurately, the
T
1
x h
initial steps taken to fulfill the purpose of the research effort.
Given below is a brief summary of applicable relations from 
mea n
0.99T
h
the Calvert program [1]. Additional relationships derived
ref
R
(4)
A
from velocity triangles and the open literature, and steps in
the derivation are not shown.
R1
0.975 A
Based on the input data given, and by using the area data giv- en at the sections, the mass flow correction factors, and the temperature rise mass weighted mean value correction factor, the following relations are useful in arriving at the Multiply-
At Maximum Efficiency:
2
T
A
A
ing factors.
mea n
x RM
x R 2
Q MF
Qmea n
(1)
0.99Tref
0.975 AR1
0.975 AR1
(5)
T MF
0.975Tref
Tmea n
(2)
At Stall:
1 ARM
A
x R 2
A
x SM
A
x S 2
A
x S1
A
x SM
A
x S 2
A
x S1
0.99Tref
8
AR1
AR1
AR1
AR1
0.975
ARM
ARM
ARM
AR 2
At Maximum Efficiency:
At Maximum Efficiency:
(6)
AR 2
0.975 ARM
x hR
hS
(3)
AS 2
0.975 ARM
(7)
IJSER © 201 2
Inte rnatio nal Jo urnal o f Sc ie ntific & Eng inee ring Re se arc h 2
ISSN 2229-5518
At Stall:
The Carter rule as applied is given by:
Tmea n
0.99Tref
m
s
(12)
T MF Q MF
Where, m, is a function of the stagger angle (between 0 and 70
degrees).
(8)
m 0.2162 0.0008595 0.00002786 2
(13)
V
D 1.03 0.4 t
w
0.7t deH
c
s .V
c
(9)
L 0.02224 0.02436D 0.05D2
(14)
By rearranging “(9)”, the tangential velocity ratio in the Ca l- vert equation (factor f) is easily derived:
“(14)” applies for, D>0.244
For D<0.244, loss parameter is held constant, and, L = 0.0193.
Stator:
2.L
V D deH 0.7t
1.03
s s
cos
(15)
V 0.4
t
.s
(10)
c
Rotor:
In Jack and Elder [2], a factor of 2 was missing in the clearance
r
s
2.L
2
equation. This was an error in the derivation based on the a v-
erage rather than, the addition of the hub and tip displac e-
ment thicknesses as specified in the Koch and Smith [3] paper. The corrected and modified tip clearance relation as a function
cos
c
cos 1
(16)
of the loss coefficients in the rotor and stator is given by:
Where, L, is the loss parameter.
t h 1 1 rV1 sV2
1 2
p
2 2 1
*
h
(11)
It has not been possible to arrive at the exact values in some of
3
2U Vw2
Vw1
the figures in the output as displayed in the Calvert program„s
[1] Sample Printer Results Sheets 1 and 2. Furthermore, in
Sample Sheet 2, the correction for mass flow rate has probably
It is proposed to use the deviation angle proposed by Carter [4] as given in the Oldham [5] paper referenced in the pro- gram. An alternative deviation angle equation with improved accuracy has been proposed by Boyce [6] but complex to ap- ply, and will take additional computing memory. It might be interesting to combine the Boyce [6] relation (since this ac- counts for the Mach number effects, and the thickness to chord ratio) with the factor, m, in the Oldham [5] relationship. The accuracy of this is yet to be verified.
been done with a specific heat ratio of “γ=1.0”. If this is an error in the program, the actual corrected mass flow is 15 per- cent above its true value. How this affects the overall results is yet to be determined.
This unpublished research effort was conducted at Cranfield University in the autumn months of 1998. The authors wish to thank Cranfield School of Mechanical Engineering for provid- ing the facilities, particularly their specilised research library materials, and computer laboratories.
IJSER © 201 2
Inte rnatio nal Jo urnal o f Sc ie ntific & Eng inee ring Re se arc h 3
ISSN 2229-5518
P Pressure
T Temperature
V Velocity
A Area
Q Flow rate
h Aspect ratio
t Blade thickness
tp Tip clearance s spacing
D Diffusion c Chord
L Loss parameter
f factor
deH de Haller number
Greek Letters
Loss coefficient
α air angle
β Blade angle
δ deviation
δ* Displacement thickness
Subscripts
s stator
r rotor
1 inlet
2 outlet
SM mean stator
RM mean rotor
[1] W.J. Calvert, “User‟s Guide to the Howell Multistage Compressor Perfor-
mance PredictionProgram”, NGTE Memo 77023, 1977
[2] T.K. Jack and R.L. Elder, “A Modified Stage Stacking Method for Axial Flow Compressors Calculation, International journal of Science and Engineering Re- search , 2012, (Pending Publication)
[3] C.C. Koch, and L.H. Smith, “Loss Sources and their ma gnitudes in
Axial Flow Compressors,” ASME J. of Engr. For Power, vol. 2, no. 4, pp. 193-218, July 1976.
[4] A.D.S. Carter and H.P. Hughes, “A Theoretical Investigation into the
Effect of Profile Shape on the Performance of Aerofoils in Cascade, R
& M 2384, 1946
[5] R.K. Oldham, “Some Stage Design Data for Double Circular Arc Compressor
Blading” NGTE Notes NT 589, 1965
[6] M.P. Boyce, “Design of CompressorBlades Suitable for Transonic Axial Flow
Compressors”, ASME 67-GT-47, 1967
[7] M.V. Casey, “A Mean Line Prediction Method for Estimating, the Perfor-
mance of Axial Compressors”, Proceedings of the Imech., C264/87, 1987
[8] A.R. Howell, and R.P. Bonham “Overall and Stage Characteristics of Axial
Flow Compressors”, Proceedings of Imech., 1950
[9] A.R. Howell and W.J. Calvert, “A New Stage Stacking Technique for Axial
Flow Compressors”, ASME J. of Engr. For Power, 1978
[10] D.C. Miller and D.L. Wasdell, “Off-Design Prediction of Compressor Blade
Losses” Proceedings of the Imech., C279/87, 1987
Tonye K. Jack, registered Engineer and ASME member, has a Masters Degree in Rotating Machine Design from Cranfield University, and is currently a University Teacher in Port Hrrcou rt, Rivers State, Nigeria, teaching undergraduate classes in Mechanical Engineering. E -mail: to- nyekjack@yahoo.com
Robin L. Elder , Ph.D, Professor of Turbomachinery Design and Engi- neering, and former Head of Turbomachinery & Engineering Mechanics Department, School of Mechanical Engineering, Cranfield University, England. He is currently a Director of PCA Engineers Ltd, a Turboma- chinery Engineering Consultancy.
IJSER © 201 2