Numerical and Experimental Investigation of Trailing Edge - - PowerPoint PPT Presentation

▶
numerical and experimental investigation of trailing edge
SMART_READER_LITE
LIVE PREVIEW

Numerical and Experimental Investigation of Trailing Edge - - PowerPoint PPT Presentation

Feb 21, 2023 25 likes •257 views

Numerical and Experimental Investigation of Trailing Edge Modifications of Centrifugal Wastewater Pump Impellers Oliver Litfin a Antonio Delgado a Kais Haddad b Horst Klein b a University of Erlangen-Nuremberg, Institute of Fluid Mechanics b

slide-1
SLIDE 1

Numerical and Experimental Investigation of Trailing Edge Modifications of Centrifugal Wastewater Pump Impellers

Oliver Litfin a Antonio Delgado a Kais Haddad b Horst Klein b

a University of Erlangen-Nuremberg, Institute of Fluid Mechanics b Sulzer Pump Solutions Germany GmbH

ASME 2017 Fluids Engineering Division Summer Meeting, July 31 - August 3, 2017, Waikoloa, Hawai’i, USA

E r l a n g e n

slide-2
SLIDE 2

Scope of the work

E r l a n g e n

slide-3
SLIDE 3

E r l a n g e n

Scope of this work

In classical textbooks (Guelich, Karassik) it is stated, that under-filing typically leads to an increase of 3 percent in head while pump efficiency stays constant or increases slightly by 0.5 to 1 percent. More recent publications (Wu et al. 2015, Gao et al. 2016) showed significant increase in pump efficiency for a mixed-flow and a low-speed centrifugal pump due to trailing edge modifications. In this work the effect of trailing edge modifications on the performance of the wastewater pump impellers is numerically investigated.

⇒ Can under-filing increase the efficiency of a wastewater pump impeller? ⇒ How can under-filing increase the hydraulic efficiency? ⇒ What are the main mechanisms involved?

FEDSM 2017 | Oliver Litfin a, Antonio Delgado a, Kais Haddad b, Horst Klein b | FAU | 3

slide-4
SLIDE 4

Description of the test pumps

E r l a n g e n

slide-5
SLIDE 5

E r l a n g e n

Wastewater pumps NS065 and NS100

Property NS065 NS100 D2 0.38 m 0.48 m D1t/D2 0.47 0.53 b2/D2 0.25 0.26

β2b@camber

29.0 degree 18.4 degree sθ2/D2 0.06 0.12 Ns (US units) 1780 2730 default (cut-off) straight-cut under-filed (incl. blunt portion)

FEDSM 2017 | Oliver Litfin a, Antonio Delgado a, Kais Haddad b, Horst Klein b | FAU | 5

slide-6
SLIDE 6

E r l a n g e n

Numerical model

  • 3D RANS (ANSYS

R

CFX

R

17.1)

  • Impeller only in rotating frame
  • No tip-gap, no backchannel
  • k-ω-SST turbulence model with

curvature correction

  • 1st to 2nd order upwind blending

scheme with β = 0.95

  • Unstructured mesh with 2.8 million

nodes (tetra and prism elements)

  • Average y+ ≈ 0.5
  • Convergence criteria 10-5 for rms

residuals of the mass and momentum equations

FEDSM 2017 | Oliver Litfin a, Antonio Delgado a, Kais Haddad b, Horst Klein b | FAU | 6

slide-7
SLIDE 7

E r l a n g e n

Dimensionless characteristics for both impellers (CFD)

NS065

0.04 0.06 0.08 0.1 0.12 0.14 0.16 0.18 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.75 0.8 0.85 0.9 0.95 1 1.05 Ψ η* Φ NS065 Ψ NS065 η* NS065SC Ψ NS065SC η* NS065UF Ψ NS065UF η*

NS100

0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.015 0.02 0.025 0.03 0.035 0.04 0.75 0.8 0.85 0.9 0.95 1 1.05 Ψ η* Φ NS100 Ψ NS100 η* NS100SC Ψ NS100SC η* NS100UF Ψ NS100UF η*

  • Φ = Q/(ω · D3

2) ; Ψ = (g · H)/(ω · D2)2 ; η∗ = η/ηmax,ref

  • Both trailing edge modifications increase the head coefficient
  • Under-filing leads to an increased efficiency for the NS100 impeller, for

the NS065 impeller efficiency is the same as with default trailing edge

FEDSM 2017 | Oliver Litfin a, Antonio Delgado a, Kais Haddad b, Horst Klein b | FAU | 7

slide-8
SLIDE 8

Loss analysis based on rothalpy

E r l a n g e n

slide-9
SLIDE 9

E r l a n g e n

Concept of rothalpy and stations for evaluation

hT = h + 0.5 · C2 − U · Cθ pT = p + 0.5· ρ· C2 − ρ· U · Cθ pT = p0 − ρ · U · Cθ

1a 1b 2a 2b 1 2 TE PA LE Outlet 2b 2 2a

  • If rothalpy is constant along a streamline (isentropic) → rise in stagnation

pressure must correspond to the change in angular momentum.

  • Impeller was divided in three sections by defining a couple of stations in

the flowpath

  • Leading edge section (LE), passage section (PA) and trailing edge section

(TE)

  • Rothalpy losses were evaluated for the LE, PA and TE section separately

FEDSM 2017 | Oliver Litfin a, Antonio Delgado a, Kais Haddad b, Horst Klein b | FAU | 9

slide-10
SLIDE 10

E r l a n g e n

Impeller section rothalpy pressure losses (normalized)

NS065

  • 0.035
  • 0.03
  • 0.025
  • 0.02
  • 0.015
  • 0.01
  • 0.005

0.005 0.01 0.015 0.02 0.025 0.03 0.035 dpT / 0.5*ρ*U22 Φ NS065, LE NS065, PA NS065, TE

NS100

  • 0.03
  • 0.025
  • 0.02
  • 0.015
  • 0.01
  • 0.005

0.015 0.02 0.025 0.03 0.035 0.04 dpT / 0.5*ρ*U22 Φ NS100, LE NS100, PA NS100, TE

  • For NS065 at BEP (Φ=0.02) LE and TE losses are almost equal

(≈-0.004). PA losses are around -0.01.

  • For NS100 similar behavior except LE losses → slightly misaligned flow at

BEP .

FEDSM 2017 | Oliver Litfin a, Antonio Delgado a, Kais Haddad b, Horst Klein b | FAU | 10

slide-11
SLIDE 11

E r l a n g e n

Rothalpy pressure losses at LE for different TE shapes

NS065

  • 0.04
  • 0.03
  • 0.02
  • 0.01

0.005 0.01 0.015 0.02 0.025 0.03 0.035

  • 0.02
  • 0.015
  • 0.01
  • 0.005

dpT,LE / 0.5*ρ*U22 dpT,LE / dpt,is Φ NS065 NS065SC NS065UF

NS100

  • 0.03
  • 0.02
  • 0.01

0.015 0.02 0.025 0.03 0.035 0.04

  • 0.03
  • 0.02
  • 0.01

dpT,LE / 0.5*ρ*U22 dpT,LE / dpt,is Φ NS100 NS100SC NS100UF

  • Normalized rothalpy pressure losses (bottom half) not affected by TE

(quite obvious).

  • Relative rothalpy pressure losses (top half) are affected by TE shape. This

is due to the increased work input (higher Ψ-values).

FEDSM 2017 | Oliver Litfin a, Antonio Delgado a, Kais Haddad b, Horst Klein b | FAU | 11

slide-12
SLIDE 12

E r l a n g e n

Rothalpy pressure losses at PA for different TE shapes

NS065

  • 0.04
  • 0.03
  • 0.02
  • 0.01

0.005 0.01 0.015 0.02 0.025 0.03 0.035

  • 0.04
  • 0.03
  • 0.02
  • 0.01

dpT,PA / 0.5*ρ*U22 dpT,PA / dpt,is Φ NS065 NS065SC NS065UF

NS100

  • 0.03
  • 0.02
  • 0.01

0.015 0.02 0.025 0.03 0.035 0.04

  • 0.09
  • 0.06
  • 0.03

dpT,PA / 0.5*ρ*U22 dpT,PA / dpt,is Φ NS100 NS100SC NS100UF

  • For NS065 impeller normalized rothalpy pressure losses are not affected

by TE shape. Relative losses are reduced at higher flow coefficients due to the TE modifications.

  • For NS100 impeller normalized losses are also affected by TE shape.

Relative losses are strongly reduced by under-filing at all flow coefficients.

FEDSM 2017 | Oliver Litfin a, Antonio Delgado a, Kais Haddad b, Horst Klein b | FAU | 12

slide-13
SLIDE 13

E r l a n g e n

Rothalpy pressure losses at TE for different TE shapes

NS065

  • 0.04
  • 0.03
  • 0.02
  • 0.01

0.005 0.01 0.015 0.02 0.025 0.03 0.035

  • 0.04
  • 0.03
  • 0.02
  • 0.01

dpT,TE / 0.5*ρ*U22 dpT,TE / dpt,is Φ NS065 NS065SC NS065UF

NS100

  • 0.06
  • 0.04
  • 0.02

0.015 0.02 0.025 0.03 0.035 0.04

  • 0.09
  • 0.06
  • 0.03

dpT,TE / 0.5*ρ*U22 dpT,TE / dpt,is Φ NS100 NS100SC NS100UF

  • Normalized losses are strongly increased by straight-cut TE → large

mixing losses. Default and under-filed TE are very similar for NS065. For NS100 under-filed TE reduces losses at higher flow coefficients.

  • For NS100 there is a strong decrease in relative losses for under-filed TE

at higher flow coefficients.

FEDSM 2017 | Oliver Litfin a, Antonio Delgado a, Kais Haddad b, Horst Klein b | FAU | 13

slide-14
SLIDE 14

Effect of TE shape on slip factor

E r l a n g e n

slide-15
SLIDE 15

E r l a n g e n

CFD slip factor for different TE shapes

NS065

0.5 0.55 0.6 0.65 0.7 0.75 0.8 0.005 0.01 0.015 0.02 0.025 0.03 0.035 CFD

  • NS065

NS065SC NS065UF

NS100

0.45 0.5 0.55 0.6 0.65 0.7 0.75 0.015 0.02 0.025 0.03 0.035 0.04 CFD

  • NS100

NS100SC NS100UF

  • σCFD = 1 − ∆Cθ2

U2

with ∆Cθ2 = Cθ2,th − Cθ2

  • NS065 σ ≈ 0.6 @BEP , NS100 σ ≈ 0.55 @BEP
  • Slip factor increased by TE modifications compared to default TE.

FEDSM 2017 | Oliver Litfin a, Antonio Delgado a, Kais Haddad b, Horst Klein b | FAU | 15

slide-16
SLIDE 16

E r l a n g e n

CFD slip factor: Trailing edge flow

NS065

0.5 0.55 0.6 0.65 0.7 0.75 0.8 0.005 0.01 0.015 0.02 0.025 0.03 0.035 CFD

  • NS065

NS065SC NS065UF

  • Flow deflection towards the circumferential direction for default TE (top).
  • With both TE modifications the flow detached at the trailing edge end of

the pressure side → less flow deflection, increased slip factor.

FEDSM 2017 | Oliver Litfin a, Antonio Delgado a, Kais Haddad b, Horst Klein b | FAU | 16

slide-17
SLIDE 17

Experimental validation

E r l a n g e n

slide-18
SLIDE 18

E r l a n g e n

Experimental results (whole pump)

NS065

0.07 0.08 0.09 0.1 0.11 0.12 0.13 0.14 0.15 0.16 0.17 0.18 0.005 0.01 0.015 0.02 0.025 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 Ψ η* Φ NS065 Ψ NS065 η* NS065UF Ψ NS065UF η*

NS100

0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.1 0.11 0.12 0.13 0.14 0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 Ψ η* Φ NS100 Ψ NS100 η* NS100UF Ψ NS100UF η*

  • For NS065 under-filing does increase the head coefficient by 7.7 percent

@BEP , while efficiency remains constant.

  • For NS100 head coefficient @BEP is increased by even 11.7 percent

while efficiency is increased by 1.7 percent. The maximum efficiency is even increased by 4.5 percent, shifted to higher flow coefficient.

FEDSM 2017 | Oliver Litfin a, Antonio Delgado a, Kais Haddad b, Horst Klein b | FAU | 18

slide-19
SLIDE 19

Conclusions

E r l a n g e n

slide-20
SLIDE 20

E r l a n g e n

Conclusions

  • TE under-filing could increase the efficiency of the NS100 impeller by 1.7

percent (1.8 in CFD). Maximum efficiency could be increased by even 4.5 percent.

  • Straight-cut modification increases head but lowers efficiency due to

increased mixing losses.

  • Under-filing does not decrease the absolute losses (except slight

reduction of mixing losses). Possible efficiency gains result from an increased impeller work at almost the same absolute losses.

  • TE modifications can prevent flow deflection around the trailing edge by

forcing the flow to detach at the end of the pressure side of the blade.

  • Rothalpy can be very useful to investigate impeller losses

FEDSM 2017 | Oliver Litfin a, Antonio Delgado a, Kais Haddad b, Horst Klein b | FAU | 20

slide-21
SLIDE 21

Backup slides

E r l a n g e n

slide-22
SLIDE 22

E r l a n g e n

Under-filing

g · H = U2 · Cθ2 (1) Cθ2 = U2 − Cm2 tanβ2b (2)

ψ = Cθ2

U2 and

φ = Cm2

U2 (3)

ψ = 1 − φ

tanβ2b (4)

∂ψ ∂β2b = φ

sin2β2b (5)

C2 C2 Cm2 U2 W2 Cm2' C2' W2' C2' 2 2'

Cslip = π · sinβ2b Z

→ ∂ψ ∂β2b = φ

sin2β2b

− π · cosβ2b

Z (6)

FEDSM 2017 | Oliver Litfin a, Antonio Delgado a, Kais Haddad b, Horst Klein b | FAU | 22

slide-23
SLIDE 23

E r l a n g e n

Rothalpy vs. efficiency

ηtotal = Q · (p02b − p00) ω · Mrotor

and pT = p0 − ρ · U · Cθ (7) p02b − p00 = (pT2b − pT0) + ρ · (U2b · Cθ2b − U0 · Cθ0) (8)

ηtotal = Q · (pT2b − pT0) ω · Mrotor + ˙

m · r2b · Cθ2b Mrotor (9)

  • 0.035
  • 0.03
  • 0.025
  • 0.02
  • 0.015
  • 0.01
  • 0.005

0.005 0.01 0.015 0.02 0.025 0.03 0.035 dpT / dpt,is Φ NS065, LE NS065, PA NS065, TE 0.92 0.93 0.94 0.95 0.96 0.97 0.98 0.99 1 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 m . *r2b*Cθ2b / Mrotor Φ / Φbest NS065 NS100

FEDSM 2017 | Oliver Litfin a, Antonio Delgado a, Kais Haddad b, Horst Klein b | FAU | 23