Research Papers: Gas Turbines: Controls, Diagnostics, and Instrumentation

Methodology to Correct the Magnetic Field Effect on Thin Film Measurements

[+] Author and Article Information
D. G. Cuadrado

Mechanical Engineering Department,
Purdue University,
West Lafayette, IN 47906
e-mail: gonza279@purdue.edu

S. Lavagnoli

Turbomachinery Department,
von Karman Institute for Fluid Dynamics,
Brussels 1640, Belgium
e-mail: lavagnoli@vki.ac.be

G. Paniagua

Mechanical Engineering Department,
Purdue University,
West Lafayette, IN 47906
e-mail: gpaniagua@purdue.edu

Contributed by the Controls, Diagnostics and Instrumentation Committee of ASME for publication in the JOURNAL OF ENGINEERING FOR GAS TURBINES AND POWER. Manuscript received July 16, 2015; final manuscript received August 4, 2015; published online September 22, 2015. Editor: David Wisler.

J. Eng. Gas Turbines Power 138(3), 031602 (Sep 22, 2015) (8 pages) Paper No: GTP-15-1336; doi: 10.1115/1.4031321 History: Received July 16, 2015; Revised August 04, 2015

Machined ferrous metal components may carry a magnetic field, which in rotation disturb the output of electrical sensors. To minimize the effect on the electrical instrumentation, the rotating components are usually demagnetized. However, even after the demagnetization process, a residual magnetism unavoidably remains. This paper presents a methodology to predict the effects of a rotating magnetic field induced on thin film measurements. In addition to the prediction of the magnetic effects, a procedure to correct the spurious variation in the readings of thin film gauges has been developed to enhance the fidelity of the measurements. An analytical model was developed to reproduce the bias on the electrical signal from sensors exposed to rotor airfoils with magnets. The model is based on the Biot–Savart law to generate the magnetic field, and the Faraday's law to calculate the electromotive force induced along the measurement circuit. The model was assessed by means of controlled experiments varying the rotor tip clearance and rotational speed. The presented methodologies allowed the correction of the magnetic field effects. The raw signal of the thin film sensors, in the absence of any correction, is prone to deliver errors in the heat flux amounting to about 8% of the mean overall value. Thanks to the developed corrective approach, the residual magnetic effect contribution to the heat flux error would be 2% at most.

Copyright © 2016 by ASME
Your Session has timed out. Please sign back in to continue.


Knauss, H. , Roediger, T. , Gaisbauer, U. , and Kraemer, E. , 2006, “ A Novel Sensor for Fast Heat Flux Measurements,” AIAA Paper No. 2006-3637.
Tipler, P. A. , and Mosca, G. , 2005, “ Physics for Science and Technology,” Electricity and Magnetism, Vol. 2A, Editorial Reverté, Barcelona.
Sohre, J. S. , and Nippes, P. I. , 1988, “ Electromagnetic Shaft Currents and Demagnetization on Rotors of Turbines and Compressors,” 17th Turbomachinery Symposium, Texas A&M University, College Station, TX, Nov. 8–10, pp. 13–33.
Schultz, D. L. , and Jones, T. V. , 1973, “ Heat Transfer Measurements in Short Duration Facilities,” Advisory Group for Aerospace Research and Development, Neuilly-sur-Seine, France, AGARDograph Report No. 165.
Mikolanda, T. , Kosec, M. , and Richter, A. , 2009, “ Magnetic Field of Permanent Magnets: Measurement, Modelling, Visualization,” Technical University of Liberec, Liberec, Czech Republic.
Ljung, L. , 2014, Matlab & Simulink: System Identification Toolbox™ 7 User's Guide, MathWorks, Inc., Natick, MA.
Garnier, H. , Mensler, M. , and Richard, A. , 2003, “ Continuous-Time Model Identification From Sampled Data: Implementation Issues and Performance Evaluation,” Int. J. Control, 76(13), pp. 1337–1357. [CrossRef]
Ljung, L. , 2009, “ Experiments With Identification of Continuous-Time Models,” 15th IFAC Symposium on System Identification, Saint-Malo, France, July 6–8, pp. 1175–1180.
Young, P. C. , and Jakeman, A. J. , 1980, “ Refined Instrumental Variable Methods of Time-Series Analysis: Part III, Extensions,” Int. J. Control, 31(4), pp. 741–764. [CrossRef]
Seber, G. A. F. , and Wild, C. J. , 2003, Nonlinear Regression, Wiley-Interscience, Hoboken, NJ.
DuMouchel, W. H. , and O'Brien, F. L. , 1989, “ Integrating a Robust Option Into a Multiple Regression Computing Environment,” 21st Symposium on the Interface, Alexandria, VA, Apr. 9–12, pp. 297–301.
Holland, P. W. , and Welsch, R. E. , 1977, “ Robust Regression Using Iteratively Reweighted Least-Squares,” Commun. Stat., 6(9), pp. 813–827. [CrossRef]


Grahic Jump Location
Fig. 1

Representation of the model with the different variables used during the calculation

Grahic Jump Location
Fig. 2

Sketch of the experimental arrangement, composed of a 16-teeth wheel disk, a micro-adjustable table, and a pneumatic engine, without MACOR block

Grahic Jump Location
Fig. 3

(a) Location of the measurement points used for the magnetic field verification. (b) Different orientations checked using a gauss meter (red line) attached to the MACOR block support (unfilled rectangles).

Grahic Jump Location
Fig. 4

(a) Magnetic field at orientations 0 deg and 180 deg. (b)Magnetic field characterization at 90 deg.

Grahic Jump Location
Fig. 5

Experimental versus modeled EMF at: (a) 0 deg, 2000 rpm, hlf  = 0.2 mm; (b) 90 deg, 1000 rpm, hlf  = 1.0 mm; (c) 0 deg, 1000 rpm, hlf  = 1.0 mm; and (d) 90 deg, 2000 rpm, hlf  = 0.2 mm

Grahic Jump Location
Fig. 6

(a) X distance input. (b) Model fitting curve with nlinfit. (c) Model fitting curve at 500 rpm and 1.5 mm.

Grahic Jump Location
Fig. 7

(a) C1 coefficient surface model result, (b) C2C3C4 coefficients model result at 1500 rpm and different distances, and (c) C2C3C4 coefficients model result at 1 mm and different rpm

Grahic Jump Location
Fig. 8

(a) Model C1 coefficient fitting, (b) model C2 coefficient surface fitting, and (c) model C4 coefficient surface fitting

Grahic Jump Location
Fig. 9

(a) Fitting result at 2500 rpm and 0.5 mm. (b) Fitting result at 2000 rpm and 0.75 mm.

Grahic Jump Location
Fig. 10

(a) C1 coefficient surface experiment result, (b) C2C3C4 coefficients experiment result at 1500 rpm and different distances, and (c) C2C3C4 coefficients experiment result at 1.5 mm and different rpm

Grahic Jump Location
Fig. 11

(a) C1 coefficient fitting for experiment, (b) C2 coefficient fitting for experiment, and (c) C4 coefficient fitting for experiment

Grahic Jump Location
Fig. 12

(a) Voltage perturbations with 20 deg phase delay. (b)Output comparison. (c) Direct subtraction error.



Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In