Relating milkfat in cream to fiber optic backscatter measurements

C. L. Crofcheck, F. A. Payne, K. S. Gillette, S. E. Nokes, C. L. Hicks

Research output: Contribution to conferencePaperpeer-review

Abstract

A fiber optic backscatter technique using single fibers to deliver and receive light was investigated for measuring the milkfat content of unhomogenized cream. Light backscatter through cream for wavelengths of 400 to 1000 nm was measured using a fiber optic spectrophotometer. Backscatter was measured for fiber separation distances from 2 to 6.5 mm and for cream containing 10 to 40 wt% milkfat. Unhomogenized cream (∼40 wt% milkfat) was mixed with skim milk (∼0.05 wt% milkfat) to yield samples with 5 different milkfat levels. Three optical responses were tested for correlation with milkfat content. One using the light intensity measurement at a single separation distance, the second using the light intensity ratio of two distances, and a third using the light intensity as a function of separation distance based on the backscatter of light in a particulate solution. The calibration equations from all three methods were able to predict the milkfat of the validation samples with a RMSE of 1.2 to 2.0 wt%. Statistical analysis indicates that the intensity ratio model may be the best suited for further testing in a dedicated fiber optic sensor.

Original languageEnglish
Pages535-550
Number of pages16
StatePublished - 2000
Event2000 ASAE Annual International Meeting, Technical Papers: Engineering Solutions for a New Century - Milwaukee, WI., United States
Duration: Jul 9 2000Jul 12 2000

Conference

Conference2000 ASAE Annual International Meeting, Technical Papers: Engineering Solutions for a New Century
Country/TerritoryUnited States
CityMilwaukee, WI.
Period7/9/007/12/00

Keywords

  • Cream
  • Fiber Optic
  • Milk
  • Milkfat
  • Sensor

ASJC Scopus subject areas

  • Engineering (all)

Fingerprint

Dive into the research topics of 'Relating milkfat in cream to fiber optic backscatter measurements'. Together they form a unique fingerprint.

Cite this