Effects of vibration on the grain morphology
- PDF / 1,211,614 Bytes
- 4 Pages / 597 x 774 pts Page_size
- 49 Downloads / 238 Views
I.
INTRODUCTION
E X P E R I M E N T S in which tungsten incandescent filaments were recrystallized by heating with both direct current (DC) and alternating current (AC) showed different grain morphology depending on whether the AC or DC was used. The filaments heated with AC had a greater fraction of transverse grain boundaries than did the filaments heated by DC. Observation suggested that vibration of the coils might be the cause of the differences. Since transverse boundaries cause serious degradation of the performance of filaments in lamps,m further investigation under controlled conditions of vibration was undertaken. II.
EXPERIMENTAL PROCEDURE
In the initial experiment, a coiled-coil strand of 50-/zmdiameter doped lamp wire heated in a DC field was compared to an identical strand heated in an AC field. The 60 Hz AC current was observed to cause significant vibration in the 300-mm length of filament, whereas no vibration accompanied the DC current heating. In both cases, temperatures of 2700 K were attained for 30 seconds. Evaluation of the grain morphology was achieved using an image analysis technique called grain shape analysis. [2j Subsequent experiments were performed on lamps with 50-/xm-diameter wire filaments which had not been recrystallized. One lamp was mounted vertically on a vibrator, and a second lamp was mounted vertically in a static holder. The two lamps were wired in parallel to a variac driven by a variable speed motor which allowed the current to be increased at a controlled rate simultaneously to the two lamps. Thus, one lamp could be heated to the operating temperature while being vibrated, while the static lamp was heated simultaneously at the same rate. Heating rates to 2900 K were varied from approximately 200 to 10 dee per second, slow enough to guarantee that the filaments would have relatively large JOHN L. WALTER, formerly with GE Corporate Research and Development, Schenectady, NY 12301, is retired. JOHN W. PUGH, formerly with the Technology Division, GE Lighting, Nela Park, Cleveland, OH 44112, is retired. DONALD B. SORENSON is with GE Corporate Research and Development, Schenectady, NY 12301. Manuscript submitted February 6, 1991. METALLURGICAL TRANSACTIONS A
grains.[3] Both lamps were then held for 5 hours at temperature to allow the formation of grain boundary grooves sufficiently deep to be easily seen by scanning electron microscopy (SEM). The frequency of vibration was varied from 50 to 300 Hz using a HEWLETT-PACKARD* function gen*HEWLETT-PACKARD is a trademark of Hewlett-Packard Company, Colorado Springs, CO.
erator to drive a Crown D150A audio amplifier which controlled the amplitude of a B&K vibration exciter 4809. Twenty pairs of lamps were tested, with each lamp of the pair being heated at the same rate. The filaments were then removed from the envelopes and examined by SEM. From 100 to 200 grain boundaries were categorized as either "transverse" or "longitudinal" for each filament. A grain boundary was not categorized unless both ends of the boundary could be s
Data Loading...
