Analysis

…Of pictures.

To analyze our pictures, we chose to use the industry-standard to measure Red-Green-Blue scales to measure the amount of light in our photographs.  We used a program called Ulead’s PhotoImpact XL and the Eye Dropper tool to determine the RGB content in the different stages of our photographs.  In the photographs, we took the color from the white square, which is in the same spot in each of the photographs (which we determined by using the XY coordinates in PhotoImpact).  This told us our brightness because the RGB scale has a measurement for the amount of Red, Green, and Blue in a picture, with 0 being no color present and 255 being full color present; therefore, a larger number signifies more “color” present, thus saying that this one spot in a picture is brighter than another spot.  Because we used different colored flashes, we decided to take the magnitude of RGB in each area for the beginning, middle, and end (red, green, and blue, respectively) flashes and use this to determine the brightness increase or decrease across time.   

File Name	Red Flash	Green Flash	Blue Flash	Total	Red / Total	Green / Total	Blue / Total
2bulb_12_bbafvp.jpg	106	147	265	508	.209	.270	.522
bulb_30_bbafvp.jpg	160	162	272	594	.270	.273	.458
bulb_32_bbafvp.jpg	162	190	347	699	.232	.272	.496
bulb_31_bbafvp.jpg	100	192	351	643	.156	.300	.546
bulb_33_bbafvp.jpg	163	185	280	628	.260	.295	.446
bulb_22_bbafvp.jpg (control)	162	177	283	622	.260	.285	.455
bulb_21_bbafvp.jpg (control)	165	196	285	646	.255	.303	.441
bulb_20_bbafvp.jpg (control)	163	191	294	648	.252	.295	.454

 

We realized that there was a light change in the control photos from red to blue to green as well, an increase in intensity, and so (as seen in our Discussion) we came up with a compensation factor from Red to Green of 3.8% increase and of Green to Blue of 15%.  We determined this by taking the difference of the magnitudes of the two colors and dividing by the total color in the three parts of the photographs.  Then, to determine the average increase in brightness for our photographs, compensated for this change, we subtracted the increase seen in the controls from the Green and Blue flash (depending on which we were working with to find the increase), then subtracted the Blue and Green flash magnitudes from the Green and Red flash magnitudes, respectively, took the average of the parts, and determined that to be the average change in brightness intensity by percent.  To get the change across the photograph, we simply added the two percentages.

…Of Video

Because every bulb explosion is different, we were unable to achieve conclusive numbers for the videos that we analyzed, only see general trends.  We used VideoPoint 2.5.0 by Lenox Software for analyzing our video frame-by-frame.  We saw three general trends:

1)      With both big and small bulbs the light was apt to go out and re-ignite later (the off to re-ignition process took approximately 45ms);

2)       With small, clear bulbs we were able to see the escape of gases, which took approximately 4.34ms;

3)      There were also bulbs that simply went out after being shot.

We could not determine what correlation there was between large and small bulbs and what each bulb did, because except for Trend Two, the results were not size-specific.  The average time for the bulb to completely cease being lit by its own power (save for when the Tungsten filament catches on fire), was 330ms for both large, opaque bulbs and small, clear bulbs.

Discussion

All of the pictures successfully taken of the BB shattering the bulb indicate the filament does not dim immediately after the near-vacuum environment is compromised. Rather, there is a marked increase in the RGB sum directly after the impact; the data shows an average increase of 19% in brightness from middle-(impact)-to-end. The 4.1% increase from beginning-to-end is much higher than the 5% increase from beginning-to-impact, thus indicating that the percent change in brightness for the impact-to-end is significantly higher than any percent change that may have been caused by mirror rotation and varying angles of light reflection for beginning-to-impact. Furthermore, this change in brightness is significantly higher than that of the controls, which show a 3.8% increase from beginning-to-impact and a 15% increase from impact-to-end. This increase in filament brightness after impact indicates that the brightness of the filament rises rather than falls directly after the impact. This could be due to several factors, including the possible combustion of the high temperature tungsten filament in the presence of oxygen. In fact, bulbs are manufactured with a near vacuum on the inside of the glass container to prevent just such a phenomenon. Another possibility is that the quickly moving BB rubs against the tungsten filament with enough friction to ignite the filament – a process that is very similar to the ignition of matches. The sharp contrast between middle-to-impact and impact-to-end brightness changes can be explained by the higher reflectivity of the blue (final) flash when compared to no flashes (at the beginning). Thus, the impact-to-end brightness changes will always be higher than those of beginning-to-impact.