User:Joseph Frye/Notebook/Physics Junior Lab 307L/NeonExcitation

From OpenWetWare

Jump to: navigation, search

Contents

Excitation Levels of Neon

SJK 06:03, 5 October 2010 (EDT)
06:03, 5 October 2010 (EDT)Overall, this is a good primary lab notebook.  Pictures and spreadsheets were very good.  Throughout, it seemed a little thin on descriptions of what you did.  I think this was because of a disconnect between the wiki and the spreadsheet while taking data.  I would like more notes about technique, or weird things that you may have noticed along the way.  Just more notes in general.
06:03, 5 October 2010 (EDT)
Overall, this is a good primary lab notebook. Pictures and spreadsheets were very good. Throughout, it seemed a little thin on descriptions of what you did. I think this was because of a disconnect between the wiki and the spreadsheet while taking data. I would like more notes about technique, or weird things that you may have noticed along the way. Just more notes in general.


This lab was done on Mondays 8/13/2010 and 8/20/2010 at the UNM physics department with my lab partner Alex Benedict. We followed experiment 6 outlined in Dr. Gold's lab manual.

Links

My Lab Summary

| Dr. Gold's Lab Manual

Alex Benedict's Lab Notebook

Equipment

SJK 05:51, 5 October 2010 (EDT)
05:51, 5 October 2010 (EDT)Excellent use of pictures and annotated schematics.  Those will go a long way in helping future experimenters with this lab.  My only criticism with this section is makes and model numbers are not recorded.  This can probably be discerned from the photos, but it's good to type it out explicitly too.
05:51, 5 October 2010 (EDT)
Excellent use of pictures and annotated schematics. Those will go a long way in helping future experimenters with this lab. My only criticism with this section is makes and model numbers are not recorded. This can probably be discerned from the photos, but it's good to type it out explicitly too.

Equipment

  • Hertz Critical Potentials tube filled with neon
  • Tube stand
  • Picoamplifier and Alarmed Meter
  • 2 Power Supplies
  • Digital Multimeter
  • 1.5V AA Battery
  • Cables

Set Up

Figure 6.2 from Dr. Golds Lab Manual

Thanks to Tomas Mondragon for redoing the text on this diagram the resolution of the original image made it almost unreadable. Mondragon's Notebook Fall '08


We first found all of the equipment we needed according to the lab manual and then connected it according to figure 6.2

we measured the current using the pico-amplifier and the accelerating voltage (Va) using the DMM.

Measurements

Download spreadsheet of our data

SJK 05:48, 5 October 2010 (EDT)
05:48, 5 October 2010 (EDT)Excellent use of spreadsheet.  Since your data were so spread-sheet intensive, it probably would have made some sense to leave some "notes" columns for recording any kind of procedure notes, or for noting quirks that happened.
05:48, 5 October 2010 (EDT)
Excellent use of spreadsheet. Since your data were so spread-sheet intensive, it probably would have made some sense to leave some "notes" columns for recording any kind of procedure notes, or for noting quirks that happened.

We did a rough scan of the voltages ranging from 0 to 30V in 1V increments on the accelerating voltage with the filament voltage set at 1.8V.

Here is the graph of that data:

Image:Frye Neon Rough18.jpg


After examining the graph we then decided to to another scan from 15V to 22V using 0.1V increments to get better resolution of the area. We did this once with Vf at 1.8V

Here is the graph:

Image:Frye Neon Fine18.jpg

We also did the same range with the same 0.1V increments but used Vf=2.1V

Image:Frye Neon Fine21.jpg


we also flipped the battery and measured Va at 1V intervals from 0V to 25V. but found this data to be less useful.

Download spreadsheet of our data

Observations

  • Looking at the data from the measurements we took with Vf=1.8V, there are two major valleys with relative minimums. One at about 18.4V and the other at about 21.3V. There were also smaller bumps at 16.0V and 16.3V
  • Looking at the data from the measurements we took with Vf=2.1V, there are again two major valleys with relative minimums. One at about 18.0V and the other between 21.0V and 21.1V. There is a strange jump in the current from 15.8V to 15.9V
  • In each case the current grows linearly with voltage after 21.5V

Calculations

According to Ritter from fall of '09, the accepted values for the first two peaks are 16.7eV and 18.65eV. There is also a peak around 20V. From the lab manual, the accepted value for the ionization energy of neon is 21.56 which is consistent with our data because for voltages higher than that our current grew linearly with the voltage as we would expect.

Our results for the first peak are:

SJK 06:04, 5 October 2010 (EDT)
06:04, 5 October 2010 (EDT)Remember to always put units on values to prevent confusion later!
06:04, 5 October 2010 (EDT)
Remember to always put units on values to prevent confusion later!
(16.3+/-0.1 + 15.8+/-0.1)/2 = 16.1 +/-0.1 SJK 05:46, 5 October 2010 (EDT)
05:46, 5 October 2010 (EDT)As mentioned on your summary page, when averaging together values from different runs, you should always ask yourself whether you have reason to believe they represent estimates with the same parent distribution.  If you have reason to think otherwise, is the average the best way to go?  Or would you be better off figuring out why the runs are different.  We'll talk about this more in future weeks.  Along the way, we'll also learn how to do a weighted average and calculate the new uncertainty after the average.  If your uncertanties here represented SEM, and you assumed normally distributed mean (which is not the case for you here, since you're estimating the uncertainty in another way), then there are methods for propagating the uncertainty.  The uncertainty in the averaged value would be less than 0.1, since you have information from two independent measurements.  In the case here, though, that would be suspicious since the two values are well separated compared with uncertainty.
05:46, 5 October 2010 (EDT)
As mentioned on your summary page, when averaging together values from different runs, you should always ask yourself whether you have reason to believe they represent estimates with the same parent distribution. If you have reason to think otherwise, is the average the best way to go? Or would you be better off figuring out why the runs are different. We'll talk about this more in future weeks. Along the way, we'll also learn how to do a weighted average and calculate the new uncertainty after the average. If your uncertanties here represented SEM, and you assumed normally distributed mean (which is not the case for you here, since you're estimating the uncertainty in another way), then there are methods for propagating the uncertainty. The uncertainty in the averaged value would be less than 0.1, since you have information from two independent measurements. In the case here, though, that would be suspicious since the two values are well separated compared with uncertainty.

error = (16.7-16.1)/16.7 = 3.6% relative error

SJK 06:01, 5 October 2010 (EDT)
06:01, 5 October 2010 (EDT)We are always tempted to report the relative error when comparing measurements with an accepted value.  However, how do we know if the relative error is small enough that our measurements are consistent with the accepted value?  The answer is to do statistical comparisons of the accepted value to the confidence interval of your measurements.  In your case, +/- 0.1 V is much smaller than the 0.6 V discrepancy.  If your +/- represented a 68% confidence interval (which it doesn't in your case), then very likely there is systematic error.  Your interval is probably much higher than 68%, and thus even more likely that there's significant systematic error.  Again, we'll talk about this more in the next few weeks of class.
06:01, 5 October 2010 (EDT)
We are always tempted to report the relative error when comparing measurements with an accepted value. However, how do we know if the relative error is small enough that our measurements are consistent with the accepted value? The answer is to do statistical comparisons of the accepted value to the confidence interval of your measurements. In your case, +/- 0.1 V is much smaller than the 0.6 V discrepancy. If your +/- represented a 68% confidence interval (which it doesn't in your case), then very likely there is systematic error. Your interval is probably much higher than 68%, and thus even more likely that there's significant systematic error. Again, we'll talk about this more in the next few weeks of class.


For the second peak:

(18.4+18.0 +/-0.2)/2 = 18.2 +/-0.1

error = (18.65-18.2)/18.65 = 2.4% relative error

For the ionization energy of neon:

(21.3 + 21.05 +/-.02)/2 = 21.18 +/-0.1

error = (21.56 - 21.18)/21.56 = 1.8% error

Conclusion

  • We found the first peak to be at 16.1eV +/- 0.1eV with relative error compared to the accepted value of 3.6%
  • The second peak at 18.2eV +/- 0.1eV with relative error compared to the accepted value of 2.4%
  • The ionization energy of neon at 21.18eV +/- 0.1eV with relative error compared to the accepted value of 1.8%


    • the +/- 0.1eV uncertainty here comes from our uncertainty in measuring the voltages since we only measured in 0.1V intervals.
Personal tools