Public Lab Research note


The user 'dhaffnersr' has been placed in moderation and will not be able to respond to comments.
  • 0

Signal To Noise Ratio data graphs on various slit width's

by dhaffnersr |

Here is an excel spread sheet graph on the various slit width's that we use:

40mm_apr3.png

18mm_apr3.png

15mm_apr3.png

12mm_apr3.png

09mm_apr3.png



spectrometer spectralworkbench desktop-spectrometer desktop-spectrometry-kit spectral-workbench tool

parent:spectral-workbench

3 Comments

Unfortunately, I'm unable to extract any meaningful observations from this collection of numbers and plots and I suspect other readers will be confused as well. There appears to be a range of inconsistent use of terminology and there is no information included to show how any of the data was obtained; eg photon, shot, flicker, flicker factor, background flicker factor, detector noise, signal source, etc. While the title references SNR, that appears to never be calculated or explained and the plots indicate an inconsistent signal amplitude making SNR measurements re. slit width questionable. Since the data appears to have been taken over 30 sec, it seems illogical that slit width would have any affect on any of the noise components of the CCD detector so I have grave doubts about all of this data. SNR vs slit width would need to be measured using a constant signal amplitude at the detector.


Well Dave, I'm just going to have to disagree with you on this one, look, it's an excel program, one of many developed by Thomas C O'Haver, a gentleman with a Ph,D., in Analytical chemistry here are references for his qualifications- https://terpconnect.umd.edu/~toh/ Ph.D., Analytical Chemistry, 1968, Univ. of Florida (J. D. Winefordner) B.S. Chemistry, 1963, Spring Hill College

So please check it out for yourself, if I make mistakes I certainly correct them but if there is knowledge to be had, I am not just willy-nilly about it either. I go to sources where they are tried and true. what about my Rhodamine B experiments or Eosin Y? I'm doing those as my first fluorescent experiments because there are NIST standards already done on them with various regents, thay way I can calibrate my own rig and fine tune my technique, and so far I'm gettting pretty close to those standards, so I must be doing something right.

I know you did a similar piece of work on SNR, and that's fine, I just think there is a lot more to spectrometry than shinning a laser or CFL down the photon tube. I'd like to see more of your work on the fluorescence side of all this, I've never seen any spectra uploaded by you, just the work you do on matlab?

Anyway Dave, have a good day.

Dave H


I've not questioned Dr. O'Haver's work, his qualifications or the tools he has developed; it is the just lack of information in your note to explain what you did, and how you collected the data, which is the basis of my confusion about the posted numbers and their units. For example, you did not describe what the plots are representing nor how that data was taken -- so, the reader is left just guessing. For example, background video noise from whole-frame data is not necessarily the same as noise from within the spectral band region alone; but that detail is not described so the reader would have to guess. The noise from external light, as diffracted, is an additional source which may be hard to differentiate from camera noise as is the noise from the camera's AGC. In other words, there's a lot going on in the system which is why it is valuable to describe it to the reader. With a CCD as detector, shot and flicker noise should be a constant and not related to slit width -- unless there is another variable such as camera AGC entering into the mix. However, your data shows these values changing but no discussion or observations related to that data which begs the questions. Perhaps you do have data to extend the reader's knowledge on this issue? If so, a reader might find that of value.

From your note's Title, I was assuming you would calculate the SNR of a spectral signal as detected by the spectrometer and then demonstrate the change in that SNR, if any, as the slit width is changed. I see no plots or analysis describing that or demonstrating, with data, which factors are correlated and which factor are not. That is not to say that information is not available or is not contained within the data you collected; I, the reader, just cannot see it. The SNR of a measurement device is a performance measurement of that device when used for it's primary purpose; in this case, the SNR of the optical signature from diffraction. How that observation is performed and how the SNR is calculated is important to describe so the reader can understand and, potentially, make use of those concepts. You can use spreadsheets and I can write code where the calculations are correct, but if we don't provide the correct input under the correct conditions with sufficient limitation on variables, and explain all of that to the reader, then we're just asking the reader to accept our work on faith, not science.

I have no wish to debate your results or mine and comparisons are limited because observations and methods are so different. I've seldom uploaded spectra because 1) the PLab interface with my browser/net is such a struggle and 2) because few of my spectra are of sufficient value to a community database whose general goal is to compare standardized measurements (the PLab database is just not there yet). Though I perform my own plotting, I always describe the source of my data; which is generally the same PLab-3 webcam as others use. I am, however, suggesting again that in communicating work (in written form) to others generally requires significant description and detail to explain the procedures, where the data comes from, how it is used and an explanation of how conclusions, if any, are supported by that data. This is how all science is performed and is a topic I continue to promote -- good science.

I hope this clarifies my previous comments.


You must be logged in to comment.