*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
*****

I'd just like to point out that I don't disagree with Jim in the slightest about the size in pixels of the Airy disk at Nyquist.  It's just that I like to put the centre on a pixel rather than in between for ease of visualization.  This means that the extremities of the 4-pixel wide disk will be on 'half pixels' which, since they are in the dark ring, will be pretty dim, so I ignore them.   I also don't see why one should represent a circle as a square, which leads to the 9 pixels I regard as containing useful information.   I do wish sometimes we could post diagrams to this list!  

George is quite right that confocals do usually scale to keep constant brightness at different scan speeds (the old MRC 600 allowed you to turn this off) but if this is done properly it shouldn't affect S/N, which is what we are interested in - you can make a digital image as dark or bright as you like.

 Another factor affecting 'brightness' (photon efficiency) is the transmittance of the lens.  Plan-apochromats often absorb quite a lot of light.  If you can, compare a plan-apo with a plan fluorite of the same NA - you will probably see a substantial  difference.   

                                   					Guy

Guy Cox, Honorary Associate Professor
School of Medical Sciences

Australian Centre for Microscopy and Microanalysis,
Madsen, F09, University of Sydney, NSW 2006

-----Original Message-----
From: Confocal Microscopy List [mailto:[log in to unmask]] On Behalf Of George McNamara
Sent: Tuesday, 1 October 2013 10:29 AM
To: [log in to unmask]
Subject: Re: image brightness on a laser scanning confocal

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
*****

Hi Julio,

Confocal microscope output is not simple. For one thing, the instrument manufacturers do things "under the hood" with the data between the PMT and the user interface you see. For example, on the Zeiss LSM510, the service engineers were supposed to tweak some setting(s) that resulted in approximately the same output value regardless of scan speed (of course you could just acquire one pixel, but then it would not be a scanner ... and you'd probably photobleach the spot before too long). 
Try scanning at the same speed (pixel dwell time) with say 16x16 pixels [which I nominate we call "Jim's box" in honor of Jim Pawley ... oops, maybe that's Guy's box and Jim's should be 25x25 pixels, assuming you have your bead perfectly centered in the middle] vs max # pixels (ex. 
8192x8192 pixels - may take a while ... of course if your Z or XY drift in that time, too bad).

Besides photobleaching, if your laser power is high enough, your fluorophores will go triplet and stuck in the triplet state(s) until either (a) the laser spot moves on, (b) you hit the off switch, or (c) they die. Amount of oxygen (which can quench excited state triplets but also can kill the fluorophore) or other triplet state quenchers. Of course trivial other things like too low or too high a PMT voltage (Cho and Lockett favor relatively low PMT gain with sensible choice of A/D converter), wrong pinhole size, scattering by the specimen (including refractive index differences between specimen, coverglass, immersion medium, lens). Not that you should expect any of your lasers to have constant power output to the specimen (try scanning to the transmitted light detector overnight for this, though the photodetector in the T path is not a PMT).

A good paper on confocal microscope calibration is:

Calibration and standardization of the emission light path of confocal microscopes. <http://www.ncbi.nlm.nih.gov/pubmed/16872427>

*Cho* EH, *Lockett* SJ.

J Microsc. 2006 Jul;223(Pt 1):15-25.

PMID:
    16872427


Ted Young, Yuval Garini et al, published a similar paper, http://repository.tudelft.nl/view/ir/uuid%3A2cdbdf0c-45be-4ae0-8daa-a48c8058e756/
Reading Ted and Yuval et al's abstract reminded me to mention to the entire listserv readership: please do not use "arbitrary units" in your manuscripts (or in your minds for that matter).

George
p.s. monitors and room lighting, LUT choice (compare blue to gray), and contrast adjustment of the image display (and adjacent white/gray/black space on the GUI) all have a visual impact on what you see: quantify the data by the numbers and graphs.


On 9/30/2013 4:39 PM, Julio Vazquez wrote:
> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> *****
>
> Dear confocalists,
>
> I have been thinking about the relative brightness of different objectives on a point laser scanning confocal microscope. The formulas one finds in books, etc, report brightness as proportional to NA^2 / M^2 for transmitted light, and NA^4 / M^2 for epifluorescence. However, I have never seen a specific discussion of single point scanners vs widefield. It is my impression that on a point scanner, the laser beam is focussed to a spot whose radius depends on the NA; on the collection side, light emitted by the spot is collected back onto a PMT, with a collection efficiency also related to the NA. At neither stage does the magnification seem to play a role. This leads me to think that the brightness of an objective on a point scanner is proportional to the fourth power of the NA, and independent of the magnification. I did some quick and dirty measurements on 200 nm beads which seemed to support this (although complicated by various factors such as the size of the back aperture, which made it difficult to get precise measurements of excitation power at the sample with high NA lenses, for example, and other things). However, I am puzzled I have never seen mention of this anywhere. Is it correct that magnification is irrelevant for image brightness on a point scanner, or am I way off the mark?
>
>
> Julio Vazquez
> Fred Hutchinson Cancer Research Center
> 1100 Fairview Ave N., DE-512
> Seattle, WA 98109
>
> http://www.fhcrc.org/en.html
>
>    


-- 



George McNamara, Ph.D.
Single Cells Analyst
L.J.N. Cooper Lab
University of Texas M.D. Anderson Cancer Center Houston, TX 77054 http://works.bepress.com/gmcnamara/26/