[RASMB] Lamp intensity fluctuations

[Tom Laue]

Hi Titus, et al.-
I am not sure what the intensity reading is you are presenting- is it 
the average value across the radial scan? Is it the reading at a 
particular radial position? In any case, it is worthwhile getting some 
idea of the complexity that is behind the XLA intensity readings.
There are two intensity readings taken with each lamp flash- 1) 5 or 10% 
(can't recall which) of the lamp flash intensity itself is monitored 
using light bounced off of a semi-reflecting mirror and a photodiode 
sensor. This signal is used to correct for the inherent ~3% variation in 
flash intensity characteristic of the lamp. 2) The light intensity at 
the photomultiplier (PMT) after it has passed through the cell and 
slit/lens assembly. I do not believe the photodiode intensity readings 
are accessible to the outside world, so your readings are from the PMT.
There are two gain settings used for each PMT reading. The first is the 
PMT voltage, which may be varied from 0 to 1100 volts by the acquisition 
system. The gain (in microamps/lumen) increases rapidly in a highly 
nonlinear manner as the voltage is increased. Beckman connects a charge 
amplifier circuit to the PMT rather than a more usual transimpedance 
amplifier scheme. The charge amplifier circuit slows down the response 
to the light pulse, but it increases sensitivity and minimizes shot 
noise. The output from the charge amplifier is a voltage spike that is 
considerably longer-lasting (milliseconds) than the light burst 
(microseconds). The second gain stage is a programmable gain amplifier 
(PGA) which multiplies the voltage from the charge amplifier by a known 
amount (1x, 2x, 4x, 8x.... 256x). The PGA output is connected to a 
voltage to frequency (V/f) converter, and the output from the V/f is 
counted for 8 milliseconds following the lamp pulse. It is this count 
(just a number between 0 - 65535), adjusted for the intensity from the 
photodiode circuit, that constitutes an intensity reading.
Now, in order to automate the XLA  absorbance system so that it can 
compensate for the variation in intensity with wavelength and optimize 
the signal to noise ratio for samples having different optical 
densities, the gain of this system (PMT + PGA) is adjusted to try to 
keep the count within a certain range. For an absorbance reading, 
changing the gain like this is not a big deal- so long as the reference 
and sample intensity readings are in a good range, the log of their 
ratio will be a valid absorbance reading. Accordingly, the XLA uses the 
signal from the reference sector to set the PMT/PGA values to get a good 
reading so that the absorbance calculation has a good signal to noise.
The XLA data acquisition is geared towards making the best possible 
absorbance readings, and not worried about what it needs to do to get 
optimize the intensity readings. So, here is what is happening in your 
system- at early times, the reference sector (which has sample in it) 
requires a higher gain setting in order to get what the XLA would use 
for an absorbance calculation. The signals increase as the sample 
sediments. The XLA does not want too high of a reading for the reference 
signal or else the detector system may saturate. Consequently, it 
reduces the gain settings. In other words, your system is working as it 
was designed to work.
The 'work around' for this is to keep the absorbance reading in the 
reference sector (and sample sector) to values < 0.4 OD. This will keep 
the gain setting circuitry from doing wild things (like changing the 
gain mid-scan).
Best wishes,
Tom


Titus M. Franzmann wrote:
> Hi everyone. 
>
> We are experiencing a phenomenon with our ProteomLab XL/I that I we do not
> understand and unfortunately Beckman is not able to address at this point. 
> Over the time-course of our experiments the lamp intensity increase
> significantly and then decreases again (please see attached files). This
> pattern is present in all runs and most pronounced in intensity mode.
> However, absorbance data seem to be compromised as well. 
> I sat down and monitored the photomultiplier Voltage in the hardware monitor
> menu in the Beckman software and noticed that the voltage settings varied
> between scans. The voltage varied by more than 100 V. My understanding is
> that the voltage setting should not change between scans of the same cell as
> this will affect the dynamics of the photomultiplier. However, I am not sure
> if that assumption is correct, or if there is some room to do minor
> adjustments during scans.
> Any comment is greatly appreciated. 
> Best
> Titus M. Franzmann
>
>  
> Dr. Titus M. Franzmann
> S. G. Walter Lab
> Mol., Cell. and Develop. Biol. Dept. 
> University of Michigan
> 4140C Nat. Sci. Bldg
> 830 N. University Ave.
> Ann Arbor, MI 48109-1048
> titusfranzmann.spaces.live.com
>
>
>
>
>   
> ------------------------------------------------------------------------
>
> _______________________________________________
> RASMB mailing list
> RASMB at rasmb.bbri.org
> http://rasmb.bbri.org/cgi-bin/mailman/listinfo/rasmb
>   

-- 
Department of Biochemistry and Molecular Biology
University of New Hampshire
Durham, NH 03824-3544
Phone: 603-862-2459
FAX:   603-862-0031
E-mail: Tom.Laue at unh.edu
www.bitc.unh.edu
www.camis.unh.edu