Y-axis unit (mAU/mV)

[jitender]

Hello Everybody

I was reading user's manual for shimadzu HPLC. Shimadzu UV detector uses 'millivolt' as y-axis unit and PDA detector uses 'milli absorbance unit'.

is there any specific reason for using different units for UV and PDA?
Probably 1 mV=1mAU; but it may be different and linearly related.

[DR]

Guessing here - you may be able to get the PDA to read out in mV or in mAU.

AU/V (1V is full scale) is adjustable, at least in their older UV/VIS detectors - probably still is... so mV and mAU may or may not be equivalent.

[jitender]

It may or may not be equivalent. What is need of adopting scale of volts for Y-axis. 'Absorbance-unit' is more appropriate for detection.

[tom jupille]

In practice, it doesn't matter, because you should be running your calibration standards on the same detector with the same settings. The y-axis units on a calibration plot are arbitrary.

[Alfred88]

Dear jitender:
We also have the Shimadzu LC. I believe that the ratio of AU/V (or milli Absorbance Unit per mV) can be modified to some values in Class VP,such as 2, 1, 0.5.
In Class VP, open Method > Instrument setup, then go to the Tab (eg. SPD-10AVP). Look for Analog output 'frame', and change the 'Aux. Range'.

We found that the Help file in Shimadzu software is not helpful for this Aux. feature. Also, the print manuals have nothing to do with the software.
Good luck.

Alfred

[jitender]

Thanks Alfred. I found it in user's manual and it specifies adjustable output scaling under Auxilarry range. By default auxillary range is set at value 2 and that corrsponds to 1.0 AU/V.

What is use of such a scaling?

[Alfred88]

That value is related to the 'gain' of the analog signal.
Alfred

[jitender]

If this value is linked with gain in signal, can we further lower the LOQ by seting higher sensitivity(setting lower AU/V?).
if I am taking the concept incorrectly, can I have a literature reference where I can read more about it.

Thanks

[Alfred88]

Dear jitender:
In general, when you amplify a signal (gain >1), you will also amplify the associated noise, so you have basically the same S/N ratio (this ratio is related to the LOQ). I quote from an old text: (Instrumental Methods of Analysis, 6 Edition, by Willard Hobarth et al, 1981, p 847)

The S/N ratio can only be improved by decreasing the value of the noise. Since each increase in the magnitude of the signal will be accompanied by a proportional increase in the value of the noise, amplification cannot improve the S/N ratio.

With the same sensitivity, when the noise is lower, the detection limit is considerably lower. There are 3 sources of noises: Thermal, Shot, and Flicker. A newer text in Instrumental analysis may have discussions of the circuits used as filters for noises.

In Class-VP, by changing the Aux. range, you can change the response factor, and the slope.

BTW, I don't work for Shimadzu.
Alfred.

[jitender]

Dear Alfred

I understand that any amplification will increase noise as well along with signal. Actually I am becoming over-optimistic by the way noise and signal are measured.

By ASTM method

1. noise is maximum vertical distance over some time interval.
2. signal is measured from the half of this noise level.
3. if noise increases along with signal, it will be beneficial to signal to noise ratio because of measurment of signal from half the noise level.

[tom jupille]

2. signal is measured from the half of this noise level.

Actually, I think this means that noise is measured from the midpoint of the noise. Increasing the gain will have no effect on S/N unless/until you increase it so much that you "flat top" the peak.

[jitender]

I would like to point an article by John W. Dolan, in LC-GC Europe, January 2006, The Role of the Signal-to-Noise Ratio in Precision and Accuracy at url:

http://www.lcgceurope.com/lcgceurope/ar ... ?id=284453

from original text:

Manual measurement of S/N can be performed, as illustrated in Figure 1. Select a section of baseline free of peaks, expand the plot scale so that the noise is easy to determine, and draw two lines tangent to the noise. The vertical distance between the two lines is the noise for the chromatogram. The signal is measured from the midpoint of the noise to the top of the peak. S/N is obtained by dividing the signal by the noise.

In figure 1, it can be seen that signal is measured from the half of distance of maximum noise (over an interval).


Copyrights of above image (Figure 1) are property of original owner.

[Dan]

The measurement of the S/N was discussed previously. See this link:

http://www.sepsci.com/chromforum/viewto ... ight=noise

Noise is electrical/electronic in its nature. Noise peaks (spikes, etc.) can swing both positive and negative from the baseline.

The chromatographic peak (for the most part) is only either a positive peak or, sometimes, a negative peak that arises from the baseline.

So noise is measured in both directions from the baseline and the chromatographic peak is measured in only one direction from the baseline.

Note that there is more than one type of noise measurement. Please see the previous discussion thread given above.

Regards,
Dan