By Bhaskarar on Wednesday, June 16, 2004 - 03:00 am:

This is for your kind attention and clarification.

With reference to Ph.Eur (Page number 30 chromatographic separation
techniques) in which it is mentioned like below

Signal - to - Noise ratio

The signal to noise ratio (S/N) influences the precision of quantification
and is calculated from the equation.

S/N=2H/h

"H" is the height of the peak and "h" is the noise of base line.


Could you pl. confirm the Formula for Signal to Noise (from Waters system)
calculation, whether
we need to multiply with "2" or not.
And also suggest the exact procedure/analysis to be performed for the same
(means, number of blanks etc...,).

Thanks in advance,

Bhaskar Reddy

-------------------------------------------------------------------------------------------------------
By henrik on Wednesday, June 16, 2004 - 12:04 pm:

I think you multiply with 2 if you use the full amplitude signal for the baseline noise, and not see it as a +/- signal.

-------------------------------------------------------------------------------------------------------
By Bhaskarar on Wednesday, June 16, 2004 - 07:55 pm:

Thanks Henrik,

But is there any specification to do so.

-------------------------------------------------------------------------------------------------------
By henrik on Wednesday, June 16, 2004 - 11:36 pm:

Ohh.. I mean the other way around

sorry I have no specification, but maybe you can find some information on LOD/LOQ, here is what agilent says:

An important characteristic of any detector is its sensitivity to a particular analyte in a given analysis. The ability of an instrument to distinguish between the actual response of an analyte and noise is expressed as the signal-to-noise ratio. The whole chromatographic system including response of analyte, composition of
sample matrix, flow and temperature, etc, influence the limit of detection. The analysis system’s sensitivity for a specific compound is called the Limit of Detection. A quantitative definition of the limit of detection is the concentration of analyte that produces an output signal three times the root mean square of the
background noise.

-------------------------------------------------------------------------------------------------------
By Bhaskarac on Thursday, June 17, 2004 - 04:43 am:

Here my question is whether I can apply 2 for my LOD calculation or not for my validation studires?

Thanks

-------------------------------------------------------------------------------------------------------
By henrik on Thursday, June 17, 2004 - 07:47 am:

Maybe I don't understand you equation, but as far as I know S/N is calculated by the signal height of the peak divided by the full signal amplitude of the noise, so don't multiply with 2.

-------------------------------------------------------------------------------------------------------
By Anonymous on Friday, June 18, 2004 - 04:53 am:

The mentioned formula in Ph eur is for "Precision of quantitaion". pl. check.
For determination of LOD or LOQ by S/N ratio method, you simply divide height by noise, need not multiply by 2.

Refer some standard books on statistical methods in analytical chemsitry to correctly interpret the formula given in Ph.Eur.

-------------------------------------------------------------------------------------------------------
By Anonymous on Tuesday, June 22, 2004 - 06:16 am:

This is written in European Pharmacopeia: S/N=2H/h
I would apprechiate, if someone(responsible for that equation)would explain this "wrong" equation.
If I have a peak height as high as the noise, I simply multiply it with 2 to get an S/N of 2.
There must be something wrong.