SBy Fahim Gouveia on Sunday, January 11, 2004 - 03:36 pm:

Hi all,
Can anybody tell me under what signal to noise ratio (S/N) quantitation is not reliable anymore? How can we relate the S/N to the precision (accuracy) of quantitation? What is the document or literature that gives specifications on this issue?

Thank you very much for your time and attention.

-------------------------------------------------------------------------------------------------------
By Anonymous on Sunday, January 11, 2004 - 09:36 pm:

Generelly, limit of detection (LOD) is set to 3:1 S/N and limit of quatitation (LOQ) to 10:1 S/N.

You should also test the injection precision at LOQ by preparing a sample that gives the S/N 10:1 ratio. This is usually max RSD% 10 or 5.

-------------------------------------------------------------------------------------------------------
By Fahim Gouveia on Monday, January 12, 2004 - 07:45 am:

Thanks for the reply. What is the source (eg.,document or literature) that specifies these values? What is the meaning of RSD% (I´m sorry, but never heard about it)?

-------------------------------------------------------------------------------------------------------
By Anonymous on Monday, January 12, 2004 - 08:23 am:

What is the source (eg.,document or literature) that specifies these values?

Try checking the USP.

What is the meaning of RSD%

RSD=relative standard deviation

-------------------------------------------------------------------------------------------------------
By syx_gf on Thursday, May 20, 2004 - 05:31 pm:

%RSD = Coefficient of variation (RSD x 100)

RSD = SD/mean
%RSD = 100 x SD/mean

-------------------------------------------------------------------------------------------------------
By Chemist on Tuesday, June 1, 2004 - 04:17 am:

Fahim, are you sure you working with LC?
Never heard of %RSD?
How do you now precision of instrument injections, sample reproducibility, etc...???