By Anonymous on Wednesday, May 21, 2003 - 03:46 am:
Greetings,
How do I calculate LOD and LOQ in gradient method.
The baseline is drifting and it must have influence on noise calculated by software. The result is that the S/N ratio is well below 2 when you can see that it is still far away from LOD and the RSD of several injections is around 1,0%.
What am I suppose to do? Any other approach? Maybe there is something I can change in the program? (I am using HPLC System Menager from Merck).
Looking forward to your suggestions.
By Anonymous on Wednesday, May 21, 2003 - 04:24 am:
take a large scale printout and determine noise manually with a ruler.
If you use the software you should understand how the software calculates the noise and if the method used is appropriate to your chromatogram (drifting baseline).
I'm not familiar with to merck software. I have only used chemstation and it takes the drift into account and has three different methods for noise determination (below). I'm still calculating noise manually in the method validation being easy to justify..
1. as six times the standard deviation (sd) of the linear regression of the
drift,
2. as peak-to-peak (drift corrected) and
3. as determined by the ASTM method (ASTM E 685-93)
By Anonymous on Wednesday, May 21, 2003 - 06:33 am:
normally you can select the area wherein the software checks for the "noise".
I assume now that your noise is pretty much the same at the start and at the end of the run. If so, all you have to do is to select a relatively straight section of your baseline and use that for the peak to peak estimate.