USP S/N ratio calculation using Empower2

<p><span>we are trying to calculate S/N ratio as per USP using empower2 software.</span></p><p><span>In processing method there is an option for calculating EP s/n ratio. By checking this option we can get value for EP s/n.</span></p><p></p><p><span>There is another option in processing method named Noise value for s/n which allows us to choose from one of the following</span></p><p><span>-Baseline noise</span></p><p><span>-peak to peak noise</span></p><p><span>-Average peak to peak noise</span></p><p><span>-detector noise</span></p><p><span>-average detector noise</span></p><p></p><p><span>By selecting each one of these individually, different value for s/n (not EP s/n) can be obtained. We are not sure which is right selection.</span></p><p></p><p><span>does anyone has experience on how to calculate s/n ratio as per USP using Empower software.</span></p><p></p><p><span>Thank you</span></p><p><span>Jit</span></p>

Answers

  • Hello

    As less familiar with these calculations within Empower as I could be - we are consulting the experts.

    We are searching for the current s/n definition in the USP, but there are a number of S/N calculations. Please bear with us.

    AS you note there are automatic calculations, but requires that you have an area where no peaks elute to take the noise segment correctly. This may be causing part of the differnce

    If needs be we will put you in touch with your local data specialist.

    There should be some information in online help - see "Calculating Detector Noise and Drift" - type into search to pull it up.

    Liz

  • Hi,

    Signal-to-noise ratio as per new USP formula can be calculated using empower software. You need to create custom field and custom calculation.

    USP formula for S/N ratio :

    The signal-to-noise ratio (S/N) is calculated as follows:

    S/N = 2h/hn

    in which h is the height of the peak corresponding to the component concerned; and hn is the difference between the largest and smallest noise values observed over a distance equal to at least five times the width at the half-height of the peak and, if possible, situated equally around the peak of interest.

    I will take routine example from regular analysis. Peak RT is at 14.122 minutes.

    Name

    Retention Time

    Area

    % Area

    Height

    Width @ 50%

    1

    Peak

    14.122

    1974

    100.00

    99.25

    0.3100

    Width of peak at half height is 0.3100mnutes. As per USP we have to select the noise around peak interest over a distance equal to 5 times of width at half height

    i.e 5 x 0.3100 minutes = 1.55 minutes.

    So I have chosen distance from 15.0 minutes to 16.55 minutes where noise is less.

    Enter this start and end time under processing method window in "Noise and Drift" .

    Create a custom field as below

    Signal to noise ratio = (2 x Height of peak )/(Peak to peak noise x 1000000)

    Note : 1000000 on denominator is because peak height in empower is measured in mv, hence 1000000 is used in the denominator.

    Once custom field is created , then empower software will calculate S/N ratio automatically.

    In above case, manual calculation is as follows

    Signal to noise ratio = (2 x Height of peak) /(Peak to peak noise x 1000000)

    = ( 2 x 99.25 ) / (0.00000774 x 1000000)

    = 25.6

    Note : Peak to peak noise obtained from Result window

    Automatic calculation from empower software is as below

    Name

    Retention Time

    Area

    % Area

    Height

    Width @ 50%

    Signal_to_Noise

    1

    Peak

    14.122

    1974

    100.00

    99.25

    0.3100

    25.7

    Hope it works for you

  • Hi there...

    Well does this seem to be a hot topic here in Milford, being on the instrument side I was not aware of the controversy! So thank you for sharing this will be of interest I believe to many other users.

    Therefore, I will be forwarding a discussion paper that is being prepared on this very subject. The programmers in Empower are reviewing your algorithm and so if you will bear with us for a little longer I will have a full answer for you shortly.

    Liz

  • Thank you for your answer.

    I tried to find USP formula for S/N ratio in current usp but could not find it.

    Can you please mention the chapter # or page # for this formula in current USP

    Thank you

    Jit

  • Hi,

    Formula for signal to noise calculation is newly included in USP 32-NF 27 through Second Supplement.
    Information in this edition of USP-NF will become official on December 1, 2009.

    It is available in general chapter < 621> CHROMATOGRAPHY.

    Regards

    Rajnil

  • Hello

    Attached is the dicussion paper and also to add to the comments already here, rather than using a scaling factor as you have done in your custom field, it is preferable to use the Scale to µVfield within your formula. The Scale to µVfield is automatically determined for each result in Empower 2 and is a value used to scale the Height value to the same units as the Noise value. This is necessary because different detectors provide signals in different voltage units. The Scale to µVfield is automatically populated with the appropriate conversion factor. This allows for automatic scaling conversion using data from any detector without any modification to the custom field's formula. The attached document provides the recommended custom field in Empower 2 for USP S/N.

    Best regards,

    Liz

  • Hello,

    I have been trying to make this work, but the "Scale to µV" field is not recognized when I enter it into the Custom field.  I've entered the formula as written in the Technical note TECN10123052 (as far as I can tell, anyway).

    Is there something painfully obvious that I'm missing?

    Anyone else had this experience?

  • Hi Dfraser,

         This error message is oracle complaining about the syntax of your variable name.  It will not tolerate spaces or "wild card" type characters.  Replace the spaces with under scores, or simply eliminate the spaces all together, in your variable name for the custom field definition.  Hope this helps.

    Eric

  • Hi

    Did you type in the Scale to uV or did you select it from the available fields in the formula? If the first try and select it from the list of available fields.

    best regards

    Rune

  • Thanks for the advice.  The problem seems to be that I've misread the technical note. "Scale to µV" is not available as a predetermined function.  It's not particularly clear, but it seems that "Scale to µV" has to be defined and entered as appropriate for each detector. I thought that it was supposed to be a value that would be generated by Empower.  Silly me.

    In a related question, does anyone know what the units for "Peak to Peak Noise" are?  Is it V?  The technical note is not clear on this - I got the impression that it's in µV, but that doesn't seem to make sense with the values I get.

  • Hi again

    The peak to peak noise units is detector units so they are different from one detector to another. If it is a UV it is in AU where 1 AU = 1V. This is why you need to use the fields scale to uV (it is a standard empower field), to get the rigth translation to units.

    best regards

    Rune

  • Hi Rune,

    If the noise units are in detector units, why wouldn't the height units be the same?  Then no scaling would be required.  An idle thought...

    Scale to uV is definitely not available in the list of fields in the version of Empower I'm using.  Nothing that even looks close to it.  Looks like I'll have to talk to Waters.

    Thanks for your help.

    Dave

  • Hi Dave

    Height units are in uV, so that is why we need the scaling factor for the noise value.

    Are you sure you do not have the Scale to uV in the custom fields wizard when creating a peak type custom field ? What version are you on ? As far as I remember it was introduced with Empower2

    See atachement for where to find it.

    Best regards

    Rune

  • Hi Rune,

    This has turned out to be a fool's errand.  We are using the original Empower.   About to upgrade to Empower3, so I didn't realize that the current version wasn't Empower2.  Explains everything.

    Many thanks for your explanation of the noise and peak height units - at least now I can manually enter a scaling factor and understand why.

    Regards,

    Dave