boolean (intersample?) CF to evaluate the System Suitability

Zonghi

Hi everyone, 
I'm trying to create a CF to evaluate my analysis System's suitability feature (pass or fail function).
Given the fact that the parameters to statisfy come from different injection of the same sample set, I'm finding difficult to create boolean CF necessary for coding the pass or fail function. 

The System Suitability Test parameters are:
- On SST injection, verify that the resolution between Impurity 4 and main component is ≥ 3.0.
- On STD injections, the % RSD Peak Area must be ≤ 3.0%
- the % Recovery of main component for any bracketing standard as compared to the average peak area of the first six (6) injections of STD must be between 97.0% and 103.0%

SST (Resolution)      SST01
STD                         S01
STD                         S02
STD                         S03
STD                         S04
STD                         S05
STD (%RSD Area)      S06
SAMPLE 1                U01
SAMPLE 2                U02
STD (Drift)               C01
SAMPLE 3                U03
SAMPLE 4                U04
STD (Drift)               C02

Regarding the first two parameters 
I created a CF (named Percentual_RSDArea) to calculate the %RSD of the standard injections.
Then I've created a boolean CF with the formula: LTE(Percent_RSDArea,3)*GTE(CCompRef1[Resolution],3) but this function isn't specific for the SST resolution.
So I've tried to insert a Intersample formula: LTE(Percent_RSDArea,3)*GTE(SST01.%..CCompRef1[Resolution],3) but it didn't work. 
Should I keep the formula as the first one or is there a way to make the second formula work? 
And how can I calculate the Recovery (Drift) for the two bracketing standard and insert it in the boolean CF? 

thanks 

Empower2019

You could alter the first formula to only display at the SST01 sample, as follows:

LTE(Percent_RSDArea,3)*GTE(CCompRef1[Resolution],3)*(EQ(Label,"SST01"))+NEQ(Label,"SST01")*-1*50000 and make sure the All or Nothing is ticked yes in the CF editor, 

Empower2019

That last part basically tells Empower that if the label is not SST01 then to return a blank cell. Empower "sees" the number -50000 as a blank cell, its a coding syntax.  And -1*50000 is the same as -50000. This will only put the result you want in the sample labelled as SST01 instead of populating it for every other sample which you are not interested in. All or Nothing must always be ticked for these values to be recognized by Empower. 

There may be a way to incorporate the other requirement in the system suitability you are looking for but can you elaborate what you mean? My understanding is that the bracketing standards are Std (Drift) in your sample set and the area of the main peak in these bracket standards must be between 97.0-103.0% of the average main peak area of the S01/02/03/04/05/06 injections, is that correct?

Zonghi

Many thanks, it works! 
But, why you put the part +NEQ(Label,"SST01")*-1*50000 in the formula?

Zonghi

thanks for the explanation 

yes,
Drift= (Area STD(C01)/Average area STD (S0%))*100 and this must be between 97.0% and 103.0%

I created a intersample CF to calculate the average of the first six STD injection with the formula: S0%.%..AVG(CCalRef[Area]), and named it Average_Area
Then i created the CF Drift with the formula (CCalRef[Area]/Average_Area)*100. But this is too general..
Because if I try with the formula (C0%.%..CCalRef[Area]/Average_Area)*100 it doesn't work, I then tried to set the sample type box value to "control only".

I don't know how to incorporate this in the bool Custom Field 

Many thanks in advance, you are super!  

Empower2019

Your formula of (C0%.%..CCalRef[Area]/Average_Area)*100 wont work because you have too many periods in the first half of the equation and the position of the brackets are wrong. Intersample CFs are Label.Injection.Channel(Field) whereas Intersample Summary CFs are Label.Injection.Channel.Function(Field), so more characters. 

Change the formula to:
(C0%.%.(CCalRef1[Area])/Average_Area)*100 and this will work. Dont restrict the Sample Type to Controls Only as Empower will then exclude the Average_Area part of the equation since that was calculated for standards. So keep the Peak Type and Sample Type to All. If you want this CF to only display at each Std Drift then amend the formula and make sure All or Nothing is Ticked:
(C0%.%(CCalRef1[Area])/Average_Area)*100)+NEQ(Sample Type,"Control")*-1*50000