Customfield to calculate Total betacaroteen

Empower2019

Thats odd that the -50000 formula didnt give the correct result. Did you definitely include the *100 part of the formula?

boesswx

I will let you know asap

Empower2019

As for the intercept, if I understand you, you want the Intercept value for a particular standard called Std3? Is this just a one injection standard? The intercept equation for a standard calibration curve is Y=Bx +A, so if you want to know what the intercept is, you could do it two ways: an intersample CF Peak, Real, Calculated, Search Order Result Set First, Sample Type STANDARDS only, Peak Type All.

Formula then is

Std3.%.(B)

Or just have the formula as B and this will show the intercept of whatever standard you are reviewing at that moment. If you want a table of this, add a Peaks Table and expand the calibration curve tree and add "B", take off the exponential notation and this will display the full intercept value for that particular standard.

boesswx

I now use  -50000 formula  for retention time calculation and it works fine.

As for the intercept i am used to the equation y=ax+b, where b is the intercept and a the slope. And b is cut off of the y axis(product of ax =0). So in your calculation is B the slope where is refers to. So in your case we need the a/height or area of std3*100

Empower2019

In order to know what CF to use for your case, I need to know what "Fit Type" you use in your processing method components tab. Is it Linear or Linear Through Zero or something else?

boesswx

Fit type is Linear

Empower2019

Ok thanks. Empower 3 slope in a calibration curve is represented by Y=Bx + A, so B is the slope. Create a peak, real calculated custom field. Leave the search order and sample types as default and make sure to tick All or Nothing. Formula is:

((B*100)/Std3.%.(Area))*EQ(Label,"Std3")+NEQ(Label,"Std3")*-1*50000

Label your standard as Std3 (or whatever suits, this is just my example). This formula will divide the slope of the calibration curve for your component by the area of Std3 and multiply by 100. The last part of the formula ensures the result only populates for the Std3 line, all other lines will be blank.

boesswx

Ok. In this case I need A for the calculation. I forgot to tell, it's 2 channel injection. And one channel(component) is calculate with height and the other one with area.

How can I indicate in the formula which channel must be calculate?

Empower2019

Empower 3 sees the equation as Y=Bx + C, so i presume you want C?

For 2 channels, you could make 2 CFs and just substitute height for the one i mentioned. Or you could incorporate them into 1 CF. You need to know the exact name of the 2 channels for the formula. View your injections/sample sets as Channels in the project and the Channel name will appear by default. If the Channel Name is very long, make 2 derived channels in the method set used to acquire the samples, and give them short names like A and B. Save method set. The formula then goes:

EQ(Channel Name,"A")*((C*100)/Std3.%.(Area))+EQ(Channel Name,"B")*((C*100)/Std3.%.(Height))

That should work. 

boesswx

I used A and the result is what I expected. In the help topic intercept= A en slope=B

I will make another formula with height for the other component and if that doesn't work

I will try the last formula you send, maybe I try this anyway

boesswx

I have send you a testreport with the calculate result. The results for A=0.18338 and for E=0.04970 are the right one. Is it possible to clear the other one(read it in second part of template from " Analyseresulaten".

For the column Intercept_STD3 I used: EQ(Channel,"W2489 Ch B")*((A*100)/S0103.%.(Area))+EQ(Channel," W2489 Ch B ")*((C*100)/S0103.%.(Height)Can I use component and then exactly use the name between the bracketing commas? I read somewhere that that isn't possible

For the column A_rel_STD3_A I used:((A*100)/S0103.%.(Height))*EQ(Label,"S0103")+NEQ(Label,"S0103")*-1*50000

For the column E_rel_STD3_E:((A*100)/S0103.%.(Area))*EQ(Label,"S0103")+NEQ(Label,"S0103")*-1*50000

Empower2019

Yes, if you only want the A_rel_STD3_A to calculate for one component and the same for E_rel_STD3_E , then pick this component in the CCalRef1 field of your processing method, reprocess and the value will only populate for that component. So you need a different processing method per channel. In this case, your 2 formulas are:

((A*100)/S0103.%.(CCalRef1[Height])*EQ(Label,"S0103")+NEQ(Label,"S0103")*-1*50000

and  

((A*100)/S0103.%.(CCalRef1[Area]))*EQ(Label,"S0103")+NEQ(Label,"S0103")*-1*50000

boesswx

The results are negative, is it enough that use ABS and where must I place that?

-0.031 and -0.160 are the right one, still values in the other fields, despite selectCCalRef1

Empower2019

If you only want values for the component in CCalRef1, then specify the channel in the CF formulas, such that:

((A*100)/S0103.%.CHANNELNAME(CCalRef1[Height])*EQ(Label,"S0103")+NEQ(Label,"S0103")*-1*50000

As for negative answers, i dont work with the "A" field, so i presume A can be a negative value, since the formula itself wouldnt produce a negative value. You can put an ABS around the argument, that might work:

ABS(((A*100)/S0103.%.(CCalRef1[Height]))*EQ(Label,"S0103")+NEQ(Label,"S0103")*-1*50000

boesswx

Hello again,

I want to check the slope of calibration, it must between to levels.

I used Boolean for the CF:EQ(Label,"S0103")*RANGE(B,CConst2,CConst3), where B is the slope and CConst2 and 3 are in the procesmethod. But it doesn't work, can you help on this one?

Empower2019

Your formula multiplies 2 Boolean expressions so your final answer is going to be 0 or 1 ie false 1st expression by false second expression (0*0) or correct both expressions (1*1), which is no good for what you want. 
If you want to test the range of B in S0103, create a peak,bool cf and use as position. Formula:

RANGE((S0103.%.(B)),CConst2,CConst3) and edit your true and false translations from the default Yes/No to your own values. Be aware that RANGE is GTX and LTEY so if your range is between 10 and 20 and you get a value of 10, Empower will exclude this value and return a False. Might be easier to round your range to the appropriate decimal place eg a range between 90 and 100 would need you to put 89 in CConst2 and 100 in CConst3.

boesswx

I can't get it to work, what I am missing. It 's the two channels results,  maybe something to do with that or do I forget something in the process method?

An other question about a composed report.Is it possible to report checks from differentinjection/results to one report. What I mean, for instance i want to check the slope of S0103 and I want the standard precision of 3 other other injected standards?

Is possible to use/make a CF for this or otherwise in empower?

Empower2019

Thats strange, it worked fine for me when I tried it on a sample set that had 2 derived channels per injection. Is your sample type definitely set to standard? Do you have CConst1 and 2 populated against the components you want tested? The way I did it was to associate a processing method with the derived channels in the method set and processed a sample set using this method set and my search order was Result Set Only. You can either use the same processing method for the 2 channels, in which case make sure to have the range in CConst1/2 set as you like for the 2 components or you can put 2 separate PMs against each derived channel, which is probably longer.

You don't need to name the channel in your formula as a channel with no name instructs Empower to perform the calculation on all channels associated with that injection, be it 1 or 10, and these channels can be instrument channels or derived channels. I presume you want the calculation done in both channels.

As for your second question, so you are looking for a peak table which will display the precision for a number of injections as well as the result for the above CF, is that right?

boesswx

Indeed it must my search order was Result Set Only. Now it works.

As for the second part, actually we have to do 4 checks each with his own injection(s)

1-slopecheck with 1 injection of S0103,

2- STD-Precision with 3 injections of S0102,

3-Rt duplicate difference with a injection of a mixture Vitamine E (label TS)

4-Resolution check of the mixture(label TS)

boesswx

I forget intercept-checks is also of S0103

boesswx

Maybe is it possible to give result for example STD-precision; must be lower or equal to 3%, a pass or fail function? applies also for the numbers 3, 4 and 5(=intercept-checks ) in the report. But must probably with a boolean-CF, equal as for the slope check?

Empower2019

boesswx said:

Indeed it must my search order was Result Set Only. Now it works.

As for the second part, actually we have to do 4 checks each with his own injection(s)

1-slopecheck with 1 injection of S0103,

2- STD-Precision with 3 injections of S0102,

3-Rt duplicate difference with a injection of a mixture Vitamine E (label TS)

4-Resolution check of the mixture(label TS)

You could put 3 of the 4 results into one All Peaks Table in your report by using suitable filter conditions. Filter by Label = S0102 and S0103 and TS, this will pull in data from those injections into report. Then add the custom field for slope check, RT difference and resolution check as well as any other identifying fields such as SampleName, Date Acquired etc. This will display the results for the 3 custom fields for those samples in the one table.

For precision, are you measuring the height/area/amount %RSD or Mean over 3 injections? You can either do this in a separate peak table by adding the Mean to the column properties of the Area/Height column or you can create a peak, real custom field with the formula:

S0102.%..AVE(Area) and put Summarize Custom Fields in the sample set before you generate a result set. If you want to further evaluate a Pass or Fail for this precision based on 3%, then instead create a peak, bool custom field with formula:

LTE((ROUND(S0102.%..%RSD(Area),0)),3)

Or if your 3% tends to vary when spec changes, change the formula to the below and make sure to populate the spec value in CConst1 against the component whose precision you are measuring.

LTE((ROUND(S0102.%..%RSD(Area),0)),CConst1)

boesswx

the composite report was successful, maybe some finetuning here and there.

Thanks again

boesswx

Good day,

Since they created each half year a new project and they copy methods, view filters and CF's from an old project in the new project. There is a problem with a part of the report-method specially where we use CF, in the report parts are missing. Are you known with those kind of issues and what can we do about it?

Empower2019

Is the CF part of the report missing? Sometimes if parts of a report is missing, it means there are CFs in the report method that are not in the project, so double check the same CFs exist in both the new project and the report method that is missing them. Also, sometimes reports get corrupt when copying from 1 project to another. Make a new simple report method with those CFs in them and try that- if it populates in that, the old report might be corrupt.

Also, check that the CFs are definitely being calculated right- labels, search order, sample type etc. And before you check the report, check that the CFs are present in the result set. if they aren't in the result set, they wont be in the report.

boesswx

For the standardprecision I now use a reporttable to calculate the %RSD of 3 injections.(See I want to use the SP.%..%RSD(Area) etc. But the results isn't what  I expected; see att.

The CF's are as follow:

For Vit A: EQ(Channel,"W2489 Ch B:)*SP.%.%RSD(Height)

For Vit E: EQ(Channel,"W2489 Ch A:)*SP.%.%RSD(Height)

SP stands for standardprecision.

What am I missing or what I am doing wrong?

The copying problem(CF was missing) of the report from one project to an new one, was error in the oracle database: an invalide identifier.

boesswx

The formula for E is

For Vit E: EQ(Channel,"W2489 Ch A:)*SP.%.%RSD(Area)

Empower2019

Hello again. It looks like the reason the vitE std precision CF is giving back 0.000 is because of your formula. Dont specify the channel name in the syntax and Empower will calculate for all available channels of an injection or series of injections. So your VitE CF should read:

SP.%..%RSD(Area). Try that and see if it gives the correct result for VitE in that channel. Dont forget to add Summarize Custom Fields as the last line of sample set. 

Or...you could just add a composite order to your report method by grouping at least 2 report items together example title and table, right click composite properties and order by Channel Ascending followed by Name Ascending. Then you get separate tables for each channel and component in that channel- add %RSD to the Area/Height.