Custom Field Calculation

[FBCena]

Hello,

I'm having trouble creating a custom field in empower I hope you can help me.

So, there was a post similar to my question a long time ago here on the forum - viewtopic.php?t=17231

It has very useful information that almost solved my problem

What I want is:

A field that averages multiple injections with the same label.

Vial --- Inj --- Label
1 ------- 1 ------ A
1 -------- 2 ------ A

This should be resolved with the following field

Search order: Results Set only
Sample Type: Unknow only
Peak Type: Found only
Formula: SAME… AVE (Amount)

But it does not work

Then I tried the following formulas:
- SAME.% .. AVE (Amount)
-.% .. AVE (Amount) (this does not specify the label but served as a test)

None worked

But when I change the Search order to Outside Only the formula works perfectly, but it is using previously processed results with the same label in the same project, which is not what I want

I wanted the average of the injections done in the current processing


What am i doing wrong ?

[DR]

You really do not need a CF for this. Just make a peaks table, double click on the column for areas or whatever you want and add a mean to your summary statistics. Group by label on the Order By tab and you're done.

[EmpowersBane]

As DR said, this is a massive waste of time to use CFs for this when you can use the report method to group a table by Label and add stats at the end of your column.
However, if you need the CF for use for further results/CFs, then:

Did you try and insert the function "Summarize Custom Fields" as the last line in the sample set? This is required for all summary CFs. Your second CF was the correct one: SAME.%..AVE(Amount), this will return the average amount of all found peaks in unknown samples with the same label.
Double check the samples are of the Unknown Type, and that you process a sample set to generate a result set, hence your search order of Result Set Only which is the norm for 99% of cases, ive never come across a case where the other 3 are used.

Another thing which may be causing it to fail is if your Amounts are calculated using several bracketing standards of the sliding with overlap type and your labels spill over both brackets. For Amount to be averaged correctly, the only standards set-up which accommodates this is either all standards injected at the start, followed by all unknowns or..standards injected throughout run to generate one calibration curve with which all the unknowns are quantitated. If you chop and change the way you calibrate your standards such that one batch of ten samples with the same labels uses several calibration curves, you will get a value for Amount but average amount wont correctly calculate for some reason..