Clear calibration during process doesn't delete old data

Hi, 

I try to understand why I have old injections data for only some of my target calibrated compounds and other are ok. I have no idea. 

First here how I usually proceed to my calibration with bracketing mode at "none".

When I check the curve, the majority of my 30 calibrated compounds have one calibration point like it is supposed to be, but some have old or more recent injection data when we look at those curve points.

How can this happen? How can this only happen to some of my compounds. 

The majority of them are ok with one point only. 

THank you for your help. 

 

  • Hello,

    I must be missing something in the description. I see 2 injections marked cal std so why would there not be 2 points in the calibration curves? I was wondering if there was an issue clearing the calibration when the peaks are not found but when I tested this for non EIC data it seemed fine. I will do some testing with MS data and extracted EICs. Unless I am able to replicate the issue, you will probably need to open a ticket so we can look at the actual result set in question. We cannot share data in the community site. 

    Marty

  •  Hi Martin. In fact, we have 4 calibration solutions. Each solutions are a mixture that contains a total of 60 target compounds. No compound in duplicate. If you check the curve of MTBE-EIC, we have 3 points coming from different injection date which make no sense since I did the clear calibration at the 1st Cal Std line injections. 

    I will place a call here in Canada and I hope they can help me. 

  • I agree it should have cleared out all the calibration data from previous injections with clear all injections. That is why we will need to look at the results. When you setup the method do you have 4 levels or 1? In a case where you have multiple standard mixes, I would have each mix as its own level so that you do not have to worry about identifying compounds that should not be in an injection. In the example below Peak A and Peak B are in one mix and the other peaks are in a second mix. 

    Marty

  • Hi Martin. I confirmed that I used 1 level calibration. 

    I used Curve calculation at "From average per level" because my internal standard are in each of the 4 standard solutions. 

    Every other compounds are there only once so I must be a one point curve to all. 

    I don't know why some compounds (maybe 6-8 compounds) had curve points from another injection date. 

    I will open a call with Agilent Canada or you think you understand the situation ?

    Here is another example today about another process. You can see that Chlorobenzene-EIC has 3 calibration curve points for unknown reason. 

    Thank you again for your help. 

    I tried to delete manually all the curve points that it is not OK, but I can't in the Bracketing mode "Overall". I will try to "none" mode. 

  • Ok, it seems that with the bracketing mode at "none" I can erase curve points that are not supposed to be there. 

    And on the report template we can still see the red point curve to show that their have not been used for the calibration. 

    I don't know why I have to do that for only few compounds and all others are ok. 

  • Hello,

    As I said open a ticket with your local support and share the results with them if possible. It is either an odd workflow interaction with your data or some corruption with the method. 

    Marty

Was this helpful?