Manual Integration Error - Index was outside the bounds of the array

Greetings Agilent Community,

I see there are other posts that relate to this mysterious error, but none address my specific issue.

My task is simple: correct the default integration applied to blank samples either by zeroing the peak or manual integration. However, I am often thwarted while attempting to redraw the integration limits for my compounds. An error is thrown that reads "Manual integration failed for compound X in sample Y. Index was outside the bounds of the array." Alternatively, in other instances I do not receive this error, but the integration limits that I draw do not stick. Instead, one limit will snap to the very edge of the integration window and refuse to be positioned elsewhere. Very strange. Not sure if both phenomena have the same root cause, but they are happening concurrently. Has anyone else had this experience or know of a solution to my integration frustrations?

As always, your attention, insight, and advice are greatly appreciated!

Thank you,


MassHunter Quantitative Analysis for QQQ, Version 10.2, Build 10.2.733.8

  • Hello  ,

    Manual integration is limited to the rules of the selected integrator and the settings of your quant method. The error and behavior you are seeing typically indicate that the peak you are attempting to create is too far outside of the bounds of what the chosen integrator could find. If you are not able to manually integrate a peak first make certain that the integrator peak filter is set to 0, either for height or area. 

    If allowed by your procedures, you can try choosing a different integrator for a compound on a sample-by-sample basis and then attempt manual integration. In the Batch table view go to the Compound Information window, right click, and choose Integration Parameters... This will allow you to change the integrator for this compound on this run only. 

    You can also try to increase the data extraction window for the compound in the RT Setup portion of your method. Manual integration still requires a certain amount of baseline information, and this can be another reason manual integration fails. 

    Another setting to adjust, if possible, is the Reference and Non Reference Window settings in Globals Setup. Reference Window sets the identification window for Time Reference compounds and Non Reference Window sets the ID window for all other compounds. If these ID window limits are set to be very narrow, say less than 0.05 minutes or so, then manual integration may not behave as desired. Also check the Correlation Window size if you are having issues manually integrating qualifiers.

    There are enhancement requests to have manual integration give better feedback about why a particular integration is rejected, but currently there is no timeline for when these features will be added. 

  • Thank you   for the excellent information.

    The integrator that I selected defaulted to filter by Peak Area (%). I selected Peak Area (count), set to 0, for a couple compounds and had a better experience. However, I cannot change peak filter parameters from Compound Information in Batch view as you say. The integrator can be changed, but the peak filter parameters are grayed out. I can only change them in the method editor. This is just a little inconvenient as I would prefer to limit my changes to the compound of interest in the selected sample, and not re-analyze the entire batch. Is there a setting that I'm missing that allows me the specificity I'm looking for?

    To your other points, I don't think increasing the data extraction window would not do much for me since I am applying a dMRM method, so extraction windows are already set to their respective limits (unless I am misunderstanding). I did not try adjusting the Reference and Non-Reference Window settings in Globals. Although, they do not appear to be very narrow. Ref. = 70% and Non-Ref. = 80%, again by default.

    Thanks again! Best regards,


  • Hello  ,

    I apologize for the delayed response. Only the integrator and its parameters, if any, can be changed on a sample-by-sample basis for a compound. Peak filtering is applied at the compound level for all samples. 

    However, the Universal integrator does have an area reject timed event that you could utilize, if you are allowed to make that change. 

Was this helpful?