I need to get the chemstation to average all calibration injections throughout a sequence, and apply the average response factor to all of the samples in the sequence. This does not seem to be an option within chemstation. I have been told that a custom chemstation macro could probably carry out this task fairly easily but I have only limited knowledge of writing macros. Has anyone else encountered this problem and found a solution?

It has been a while since I have used Chemstation, but if I remember right, when you create your sequence there is an option as to whether to overwrite standards or average them. I used to overwrite the standards for the first injection of each concentration, then average any subsequent injections. If you don't overwrite the first set of standards, the software will average in any previous data that is in the curve. Hope this helps.

Thanks Michele.
You are right in what you say, but my analysis consists of the following structure; standards/samples/standards/samples/standards/samples/.....etc and I need to average ALL of the calibration standards prior to applying the single response factor to ALL the samples. I realise that this means no results will be available until the end of the analysis. Currently, we manually calibrate with all the standards at the end of the sequence and then reprocess the sequence. We need to automate this process.

Michele is correct, we usually have first line in the sequence as Replace, then second line as Average. Michele's way uses the average of her last injection on the Replace line and all the injections on the Average line to calculate results for all subsequent samples. John, there are two ways to do what you want. You could use the Chemstation bracket function, which can be a little confusing, follow the directions carefully. Or you can simply reprocess: first reprocess all the standard injections only using Partial Sequence (the average of all will then be in the computer's memory); then reprocess the sample injections only using Partial Sequence. Remember to change filename of sequence summary report so the file won't be overwritten, or rename the file using File Manager/Explorer. We mostly use sequence summary report with Compound Summary and %ESTD to get results directly in Percent.

Hi !!

I'm new on the Chemstation and I'm tryin to use the chemstation software to quantify pharmaceuticals samples directly un the same way as USP monographs said.

We quantify the PAI with two different weighed standards and two different samples. Three injections of each and finally we make the average of all the results.
The problems is that the system doesn't understand that I could have two different standard amount with the same calibration level, and then only uses the first calibration set for all the samples.

Can you give me an advice

Thanks and sorry for my english expression

Sounds like you need to set up a level 2 calibration.

Thanks for the replies. I like the partial sequence idea, although I have just tried it and realised that it gives me a rolling average of all standards and not a true average of the response factor (eg av of 1,2,3=2; rolling average=2.25!). Shame, it sounded interesting although would still not be fully automated of course.

It sounds like the other problem does need a 2nd level set up in the calibration table. This can be set up even if the second standard is similar in concentration to the first. You will effectively end up with an average factor for the two standards as long as you force the line through the origin. Hope this helps.

If anyone has more ideas on how to get the chemstation to give a true average of response factors automatically, please let me know.

Why the need for so many stds? Why wouldn't the bracketing with 2 stds be acceptable?

We have long sequences and have found that averaging all standards gives us the best results. If a slightly spurious standard injection occurs, bracketing can lead to poor results for samples within that bracket. Averaging helps to accomodate this. Thats what we have found. Bracketing can be good sometimes, especially if there is a noticeable drift through the sequence. However our analysis does not show drifts. Just the normal random fluctuations.

Hi John. An idea to completely automate your sequences could be:
1) prepare a calibration table with different levels, one for each standard (in this way you will get a true average)
2) prepare the sequence with all the standard at the beginning, setting them as calibration samples and using "replace" for the option "Update RF". At the end of the standards injections, your calibration is ready without further participation and you can continue with the samples injections.
3) inject again the standards during the sequence, but setting them as "control samples" (just to check that you really don't have drift).

Good work, Olivia

Thanks Olivia,

That's a very interesting solution to the problem. I still cannot believe that the function I wish to perform is not standard in the chemstation pacakge. Maybe it will be addressed in later revisions of the software.