Test method transfer

[singil]

I am transferring a validated GC Test method from a site the uses a Perkin Elemer to one that uses Agilent. I need to change some operational parameters - injection time, split ratio to execute the method on an Agilent GC. Do I need to do a robustness study prior to transfer?

[rb6banjo]

Are you operating under any specific guidelines that require you to do so? I'm assuming not since you're asking the question. I assume you're using the same column for both instruments so the separation part should be the same, independent of the instrument.

For me, it would be sufficient to set up the instruments, calibrate them, run a particular sample a number of times on both and make sure that you're getting:

1) the same answer on both instruments (withing statistical variation)
2) the precision is similar or at least acceptable on both instruments.
3) acceptable signal-to-noise for your analyte at the limit of detection and limit of quantitation.

Accuracy, precision, and good detection limits is all I really need out of an analytical method.

[Consumer Products Guy]

I need to change some operational parameters - injection time, split ratio to execute the method on an Agilent GC. Do I need to do a robustness study prior to transfer?

I think that robustness study should have been part of the validation, so you'd know if stuff like injection time is critical. Can't the Agilent be programmed to match up?

[James_Ball]

I need to change some operational parameters - injection time, split ratio to execute the method on an Agilent GC. Do I need to do a robustness study prior to transfer?

I think that robustness study should have been part of the validation, so you'd know if stuff like injection time is critical. Can't the Agilent be programmed to match up?

Unless it is one of the newest Agilent 7890s I believe you only have the option of Fast or Slow injection speed on their autosamplers.