GC Instrument Calibration

[surinzp]

Can anyone tell me how to calibrate a Shimadzu GC instrument witha FID detector please.

[analej]

Can anyone tell me how to calibrate a Shimadzu GC instrument witha FID detector please.

what do you mean with calibrate?
only you need to check flow with flowmeter and there is alowance that is acceptable, also T in SPL, oven and FID and performance test

[chromatographer1]

analej is correct. Please explain more clearly.

What do you mean, your question is unclear. A GC is not 'calibrated', there are no adjustments or measurements against another GC for which you would perform a calibration. That being said, you can certainly check the response of a FID detector and determine if the sensitivity of response is appropriate.

But subtle changes in gas flows can make sizable differences in response to an analyte, but this isn't bad or necessarily wrong, or 'out of calibration', as it may be necessary to perform an analysis on a certain matrix for a certain analyte.

Would you be more forthcoming?

Rodney George

[surinzp]

Hi analej and chromatographer1,

thank you for your reply. I want to perform a 6 monthly or a yearly calibration checks for the GC.

I just want to know the list of the parameters i need to check and how do i perform them?

Parameters like Flow and FID response, Temperature,Inj Volume etc...

Thanks

[chromatographer1]

When the GC was purchased the delivery and installation was by a technician supplied by the manufacturer. They set up the GC and performed a series of tests which should be in the manual for the GC.

Perform these tests and you have a basis for determining the proper operation of the GC.

Find a test unit to measure temperature, a thermocouple, etc and check the temperature settings are actually giving the temperatures set for the unit at the oven, the injector, and the detector block.

Good luck,

Rodney George

[surinzp]

Thanks George.I will go through the manual and contact my instrument installer.

Thanks

[HW Mueller]

Why would one do this except for troubleshooting? One has to calibrate and continuously check any analyses performed on the apparatus, anyway.

[chemwipe]

I want to perform a 6 monthly or a yearly calibration checks for the GC.

Thanks

I wasn't going to say anthing, but...

Annual and bi-annual calibration checks? I run calibration checks daily.

Are you sure you're asking the right question?

John

[chromatographer1]

Hans,

In a regulated environment for labs performing work reviewed by regulators, the FDA for example, the sensitivity of the detectors and temperature accuracy of the GC oven are concerns that have to be addressed.

A research lab may have different concerns or not be concerned at all if their oven at 200°C is really 198°C, or if the sensitivity of the FID has dropped by 30%.

best wishes,

Rodney George

[HW Mueller]

The point I tried to make is the following:
To stay with your example, Rodney, if the sensitivity of the FID drops by 30% one will see this change happening if the analyses are properly executed. . . . .

[chromatographer1]

Yes, I understand your point, Hans. It has great merit.

But my point unclearly made, is that a 1% a week drop in sensitivity in a busy production lab might not be noticed over several weeks by harried techs as they try to keep up their output in QA analysis of the drug product.

This is why the FDA requires a yearly or semiannual validation, not calibration, of instruments employed in a laboratory.

Indeed, even in a research situation we had GC serviced monthly in order to keep their optimal sensitivity and to maintain strict conformity with the continuing work. But production people don't always have the time to be attentive to the details like a researcher might. The real world is the real world. Thus the regulatory requirements.

Rodney George

[Don_Hilton]

And one should note that with losses in detector sensitivity, calibration checks may come in looking OK while the LOD and LOQ are creeping upward, reducing the reliabilty of data on the low end. And whild a calibration check is easy, a check of LOD is a bit more work. (Not part of the instrument's data processing software.)

[HW Mueller]

Well yes, I see the reason for all this regulation, my dislike of it and worries about it just break through once in a while.