I am setting up some gas analysis for our fixed bed reactors. I understand the gas analysis naturally has bigger error bars, especially on the sampling loop pressure control. I'm currently have a BPR there, to control the sampling loop's pressure at 30 psig, but the pressure was not controlled consistently overtime, mostly because the input pressure was not consistent. I tried to add in pressure reducing regulators with other flow restrictions, it did not go well. I checked some of the swagelok setup (
https://www.swagelok.com/en/blog/back-p ... -engineers) but the needle valve might need to be tuned relatively regularly.
I know people suggested atmospheric injection (stop flow, water seal, etc.), my biggest concern is if there is a pressure difference in the GSV, even 0.5 psi out of atm, that will be more than 3% difference (0.5 psi/14.7 psi). 3% could lead to bigger error if you calibrate multiple compounds.
Is my understanding on the atmospheric injection correct or it is overthinking?