Ummm...why? Far easier to store data than samples...

CR

"As iron sharpens iron, so one person sharpens another." [Proverbs 27:17, NIV]

We routinely do this in custody transfer gas metering systems - a automatic gas sampler takes in a measured flow of the flowing gas into a gas sampler bomb for later offline lab analysis in order to obtain flow weighted physical properties such as molecular weight, N2, CO2 or H2S content etc. One company that designs and installs such systems is Welker - many instrumentation engineers in the oil/gas or downstream refining business may be familiar with these, so talk to your plant field instrumentation engineer to start with or else talk to Welker direct.

You can request this sampling to be done in the way you want it to : fixed time interval, or flow or mass weighted, etc. Be careful with what and how you are sampling and make sure it is exactly what you want :- this is often affected by the nature of the fluid flow ( single phase or 2phase vapor liquid etc), the type of sampling probe, how the probe is oriented, and where it is located in the gas stream to be sampled. Not to make too fine a point, in applications where millions of dollars are involved in operating revenues / day at this sampling system, it is better to get your plant senior process engineer to supervise this exercise/ carry the proverbial jerry can - the devil may be in the details on this one.

To second georgeverghese's post, the devil is absolutely in the details. A tiny bit of swirl and a wall sample will tend to be rich. You may get different results when the piping around the sampler is in the sun or the shade (with no actual change in the flow mixtures). Do you want the sample duration to be less than 1 second so you can get a new sample at the start of the next second? or do you need multiple probes so you can flow for more than 0.9 seconds? If you need multiple probes how will you place them to prevent the act of sampling from impacting the next sample? How many samples do you need to take without reloading your sample bombs? How will you keep the sample bombs warm? How much gas does the lab need for the analysis you want performed? The details are really important.

David Simpson, PE
MuleShoe Engineering

In questions of science, the authority of a thousand is not worth the humble reasoning of a single individual. Galileo Galilei, Italian Physicist