I am saying that its unrealistic to get an accurate reading using this method... if you read into this further you will see they are actually doing this for generalities, because of a lot of iissues.. first is that laser diode readings are incredibly voltage sensitive... well most analyzer types are. it causes analyzer drift. several factors account cause changes... from as simple as dirt on a lens, to temperature changes, to pressure changes (which in this case aren't likely to be happening in a closed environment) to actual changes in voltages from degradation of electronics, ect. To remedy this readings have to be constantly checked against a standard and the voltage readings adjusted. its not a constant. they are doing marvelous things with cells for calibration instead of actual verified gas standards, but I haven't seen a model that doesn't have a pretty good +/- on it yet and that's when testing gas in a reaction cell that's a few inches long... we are talking about a laser reading miles in length. Moreover its been standard practice in some previous projects like this (though I do not know about this one exactly yet) to use expected readings to calibrate rather than known concentrations, in other cases there was a simple expectation that the readings would stay stable, which over time, they do not.
We are talking about not removing interferants before testing... other things will reflect on the same bandwidth. For that reason for accurate readings down here, gas is typically either dried by dropping out moisture or by diluting it with known quantities of dry air that renders the moisture insignificant. What they are doing is going to give general numbers... which would be fine if they were reading a couple % Co2... but they are looking for numbers in the low parts per million... that's why you see the numbers always given at a nice round 400 ppm or 200 ppm, I have an issue with using tech that isn't as precise as the data they will be using it to validate needs it to be.
I just think too many are taking data pushed out by this kind of method and assuming its accurate to a minute level because "NASA" did it... there are limits to what we can read. Rules and regs oftentimes utterly ignore this, and many scientific studies are done with instruments reading below their MDL. its just not as accurate as we need for the levels it is at. legitimately bias adjusting it could realistically be totally changing the data based on a few rounded percentages of something like moisture, which isn't going to be exact anyhow. Its fairly accurate, but as I keep telling you, the world is a very complex chemistry equation. its not like testing in a lab, and its not even as easy as testing in a confined space. there are far more things to account for
We are talking about not removing interferants before testing... other things will reflect on the same bandwidth. For that reason for accurate readings down here, gas is typically either dried by dropping out moisture or by diluting it with known quantities of dry air that renders the moisture insignificant. What they are doing is going to give general numbers... which would be fine if they were reading a couple % Co2... but they are looking for numbers in the low parts per million... that's why you see the numbers always given at a nice round 400 ppm or 200 ppm, I have an issue with using tech that isn't as precise as the data they will be using it to validate needs it to be.
I just think too many are taking data pushed out by this kind of method and assuming its accurate to a minute level because "NASA" did it... there are limits to what we can read. Rules and regs oftentimes utterly ignore this, and many scientific studies are done with instruments reading below their MDL. its just not as accurate as we need for the levels it is at. legitimately bias adjusting it could realistically be totally changing the data based on a few rounded percentages of something like moisture, which isn't going to be exact anyhow. Its fairly accurate, but as I keep telling you, the world is a very complex chemistry equation. its not like testing in a lab, and its not even as easy as testing in a confined space. there are far more things to account for