4 replies to this topic

[voxmpg]

Hi everyone,
I am an engineering intern and I am working on Carbon dioxide sensors. I am going to have to test several sensors for accuracy by running mixtures of various concentrations of CO2, N2O, and O2. I am having trouble thinking of a way to make mixtures of known concentrations of each gas to run through the machine. I had one idea which involved using the Ideal Gas Law and an industrial gas bag. Basically I would use the equation PV=nRT and use 15L for my volume (for example because the gas bag has a volume of 30L, so let's say this mixture will be 50% CO2 and 50% O2), keep each of the gases at the same pressure (the P), measure the temperature, find the number of moles, and convert it to molecular weight. Then I would place a gas bag on a scale and fill it until the desired molecular weight was reached.
While theoretically this may work, it seems rather inaccurate and I'd like to have a more dynamic system. Does anyone know of a way to take a Y-tube, connect it to two cannisters of different gases and vary the pressure so that you can obtain a desired concentration of each? I hope I did a decent job explaining this, hopefully it's not too wordy. Thanks for your ideas!

[gvdlans]

It seems to me like you are trying to re-invent the wheel...

Can't you just buy mixtures with known composition at your industrial gas supplier?

Another obvious way is to have each pure gas flow through an accurate flow controlling device (e.g. rotameter with needle valve) before the mixing point.

Question is how accurate the composition needs to be?

[voxmpg]

It seems to me like you are trying to re-invent the wheel...

Can't you just buy mixtures with known composition at your industrial gas supplier?

Another obvious way is to have each pure gas flow through an accurate flow controlling device (e.g. rotameter with needle valve) before the mixing point.

Question is how accurate the composition needs to be?

I am going to be testing about 20 to 30 different combinations of gas, so I can't really buy that many different mixtures from a gas supplier. I had thought about using a flowmeter, but do you know how I can determine the concentrations of each gas in the mixture based on flow rate?

[gvdlans]

Well, this is very simple. For example, if you have a flowrate of 1 ml/s of CO and 1 ml/s of O2, measured at the same pressures and temperatures, your mixture will have 50 volume % CO and 50 volume % O2. If you're pressures and temperatures are different (and your flow measurements are in actual volume units per time unit) you have to make a correction for this based on the Ideal gas law.

[voxmpg]

Ok great. Thank you very much!