GC SHOWING LESS VALUE FOR ONLY ONE COMPONENT

[zahid]

We have gas chromatograph measuring 5 component in a sample. H2, Ar, N2 NH3 and CH4, with 2 TCD.(separate TCD for only NH3). Sample line length is aprox. 50 meters.
Now the issue is when we put the analyzer in calibration it reads perfect value(20% NH3 as per cylinder standard gas) But as we put it into service it reads contiguously 9% NH3 whereas the actual value in sample is 19%.
I dont know what chemistry take place in sample line????
I can provide more information if anyone need about this issue. THANKS

[Don_Hilton]

Sounds like sampling a process? Some questions:

1) Is this a new installation or is this something that has worked in the past?

Assuming new:

2) what is the material of the sample line?

3) How do you know the composition of the sample to be 19% ammonia?

4) What pressure and temperature is the sample line? And how does that relate to the temperature and pressure of the process stream? (my thoughts wander off in the direction: Is the pressure in the sample line is high enough ammonia can partially condense and be running as liquid droplets in the line and not split off into the sampling loop for the GC?)

[zahid]

Thanks for reply dear first.
1. Its new installation.
2. line material is ss 6mm tube
3. 19 to 20% is designed concentration and we often check it through our main lab.
4. I am also thinking about the 4rth point. Main line pressure is 197 bar (temp 92c) and we are reducing it to 2 bar through simple regulator (matching pressure requirement but it is not steam heated) So i feel there some cooling effects. Secondly sample tubing are not insulated or hot insulated and you can say now our sample temperature is 30 to 50c as according to atmosphere.

Waiting for your kind reply.

[Don_Hilton]

I can only offer the suggestio that you look at how you sample the mixture for taking a portion to the main lab and then look at the process of dropping the pressure and temperature into the sample line. The vapor pressure of ammonia is high enough that I would not expect it to condense in the sample line, but if the gas mixture is allowed to cool before or in the regulator, I would wonder about some condensation - and segregation before passing through the regulator. I do not have a temperture/vapor pressure curve for ammonia to have an idea if this is likely.

If the regulator can take the temperature, it may be worth warming the line to the regulator and the regulator to see if the measured ammonia level changes.

And, if you can draw a sample from the sample line to the GC, you can take that sample back to the main lab and check for agreement between instruments. If the instruments agree on what is in the sample line, there is a difference in the samples between sampling points.

[zahid]

exactly dear, we do so. we arrange a portable NH3 meter and measure at the analyzer inlet. It was also showing less value near about same to analyzer. So it is clear that analyzer is ok. Now i have decided to do the hot insulation of sample line and also regulator because i have noted anther point is Nh3 reading on analyzer rises as atmospheric temperature rises.