Using NOx for performance tuning.

[MadBill]

Hello Bob; Yes it may seem a math problem however NOx is not just NO , it can be any number of Os; up to 4 I think it is with one N, so I think it might be reasonable to raise the O2 reduction level a little bit. And I agree its not the primary concern as to power reduction but it needs to be mentioned. ...

Just to nit-pick, can someone verify whether the various nitrogen/oxygen reactions are exothermic or endothermic? If the former, the oxygen consumed contributes at least some heat and thus cylinder pressure. If the latter, it subtracts more than just the % O2 loss...

[rschrader]

Hello Bob; Yes it may seem a math problem however NOx is not just NO , it can be any number of Os; up to 4 I think it is with one N, so I think it might be reasonable to raise the O2 reduction level a little bit. And I agree its not the primary concern as to power reduction but it needs to be mentioned. ...

Just to nit-pick, can someone verify whether the various nitrogen/oxygen reactions are exothermic or endothermic? If the former, the oxygen consumed contributes at least some heat and thus cylinder pressure. If the latter, it subtracts more than just the % O2 loss... :-k

Two things:

1. Most (80% or more) of the NOx measured is NO. (Different combinations of N and O are generated in the heat of the combustion process, but they decompose into NO by the time they reach the gas analyzer. Just how much Oxygen is temporarily tied up is unknown, but it is pretty small - a fraction of that tied up in NO.)

2. This is an endothermic chemical reaction (takes heat to produce NOx) - so it actually reduces the heat and cylinder pressure.

[TRN]

This topic is of interest to me, I have been using a GC-TCD/FID to analyze CO, CO2, and <C4 hydrocarbons in exhaust. I have taken measurements from a particular engine, and was planning to retake measurements after reducing the squish clearance and reducing the piston dome. Recent changes in my employment status have left me without access to the GC, however. I have no idea how you would set up a GC to measure the nitrogen oxides, but TCD would be sufficient for concentrations down to 100ppm range, I would think (haven't looked up the thermal conductivity). I was looking forward to the data from this experiment providing some insight into combustion conditions with tighter squish. Older GCs with TCD/FID are not expensive, and I may soon acquire one to continue. If you have any suggestions regarding GC analysis of NOx I would appreciate them very much.

If I can get this experiment resumed, I may swap thick head gaskets back on the engine and retake samples to include NOx concentrations. Compression ratio will be different, the pistons are aleady machined.

[Shrinker]

Hello Bob; So if the NOx decomposes to NO does that happen in time for the O's to be used for the combustion or is it too late?

[rschrader]

This topic is of interest to me, I have been using a GC-TCD/FID to analyze CO, CO2, and <C4 hydrocarbons in exhaust.

Did you really mean <C4? Butane or smaller? Seems pretty specific analysis to me - curious.

I have taken measurements from a particular engine, and was planning to retake measurements after reducing the squish clearance and reducing the piston dome. Recent changes in my employment status have left me without access to the GC, however. I have no idea how you would set up a GC to measure the nitrogen oxides, but TCD would be sufficient for concentrations down to 100ppm range, I would think (haven't looked up the thermal conductivity). I was looking forward to the data from this experiment providing some insight into combustion conditions with tighter squish. Older GCs with TCD/FID are not expensive, and I may soon acquire one to continue. If you have any suggestions regarding GC analysis of NOx I would appreciate them very much.

Can't help you much here - as I know of no one using GC for NOx - seems like it may not be practical to measure NOx this way.

By far the most common method is to use a NO sensitive chemical sensor - that is what almost all exhaust gas analyzers use, and it is a pretty cost effective solution - with resolution down to the ppm range, but with response time about 15-30 seconds.

It is by far the most cost effective method - providing that you can live with the response time.

[rschrader]

Hello Bob; So if the NOx decomposes to NO does that happen in time for the O's to be used for the combustion or is it too late?

NOx is generated in the first half of the combustion process (about) and will decompose somewhat as the piston expands the volume and oxygen gets depleted by the combustion process - so the answer is that some of the early NOx generated will release its Oxygen for combustion.

However, no one measures the NOx in real time (that I know of, anyway), so the NOx being measured is the NOx remaining after the combustion process is over - pretty much. Sure, some NOx decomposes as the gas travels to the tailpipe - and that oxygen is not available for combustion, as the process is over - but its absolute concentration is small.

In essence, measuring NOx a few seconds after the combustion event can be assumed to represent how much NOx there is in the gas leaving the combustion chamber.

I have seen no indication that header NOx is substantially different than tailpipe NOx - so it seems the decomposition rate is low.

[TRN]

The range of hydrocarbons is limited by current column configuration. I have been using Molsieve 13X and HayeSep N for a different application, and tinkering with exhaust gas analysis during slack time. The main goal I hoped to achieve was determine the extent of combustion quench effected by different squish clearances.

The analysis of NOx by GC is not common, but I learned a few details last night. Most have used Molsieve 5A, very well conditioned, with TCD. Helium carrier gives excellent response. This has been most successful measuring NO2, since NO reacts with O2 (if present) to form NO2.

Used GCs are available for less than $1000, making them very cost effective relative to some other sensors. They have a serious problem with response time, my analyses were about 15 minutes. Extending the range of hydrocarbons would probably add about 2 minutes per carbon.

How is it that the analysis of NO by GC has failed due to the conversion to NO2, but you claim that NO is the predominant NOx? I am not a chemist, chemistry is a fun hobby for me. Is the Molsieve acting as a catalyst to this reaction? The use of helium carrier would limit the exposure to oxygen, and the texts I read warned that the column must be well conditioned to remove moisture. Any thoughts?