Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Exhaust gas recirculation

  1. Aug 2, 2016 #1
    i have done tests on Exhaust gas re circulation on my engine test bench and these are the results, i don't know what's wrong? but at 8% loading conditions i am getting some weird results
    the trendlines in the graph shows an increase in NOx with EGR
    is there any explanation for this ? thanks

    Attached Files:

  2. jcsd
  3. Aug 8, 2016 #2
    Hi Ali,
    I am not an engine tester, just trying to give you a bit of help. Have you spoken to your lecturer / supervisor about this?

    The graph might be better represented by data points and a logarithmic curve. The curve of best fit does suggest where the datapoint might be, showing a bit of scatter there, but in practice, such things usually follow the natural logs / exponents.

    The test rig? The first thing I think of off the bat, if you are just reving the engine, measuring exhaust NOX, and getting that kind of scatter, is the quality of the rig and the quality of the measuring device. Perhaps you can try calibrating the device and 'raving' up at different speeds, like calibrate and rev straight to 4000, calibrate (reset) and rev straight to 5000, etc. Also, check your NOX instrument is fixed, and also at the end of a section of pipe that is straight for at least 5 diameters, to eliminate the effect of turbulence (vortexes) on the measurements. No holes or loose parts vibrating around the engine / exhaust pipe.

    The recirculation mechanism? I am not sure if you guys use an industry standard recirculation. Recirculating through the engine, or through the catalytic converter, or something else, but perhaps there is some aspect of that which affects the cleaning performance at different rpm.

    Based on the information at hand, perhaps do some troubleshooting: check your rig and calibrations, and run a few more tests.

    Would be great to see how you get along.
  4. Aug 10, 2016 #3
    Bro i am not good at making graphs so i did not know which plot would suit better
    as you said about the calibration, so this i think is the problem we did not calibrate it, because we don't have any facility of gas analyzer in our labs, we did it in a gas station and the person told us its fully calibrated, also there were alot of vibrations around the exhaust pipe
    thanks a lot for your useful information i have already submitted my report and its been accepted :D and i don't know why have they accepted my report.
  5. Aug 11, 2016 #4
    Up and down results like that are probably not the results of calibration, but the vibrations. Either they looked at your chart and saw the value was less than target, or they just don't care!
  6. Sep 5, 2016 #5
    Yes that was the target values that we has achieved, maybe because calibration would have just shifted the graph in one direction, either up or down, But i didnot understand one thing, how is the turbulence or vortex going to affect the measure of exhaust gases? after all the same amount of gas i going to pass through the probe, the same amount is going to pass through the chemical sensor then why is turbulence effecting it so much? and as far as vibration is concerned every real engine exhaust assembly has vibrations then why are my results so different?
    thanks in advance
  7. Sep 7, 2016 #6
    Congratulations on your pass.
    All very interesting questions.
    You need to do the troubleshooting ... move your 'probe' and isolate vibrations ... and let us know what you find.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted