my reaction (as noted in original thread) is that retarding timing to reduce NOx will tend to increase BSFC.

 

I suppose a point to keep in mind is that if you're changing timing by upgrading from a mechanical injection system to an electronic one, you may be able to improve BSFC in parts of the operating range where you don't measure NOx, while improving NOx elsewhere.
 

PM will probably increase.

opsguy,

Reducing NOx basically requires reducing peak combustion temps.  Retarding injection timing from MBT will usually result in lower NOx, but higher HC and SFC.  

Charge air cooling and EGR are also effective approaches to reducing NOx.

Hope that helps.
Terry

The tradeoff is about 30:1 NOx vs Fuel economy in g/km. I.e. reduce NOx by 30% from retarding timing alone from a nominal setting and you will increase fuel economy by 1%.

PM will also increase with retarded main injection timing (About 20:1 PM:CO2).

Max 4/5 degrees from nominal setting.

NOx vs CO2 g/km by the way.....

Tradeoff ratios vary by specific application, combustion development and where you are along the trade-off curve.  It's not a linear curve.

in this case the application seems to be a medium-speed diesel engine burning 180 cst heavy fuel oil.  This could be in a marine propulstion, marine aux, or land-based power gen application (not clear from post).

 

Pilot injection and other strategies for injection rate shaping could also reduce NOx by reduce the peak rate of heat release and pressure rise without the disadvantages of retarded start of injection and EGR.

the application is a land based power plant

@Tdi, a common problem with such a strategy is that your end-of-injection timing becomes critical due to oil sooting and piston life concerns.  If you then swap injectors to give higher delivery rate, you hurt your smoke & pm.

 

Changing injection strategy and cut-off ratio will of course affect the NOx and PM outputs as well, but as a rule of thumb those are the tradeoffs when retarding/advancing main SOI whilst leaving separation constant for HSDI engines within that timing range (i.e. around an already optimised combustion).

@70btdc: I don't get the impression from the OP post that the engine in question is either HSDI or in an optimised-state.

@ivymike: That's why proper combustion development and any changes to injection strategy should be only one part of a holistic exercise, which may include reshaping of piston bowl geometry.  This latter factor in itself also has a huge influence on the aforementioned NOx/PM/NVH/SFC tradeoffs.

I get the impression that this is an after-sale effort, wherein the desired scope of change is in software (or mechanical) settings only.


 

The proposed modification does include changing pistons for a different design  

Anything you do to reduce NOx will reduce fuel efficiency. After all N2 combined with oxygen to create NOx releases energy, so N2 in this case is fuel. There are a lot of band-aides to try to fix this but the best they can do is get you back where you started at the cost of making things more complicated. Two things that I found that reduce NOx generation and at the same time improve fuel efficiency is to reduce the charge air/ air box temperature but there is a limit to that as you start getting white smoke. The other thing is to raise the jacket water temperature and this has a ignition delay feature that they are trying to acheive with electronic control. I would like to complete the experiment on this idea. Anybody have a locomotive I can borrow?

L4189, thats very interesting.  What is the mechanism to reduce NOx by reducing the temperature gradient?  I would have thought that promoting higher temps to be kept within the chamber by reducing the gradient between charge and chamber wall would have had the opposite effect - needing less advance and creating higher peak temps?

TDImeister, being already optimised was an assumption, but if we dont assume some constants at least, the OP question is to broad to answer.   I only have experience of HSDI engines, from 1000-2500rpm steady state these tradeoff ratios are correct for a good range of engine displacements.   

L4189 said: "Anything you do to reduce NOx will reduce fuel efficiency."
Don't tell that to the people working on HCCI, PCCI and other forms of LTC.

@70btdc: Granted.  A linear proportional assumption is OK when you don't deviate far from the original point.

 

70btdc,
Advance doesn't directly cause NOx, temperature causes NOx. The rate of the NOx generating reaction is exponential with temperature (as are many reactions). The peak temperature within the chamber virtually defines the amount of NOx created, because all the lower temperatures in the combustion chamber (in time or geometry) do not contribute very much to the total NOx generation.

Went thru this several years ago with engines running fuels from distillate to 240cst HFO.  CAT, Mak, Alstrom and a couple Wartsila units from 3 to 8 MW. All were gensets, island and grid paralleled units.

In general timing retard resulted in higher fuel consumption, in the HFO engines actual change was hard to track due to batch to batch fuel variations.  Most testing was done with a nominal 4 degree retard.  In almost all cases timing was returned to original as NOx reductions were lower than expected and increase in CO, PM, visible smoke and fuel consumption high enough to cause concerns.  Exhaust port temps in all cases increased, as did turbo speed.  In a few cases we experinced turbo surge.

Are you doing this as a result of regulatory requirements?  What are your expected results from the changes you're making?

Mike L.

What are the piston changes?  Change to compression ratio? Bowl geometry? Are you changing injection nozzles as well?  I'd expect with a piston change that a change in spray angle might be needed to get optimum performance.

I am aware of the mechanisms that create NOx, I was just wondering how having HIGHER combustion temps (due to RAISED coolant temps) would increase ignition delay?

I could understand if the idea was to LOWER coolant temps.

The magnitude of increase in coolant temperature will have an insignificant bearing on peak in-cylinder cycle temperatures.

TDI, yes, I would have also thought so, but isn't L1489 suggesting that raising coolant jacket temperature (assuming to the max limit of the design) by a few degrees will not only have an affect on ignition delay, but also that it will actually increase it?

Or maybe I just read it wrong...

This is meant to be completely constructive criticism and not a personal attack on L1489, but I cannot take somebody's advise seriously, who suggests that "anything you do to reduce NOx will reduce fuel efficiency" because that nitrogen in the charge air is itself fuel and did not combine with the oxygen in the same and release its contribution of energy of oxidation doing so.  The ppm order of magnitude of NOx emissions of the worst polluting engines guarantees that this exothermic energy contribution is absolutely insignificant within the amount of actual combusted fuel energy.

What's next to reduce charge air temperature - introduce an air conditioner into the airbox or charge air cooler?  

cannot take somebody's advise seriously... The ppm order of magnitude of NOx emissions of the worst polluting engines guarantees that this exothermic energy contribution is absolutely insignificant within the amount of actual combusted fuel energy

my thoughts exactly

The reaction of oxygen and nitrogen is endothermic. This is why air does not spontaneously burn and nitrogen oxides only form at high temperatures.

@Compositepro - you're right, I goofed.  Oxidation of nitrogen in air is (very) slightly endothermic; I verified it running GASEQ for only air as the reactant @ 2000 K, 100 bar, adiabatic and constant volume.

According to GASEQ, at these conditions, the mole fractions of NO and NO2 are 0.00747 and 1.24e-04 moles per mole of air, respectively.

However, that air does not spontaneously combust upon itself and that NOx form only at high temperatures are not in and of themselves factors that make reactions endothermic.  Spontaneous oxidation of any oxidizeable material is a function of its autoignition temperature (analog of the energy required to reach and exceed the reaction activation energy).  Endothermic reactions can and do also take place at- and below ambient temperatures.

"Anything you do to reduce NOx will reduce fuel efficiency"

I would agree with TDIMeister.  EGR will reduce NOx without much effect on SFC, until the EGR rate exceeds about 8% to 10%.  Exhaust gas is mostly inert (CO2 and nitrogen), and it reduces combustion temps by absorbing heat during combustion while not contributing oxygen to the combustion event.

Intake charge cooling is also effective at reducing NOx without too much of an SFC penalty.  Charge cooling reduces NOx by reducing the (T1) temperature at the start of the combustion cycle.  The SFC penalty with charge cooling is only due to the heat rejection loss and intake flow losses at the intercooler.

NOx can also be reduced by improving the injection mixing and combustion rate, which would allow a BOI closer to TDC.  Higher injection pressures with a greater number of smaller diameter nozzle holes will give a better fuel spray distribution and mixing rate, leading to less ignition delay and more rapid combustion.  But such a modification would be expensive.

opsguy says a piston change is planned and this is a land based engine, so I would recommend reducing CR and increasing boost with lots of charge air cooling.  This will keep your peak cycle pressures at acceptable levels while decreasing NOx.

Good luck.
Terry

Low SMD fuel particles are key to lowering nox with less specific fuel consumption penalty.  Often smd's are too high or too poorly distributed and they transport beyond the desired reaction zone into quench layers.  Higher coolant temps can reduce some quench but the higher surface temps radiate heat back into the working fluid..  If the chamber surfaces were less efficient at heat transfer then the coolant temps would need to be higher but that is not the case with any engine ecxept for idling diesels.

Apologies if it was already mentioned, but cooling of EGR offers additional benefit especially at high EGR rates. Water injection anyone?

Reduce Heat, pressure and or time along with reduced re-entrant lean residuals reduces nox.