Emissions: white 'smoke' vapors, nite haze, black smoke

gmctd · **Posted:** Tue Dec 09, 2008 2:23 pm

White vapors, usually referred to as 'white smoke' by the uninformed, are completely normal in cool\cold weather when starting an infernal-combustion engine, be it gasoline-fueled or Diesel-fueled:

a - when a running engine is turned off, as the engine\exhaust system cools, raw atmosphere with included water vapor (humidity) is drawn into the intake, also the exhaust system - start the engine from cold-start, and the hot exhaust evaporates the moisture in the exhaust system, giving white water-vapors - gasoline engines, with hotter exhaust gas temperatures at idle, create denser vapors over a shorter period of time - Diesel engines, with lower btu output at idle, take longer to heat up the exhaust system so the vapor is lighter - the serpentine belt-driven viscous coolant heater for the cabin is proof enuff for that - yer CRD don't warm up very quickly, even in warm weather

b - infernal combustion engines require increased advance and fuel rates at cold start -
- the automatic choke on carburetors is proof enuff of that, and EFI systems have their own methods, with the Idle Air Control and Computer Controlled Spark referenced to the Engine Coolant Temperature and Intake Air Temperature sensors and the heated 02 sensor - gasoline is highly volatile, burns rapidly, and is easily ignited by the ignition spark, so combustion temperatures are easily managed - result: minimal un-ignited fuel vapors out the exhaust
- Diesels have their own systems, with increased injection advance and injected fuelrate referenced to ECT and IAT, with glowplug system response referenced to those same sensors for increased glow duration - Diesel fuel is ignited by high compressed-air temperatures in the cylinder, so when the intake air and engine coolant and sump oil and cylinders and heads and injectors and injection pumps and lift pumps and fuel lines and fuel in the tank are all cold, cylinder temps can easily drop below threshold of combustion - result: unignited fuel passed into the exhaust system and out the tailpipe = raw white fuel vapors, mixed with the normal water vapors in the exhaust system - glow plugs and heated fuel serve to alleviate this problem, as does the engine coolant heater in the cylinder block - other Diesel systems use an intake air heater along with fuel heating and the coolant heater

Black smoke, or unburned raw fuel where the fuel was at or above combustion temperature and the flame went out, is caused by two conditions:
a - too much fuel, either from excess injection from faulty sensor feedback ,or reduction of air\fuel mixture, as from dirty air filter or ruptured intake ducting, burns until oxygen volume in the cylinder is depleted, flame goes out, remaining unburned fuel component is carbonized - result: black smoke out the exhaust - EFI CRD systems manage this better than the mechanical systems, so the initial symptom from restricted air intake will be more haze, easily recognizeable at nite in the headlites behind the rich emitter, and\or, particularly from ruptured intake ducting, black smoke out the exhaust
b - water in the fuel\contaminated fuel quenches the flame B4 the fuel charge is depleted, or cannot sustain combustion - flame goes out, remaining unburned fuel component is carbonized - result: black smoke out the exhaust - EFI CRD systems do not manage this condition any better than mechanical systems, as watery fuel is not conducive to efficiently converting fuel to btu's

White vapors are water or\and unignited fuel, usually increasing in cooler weather, but also from failing or defective glow plugs

Black smoke is unburned fuel, usually any season

nursecosmo · **Posted:** Tue Dec 09, 2008 3:59 pm

Nice write up GM. I would just like to add a caveat to your article. In a 100% ECU managed Fueling system like in our CRD, there is no black smoke from a plugged air filter. The computer adjusts the fueling solely based on MAP pressure VS RPM and TPS signal (other very fine adjustments are made based on the MAF signal on some modern turbodiesels but not on the CRD). If the Air filter is plugged enough to affect MAP pressure the computer will drop the fueling rate to an appropriate level for efficient combustion, so, no black smoke or haze, just reduced performance because of longer turbo spool-up.

nursecosmo · **Posted:** Tue Dec 09, 2008 5:25 pm

mackruss wrote:

Does that mean that the high flow air filters will also cause the MAP to increase the fueling rate due to the increased flow rate and stuff up the MPG?

No. The ECU regulates the charge air pressure by adjusting the boost pressure with the VGT. The ECU will only allow the turbo to boost to a predetermined limit programed into the air-fuel computer map. Performance chips get higher boost by altering the signal coming from the MAP sensor thus fooling the ECU into giving more angle to the VGT, making it spin faster. This can also improve the economy because as opposed to a mechanicaly controlled turbo diesel system which finds it's own air/fuel equalibrium, an electronicaly controlled system like our CRD has can control the air/fuel mixture to run slightly rich which achives beter NOx emmissions but increases the output of CO, CO2 and Particulates.

nursecosmo · **Posted:** Tue Dec 09, 2008 5:26 pm

That is wierd. Where did your post go Mac?

mackruss · **Posted:** Wed Dec 10, 2008 3:32 am

nursecosmo wrote:

That is wierd. Where did your post go Mac?

I was busy editing it and you must have sent off your reply but my question was basically as you quoted me thanks.

gmctd · **Posted:** Wed Dec 10, 2008 12:52 pm

Well, it's cold out, and nasty out, and I'm bored, altogether putting me in contentious mood - so, better 'splain me dis, Lucy: if the CRD system is so efficient at managing air\fuelrates based on MAP sensor etc feedback, then why the black smoke when the charge-air ducting blows off or is ruptured, low power and black smoke being the common complaint - those are definite cases where fuelrate management would be required to meet emissions criteria - even my totally-mechanical FI '96 Dodge Cummins reduces fuelrates with resultant low power but no black smoke when Boost fails, whether from ruptured ducting, ruptured tubing, ruptured wastegate diaphragm, or failed turbo - my primitive EFI\mech 6.5L TD doesn't handle it any better than the KJ CRD, blowing black smoke with no power at loss of Boost.

Increasing haze, initially seen at nite, and increasing puff of black smoke when nailed are symptoms of increasingly dirty air filter - the MAF could have been used to sense any such 'minor' reductions in mass airflow to reduce fuelrates, but apparently was not - ECM responds to TPS, and MAP has more weight in controlling max Boost pressures at various operating conditions - sadly, mass airflow, as sensed by the MAF, is used only for EGR

nursecosmo · **Posted:** Wed Dec 10, 2008 4:54 pm

gmctd wrote:

Well, it's cold out, and nasty out, and I'm bored, altogether putting me in contentious mood - so, better 'splain me dis, Lucy: if the CRD system is so efficient at managing air\fuelrates based on MAP sensor etc feedback, then why the black smoke when the charge-air ducting blows off or is ruptured, low power and black smoke being the common complaint - those are definite cases where fuelrate management would be required to meet emissions criteria - even my totally-mechanical FI '96 Dodge Cummins reduces fuelrates with resultant low power but no black smoke when Boost fails, whether from ruptured ducting, ruptured tubing, ruptured wastegate diaphragm, or failed turbo - my primitive EFI\mech 6.5L TD doesn't handle it any better than the KJ CRD, blowing black smoke with no power at loss of Boost.

Increasing haze, initially seen at nite, and increasing puff of black smoke when nailed are symptoms of increasingly dirty air filter - the MAF could have been used to sense any such 'minor' reductions in mass airflow to reduce fuelrates, but apparently was not - ECM responds to TPS, and MAP has more weight in controlling max Boost pressures at various operating conditions - sadly, mass airflow, as sensed by the MAF, is used only for EGR

Here is the deal. In the CRD the turbo boost is controlled exclusively by the vane control actuator. This little unit is a vacuum motor controlled by an electric solenoid stepper valve which will make one step, tell the ECU that it has made the step, wait for the ECU to give the instruction to go one more step, make the next step and so on, this pattern continues through the whole range of it motion. The solenoid receives it's instructions from the ECU. The lag period period from 0 boost arm extension to full boost arm extension is about one second.

Lets move on the the fueling loop. The ECU also controls the injection timing, duration and pressure in steps. The ECU takes hundreds of sensor samplings a second but it cannot control the metering to go from full fuel to zero fuel in one jump. When the foot is taken off of the accelerator it senses this from the TPS. reduces the injector duration one step, waits for the return signal from the injector that this has indeed happened, then reduces the metering valve in the IP one step, waits for confirmation of this event, checks this against the rail pressure sensor, opens the cascade valve one step, waits for confirmation of this event, compares this to the MAP reading, decides to reduce injector timing by one step etc... The time required for the ECU to take fueling from max to zero is a little over one second and longer to get from zero to max.

The sudden black smoke from a broken hose comes in from a feed back oscillation loop caused by the delays in the electronic controls. As boost pressure builds under acceleration, the split CAC hose is able to hold a certain degree of pressure but as pressure builds the split opens wider thus releasing more volume of charge air. The ECU senses this via the MAP sensor and sends the signal for more turbo vane actuation which the turbo obliges to do, when all of the sudden the Split opens to full width and spills most of the charge air out through the tear. In the mean time the fuel loop was geared up for max fuel delivery. It senses that there is MAP vs fuel rate mismatch and trips a CEL, but there is still 24000 psi of fuel pressure in the rail and the stepping circuit is trying to reduce timing and duration as fast as it can and opening the cascade valve while telling the turbo actuator to go to the next step down in boost pressure. All of that fuel pressure has to go somewhere and that somewhere is into the cylinder but because the boost is now very low there is not enough O2 for good combustion and black smoke occurs. It will continue in these approximately 2 second oscillations as long as one is stepping on the pedal hard.

The old P-pumped Cumins will not go through the funny RPM oscillations because their fueling rate is controlled by the Aneroid Fuel Control valve and the turbo is controlled solely by the exhaust pressure and has no electronic interference. The response to changes in manifold pressure is virtually instantaneous in the AFC (unless you have altered or removed the fueling cam in the AFC housing or adjusted the star wheel for smoke).

After some thought I did come up with a possible if unlikely cause for increased haze when the filter is dirty. If the air filter is plugged enough to increase vacuum in the intake, it could possibly suck some oil past the turbo charger shaft seals or PCV if connected, however by the time this occurred the performance of the engine would be noticeably reduced. Enriched fueling from a plugged filter is impossible with the way that the ECU progressively steps up fuel rate based on MAP.

gmctd · **Posted:** Fri Dec 12, 2008 4:53 pm

Holy insider-trading info, there, Batman! Good thing we're talking motors, and not stocks'n'bonds'n'stuff, eh...........

Alternately, consider this:

- the Variable Geometry Turbo functions to vary the effective A\R ratio of the turbine snail, such that, at full vacuum
applied and the vanes fully extended the turbine responds at low-rpm low exhaust-energy conditions as if it is a
small turbo - then, by reducing applied vacuum and retracting the vanes as necessary as exhaust energy increases,
to act as an increasingly larger turbo so as to continue to generate Boost without being restrictive to the increased
exhaust flow rate\volume energy - this balancing act is accomplished by electrically regulating the vacuum source
as applied to the vacuum-actuated spring-loaded diaphragm-type servo actuator on the turbo - thus, the vanes can
be held in any position while the engine is running, indefinitely, if required - the vanes can be quickly retracted in order
to quickly dump Boost

Last pic (from an EURO right-hand drive KJ) is turbo controller and vacuum regulator - auxilliary solenoid upper right, pulse-modulated vacuum regulator solenoid, to the left, filter lower left, both solenoids mounted atop vacuum reservoir - hose-routing is clear, with vacuum supply coming up from the engine and in from the right - the plastic tubing goes to the vacuum-actuated servo on the turbo -
http://www.lostjeeps.com/forum/phpBB3/vie ... 542#405542

- the vacuum regulator consists of two 12vdc solenoid valves located on the vacuum reservoir, that large assembly
on the turbo-side fender just behind the airbox : the auxilliary vacuum solenoid, being a coarse control for quick
dumping of applied vacuum, and the solenoid-operated vacuum regulator, being the actual electrically-modulated
vacuum regulator valve - vacuum is supplied to the reservoir from the engine vacuum pump, with a one-way valve in
the supply circuit to prevent loss of vacuum at engine shut-down, then to the valves from the reservoir

- the auxilliary vacuum solenoid is a two-way solenoid-valve with three ports: one to the vacuum reservoir source,
the second to the solenoid vacuum regulator, third open to atmosphere - power off, the valve seals the vacuum source
while porting solenoid regulator valve port to atmosphere - applying +12v power connects the vacuum source to
the vacuum regulator valve while closing the port to atmosphere - this is a simple valve intended to quickly dump
ported vacuum, allowing quick vane response - it can also be used to prevent vacuum loss in event of regulator
failure, and to prevent turbo response in Limp Mode

- the vacuum regulator solenoid is a solenoid-actuated diaphragm-valve type regulator with three ports - it is a real
vacuum regulator, adjusted electrically by pulse-modulated solenoid-plunger rather than the typical screw-handle -
the vacuum-supply port connects to the vacuum source, the second is to the vacuum-actuated servo on the turbo,
third is bleed, open to the atmosphere via that small filter that everyone asks 'whut it is?' - power off, the turbo servo
port is connected to the vacuum supply port while sealing the bleed port, allowing vacuum to build, extending the vanes into the turbine snail - applying +12v closes the supply port and connects the turbo control port to the bleed port, reducing vacuum on the vane-servo actuator allowing the vanes to retract from the turbine housing - by pulsing this valve on and off rapidly between supply and bleed, regulated vacuum is maintained on the vane actuator diaphragm, holding the vanes in any position as required by operating conditions as determined by ECM

- only 'direct' feedback for this circuit is from the Manifold Air Pressure (Boost) sensor with it's included Intake Air Temperature sensor - Boost is also regulated by air temperature, being reduced as compressed air temps climb above prescribed limits in order to prevent overly-high EGT's

Now, if ECM can control Boost and fuelrates in Limp Mode to reduce power and prevent black smoke, then it should
be able to do so at increased intake restriction, also at loss of turbo output, whatever the cause - required will be a
litmus test to determine extent or limit of that control - stuffing a shoprag or two across the air filter should reduce
intake airflow, while pinching off a vacuum hose or removing a Boost hose should do the other - we shall see

Also, correct on the P-pump Cummins, with exception that turbo is controlled by manifold air pressure (Boost) on the spring-loaded diaphragm-type wastegate actuator - ruptured diaphragm would result in max Boost, but leak also prevents manifold pressure buildup, thus preventing AFC from increasing fuelrate - result: no power, no black smoke - solution: plug pressure hose to wastegate actuator to seal the leak, let your foot control Boost - turn on your brain, let it control your foot - works for me.................

nursecosmo · **Posted:** Fri Dec 12, 2008 8:09 pm

Very nice writeup of the Boost control system GM, it is amazing how many people have a hard time understanding how it works. You did a nice job of simplifying the Manual.

As to your litmus test to try and produce smoke by restricting the air flow, let me warn against putting a shop rag (or plastic bag) across the air filter to restrict it(sucked up filter). Let me also warn against trying to restrict it by putting your hand across the rubber intake pipe going to the compressor(collapsed pipe). also let me warn against using your hand as the restriction inducer over the mouth of the compressor (massive hand hickey). Ask me how I knew positively that filter restriction does not produce black smoke on day one of CRD ownership.

gmctd · **Posted:** Fri Dec 12, 2008 11:21 pm

Unfortunately, that info is not in the shop manual - would help the untrained Jeep techs if it were, tho, as would some of my other not-in-the-manual posts - some of the stuff that is in there is drastically wrong - even basic critical info like dropping the trans for a torque convertor r&r is not in there, and this Diesel engine is a world of different assembly technique than those other engines

Just FYI, I've been 'playing' with turbochargers since back in the '60's, when most people couldn't even spell turbocharger, much less know what one was for - it was all scary and difficult to manage until more wastegated versions began showing up oem, and knock-sensors 'n'stuff - the current engine technology with microprocessor control and crystals and ceramics is simply wondermous, to an old-timer like me

Also, nice write-up of the Dodge 6.7L VGT operation - I have one waiting for install into my '96 - I was also gonna use the Cummins Jake-brake, but it is an abortion, as-is in that exhaust elbow - they shoulda stuck it in a straight-section casting, so as not to block the exhaust in the open position - I'll modify it to that straight-section configuration when I install it

Emissions: white 'smoke' vapors, nite haze, black smoke

Who is online