I found this article a while ago which lead to my interest: Sadly I do not have the source:
“I've been mulling this around for a while now and its definitely an area to look into. How can we use EGR to increase our FE?
EGR could allows us to decrease pumping losses for the same power output, somewhat similar to lean burn. However, it doesn't have the NOx problem that lean burn does since there is not an overabundance of oxygen in the air/fuel mixture.
Lets take a look at how the OEMs utilize EGR. From
this Toyota article, it seems they are using it solely as a means to decrease NOx emissions. I think this can be backed up by noting that the 96 Paseo had EGR vs the 97 which did not, and the 97 has a higher EPA mpg rating. So, we need to figure out how to alter EGR for our uses.
Quoting the article:
Quote:
High EGR Flow is necessary during cruising and mid range acceleration, when combustion temperatures are typically very high.
Low EGR flow is needed during low speed and light load conditions.
No EGR flow should occur during conditions when EGR operation could adversely affect engine operating efficiency or vehicle driveability (engine warm up, idle, wide open throttle, etc.) |
Why would we not want high EGR flow at idle and light loads? Pumping losses are the greatest when the throttle is closed or close to it. Low speed city driving mileage would be noticeably improved.
I also think real gains could be had if EGR flow is increased at cruising loads to further decrease pumping losses. This will have a downside, loss of power, but I think many here are willing to give some up. Also, if we again make it similar to learn burn, we could simply make some override switch that pops us out of 'EGR mode' and back to normal mode for power.
Now, I don't think any of this would be real easy to implement, but I really think its worth considering. There are definitely problems with EGR. First off it is hot, and its also messy (carbon build up). There are additional things such as ignition timing that would need to get looked at to make it a perfect system since an EGR rich mixture will burn slower and require additional ignition advance. There is also a point at which you can no longer dilute the intake charge without causing misfires (same as lean burn), so you can only take EGR so far.
I'm really looking for input, and ideas on how we could use this, the different systems out there, and how control them. I really only know Toyota. Their setup is described in the PDF I linked to. I also know Honda's Insight uses EGR, but it has an electronic actuator that opens and closes the valve. What do other mfgs use?
PS: anyone planning to mess with EGR will need an MPGuino or equivalent to properly measure the effects on fuel consumption.
The ScanGauge will not report fuel consumption properly under high EGR concentrations for the same reason it doesn't work well with lean burn.
A relatively simple way to test the effect of diluting the intake charge to mimic EGR (without actually going through the process of modding it) would be to get a bottle of some inert gas (CO2?... what gas do MIG welders use?) and feed the intake through a tube.
I think we should get an idea of how much EGR flow we are looking at for doing something like this. I'm going to use a 1.5L engine example because that is what is in the Paseo and thats likely what I'll be doing testing on. Please correct me if you see any errors in my methods here.
The at 55mph, the Paseo is geared to be at 2600 rpm. At 2600 rpm and WOT, we are looking at the engine sucking in around 60 CFM (assuming 85% volumetric efficiency).
Now, from reading I've heard that a 40% EGR mixture is the absolute maximum that an engine can handle before misfires occur and things start to go downhill. Darin thinks he read that it is more like 20%. So, I'm going to go with the lesser of the two just to be on the safe side. Again, feel free to add any info/corrections.
20% of 60 is 12 CFM at highway cruising speeds. 40% would simply be double that, 24 CFM.
12 CFM doesn't sound like a ton to me. This is probably good because I pulled the EGR valve off the 96 Paseo I have and the hole going from the exhaust side to the valve is pretty tiny. I didn't clean it all up yet, but it looked to be around 3/16" (5mm) or so. There is no way that an orifice that small can pull 12 CFM through it without some pretty hefty pressure drop. So, it looks like the stock sided EGR valve is going to be undersized. I'm also going to assume this is similar for most vehicles out there. I sure would be nice to know what the stock EGR systems are designed to flow.
Alright, so from all that reading, it looks like we have two issues with running large amounts of EGR.
1) Fuel mixture dilution causing misfires and/or incomplete burn.
2) Hot EGR gas causing detonation.
#1 is going to be unavoidable. A high swirl/tumble head will probably help to some extent (or direct injection).
#2 is avoidable with some sort of EGR cooler. I know they use these on diesels. Can any TDI owners pipe up with any info on these systems? It looks like they run coolant through a heat exchanger to cool down the EGR charge.
Here is a picture of a TDI EGR cooler. It looks like they use engine coolant to cool the EGR. I'm guessing this is due to space constraints mainly.
At part-load operating conditions, it is advantageous to dilute the fuel-air mixture, either with excess air or with recycled exhaust gas. This dilution improves the fuel conversion efficiency for three reasons: (1) the expansion stroke work for a given expansion ratio is increased as a result of the change in thermodynamic properties of the burned gases; (2) for a given mean effective pressure, the intake pressure increases with increasing dilution, so pumping work decreases; (3) the heat losses to the walls are reduced because the burned gas temperature are lower.
I read through this thread but did not delve into all the links. EGR in my understanding reduces peak combustion chamber temperatures, to avoid pre ignition, and allow more timing advance and higher compression, without exceeding NOX emission standards.
Adding EGR will decrease peak combustion chamber temperatures, which will directly affect the power produced.
Most EGR systems are designed to allow EGR introduction when manifold vacuum is low to non existent. They work with a combination of ported and manifold vacuum working against each other. Modern systems use computer controls to accomplish basically the same thing.
Peak combustion chamber temps only occur when timing is close to max and effective compression is also close to max. This is when NOX levels really jump.
EGR's purpose is to reduce peak temperatures from 3500 to about 3200 degrees. This is one of the main reasons why engine performance has increased so much since the first anemic emission choked engines of the early 1970s.
Think of it this way. If you eliminate EGR you would have to retard ignition timing or reduce engine compression to prevent pre ignition.
In my opinion adding significantly more EGR under conditions other than very low manifold vacuum and very advanced timing will reduce peak combustion chamber temperatures when high temperatures are necessary for best power for fuel consumed.
You may find a better balance, with a slight increase in efficiency, by tweaking the factory systems, but it will probably be insignificant. There may be a balance point that is more economical, but it would have to be where the factory specs were more performance based.
As always these are just my opinions, take them with that in mind.
Understand maximum effective compression. This is when you have almost no manifold vacuum. It means your engine is creating very close to the mechanical potential compression. Maximum effective compression is the main reason why you have best BSFC when the difference between the compression created and the pressure increase when the fuel is ignited you have the best "leverage" and creation of power. Maximum effective compression can occur at very low throttle positions, which a vacuum gauge will confirm. When you combine best compression and lower engine speeds, you have loaded the engine so it produces the best power for the least fuel.
The Prius runs close to 13.5 to 1 compression ratios, so that level of EGR would be possibly an essential component of that compression level. The same situation will not apply to any engine that has significantly lower compression. The reason why the early 70s engines were so gutless was they dropped the compression ratios and added EGR. When people tried to disconnect the EGR pre ignition was the consequence.
When oxygen sensors appeared and computer controls made engine management much more precise, compression ratios rose back to pre emission premium fuel levels. This happened beginning in the early 1980s. Nissan bumped the compression in the Z Car from 160 gauge to 190 gauge and power levels climbed.
My Echo runs 10.5 to 1, same engine as the Prius (up to 09). I doubt that you would see that percentage of EGR on the same engine with 10.5 to 1. Not sure if the Echo uses the Atkinson cycle. Atkinson developed his cycle to try to get around Otto's patents. (think that is right might be Benz).
If 50% of the intake charge is cooled EGR at half power, there is nearly twice the total mass of gasses to heat with combustion and expand for power.
Putting large amounts of cEGR at low outputs presents an ignition problem. If the Accel multistrike box and the stock coil peter out, speedshops are full of hotter coils.
with the exhaust gasses around 1000 F, reforming ammonia for hydrogen would give an easily measurable amount. There are claims that small amounts of hydrogen help combustion, so maybe we will see. If we got really lucky, the refrigerant effect of expanding liquid ammonia would do much or all of our exhaust gas cooling. This would be dead loss refrigeration, and one must expect issues of suspicion of methamphetamine synthesis when seeking to occasionally buy a gallon of liquid ammonia. Still, it's the safest storage of hydrogen I know, as well as a topnotch refrigerant. The unreformed ammonia burns fine when there is enough hydrogen around to get it lit.
Bottom line, these very lean, or very diluted mixtures are hard to ignite. One possibility is hydrogen, possibly from catalytic reforming of ammonia in the mixed gas stream toward the intake.
To adapt this to a G10 Metro will require megasquirt control, and an appropriate MAF sensor, as the speed density calculation is thrown off by the dilution.
How will I ever get it back through smog after this?
EGR dilution as a way to increase FE at part-throttle on an ignition-type ICE has fascinated me for a long time. Some work I did as an engineering student way back in the late 70's was working towards that. Problem was, we didn't have anything close to the electronic controls that are available today.
OK. Some of the control problem you're going to encounter has to do with the contradiction that your intake will draw through whichever has the least restriction. You experience this when a car with an intake gasket leak runs like cr@$ with the throttle closed, but can run just fine at quarter throttle or more. That's becase when the throttle plate is closed, you are developing high intake vacuum, which can suck a lot of air through a little gascket leak, but when you open the throttle, lowering the vacuum, the amount of air drawn through the gasket leak is much less, AND the metered air flowing past the throttle plate is much greater, and the cpu/metering valve puts in fuel to match this air across the throttle. So, under high vacuum, if you use a monster egr valve, it will be next to impossible to adjust it in such a way that is meaningful. You see, the point of egr is % dilution, and at small throttle openings, that means very little exhaust recirculation, or the car don't run.
Here's a thought:
Ford used variable cam timing on some of the Zetecs (Contour, Cougar, ZX-2) to advance the exhaust timing in such a way that the exhaust stroke wasn't completely scavenged. They were able to use this instead of egr to meet nitrous standards. So how much the intake charge is diluted has much less to do with intake vacuum. The smaller the throttle opening, the smaller the charge that went in on the previous cycle, and so the easier the subsequent exhaust gasses have of getting out during the exhaust cycle, so the lower the total number of exhaust gas molecules left. Now, this isn't exactly linear, but a whole lot more linear than egr.
So, I'm thinking of putting a Zetec /exhaust VCT in my Escort. MegaSquirt the intake, control the VCT solenoid (it's PWM, and I think 12v, but it might be 5v. Have to play with it, 'cause there is limited info on the solenoid.) with an Arduino board if I can't build a pwm circuit in the prototyping section of a v.3 MS board.
But, for those engines that can't alter the exhaust timing, how about an exhaust throttle? You know, they used to call them heat risers? I think GM used them as late as '81 on their V-8s. They used a vacuum dashpot, but you could hook it up to a manual push cable, hook up a pressure gauge to read exhaust pressure, and experiment away.
__________________
Rich
