Cooled EGR vs Water Injection vs Super Lean
 

So I'm increasing my CR from 7.5:1 to 9.5:1 in my 72 VW Bug. I'm going to try to use cooled EGR to control detonation for now. I'm making the valve so that it operates with the throttle so that it opens as much as the throttle valve does only it's about 10% the size of the throttle valve. I'm pulling it off further down the exhaust so that there's less pressure in an attempt to get around 10% to 15% EGR even at full throttle. I also have a fully adjustable digital CB Magnaspark distributor.

I was also contemplating water injection. But it seems WI usually decreases fuel economy. I suspect that it has to be finely metered on a gasoline engine in order to increase fuel economy.

One problem with both EGR and WI is that they can reduce fuel combustion increasing CO and HC emissions while decreasing fuel economy. One thought that I had was that if I used enough, then I could run lean, even at WOT. But that would likely cause even less fuel to actually burn, perhaps even causing misfires.

But then it ocured to me. EGR and water can be used because they are inert. An enriched AFR can also do the same thing since fuel without oxygen is inert and so boils off and lowers combustion temps. So why can't a very lean AFR be used. The excess air without fuel would be inert just like EGR only that a very lean AFR would provide more than enough oxygen for complete combustion. I've read where after a certain point (about a 16:1 AFR) combustion temps and NOx production are reduced dramatically. Since a gasoline engine can run on as low as an 18:1 AFR why not use that to reduce detonation? Of course that might not be enough, so it might have to be used along with water injection at WOT.

Any comments will be appreciated.

This is what I'm talking about.

Could lean buning at WOT be possible with either the addition of cooled EGR or water injection?

Any thoughts would be appriciated, including the criticizm of those wondering why I'd even think of taking an air-cooled engine and increasing it's CR to way too high and then leaning it out to way too lean.

You can try it, but I don't think you'll have that great of success. If your engine can only run 18:1 in lean burn, you're not even going to get to 18:1 if you start injecting other stuff. The reason that the engine starts to miss at 18:1 is the charge is diluted so much that it simply doesn't propagate the flame front smoothly and evenly. There simply isn't enough fuel to support smooth even combustion. If you start adding water or EGR to the mix, its going to be even further diluted and/or quenched.

There are ways around this. Honda did pretty much all of these with their lean burn engines. Designing the head and combustion chamber for lean burn is one. This means lots of swirl in the cylinder and a higher compression ratio.

Water/EGR will definitely help if, for example, you're running 16:1 and hitting high EGTs. At this point adding water or EGR will help because you've still got good flame propagation.

Thanks! That makes plenty of sense. So basically if an engine has a lean limit of 18:1 then that's an air+EGR+water to fuel ratio. In other words an 18:1 AFR plus 15% EGR is actually a 20.7:1 of everything to fuel ratio, which is beyond the limit. Interesting. Whereas a 16:1 AFR plus 15% EGR is 18.4:1.

Well, I'm going from a 7.5:1 CR to a 9.5:1 CR. That should help (or hidder). Also I'm shooting for a tight quench. I can put dual spark plugs, that is, two in each cylinder. Would that help? How do I get more swirl?

Dual spark plugs will help.

As for increasing swirl, I'm not sure there is too much you can do. I assume your engine only has 2 valves per cylinder. I believe what Honda did on their lean burn engines was close one intake valve before the other, thus making the intake charge swirl a bit more in the cylinder. Really, all you really want is an evenly mixed charge in the combustion chamber. You don't want any areas where the fuel is not dense enough to burn properly.

Well then I better get things ready to tap those spark plug holes from the bottoms of the heads. In that case I'll be using two Ford TFI moduals to drive two Toyota 1.5 ohm coils. I've heard that that particular combination really works well with lean AFRs on a single spark plug. I'll need to figure out how to adapt a Nissan dual spark distributor cap and rotor to my distributor.

I'm sure this whole thing is going to really throw the ignition timing for a loop. Good thing I got a fully adjustable distributor.

I'm really working to get a well mixed charge with a carburetor. I supose that if push comes to shove I'll have to change to FI.

Oh and yes. It only has one intake and one exhaust valve per cylinder. I suppose the only thing I could do would be to add aluminum around the valves in order to shroud them. That's a bit more than I'd like to do right now.

Around 16:1 burns valves.

Isn't stoichiometric the hottest?

Nope. NOx increases as combustion temperature increases.

http://ecomodder.com/forum/attachmen...1&d=1434383825

You might want to look up some old SAE papers, that discuss cylinder head shape, spark plug location, squish angle, and so on.

The squish angle is something that can direct the charge from the outer cylinder wall area towards the spark plug when the gap gets tight.

When I was really interested in two stroke engines, Gordon Jennings from Cycle magazine was a great source for this type of research.

Yamaha submitted numerous papers on cylinder head squish bands, and Mcculloch (chainsaws) posted a paper or two on two-stroke squish bands as well.

There was a series of research papers from a professor at a Belfast University that was using computer modelling to predict engine tuning characteristics back in the 1970's, and years later I came across another research paper that showed he was selling a computer model to large companies for modeling the same topic. I don't know if this is something that would help you? The software might be very expensive now.

I do remember coming across one research paper that utilized 17 very small spark plugs in one cylinder! Haven't looked at those in years, but it might lead you down some different paths in your noble quest!

Jim.

stoich would be hottest with perfect mixture/atomization characteristics, but that isn't the case with any engine that I'm aware of(or at least, not outside of a very narrow RPM/load range). you'll always have some fuel somewhere in the chamber that isn't quite mixing and combusting at the same time as the air/fuel charge around it. essentially delayed combustion, which lowers the overall flame temperature a bit from it absorbing heat but not combusting.

with a little less fuel(say 15.4:1 vs 14.7:1), that pocket of fuel may not exist, preventing it from effecting the overall flame temp, causing higher overall combustion temps. now you just have air that didn't mix with fuel, which isn't nearly as detrimental to the flame temperature as air that didn't mix with fuel AND fuel that didn't mix with air.

either side of whatever ratio your engine burns a charge the fastest at will burn cooler. if you were really curious and happened to have control over commanded air/fuel ratio, you could install an EGT probe pre-catalyst and see what ratio causes the highest EGTs. there is some variance to be accounted for, different ratios require different amounts of spark advance to make best power per fuel unit, so being able to control air/fuel ratio and spark advance on the fly would be advisable. overly rich AFRs and retarded timing can do some really interesting things to exhaust temps though, since that is a recipe for sending a still burning air/fuel charge out of the exhaust port, wreaking havoc on everything along the way.

I've run everywhere from stoich up to 17:1(on E10) with no EGR, normal amounts of EGR and HUGE amounts of EGR(as much as the valve could flow) up to the point of lean misfires, EGR and lean burn do accomplish much of the same effect when used correctly as they both allow the introduction of excessive amounts of air(and "air") that isn't used for combustion. for reasons of cost, complexity and emissions requirements, EGR has been fairly universal and lean-burn much less so until recently.



can't speak for water injection though, never tried it. I see the appeal for certain circumstances though.

Thanks for the posts! Lot's to think about. It seems that for this particular engine most see the hottest EGTs around an AFR of 15:1. They usually see quite a bit less by 16:1.

One question, will 18.4:1 AFR run as cool as 16.1:1 + 15% EGR? I mean, at that lean is EGR even necessary as the air is just as inert as the Cooled EGR gasses?

going the lean + EGR route as opposed to even more lean - EGR, it would allow for a small drop in backpressure, would also heat up the intake charge a bit as well, which has its compromises.

being a less dense intake mass, there should be less heat overall. then you go back and forth between needing more throttle to produce the same amount of power(which also has its compromises), which would probably bring the cylinder temps close to where they would be with just the super-lean ratio.

six of one, half a dozen of the other.

Yes, NOx increases with temp. But doesn't it increase with more available oxygen too? All I know is that a lot of guys report lower temps at 16:1 AFR and leaner. Please correct me if I'm wrong. Of course there's more to it then just AFRs. There's load and timing too. The guys reporting lower EGR's while running 16.5:1 AFRs are for the most part doing that at part throttle, but enriching it to 13-12.5:1 for full throttle.

Thanks for the posts, because I've learned a few things so far.

First of all, hottest EGRs occur just lean of stoichiometric (14.7:1). But after leaning out even more combustion temperatures do subside just as if it were running on EGR. At which exact AFRs it does this on this particular engine I don't know, but others with this engine have indicated about 16:1 and leaner. I might get an EGT gauge in the future, but being it's an air-cooled engine the CHT gauge I have might be a pretty good indicator of what's going on.

I guess for now I'll shoot for a 16.5:1 to 17:1 AFR while cruising. That seems safe enough for this particular engine. It's the WOT I'm not sure what to do with yet. I really don't like the idea of running a 13:1 or richer AFR for high load. I could try to go even leaner than 17:1 in order to avoid detonation and ultra high EGRs. Looks like cooled EGR would have about the same effect anyway. Water injection would do the trick but that's not that easy to work with.

I guess EGR has been so popular not because they couldn't get a lean burn to do the same as far as combustion temperatures go, but because that way the exhaust gases are always depleted of oxygen, which is necessary in order to reduce NOx in the common three way catalyst (which I don't have).

One assumption, being this is a carbureted engine, is that it'll be very important to control intake charge temperatures. If I set it lean on a cool time of day if it gets hot then it will be running richer. I have a thermostatic air cleaner, but it's vacuum operated, so it stops regulating heated air at WOT. Maybe I'll need some sort of vacuum pump, unless I go with EGR or water injection as WOT detonation control.

One more question. What would be more fail safe? The popular method has been to enrich the mix to around 12.7:1 or so. But obviously a clog in the carburetor or faulty fuel pump can make it go lean. EGR can get clogged over time, or the linkage could fail. A lean burn could go rich if an air bleed got clogged or with a change in altitude. Water injection could cause lots of problems if water ran out or a pump stopped working or if it kept working when it shouldn't.

Yes exhaust gas temperature drops off pretty rapidly once you go leaner than 16:1.

It appears that some where around 16:1 is the best mix of the highest temperature and most free oxygen required to burn exhaust valves.

About 13.2:1 makes the best peak torque, 12.5:1 to one makes the highest horsepower in a naturally aspirated gasoline engine. (turbocharged or supercharged gasoline engines use lower air fuel ratios)

I try to stay between 12.5 and 13.2 when the gas petal goes to the floor. But this doesn't always happen.

Does this carburetor you are going to use have a primary and secondary setup?
Or is it single barrel?

If its just single barrel with a metering screw adjustment then you are only going to be able to tune your idle air fuel ratio a little and you will be stuck with one air fuel ratio that stays fairly consistent through out your entire operating range. Unless you get creative. If the single barrel carburetor has a metering rod and spring setup then you will have some ability to tune cruise, part throttle and wide open, but you wont be able to run lean cruise, then go a lot richer for WOT like you want to.

The primary and secondary carburetors give you more built in ability to tune cruise, part throttle and then accurately dump much larger amounts of properly portioned fuel and air mix into the engine at WOT when you need power.

Basically the more excess oxygen going through the engine the less effective the catalyst becomes. So a lower and lower percentage of that NOx produced gets treated.

16:1 is about where the flame temperature is hottest, but combustion temperature and exhaust temperature are very different. Under heavy load the exhaust temperature at slightly lean conditions will be quite high.

This is a single barrel Solex 34PICT-4 carburetor with a 26mm venturi and a 34mm base. Besides adjusting the idle mix I can get very finely incremented pilot jets that affect off idle AFRs. Also there's a wide variety of main jets to choose from. I can also possibly source three different sized acceptable used air correct jets for this carburetor, which basically are a replaceable air bleed for the main jet. There's also an "auxiliary jet" which in reality is a separate jet that runs the idle circuit while the pilot jet mainly runs the off idle circuit.

The power fuel system is quite different on this particular carburetor. It is designed to run at stoichiometric at part throttle but then enrich to around 12.7:1 at full throttle and high RPMs. It simply consists of two "high speed enrichment" jets (actually three if you include the accelerator pump outlet as a third jet) that are situated above the venturi. They're not activated by any mechanism off the throttle shaft nor from a vacuum diaphragm. They simply get their own venturi effect at high air speeds.

So yes, it seems like this carburetor is better suited for one AFR across the board. Most VW guys usually jet these to run 13:1 or rich across the board. A few ecomodding types can get them to run around 16.5:1 to 17:1 at low throttle and then enrich to 13:1 or so at high throttle. What I'd like to know is if I can get it to run 17:1 to 18:1 across the board and use something other than extra fuel to control high EGRs and detonation at high load.

You can certainly run 17:1 across the board, you'll have a lot less power but it'll work and not burn valves. (maybe hook up an aftermarket methanol/water injection kit to mist the engine with extra gasoline at full throttle?) At high rpm you'll have a lot of problems though since it'll probably lean out even more and lean mixes do not burn well at all.

Perhaps you could try shaving the heads to raise the compression ratio even more.

The engine can't do over 4,500RPM. I'm some what afraid to go any higher with the compression ratio. According to any VW guru, I'm nuts for having more than a 7.5:1 CR. (This is what all your typical air-cooled VWs have become thanks to all those VW gurus, 7.5:1 CR or less (one of them recommends 6.5:1 or less) with a 13:1 AFR or richer across the board along with a mechanical advance distributor.) I think I can pull off a 9.5:1 CR. If I can control pinging good enough at that CR I might take the heads off and have them fly cut a little more than what I've cut them so far (on air cooled VWs you fly cut them to raise CR).

Sounds like a lawn mower.
I say think out side the box. Bump the compression up to whatever you think it will take make you smile when you punch it.

when in doubt, emulate. older Porsches are easily the closest design to the VW boxer(and for good reason), but Subarus are a close second, they certainly had fun options.

LOL! Since the old VW engines are air-cooled there are a lot of myths about them. For an example, that you should only run single grade oils in them, or 50 weight because of the belief the oil runs much hotter than in a water-cooled engine. Interestingly enough, this myth is untrue since the typical running oil temperature in an air-cooled VW is between 180 to 210*F (less than 100* C) which is the same or even lower than many and most water-cooled engines.

I wish Mythbusters would do an episode on air-cooled VWs.

I think I can do fine with the VW engine. I'm not in this for power. The magnesium case on a VW engine is 2/3 the weight of aluminum. Calling it alawn-mower engine isn't too far fetched. I can pick it up by myself with just my arms and hands. The whole vehicle has a curb weight of 1,900lbs. I might cut a lot of that off if I go with fiberglass fenders and hoods someday.

They are really running out of ideas. I think they make one episode about every 4 to 6 weeks now.

Oh lol, 4500rpm max, that sounds like a good candidate for a compression ratio increase + lean burn haha. 9.5 plus a detonation prone design will help the mix actually burn at 17 or 18+ AFRs and advance should be easy to dial in since there's only a small range of speeds.

According to the owner's manual the engine shouldn't be run at that RPM for over a minute. 3,800 is about the highest constant RPM. Not that these engines can't be modified to do more. With the right modifications, this little engine can push a Bug down the 1/4 mile in less than 10 seconds! I think the record in a Bug is something around 6 seconds! But the original design was "the car that costs the same as a used car new." It also was rated to get mid 30's in MPG. Not bad for a 1930's design!

When you say "detonation prone" do you mean it'll be pinging all over the place or do you mean it would be pinging all over the place with a rich AFR?

Of course I have that fully adjustable distributor with a 31 x 31 fully custom spark table. I plan on doing a very thorough tune on both jetting and ignition timing, basically as close to a dyno tune as possible.

I thought it was interesting in that article I mentioned earlier how some pilots run wide open throttle and then regulate power by leaning it back, which puts it leaner than peak exhaust temps. Interestingly enough, that's a lot like a diesel. Of course the problem with a gasoline engine is you can only lean back so far, and then you have to cut back air too or it'll missfire.