small engine primitive EGR

[KevinKoss]

I am working on air-cooled single and twin cylinder engines. As of next year, our friends at CARB tighten the regulations on these smaller engines such that a catalyst is needed (for a reduction of about 38% HC+NOx).

I have tested catalysts and they will all work just fine, at the cost of about $12-15 per engine (which on manufacturing costs alone is about a 10% increase - not jumping for joy here).

I have also "tinkered" with EGR and can attain nearly the same reduction without tweaking the fuel mixture (as we could run a little leaner with cylinder temperatures coming down). Average lambda in this market is .85-.90, because they are air-cooled. With EGR we could probably lean it out and get even more benefits.

Here is my dilemma: EGR may cost more, so I was thinking of simply adding some calibrated plumbing directly between the exhaust and intake ports. Do you think this is possible as a cost reduction? What is the average OEM costs of a vacuum EGR valve? Aftermarket is like $70-90, so my intuition tells me it would be at or below the cost of a catalyst, but it might take a bit more development time.

 

[franzh]

I played around with an EGR system on small 4 cycle engines and had nothing but problems. Since the volume of EGR was so small (about 8% of total exhaust flow at max flow), any tubing connecting the exhaust to the EGR valve, then to the intake would plug in short order. Any condensation would quickly settle in the tube and trap both unburned fuel and exhaust residuals. Direct flow with no EGR valve between the ports played havoc with idle speed and mixture control.

I even removed the EGR valve and had the exhaust constant flowing into the intake, above the throttle body, but any attempt at idle control was almost useless, the exhaust flow was too high a percentage.

I also ran into the problem of pulse reversion, intake air-fuel charge was being pulled into the EGR tube and condensing, then hardening (carbonizing). I ran the EGR tubing in a cooling water bath, no luck, then wrapped the EGR around the exhaust to super heat it (no condensation) but all that did was to super heat the exhaust flow into the intake, and a resultant loss in power (noticeable at load).

We played around with a mechanical linkage to the EGR valve opening to keep it closed at no and part throttle, but after only a few hours of running the EGR passage tube would occlude.

Engine particulars:
<26 bhp, V-twins
< 12 bhp, single cylinder
Propane and natural gas for fuels.
Tube diameter = about 3/16" od, various orifices tried.

When it was working, it really had an impact on NOx, which let us play around with HC and CO, plus we could bump compression without detonation, a nice side benefit. BUT, we could watch the NOx number climb over a 5 hour steady state 75% load (generator set package).

Franz

eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[KevinKoss]

Thank you Franz,
You mentioned you had exhaust flowing above the throttle body, did you try going behind the throttle plate (where more benefits could be attained).

Currently I am working on a gasoline 6HP single. What I have noticed so far is that WOT power drops 5-10%. At all load points HC goes up, NOx goes down. I could hold constant power down to 10% load, but 0% was painful.

I tapped into the first chamber of the muffler, ran through steel-braided line to a ball valve, another line, then to the intake manifold after the carb. From exhaust to intake valve, probably 30-36" of travel. I would like to shorten this path, because the exhaust cooled down to only about 100 degC to the intake. From what I know, the heat is an hidden benefit helping to vaporize the fuel prior going into the combustion chamber.

I think my next step would be tapping the head closest to the valves and running steel line with different orifices - but I haven't done testing to see if there is going to be significant contamination - it would be nice if off-idle EGR could be controlled by a simple $2 check valve based on deltaP. I would also like to do possibly tune the EGR piping length based on pressure pulses and valve timing.

I already have a working catalyst system nearly complete for compliance with CARB, but I want to keep my options open for the future.

 

[Fabrico]

If "calibrated plumbing" means a constant opening, it's a big compromise. Too little to do much good, and too big to idle properly. There's also starting, throttle response, and back firing to consider.

In addition to vacuum, it might also be possible to control an EGR mechanically by throttle opening, or electrically by RPM.

 

[KevinKoss]

The "calibrated plumbing" is to time the exhaust pressure pulse to hit the intake runner at nearly the same time the valve opens (this application is a constant RPM engine).

The mechanical controlled EGR based on throttle position is probably the cheapest and easiest to employ. For this 6-mode emissions test: 100,75,50,25,10,0% The points 100%, 10%, & 0% sum to a total of 21% of the weighted result...

So closed at WOT, and closed at around idle would keep power and starting close to the status quo. Then the most benefit would come for the remaining part throttle loads, and we would just have ensure a smooth transition between EGR vs non-EGR to the engine doesn't end up "searching" for a throttle position in those modes.

Excellent idea Fabrico, I think I have something to work on during the upcoming slower winter months now. I may try and grossly fail, but at least I know I tried. We all know EGR works, it's just a matter of employing at on a single-cylinder (pressure pulses galore), air-cooled (very cheap) engine.

 

[franzh]

When "tuning" the length of a pressure wave pipe, it is ideally synchronized at a certain rpm / pulses per second or minute. I had all of my versions tested except the one I described earlier below the throttleblade, exposed to "manifold" vacuum, however erratic it was. I didnt spend any time tuning the tubing length. The ball valve I used siezed due to heat, the replacement valve began to have problems with plugging. I didnt have the resources to develop an electronically driven pintel-valve automotive style EGR valve, and would outpace the realistic cost benefit of the system.

Does it work? Yes. Is it robust enough to allow CARB emission lifetime cert? I am not certain enough to have a good warm and fuzzy feeling.

Franz

eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[mpgmike]

I know that OEM types don't like soliciting outside patented ideas, but something you really ought to study is

http://www.somender-singh.com

With a little creativity, you might spark on something that can solve your problems.

Mike

 

[KevinKoss]

Thank you for the inventor's website, maybe something to tinker with, but there is no hard evidence of it working, at least as far as emissions.

 

[Flamefront]

In the battle for reduced HC+NOx there are a couple of other tricks you can play that are very reliable.

While consulting with one of the larger 2 stroke mfrs over the past years in helping reduce their HC numbers (NOx is not a concern in 2 strokes), I found some mettalurgical games you can use in the EGR lines to assist coking reduction that might help with the check valve plugging issue, which can get your EGR to make it to the 300 hour "commercial life" milestone.

One other way to reduce HC is to us a more aggressive spark plug electrode, believe it or not (as proven on projected nose plugs by KevinKoss above). You can improve on projected tip spark plugs by using combining a few old racing tricks in the spark zone.

I worked with the EPA during the rulemaking in validating an improved electrode that reduced HC emissions by ~6% in 4 strokes, and improved bsfc by a similar amount. The EPA classified the spark plug as a 'supplemental emission control device' and put it as a leading technology in the two Off Road Rulemakings. The main benefit came from an increase in IMEP through reduced flame travel time and a corresponding reduction in COV over all operating conditions. We did this by carefully tailoring the electrode pair interface.

The invention lead to the EPA "support", an expensive patent, an SAE paper, some CARB Workshop presentations, widespread academic support, and a long road to commercialization, which is another soap opera in itself. Along the way we became very close with a number of labs and OE engine makers during and since that time as the 2007 regs approach (here now, as far as production is concerned).

There are a few more ideas that are not covered by pre-existing NDAs that could be of use.....if someone would implement them. I found that it was often hard to get Engineering Directors to perform basic emperical research or to implement what was proven by testing even if the cost was low. Worse than NIH syndrome, it was a lack of intellectual curiosity on some critical matters...

 

[EHudson]

Check out an early Ryobi 4 stroke trimmer, they just drilled a hole between the two ports. Similar to heat riser to manifold ports on late 70's cars before the fancier EGR controls were put in place.

 

[KevinKoss]

FlameFront:
I used to work for a spark plug maker, of course a spark plugs could improve combustion if placed right... but a lot has to do with fighting for space with 4 valves and overhead cams. Plus, a double fine wire projected tip 8mm plug is about the edge of what can be made and even though its the best for combustion - In OEM cost-benefit analysis, it won't be worth it.

On a side note, I got approval to try the head scoring and actually get some data on power and emissions, at least on a small engine.

EHudson:
I read about the Ryobi Pro4Mor with the EGR, short of buying one and looking at it. It sounds to be slightly more complicated then a simple hole (which I thought it was as well). That is what got me to try EGR in the first place, if it were as easy and cheap as a calibrated drill bit. But that is a screamin' 50cc engine, I don't know RPM buy it is probably significantly higher than my 3600, meaning timing and pressure pulses maybe are more attuned to a simple hole drilling.

I already have a working solution to meet upcoming CARB standards. I would just like to meet them without as much cost.