Lean burn on automotive natural gas engines 3

[msantiago]

Hello All,

I would like to know if anyone has experience on the increase of efficency and emissions with natural gas automotive engines running on lean burn.

Also if any advice for vendors that can supply GM vortec 8.1 turbo or Ford 6.8 turbo on lean burn.

 

[catserveng]

I rummaged thru some old notes from that era.

Our major issues were,

Getting enough air into the engine to get the required excess air, as Ivymike pointed out above, to get a true lean burn and meet the emissions targets we were shooting for we needed about 8% O2 in the exhaust. Finding turbo matches was very difficult.

Fuel management, we did most of our work around Impco mixers, a lot of custom fuel valve shapes got made and tried. Continental Controls was new about then, we tried some of their early stuff, never met our needs. We also played around a lot with gas regulators and trim valves, idle air bleed valves, and some other things that got a bit crazy.

Ignition systems, trying to fire a plug in a lean mixture takes a lot of energy, the Altronic DISN came along and helped that out pretty good, but we had terrible trouble with spark plug life, either due to electrode erosion or flash over at the insulator. Plus it was a pricey fix at the time.

With some Ford engines we got some help from a local race shop, one of the guys there came up with a cam grind that helped, but overall the engine got real expensive to keep running as a lean burn.

Air fuel ratio controls, mostly issues with early exhaust oxygen sensors. CAT had one, but it was REAL expensive and had poor service life, plus it needed a fairly complex signal conversion. Woodward came out with the GECO Control near the end of the project, we worked thru quite a few bugs with them and got a couple working quite well, but the overall price for a unit at 75 kW was awfully high at that time.

We did bust up a couple of engines pretty good, they had high boost pressures and high compression ratios, and at 8% O2 ran great, but if they got below 5% O2 they detonated and hammered themselves apart fairly quick.

Mechanically if we could keep the engines out of the detonation areas life was good, bottom ends were never a problem during this exercise, top ends had good wear rates compared to running rich burn. Some cam and lifter combinations didn't like the ultra low or nil ash oils (scuffing and spalling), some piston and ring pack conbinations didn't like medium ash oils (deposit rates).

The project wasn't a total bust, we learned alot about natural gas engine performance, and about ignitions and AFRC's that we turned around and applied to the larger engines with some benefits. But the objective to setup a viable and cost effective lean burn auto derived engine for generator or chiller service in our required emissions levels was never met our targets.

 

[franzh]

It is extremely difficult to have an engine start and run at low and part throttle modes during lean burn, most apps shift to lean burn at cruise and WOT but start and part throttle ramp to load at stoich. This is what is experienced in transit applications.

If I were to guess the number of lean burn applications on an over the road use, they would be only a small fraction of what I see in stationary steady-state genset use.

Early spark plug failure, piston damage, and exhaust valve torching are common lean burn problems. I seldom see this occur in steady-state, but much more frequently in transient mode applications. The damage can happen very quickly, often too fast to stop the engine. Interestingly, similar damage can occur from overfueling as from lean burn.

In the early 90's, I participated in several lean burn projects, some worked well, others ate up too many pistons.

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.

 

[dcasto]

just google up industrial lean burn engines from Waukesha, Catapiller, white-superior and look at what they do.

http://www.waukeshaengine.com/index.cfm/go/product-detail/product/VHP-12-Cylinder-Gas-Compression/

 

[turbocohen]

When I visited customers who were working on lean burn LD engines using the first few iterations of cng injectors we were developing often I would get to see what did not work or deep six'd. Lean burn, like steroids offer some gains but at a higher economic risk. A decade + ago we sponsored an individual in Ontario working on an unthrottled chrysler 4.7 V8 with one intercooled solenoid electronic EGR per cylinder mfg'd by another division of the company I worked for at the time. Using our cng injector and egr valves along with a cam ground to help reduce cycle to cycle deviation and a post cat exhaust throttling valve (normally open) used only to provide more egr at higher stoich loads, the guy was able to demonstrate a range of loads ranging from lean to just slightly rich of stoich idle as well as mid range loads at most rpms with up to 70% egr. Full load required stoich but he was able to maintain 15% egr @ WOT and kept precat nox remarkably well under control except at cold part load. If I read the above I would question it, as I did and my boss did but we visited and took a weekend of measurements.

The project was quite interesting but it was risky to presume that multiple egr solenoids could be economically produced within the needed dynamic range. And as expected there was very little misfire tolerance.

Turbo

 

[Lou Scannon]

"at most rpms with up to 70% egr"
turbocohen/b], that sounds a little extreme; care to elucidate?

 

[turbocohen]

Lean low load and idle the 12:1 cr engine ran reasonably stable wot using o2 laden egr dilution. I thought BS too until after the visit. 70% egr is not 70% manifold co2.. calibrated the stand that day myself and saw about 50%. No, aliens did not invade my body but when I saw the um, lets say.. enhanced coil on plug ignition I could imagine they were picking up rfi. This engine was not fit for prime time over the road but a testbed for an ME that my former employer sponsored. I built injectors for and visited over 50 univerities with alt fuel projects and few were as out of the box as this one. If I am not mistaken the guy working on the project went to work at Navistar or a diesel r&d outfit in colorado. An SAE paper was in the works that I reviewed but I am certain it was not published.. if you want to know more I can let you know.

 

[Lou Scannon]

OK, so we're talking cutting (bleeding) edge here...
I'm in Colorado so which R&D outfit are you referring to?
If you have more to share perhaps a new thread would be in order...

 

[franzh]

The company Turbo was formerly associated with was indeed cutting edge which is how I became familiar with him. Whatever one reads as published research material, there are several layers of unpublished material behind the scenes, and THAT is where the Skunk Works does their dirty work.

Even though I was a contractor at an engine R&D lab in the early 90's, some areas were off limits to me. Windows papered over, cardboard blinders over the instrument panel, one cell even had a military guard stationed there. Sometimes, weird sounds or smells would be a give-away, and sometimes just the absence of sound could be just as telling.

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.

 

[turbocohen]

In this case it was not rocket science, nothing automotive is... The project was of interest to my employer due to the requirement for many solenoids from one mfg. There was a cng fuel rail and an egr supply rail. The cooled egr was delivered through the existing port fuel injector ports on the stock plastic manifold and the cng was injected through the sidewall of the egr pipe right next to the port maybe 10 mm above the stock location on the same plane. The egr dilution strategy, I presume, was to maintain enough N2, co2 dilution to limit the %o2 and afr on an unthrottled engine without using vvt. The toughest part was aquiring reliable feedback and the mfg cost would be higher and much riskier than throttless VVT and DI.

The vvt/di approach won for good reason and in the future I believe vvt/di stoich will work better than lean port cng as well. We also supported a few di cng projects over a decade ago.

Going back to the OP, a GM 6L v8 low pressure (~60 bar) DI CNG lab rat was on the road some time ago that showed some promise but thelack of funds, lower cost of port injection and little interest at the time pulled the plug. I built a few sets of DI capable cng injectors for former clients and another group of students who were competing at an sae competition. Mileage improvements and co/hc reductions were there in spades but Nox was horrific due I beleive a lack of proper chamber shape, inadequate spark heat and less than optimized placement of less than optimum injectors. The port cng injectors we built then (substantially identical to todays continental cng injector) were equipped with modified armatures, coils and pole pieces shared by an earlier DI gasonline application. Opening under high inlet pressure was not an issue with enough amp turns and drivers.. the problem imo was poor control over fuel/air preparation in the cumbustion chamber.

Just my 0.02 dinars.. Respectfully, Turbo Cohen

 

[dcasto]

hemi, CSU has one of the largest natural gas engine research centers in the US.

Colorado State University's Engines and Energy Conversion Laboratory (EECL)

 

[Lou Scannon]

Yes, I'm aware of CSU's lab but I haven't had the pleasure of a visit, yet. Haven't heard of an "extreme EGR" project there, either. Is it hush-hush?

 

[turbocohen]

For more info feel free to read

http://www.flexdi.com/flexdi-solution-gaseous-fuel.asp

The company was partners with Siemens long ago and this group broke off after I left. They were in the same facility I worked at but they were not involved with cng until after I left..

The technology aint perfect yet, nobody has perfected gasoline or diesel injectors either, but this may be a good place to look if you are considering a lean burn gm or ford cng di application.

Good luck with your search Msantiago