Atkinson cycle > how much to get

[Jann353]

Good morning everybody (it's morning here in Poland, so let me start like that), I'm sort of new here. Although I've been reading the forum for quite a time, it's my first post.

I'm on a very early stage of designing an Atkinson 4t engine, the design will be most likely followed by making that engine. Therefore I have a few questions...

1. What efficiency is really possible to get with compression/expansion ratio? I thought of ethanol driven engine with CR 12:1 and ER 25:1. The engine will be anyway small, the issue is maximum achievable efficiency - not gonna be a useless model. I counted theoretical efficiencies as Mr. Heywood advised and did my own real cycle calculations. - The thermal efficiency values seem to grow together with expansion pressure reaching 2.5-3 bar in real cycle. On the other hand, my friend who works as a professional engine constructor suggests much lower ER, claiming that in true engines we don't operate under the pressure of 5 bar. Could you advice me sth?

2. I'm collecting fuel data for ethanol. I know the efficiency-to-AFR curve for petrol. Do you have such curve for ethanol? Where to find it?

Thank you in advance for any tips and hints.

 

[Lou Scannon]

I strongly suspect any efficiency to AFR curve will be strongly dependent on the engine in question, particularly its ability to combust dilute mixtures efficiently. In addition, I expect there would be an rpm dependency. Also, ignition timing will play a strong role; how is that accounted for in such a curve?

Can you direct us to your reference for the curve you have for petrol?

The problem with an extreme expansion ratio that looks good on paper is that it encroaches on the part of the cycle needed for exhaust blowdown; this will impair volumetric efficiency, thus reducing power density. For your project you might consider building a model in GT Power to evaluate these and other trade-offs.

 

[Jann353]

Hi!
AFR to FC curve I thought about sth like this one:

http://www.clubmr2quebec.com/photo_upload/tech_ref/AirFuel_SRFCurve/Stoich.gif

Not getting into details, i thought about this general principle that burning lean gasoline mixtures with lambda 1,1-1,3 might bring some efficiency increase. For ethanol the boundaries of mixture's combustibility are different, that's why I'm not sure about the optimum ratio. I plan to use two spark plugs to improve ignition.

Exhaust blowdown - sorry, until now I have been mainly calculating 2S engines. I just assumed that in case of having the whole stroke for expelling the burned gases out from the cylinder it's not such a big deal anymore. Especially that the engine is to work on ethanol (I guess VE would go up) and the Atkinson cycle is gonna be achieved by late IC - shouldn't this also keep the VE relatively high?

 

[BrianPetersen]

It is not worthwhile to expand the gases all the way out to atmospheric pressure, at least not at full load. The extra friction in the piston rings and the extra heat loss associated with the extra surface area due to the larger-than-otherwise-needed cylinder to achieve a specified power output will cost you more than you're gaining. I suspect that with an expansion ratio basically double the compression ratio, that you are going to be beyond the point of diminishing returns.

If you have variable intake valve closing timing, you can use that to vary engine torque output by varying the amount of charge that stays in the cylinder, in place of a throttle. Many production engines are already doing this to some extent. Pumping losses are a little lower than by throttling if you do this.

 

[moon161]

Check out

http://www.pattakon.com.

The guy looks at lots of variable cycle engines.

 

[Jann353]

The engine is very small, supposed to work @full load in relatively narrow range of rpm. I know about the issues of friction. So far I never had to build an engine I calculate. Of course, I will launch some optimalization, most likely with AVL soft, but wanted to begin with assessing some rational dimensions i could refer to later on in my work...

That's what my questions are about.

 

[patprimmer]

Very late intake valve closing decreases volumetric efficiency by pumping charge back into the inlet manifold. As this air will already contain fuel unless you plan on direct injection after the inlet valve closes, unless this is well reasonably homogenous, it will upset a:f.

Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[BrianPetersen]

If it's fuel-injected, it can be mapped to account for that. Toyota Prius uses late intake valve closing, and it uses ordinary port fuel injection.

 

[turbomotor]

Jann353,

Have you considered closing the intake valve early to limit effective CR?