[dijinj]

I have invented a new rotary engine with more than 80% efficiency please help me to improve this concept and cfd modelling of the engine. Technical details are given in site http://sites.google.com/site/anyoonrotaryengine/

[auricom]

hello, I have some experience with Wankel type rotary engines. This concept is interesting, but I have a hard time believing an efficiency of over 70%. What is your peak cylinder temperature? Are the "stoppers" in your diagrams the seal between cycle chambers (compression seals)? These appear to be linked to the stator or "rotor housing" as opposed to the Mazda strategy of attaching them to the rotor. I suppose this means that they are to slide in and out as the rotor face approaches and moves away from the rotor housing? Does this relate to the integrated cam somehow? Does this engine regulate torque output with an air throttle, or by some other means? The linked website describes the cooling as taking place via water injection, is this the only cooling path, or is there an additional coolant gallery? What sort of skin temperatures on the rotor / rotor housing are you predicting? Does the combustion take place in the same location with respect to the rotor housing for every cycle? Do you require lubricant injection to the combustion area for the sake of your compression seals (the "stoppers"?).

Please tell me more about this idea.

~Max

[dijinj]

it achieves this efficiency by using thermodynamic cycle advantages and new cooling system. stoppers seal cycle chambers you can view or download animations from download page for reference. cam is integrated to rotor. additional air cooling path is given. around 2000k.yes combustion takes place in combustion chabor in casing. mist lubrication is used

[auricom]

if you are claiming 80% thermal efficiency, your peak temperature would have to be around 7500 *K assuming it is running in a room temperature environment. And that is also assuming you have created a perfect Carnot cycle engine.

So in reality (non-Carnot cycle) you would have to be well over 7500 *K (closer to 10,000*K) peak temp. This would be very difficult in an engine that has a single area of material that is close in proximity to this level of heat most of the time. The section of rotor housing near the combustion event never cyclically gets the cooling intake air moving past it. With a piston engine at least every part gets a dose of cooling air once a cycle (even the exhaust valves can be allowed this via valve overlap) Perhaps you could get away with this sort of thermal loading with a clever air cooling strategy. I am reminded of cooling holes in the leading edges of high pressure turbine blades.

I hope most of this engine is somehow made of ceramic materials...

[dijinj]

theoretically an engine can have 100% efficiency if exhaust temp is same as intake temp; i mean no heat is rejected all heat is converted to mechanical energy. normally a engines have around 30% efficiency. and 30% is lost through exhaust and another 30% is external cooling loss. I attacked both problems by implementing Atkinson cycle and internal cooling by water injection respectively. to maintain power and also improve efficiency constant volume heat addition is also implemented. so peak temp is around 2000k. compression ratio is around 10:1

[auricom]

Do you disagree with this statement below (copied from Wikipedia, but can also be found in any thermodynamics book, such as "Fundamentals of Engineering Thhermodynamics" by Moran / Shapiro which I am looking at right now):
[edit] Carnot efficiency

The second law of thermodynamics puts a fundamental limit on the thermal efficiency of all heat engines. Surprisingly, even an ideal, frictionless engine can't convert anywhere near 100% of its input heat into work. The limiting factors are the temperature at which the heat enters the engine, , and the temperature of the environment into which the engine exhausts its waste heat, , measured in an absolute scale, such as the Kelvin or Rankine scale. From Carnot's theorem, for any engine working between these two temperatures:[4]
This limiting value is called the Carnot cycle efficiency because it is the efficiency of an unattainable, ideal, reversible engine cycle called the Carnot cycle. No device converting heat into mechanical energy, regardless of its construction, can exceed this efficiency.

[auricom]

Now, If you can get your peak temp up WAY higher, then I can see where you might have a shot at approaching the efficieincies you are talking about. I'm not saying what you are claiming is impossible, but based on what you have told me, it does not add up.

[dijinj]

engine working between 2000k and 270k can have efficiency 86.5% according to Carnot limit. this 270k can be attained by higher expansion ratio and water injection

[dijinj]

even engine working between 2000k and 300k (room temp) will have Carnot limit of efficiency around 85%. the 2000k is achieved by implementing constant volume heat addition without any mechanical complexity and the 300k is achieved by higher expansion ratio and water injection for cooling.

[auricom]

So you are saying that T_hot is actually much greater than 2000*K, but the peak temperature in the chamber only reaches 2000*K for various reasons?

I'll have to go back and check on what water injection means to this process. Thermodynamically all water injection means is that the ratio of specific heats of the mixture will be different than with just air. Also, there will probably be a phase change of the water from suspended liquid dropplets to gas. I'm not sure if this will make a difference to the carnot cycle, but I would suspect not. This is becasue I believe carnot cycle limits are not constrained to mixtures of a certain specific heat capacity (air), you could use xenon and liquid mercury and I don't think it would make any difference.

I am quite sure that the over expansion has no effect on the Carnot cycle. This cycle is an idialized internally reversable cycle. That already implies that the fluid has done the maximum amount of work onto the machinery. Of course if you try to expand the gas in the expansion stroke beyond atmospheric pressure, then it will take energy to do this, so you have a limit there.

Please correct me if I am wrong.

[auricom]

Also, your comments about constant voume heat addition seem misleading. The only reason why constant volume heat additions is preferable is becasue for air Cv<Cp.

This means that it takes less kJ to reach the same delta*K if you are doing it Cv than with Cp. Again, the point of this is getting the highest *K possible with as few kJ as possible.

[dijinj]

please see the pv diagram in the link http://sites.google.com/site/anyoonr...ine/pv-diagram. constant volume heat addition can provide more work than piston engines engine. in piston engines spark is given before TDC and combustion continues a few 10s of degrees after TDC. this is clearly given in PV diagram

[auricom]

Well that is all well and good, and the principal is not incorrect. I don't know about the accuracy of your PV diagram, but they are clearly just for visual aids so I will not dwell on that.
If I assume you will still have a single moving flame front in this engine, then I must assume that you pull off your constant volume heat addition by holding the volume at TDC nearly constant for some reasonable amount of crank angle. Have you considered what this will do to your average combustion wall temperature?
I think you are correct in assuming that holding the volume constant at TDC while the mixture burns will on average produce desirable results, but there will be a drawback that more heat will escape through the chamber walls. More compared to a quick stay at TDC assuming water injection / materials are constant. Anyway, as I said this is probably a small price to pay, but you should consider it. Also, how is the cycle effected at drastically different engine speeds? The fuel will take roughly the same time to burn, but that will equate to drastically different changes in crank angle. This would mean that at low speeds if you hold TDC constant for 20 degrees, it might only take 5 degrees for the fuel to finish burning. Now you have 15 degrees of crank angle where you are just leaking heat out of the engine and doing no work. Again, this might not be a huge deal, but these sorts of little inefficiencies start to add up.

Additionally, it is quite possible for a "piston enigne" to follow the same type of curve, just not with a conventional con rod / crankshaft assembly.

Do you happen to have a T-S diagram to go with that P-V diagram, I would be very interested to see that. Please indicate a few termperature values on the y-axis if possible.

[dijinj]

Fuel burning crank angle will be almost same irrespective of rpm even in piston engines because of turbulence also increase at high speed and in turbulent flow flame propagation will be at very high speed

[auricom]

That is most certainly not true, why do you think spark angle is advanced as engine RPM increases in virtually every spark ignition engine ever made?

[dijinj]

what i am saying spark angle is not advanced because turbulence also increase at high speed and in turbulent flow flame propagation speed will be very high. so at higher rpm fuels burn at faster speed than lower rpm. angles rotated by crank will be almost same for complete burning of fuel because of turbulence. you can refer engine design books for confirmation

[auricom]

Is this phenomenon something specific to your engine? It is certainly not normal. This is true that a flame front will in gernaral propogate faster with more smaller turbulent eddies in front of it. There is no perfect equation for predicting the speed of a moving flame front in the "flamelttes-in-eddies" regime, but you can bet good money on the speed of the flame front not keeping up with the engine speed change. Again, I ask: Why does every spark ignition engine advance the timing as engine speed increases? Giving you the benefit of the doubt, it appears from your answer that the engine you have designed has some unique charecteristics that cause the flame front to not behave in the same manner as every other spark ignition engine in the world. Is this the case? Is the engine even spark igntion (I just realized that I never asked that) What is the designed speed range of your engine? You would not have to worry about this problem if the engine is only designed to run at a single speed (like in some gen-sets).

[dijinj]

spark timing is advanced in SI engines to compensate for ignition lag. A turbulant motion of mixture intensifies the process of heat transfer and mixing of burned and unburned portions in the flame front (diffusion). these two factors cause the velocity of turbulent flame to increase practically in proportion to the turbulence velocity. the turbulence. swirl chambers and and intake flow modifiers are based on this theory. rpm limit is 8000 and because of special shape of rotor in expansion side torque will be high even at low rpm.

[dijinj]

You have very high theoretical knowledge and being on CFD-online ; I assume you are a CFD specialist with mechanical engineering background. Am I Correct.

[auricom]

The primary purpose of increasing advance as engine angular velocity increases is not to compensate for ignition lag. Where did you hear this information? A flame kernel will be created a VERY short time after an isolated high temperature region is created nearby (a spark). This very small time is what is known as "ignition lag". This amount of time is dwarfed by the peak pressure time versus crank angle delta that will be created if igntion advance remains static while rpm increases. calibrating spark advance is one of the two fundamental parameters necessary for a well running engine. That being said, "tuning spark advance" is the first or second thing anyone who tunes engines will study. The tuning of spark advance is typically based on engine load and engine SPEED. Advance 99.9% of the time will increase with speed, sometimes as little as an additional 10* in forced induction applications (partially due to the non-steady manifold pressure at WOT) and sometimes 25* or more for naturally aspirated engines. Unless I am missing something, there is nothing about your engine that will result in drastically different turbulence charecteristics, so I would assume that your engine would need to follow these general guide lines as well.

To answer your second question, I am a mechanical engineer working in the heavy duty diesel industry specializing in turbo machinery. Beyond that I am also building and calibrating a control system for a Mazda 13b turbo rotary engine with my electrical engineering friend which we work on in our free time. I am only very new to CFD, but have a stong background in fluid dynamics which was the primary subject of my focus during my last couple semesters at undergrad. I am also designing a unique engine concept, so I have likely recently done similar research that you have been doing for your engine.

What is your background?