The Star Rotor motor 2

[SBBlue]

Okay, here's a fun site to visit that has nice graphics and presents an interesting concept;

http://www.starrotor.com/indexflash.htm

Here's everybody's assignment;

Aside from the fact that the actual star rotors device might not function as hoped, what problems will probably keep the entire concept from working as advertised?

I'll put in my two cents in a couple of days.

 

[SMOKEY44211]

The engine described in the drawing can't work. The pressure at the compressor discharge is equal to the pressure at the expander inlet. I'm guessing that the compressor and expander are on a common shaft, although its not apparent from the drawing.Also blow by on the expansion side I think would be high. How is this baby lubed? Expansion ratio #'s don't look too good. In the end I think an axial flow turbine does a much better job.----Phil

 

[Warpspeed]

It should work, its just a small low speed gas turbine. The volume of gas discharged will be much higher than the volume compressed. So the turbine would have a much larger displacement or run at a higher higher speed than the compressor.

It would suffer all the same problems as a gas turbine, and with a few more besides. Most probably it would be very inefficient at less than full throttle. I would also expect the speed range might be quite limited.

Also the positive displacement expander might tend to clog up with combustion products and come to a grinding halt pretty quickly!

 

[maniacle]

Ive been following this engine for some time. First analysis of thier claims seem a little off. They origionally claimed efficiencies approaching 80%, but in the past several months have added new material and placed more realistic expectations. Still, compression to 6 atmospheres only works out to an equivalent compression ratio of 3.59. that means thier efficiency should be sitting somewhere around 40%, not 44 to 46. acknowledging they plan on preheating thier charge with a heat exchanger, it still seems hard to obtain those high efficiencies. For starters, its hard to convince that a posotive displacement unit thats expandning the ignited gases is going to be adiabatic or isothermal, so how much heat will they be able to extract from the exhaust gasses for the pre-heating stage?

a second issue to take up, is they claim low rpms, or at least lower than typical blade turbines. At low rpms, with no sealing to speak of, it will be difficult to maintain compression. To improve that the engine must be run at higher speeds. NSU had an engineer who worked with Felix Wankel named Walter Froede, and he designed an engine called the DKM-54. This was a similar design, of two cyclic rotors, one within the other. It ended up running most effectively at higher rpms, 11,000 to 24,000 or so. The issue at those speeds was the inertia alone of the outer rotor made the engine difficult to accelerate, and best off running steady state.

i could keep spouting off information, buts its probably better left alone. if im wrong on any of this stuff, just point it out.

 

[SBBlue]

I think their thermodynamic calculations are correct -- I tried running them and got essentially what they claim.

But. . .Here are the "pinch points" I see in the design:

1. The engine uses "inlet fogging". A 155 horsepower engine would require 23 gallons of water an hour for fogging. Besides filling up with gas, you would have to fill up with water too.

2. The heat exchanger to preheat the air going to the combustion chamber will have to be rated at 13,750 BTU, and will have to handle high temperatures (1500 deg F). This is similar to a "recuperator" used on turbine engines. A heat exchanger of that type will probably cost at least a dollar a BTU/min -- hence you're looking at $13,750 for the heat exchanger.

If you want to put one of these heat exchangers together yourself, you will need about 4600 feet of 0.6 inch outer diameter stainless steel tubing.

3. The storage tank will be able to boost the engine output by 30% for about seven seconds.

4. I wonder just how efficient a gerotor gas expander will be. It's noteworthy that they have announced that they have a compressor working, but don't make mention of the expander portion of the engine.

5. I think they are being a little bit deceptive in their efficiency claims. A gas turbine is not an especially efficient engine; diesels are more efficient. Diesel's do have relatively good durability.

 

[motorsportsdesign]

A piston engine with the same concept was made in the early 1900s by a guy named Brayton. There is an example in the Henry Ford Museum.

Interestingly it is displayed next to an otto engine.

I think this concept was also used by the Germans in WW2 to power cars with wood IIRC.

The weak point I see is the length of time that the expander surfaces are xposed to high temperatures.

Turbine engine handle this by internal cooliing on the blades and cold air barrier layers in the combustion chambers.

I think the four stroke has an often under appriciated feature in that the high negative power compresion work is done at high mechanical advantage.

Jonathan T. Schmidt

http://www.motorsportsdesign.com

 

[Pancholin]

It looks to me like someone was bored and looking at a Ford oil pump and said "Hey" Then maybe noticed that an air compressor at either end would really make it spin.
Problem: The compressor volume would have to be 6 times the volume of the expansion stage and at the moment of combustion without some sort of check valve the compressor with 6 times the exposed area will try 6 times as hard to reverse its direction than the expansion side to turn in a forward direction.
Problem: In an oil pump temmperature is stable throughout. With combustion there will be uneven heating and expansion that is bound to cause problems.
I have done the first part of this myself and gave up when residual lubrication became inadequate and the spinning pump seized. I can still imagine it dancing on the garage floor while I tried to stop the bleeding from my finger. There are some sharp places on those castings!

Happy holidays!
Pancholin

 

[StarRotorMan]

I am the president of StarRotor Corporation and have read with interest the comments in this forum. Because it is important that engineering opinions be based upon facts, I would like to clear up some issues.

Issue 1: The engine cannot work because the pressure at the compressor outlet and expander inlet are the same.

The StarRotor engine is a Brayton cycle engine. In all Brayton cycle engines, the compressor outlet pressure is nearly identical to the expander inlet. In Otto cycle engines, the heat input causes the pressure to rise because the gas is in a confined volume. In Brayton cycle engines, the gas is not confined to a fixed volume, so the heat input causes the gas volume to increase. Net work is produced because more work is produced in the expander than is required by the compressor.

Issue 2: Blow-by

Blow-by is our biggest challenge. We are addressing it using a variety of tricks such as conformal coatings, labrythian seals, and high speeds. Our testing protocol is designed to verify that we can reduce leakage to acceptable amounts. It should be noted that jet engines have blow-by past the blades, but their performance is acceptable. Because our rotor components do not touch, we eliminate friction and wear, but replace it with some blow-by. Our calculations indicate that blow-by losses should be minor, but we have not proven that yet.

Issue 3: Lubrication

We lubricate our engine in regions that are not subjected to high temperatures.

Issue 4: Low expansion ratios

With a recuperated Brayton cycle, high expansion ratios are not necessary to obtain high efficiencies.

Issue 5: Expander is larger than compressor

As described in Issue 1, the Brayton cycle relies on the expansion of gas volume. The expander is about 3 times larger than the compressor.

Issue 6: Efficiency at part throttle

Axial flow turbines have poor performance when throttled. They are well know for being efficient only over a narrow speed range. This is because the blades are designed for a particular speed and gas density... deviations from the design condition are not efficient. This is a characteristic of all such "dynamic" devices. In contrast, the StarRotor engine is a positive-displacement device. It will operate efficiently over a much wider range of speeds. Because it is inefficient, we do not plan to throttle the engine to vary power output. Instead, we will vary the compression ratio of the engine, which is much more efficient because it eliminates the irreversibility of the throttle.

Issue 7: Combustion products

It is true that combustion products could potentially coat the expander and heat exchanger. To eliminate this problem, we plan to use a tubular combustor. With over 20 years of experiments by Stuart Churchill at the University of Pennsylvania, he has never detected unburned hydrocarbons exiting his tubular combustor. All the fuel must pass through the flame front, so it is completely combusted.

Issue 8: Claimed efficiencies of 80%

We have never claimed engine efficiencies of 80%. We do claim that the compressor efficiency could be 80%. Solar Turbines claims that some of their compressors exceed 85% efficiency, so we do not think our claim is unwarranted.


Issue 9: Unreasonable efficiencies

We have done extensive modeling of our recuperated Brayton cycle engine. Numerous engineers have checked our calculations, and have invested in our company. Our high efficiency claims result because we keep the compressor nearly isothermal by spraying atomized liquid water into the compressor, keeping it cool and reducing the amount of parasitic shaft work it consumes. Water spray is a well-known trick with Brayton cycle engines. It is rarely used with axial blades because the liquid water can pit the blades due to high-speed impact. Our speeds are much lower, so we think we can tolerate liquid water in our system.

Issue 10: Inertia

Likely for automotive applications, we will hybridize our engine with an electric motor. This offers a number of benefits. For example, the automobile will start immediately so the driver does not have to wait for the combustor to heat up. Also, the engine can be very small so it will have a low inertia. For truck applications, rapid start and low inertia are not as important.

Issue 11: Water consumption

At full power, the engine uses 2 gallons of water for every gallon of fuel. However, automobiles are rarely operated at full power. Cruising down the highway takes only about 15 hp. At this power output, we need a compression ratio of only about 1.5 (not 6). At this low compression ratio, the compressor temperature does not increase significantly so water spray is not needed. Likely, water spray will be used on stationary applications (e.g., distributed power), which are running at full compression. Obviously, it will be easy to supply water to a stationary application to get the extra efficiency.

Issue 12: Heat exchanger cost

Likely, in early production, the heat exchanger will be expensive because it is will be produced in low production volumes. However, when mass produced, there is no reason to believe they will be prohibitively expensive...they are made of folded sheet metal. High exhaust temperatures will require ceramic heat exhangers, which will be more expensive. These would be used only where high efficiency is really important.

Issue 13: Gas turbines are not efficient

It is true that gas turbines are not very efficient, particulary when no recuperator is employed. However, with a recuperator and water spray into the compressor, the cycle approaches the Ericsson cycle, which has the same efficiency as the Carnot cycle (the highest nature allows for a heat engine).

Issue 14: High temperature of expander

We plan to make the expander of ceramic so it can tolerate high temperatures. Because our speeds are much lower than axial turbines, ceramics can be used in our application without flying apart due to centrifugal forces. Minimal cooling will be required because ceramics can withstand high temperatures.

Issue 15: Check valves

The StarRotor engine requires no check valves. When the compressor and expander are sized properly, there is no need for them.

Issue 16: Uneven heating

Unlike the Wankel engine, which does have uneven heating and resulting distortion, the StarRotor rotors see an average temperature of the inlet and outlet. They are constantly spinning which averages the temperature variations.

Final Point:

We are still in development and do not claim to have all the problems solved. However, we have thought through the StarRotor engine very carefully and believe that we will be able to solve every problem that confronts us along the way. We have an outstanding engineering team that has solved every problem so far. We expect to continue that tradition.

 

[mackerm]

I have some questions about StarRotor too.

Why doesn't

http://www.starrotor.com

work? I have to go to

http://www.starrotor.com/indexflash.htm

Why do you need seperate rotors for compression and expansion? Couldn't you use one side of a gerotor for compression, and the other for expansion? (You'd probably need more "teeth" though.)

Gerotors have two features which make them particularly suited to the Brayton cycle: they compress internally, and they don't need valves. But there are other devices with these features: screw compressors, scroll compressors, and Wankel rotary engines. Screw compressors have a complicated shape and could be challenging to manufacture, but do gerotors have an advantage over scroll compressors or Wankel rotary engines adapted to the Brayton cycle?

How about using a StarRotor as a turbocharger?

Mike Ackerman

 

[SBBlue]

StarRotorMan:

Thanks for your answer!!

One of the questions I do have is about your heat exchangers. While your actual Georotor engine may be compact, it looks to me that a heat exchanger (recuperator) that will handle the designed amount of heat will be reasonably large.

I haven't triple checked my calculations, but several months ago it appeared to me that a shell and tube stainless steel heat exchanger for the 150 hp engine you describe in your web site would require 4600 feet of tubing with an 0.6 inch outer diameter. If the tubes were five feet long you would need 920 such tubes.

If you were able to achieve and extremely dense packing of such tubes -- with the area between the tubes equal to the area of the tubes -- you would need to have a "shell" a little over two feet in diameter. I will grant that you could improve on the tube and shell design with baffles and such.

I haven't calculated out the weight of the stainless steel tubes and heat exchanger, but it must be non-trivial. And you would have to make the tubes of stainless steel, because of the temperatures you are working at. And stainless steel is not the easiest stuff to work with and weld.

Do you have any projections on the weight, size, and cost of such a heat exchanger?

 

[StarRotorMan]

Thanks for your questions. Here are the answers:

Question 1: Problems with StarRotor.com

Every computer that I have ever used responds to StarRotor.com and sends me to our homepage. Who can explain why computers do what they do? Certainly not this chemical engineer.

Question 2: Combined compressor/expander

Some of our designs do combine a compressor and expander on a single shaft. However, it is not possible to combine the compressor/expander functions into a single gerotor. The temperatures and volumes of the compressor and expander are completely different.

Question 3: Other geometries

We have considered screw compressors, but the geometry is very complex. It must be accurate in 3 dimensions whereas the StarRotor geometry needs to be accurate in only 2 dimensions. Our geometry should be much less expensive. Both scroll and Wankel geometries actually have an orbiting motion, which puts huge loads on bearings.

Question 4: Turbocharger

We have considered applying our technology to both turbochargers and superchargers. We have already started a design effort for a supercharger.

Question 5: Heat exchangers

A shell-and-tube heat exchanger is NOT the way to go with a recuperator. A compact heat exchanger is the way to go (see Kays & London, Compact Heat Exchangers). We estimate that a compact heat exchanger for a 100-kW engine would measure 0.43 meters on a side assuming a 50 K temperature difference. We do not have a cost estimate yet. It would be made from folded sheet metal, so it should not be too heavy.

 

[tbuelna]

StarRotorMan:

Your engine concept suffers from the same problem that most rotary engine concepts suffer from: lack of an efficient, durable gas seal. Until you solve that issue, your engine's efficiency will never exceed that of a gas turbine. Think about it- a gas turbine would be a very efficient engine if it could achieve high compression/expansion ratios while maintaining good compression efficiency. Why can't it do that? Simple, leakage!

Good luck and keep up the hard work. Who knows, maybe you'll solve the problem and produce a world-beater!

Terry

 

[StarRotorMan]

Terry,

There is no question that seals are our major challenge; however, we feel that we can meet that challenge.

Screw compressors have an even more complex geometry than the StarRotor geometry, yet they are able to achieve very high single-stage compression ratios (6:1 or better). Further, they are very efficient. We will borrow some sealing techniques from the screw compressor folks, plus develop some of our own.

Seals have nothing to do with the inability of gas turbines to achieve high single-stage compression ratios. Their problem is that they use "dynamic" compressors and expanders. Because gas densities are low, it is difficult to impart a significant momentum to the gas, so it cannot develop significant pressure in a single stage. According to the Solar Turbine website, they have some large compressors that are >85% efficient. If such efficient compressors (and expanders) were used in a recuperated Brayton cycle, the engine efficiency would be about 50%.

 

[GraviMan]

Nice design StarRotorMan - I've been wondering for years about this. I have a few suggestions to improve the design, but will post when I have time.

Need any help, analysis etc?

Mart

 

[StarRotorMan]

Graviman,

Thanks for your kind words.

We are always open to suggestions and ideas, so please let me know when you get a chance.

StarRotorMan

 

[mackerm]

StarRotorMan,

It is possible to construct a scroll compressor without an orbiting motion as shown in the second animation on this page:

http://user.ecc.u-tokyo.ac.jp/~tmorisi/morisita.html

This isn't saying that one geometry is necessarily better than the other, but I suspect the same could be done with Wankel rotary engines-- putting the housing on a separate offset shaft. And the sealing problem has already been solved with Wankels.

By the way, I thought of a better reason to have separate rotors for compression and expansion (as shown in your illustrations): so you con't cool the expanding air.

 

[StarRotorMan]

Mackerm,

Thanks for the website address for the cool animation of a dual rotating scroll compressor. It does eliminate the problem of slinging a mass in an orbiting manner, but it introduces another problem: getting high-pressure gas out of the scroll. This can only be done through the rotating shafts, which adds complexity. Further, it is not easy to alter the compression ratio on the fly, which is possible with the StarRotor engine.

The orbiting motion of the Wankel geometry can be eliminated by rotating both the inner and outer rotors. If the Wankel geometry is used in a Brayton cycle engine, it must function as both a compressor and an expander. Wankel compressors do not completely compress the gas when the tip sweeps past the discharge port; therefore, they require a check valve. This is fine for a compressor, but not the expander.

Thanks,

StarRotorMan

 

[GraviMan]

Been thinking about this for a couple of days now. It is a very good concept, and I have no doubt that the thermodynamics work out on paper. I've read some papers on the Rover gas turbine work that went on. Interesting that they opted for rotating ceramic disk heat-ex, although copper or stainless tubes should also be ok.

My main concern is that the design is wholly dependant on water injection. As somebody who usually goes about a week before I get around to filling up my windscreen washer tank, I can see this as a possible major source of premature engine failure! The other concern is the need to make available extra package space, and weight protection for between 50 to 100kg of water in automotive applications. I work with 100 ton off-road trucks, so the additional ton of water would wipe out any attempt by me to save weight! I can see real acceptance problems on this point - even though the thermodynamics holds true.

Must admit I had reached the same conclusion about helical compressors. Even worse, in my case, since I was considering a design based on the Eaton supercharger. I saw a similar concept for gas turbines in the Duetches Museum in Munich. I preferred a non heat-ex Brayton cycle, requiring high pressures to keep exhaust temp sensible. The spiral got tighter towards the HP end, to keep the pressure across each seal approx constant. I had considered a sliding outer case to cope with variable demand, although in my case PR was to be constant.

In the end I decided the pressure requirements and sealing requirements would give compressor efficiency of about 60% at best. This leant more towards a turbo 2-stroke solution for hybrids, so I let the idea drop. You patented yours, and well done - I wish you every success in it's implementation.

You might like, as a contingency, to consider a non water injected version. Since this will push exhaust temp up, a heat-ex becomes less practical, so high PR is necessary. The end result would be a staged engine - ie LP, IP and HP. A turbo could be LP, with this unit certainly doing the IP. There is nothing to stop you ganging up these units on a single shaft, since naturally spiral castings are hard to tolerence. There will still be the advantage of seperating inlet and exhaust temperatures - effectively it will be your unit running in a higher pressure environment!

I am curious about the bearing design. Shaft alignment is critical, but wouldn't an outer bearing for the outer gerator be more compact. This will be vital if the compressors and expanders are ganged...

Have you considered fuel yet? I'm guessing that it will be Diesel or Kerosine. Cooling flows will be very important - both water and air. I would use the engines simplicity and lightness as it's selling feature, rather than efficiency. It would fit very well into the aero rotary market, or even smaller GT for helicopters!

Finally I wish you every success. For my own part I am quietly persuing an interest in my own fuel cell design, which I feel may well address most of the current problems. I may well be wrong, but will have fun anyway! If you wish, at some point, I am on MartinGarrish@AOL.com.

I look forward to seeing the prototype...

Mart

 

[StarRotorMan]

Mart,

Thank you for your kind words and wishes.

Below, I have responded to your points.

Point 1: Water injection

Water injection improves the engine efficiency by maintaining near-isothermal conditions in the compressor, which lowers the parasitic shaft work that must be invested in the compressor. In the case of an automobile, most of the time, a low compression (1.5:1) will suffice. At low compression ratios, the compressor outlet does not get very hot anyway, so the benefits of water injection are minimal. Only occassionally do you need the full power output of the engine, so full compression (6:1) is infrequent. Full compression would be made more efficient with the water spray. You could leave water spray out entirely and say it is not worth the hassle. If you do use the water spray, then the tank can be pretty small because it is not used much.

In the case of engines the run at full power most of the time (e.g., electricity production), then water spray should be considered. For stationary applications, it's not a hassle at all.

A locomotive could pull a water car (they used to do that for steam engines). On a percentage basis, the water car would not add much weight to the train but it would save a significant amount of fuel.

So, the bottom line is that we have considered both dry and wet compression and will select which ever is most appropriate to the situation.

Point 2: Bearings/shaft alignment

At the moment, we are using roller bearings. They can handle the high side loads and are reasonably efficient. For large engines, we might use journal bearings.

Some of our designs have low side loads, so ball bearings could be used.

Shaft alignment is an important issue. In some of our designs, the precision alignment of the rotors occurs through the shaft. In other designs, the precision goes through the housing. As a general rule, it is easier to make a precise shaft than a precise housing.

Point 3: Fuel

At the moment, we are focusing on the compressor and expander design. We have not invested much effort into burners and fuels. We are forming some relationships with partners that have expertise in that area, so we may benefit from their knowledge. In any case, Brayton cycle engines are known to use a wide range of fuels. The tubular combustor we hope to use has been tested with a wide range of fuels, even including powdered coal!


Thanks again for your nice comments. We look forward to seeing the prototype as well. As any engineer knows, turning an idea into reality is a challenge, but there is a great deal of satisfaction when the hardware works properly. We are looking forward to that day!

StarRotorMan

 

[SBBlue]

Okay, I'm puzzled here. . . .

If you're going to run the system with a compression ratio of only 1.5:1, and assuming a compressor and expander efficiency of 80% (I think that's a generous assumption), a heat exchanger efficiency of 90% (again, a generous assumption for a heat exchanger that can handle such high temperatures) and a combustor temperature of 2002 deg K (3144 deg F)---

Your best efficiency will be 33%. That's considerably less than what you can get from a state of the art diesel -- which will weigh in at about 40-45%, and 50% for a state of the art stationary diesel.

And these numbers (for the starrotor) assume a completely adiabatic engine (won't happen) with zero friction loss (also won't happen).

Do you really think there will be a market for an engine that is this inefficient -- at least compared to a diesel?