Q: You say there are narrow gaps between the rotors. How narrow must these gaps be to keep the leakage neglectable?

A: The relative loss of gas through the gaps depends on the rotational speed of the RPA: the faster the cavities move from one end of the RPA to the other, the more gas they pump, but the less time the gas has to get lost through the gaps. The percentage of gas that gets lost decreases quadratically with the RPA speed. Doubling the rotational speed will reduce the losses by a factor of four. For any given RPA with a certain gap width there exists a certain minimum speed, above which the losses can be neglected.

In practice, one would manufacture the air gaps as narrow as possible and then find out at which speed the RPA performs best. I have no actual figures available. But I am sure that the RPA will perform well far below the speed of a conventional gas turbine, which is highly desirable.

I think that for an RPA with 100 mm diameter rotors a gap width of 0.5 mm would be a good choice.

Q: What is the estimated cost of your RPA solarthermal power generator? Is there anything else similar from an engineering standpoint working in the market place - so that it is a proven concept? Please confirm that it will easily adapt to ... our solar dish collector system - and how? Is your system scalable - meaning if we need a smaller or larger dish - can you produce a smaller or larger RPA?

A: It is hard to give a cost estimate before having built a prototype. I can only guess that it may cost more or less the same as a Diesel engine of the same power.

However, I am sure that the RPA technology is scalable, because all displacement machines are - like the Diesel engines - in contrast to gas turbines, whose efficiency is dominated by aerodynamic losses at the small scale.

Though, the RPA heat engine is, from a thermodynamic view, a gas turbine, and thus a proven concept. Just the compression mechanism is different. It's behaviour can be predicted from the same Brayton/Joule cycle T-s-diagram, with an expected conversion efficiency of 40% to 60%, perhaps more. Compression by displacement works even at low gas speeds where the aerodynamic losses are small, but which is prerequisite for heating the working gas in a solar collector.

I am sure that the RPA heat engine can be adapted to an existing solar dish collector system. It will need its own absorber box, which heats up the RPA working fluid directly, i.e. atmospheric air at a pressure of 10 to 25 bar, and preferrably at a higher temperature.

One cannot have a new technology without taking a certain risk, but scalability allows trying the RPA concept at a smaller and cheaper scale. Once having successfully tried the small prototype, the extrapolation of costs should be easy.

Q: Has the RPA heat engine ever been funded by ... before? Why has no one else either already developed the RPA or is currently pursuing it?

A: I doubt that the RPA has ever been funded by ..., though I am not absolutely sure. The only references I have are a few old patent numbers (e.g. DE19738132, FR1199521, US2410341). But if, then the RPA was certainly not intended as a solar-driven slow gas turbine as I propose.

The latter may also be one of the reasons - I can only speculate - why the RPA has not yet been tried: the mere use as a simple compressor was probably not interesting enough. Other reasons are various technical problems of earlier designs, most of all the large number of expensive gears that were needed. But these problems are now solved by my inventions. Finally, the communicating of the RPA concept to investors has probably been too difficult in earlier days without the computer animations.

Q: Your graphics are great - did you create those or a 3rd party?

A: I did.

Q: Why do you think is the RPA heat engine more efficient than a Stirling engine e.g. in solarthermal power generation?

A: The efficiency of a heat engine depends on the temperature difference, the compression ratio, the shape of the thermodynamic cycle, but also on friction and other losses - the RPA heat engine can be in all aspects better than the Stirling.

The Stirling engine has a limited temperature range because it is a reciprocating piston engine. The piston inevitably has sliding contact with the cylinder wall, with a graphite sealing between. Abrasion becomes a problem at high temperatures. The lubrication with oil is here not possible. Besides, the Stirling has two special components, the regenerator and the cooler, which both "consume" a bit of the available temperature difference for themselves, and thus effectively diminish it.

In contrast, the RPA rotors do not touch one another. There is no abrasion at no temperature. The RPA can be operated at a much higher temperature, and it can use the full temperature difference to the ambient air. The cooling or thermal insulation of bearings and other heat sensitive parts is easy.

The Stirling engine has a considerable clearance volume, mostly inside the regenerator and cooler and all the tubes, so that it cannot achieve a high compression ratio, which is typically something like 1.5 to 3.5. The RPA heat engine, on the other hand, can be expected to have a compression ratio like that of a gas turbine, which is among the highest of all heat engines, typically 10 to 25.

Below are pictures of my latest solarthermal heat engine, which is specialized for very high temperatures. All parts in contact with hot compressed air are now covered with heat resistant ceramic plates. The rotors are made of ceramics and have cooling channels inside. The bearings are also air-cooled. The "trumpet" draws the stream of cooling air through the bearings and channels. The rotating valves have been avoided, because they cannot be made fireproof.

Q: It seems the machining process for many of the components needs to be very precise. Has an actual working prototype been manufactured? If so, how many hours are on the machine?

A: No, I have not yet built a prototype. But I have prepared a very detailed parametric CAD model with all the necessary parts and all the gaps and plays between them. I checked that the parts fit well together and do not intersect, and for every part I can propose a method to manufacture it with common equipment. A well-equipped company could immediately start with building the prototype. Only the model parameters need to be determined for each specific application.

The synchronization rings can quickly be produced on a lathe and a few other tools with micrometer precision.

Surprisingly, the rotors do not need to be machined with an exceptional high precision. Between the rotors must exist narrow gaps to avoid wear and friction, even when the rotors bend under pressure. A machining precision of, say, 10 percent of the gap width should be enough. For a rotor diameter of 100 mm I would suggest a gap width of 0.5 mm, and thus a precision of 0.05 mm. Most present-day manufacturing methods can do much better.

I would avoid milling the rotors. For a temperature range of -40 to +180 degrees Celsius I would mold the rotors of fibre-reinforced epoxy resin. This is sufficient for quite a lot of applications including pneumatic motors, compressed air energy storage, waste heat recovery, power generation from air-sea temperature difference.

For higher temperatures I would compose the rotors of metal or ceramic plates along their axles. The plates can quickly be cut out from a raw plate with a laser beam, with the proper angle at the rims, and together with air cooling channels and the quadratic axle hole. A slightly specialized laser cutter is needed that can cut at a computer-controlled angle.

Q: One question that comes to mind is reliability and maintenance. There are many moving parts. Quite a few bearings. Has a reliability analysis and corresponding cost analysis been done?

A: There are many moving parts, yes. But compared with a reciprocating piston engine, or even a turbine with all the blades, the RPA parts are simple in shape and highly repetitive, and they are relatively small, even the rotors. They can be produced with only a moderate number of relatively small tools by a relatively small company.

There are quite a few bearings, yes, which increases the probability that one of the bearings may fail. But on the other hand, the forces that act on the bearings behave nicely, without sudden jerks or pulls. And the bearings run relatively slow because the RPA pumps quite a lot of gas per turn.

The RPA will probably achieve a much higher power density at moderate speeds than any other principle I know. For some applications - such as the electric power generation in hybrid cars - the number of parts or bearings will not matter if only the machine is light and compact enough.

Q: A Brayton/Joule cycle heat engine can be constructed with all kinds of compressors. Why do you think is the RPA the best choice of all?

What I propose is a "slow" gas tubine: one that produces a high torque at a low rotational speed, and especially with a slow flow of the working gas. At least with respect to a normal gas turbine. Slowness is important in the proposed applications, because it allows the gas stream to be heated in a solar absorber or heat exchanger. Slowness avoids that the generated power is output as thrust, and it avoids most of the aerodynamic losses that a gas turbine normally has. This all is achieved by replacing the turbines with displacement compressors.

Among the displacement compressors, the RPA has the following advantages.

Unlike reciprocating piston machines and quite a few others, the RPA has no sliding contact between moving parts, which makes it applicable at very high temperatures, without being constrained by the properties of heat resistant seals or lubricants, and without wear and friction.

Unlike many other rotary compressors, such as the screw compressor, the RPA is much easier to produce. The RPA has only one type of rotors of a simple geometrical shape – with only circular arcs in the cross section - and no housing, whereas the screw compressor has two different rotors of a very complex shape in a precise housing. Moreover, the RPA rotors rotate in the same direction and at the same speed, which allows for a design without gears. In contrast, the rotors of the screw compressor and quite a few others rotate in opposite directions.

Last but not least, the RPA seems to achieve a much higher power density than all other compressor types, a bigger volume of working chambers for a given total machine size, which does not only save space, but also material, weight and costs.