Thrust on marine steam turbine casing

Question

I attended a lecture today given by the steam turbine design expert Geoff Horseman. He discussed the turbine installations on ships such as the RMS Mauretania. Among other things, he remarked that an axial pressure gradient inside the turbine casing contributed around 100 tonnes of forward thrust to the vessel. He described himself as having been surprised by this discovery, but as having verified with a colleague that it was valid and known to the original designers.

I'll probably be able to come back to this with more information at a later date, but right now my intuitive understanding is as follows:

A solid object immersed in a gas which has a linear pressure gradient will experience a force in the direction of decreasing pressure. Similarly, the walls of a rigid cavity which contains a gas with a linear pressure gradient will experience a force in the direction of increasing pressure. This latter force is the one of interest here. Simply by arranging the direction of steam flow in the turbine to be toward the stern, a favourable high pressure to low pressure gradient is established (through the natural operation of the turbine). The "forward wall" of the turbine casing experiences a higher outward pressure than the "aft wall".

I assume this is really just a case of the conservation of momentum in disguise—and like Horseman I find it to be pretty counterintuitive. Is this a widely known phenomenon?

I am aware of unexpected sources of thrust in jet engines, such as the large contribution of the variable geometry inlet of Concorde's engines to the total thrust exerted on the airframe. This feels like it might be a similar kind of thing.

EDIT October 30 2024: The slides the talk I attended (given by Geoff Horseman) have been made available: https://www.birrtheatre.com/_files/ugd/b4096e_b0283af6ae0b4af0bd2a0632d97641b7.pdf

In the relevant slide (actually about Titanic's LP turbine) he states:

"The ship was was propelled by the stationary turbine casing! There was a net force on the casing of ~100 tons due to steam and atmospheric pressures which pushed the ship forward. The thrust from the propeller was restricted to the limit of the thrust bearing. The remainder of the propulsive force came from the casing."

I believe that the accepted answer addresses all of this satisfactorily.

Answer 1

Parsons knew about the axial thrust issue in his turbine designs, and how difficult it was to design shaft support bearings that could take up that load while the turbine was running. It was for that reason that he designed turbines in tandem (two mirror-image turbines on one shaft) to cancel the thrust loads within the turbine. Steam flowed into the center of the disc array, flowing forward through one set of discs and aft through another on the same shaft. Result: zero thrust load on the output shaft. I do not know if the Mauritania had tandem-wheel turbines in it.

Answer 2

The rotors needn't experience thrust. Just arrange things such that the axial momentum doesn't change through the rotor disc. Everything else is fair game, so plenty of design space to accommodate whatever needs doing - expansion, redirection, etc.

However, from a system point of view, the propulsion system was a closed fluid system, so there could be no net thrust due to the fluid itself. If the turbine base had a net thrust, the condenser or boiler must have an equal and opposite thrust. The only way to get thrust would be for the rotor torque to somehow be causing a thrust reaction. But this doesn't pan out either. This violates the superposition principle of angular momentum and linear momentum. The rotor can only produce a balanced force couple - it can not interact in any way that produces a net force in any direction.

So let's look at this from a control volume standpoint. You have a ship with four props. The shafts have thrust bearings to absorb all the propeller thrust. The shafts inboard of the thrust bearings are in pure torsion. Draw your control volume around the ship cutting the shafts at the thrust bearings. The thrust bearings are pushing against your control volume boundary. Let's say the shafts are counter-rotating and carry equal torque, so the net torque being reacted by the ship is zero. The net thrust from the shafts inside the ship is also zero. There are no mass fluxes with axial momentum cutting the control volume, just a bit of combustion air up and down the stacks. The only things entering the control volume are the four torsion shafts; and from first principles they can't convert torque into thrust.

Have you ever heard of a transmission attached to a conventional driveshaft that manages to push a truck on it's own mounts? Of course not.

I guess the shortest answer is that heat or work flux into a closed system (wrt mass) doesn't affect the system's two momentum conservation laws. Which is why they are conservation laws.

Hopefully, the paragraphs above convince you that "I assume this is really just a case of the conservation of momentum in disguise" isn't just counterintuitive, It's impossible.

So what does that leave us with? On Mauri the propshaft thrust bearings were mounted on the aft side of the turbine housings. So the propeller thrust was transferred to the turbine housing, and the turbine housing pushed the ship. Now we have an opportunity to use rotor thrust to lessen the load on the bearings. We can't change the load on the housing, but we can change the load on the bearing. If the rotors can be designed to produce thrust that counters the propeller thrust, the bearings see less load. Of course this means that the stators have to see a load equal and opposite to the rotor, and they transfer the load to the housing and the ship. It's a way to unload the thrust bearing by coupling the thrust through the fluid.

https://earlofcruise.blogspot.com/2016/09/starting-rms-mauretania-from-cold.html