Shrouded submarine propulsors - conflicting claims?

[totoro]

I tried looking into the whole issue of pump jets; shrouded propellers; shrouded propulsors and open skewback props for submarines. What I expected to find is not a lot of data. But what I did not expect to find is conflicting claims.

First of all, there seems to be a pretty blurred line between shrouded propulsors and pump jets. There are various claims around using both names for what essentially looks like the same thing. Basically a converging shroud of considerable diameter (rivaling the diameter of open props of previous subs) which has several stator blades and which then usually has around 10 or so rotor blades. They're basically almost touching the shroud and they're of constant chord, with a fairly gentle curve.

Now, most of the western subs seem to have that design. And perhaps even the only eastern sub (borei) may have something similar. Though, that one youtuber/ex submariner Jive did claim in one of his videos that borei uses a different design, where the shroud is smaller as the blades are shorter in the vertical axis but much longer in the horizontal axis. basically forming long vents. now.... i am not sure that actual photos of the borei shroud support that claim. Does anyone know more about it?
And if i am not mistaken, IF that's true, that'd be the only such example of a pumpjet on subs - as western subs, at least as per open source drawings, seem to use propulsors that are closer to regular short blade pumpjets.

What's also confusing is the fact there are papers and articles claiming that pumpjets are usually becoming more efficient (power wise) than open props somewhere in the double digit knot speeds. Anywhere from teens to 30 knots and above.

But if that was true - then why would a SSBN like Borei, which likely wants to be optimized for slow patrols - use a pumpjet?

I can understand pumpjet usage on SSNs like on US and British SSNs if it's true. Basically, slow speed power efficiency is sacrificed for medium/high speed efficiency, while possibly retaining noise advantage at any speed. Given that SSNs would usually operate at higher speeds than SSBNs on the average mission.

But then there are also articles which claim basically the opposite. And say pumpjets are actually noiser, at high speeds than at lower speeds, compared to open props. Which... I guess even makes some sense if we look at borei SSBN and severodvinsk SSN. SSBN would want to be quiet while going slow. While SSN might want to trade off some of that low speed quietness for more quietness at high speeds - since it's going to be using those high speeds more often.

To sum it up - is there an actual scientific paper or article of half decent credibility that talks about the issues of pumpjets/open props for various speeds of submarines? Or are we destined to use dubious google result articles written by unknowns?

 

[Foo Fighter]

Interesting and thought provoking post, than you.

 

[Scott Kenny]

To sum it up - is there an actual scientific paper or article of half decent credibility that talks about the issues of pumpjets/open props for various speeds of submarines? Or are we destined to use dubious google result articles written by unknowns?

I haven't seen any scientific papers, no.

Doesn't help that most places don't care about how noisy a given propulsion method is, only how efficient it is (which may or may not impact noise levels).

As to Jive's analysis of the Borei's propulsor, the Ohio class torpedo ejection pump is basically an air turbine driven long chord propulsor. They're quiet.

 

[TMA1]

Yeah I have to get any of this answered. So far thr yt experts I have similarly listened to have not satisfactorily answered these questions. One that sticks out as you said is the yasen class subs with their skewback propellers and yet yasen is described as very quiet even by western standards. Interestingly it sounds like the yasen is actually a lot more expensive to make than the borei subs. I have always heard it commented off hand that pumpjets are superior. They indeed must have major advantages. That said I have heard and read the same sources and am equally as dumbfounded by it.

 

[Scott Kenny]

Yeah I have to get any of this answered. So far thr yt experts I have similarly listened to have not satisfactorily answered these questions. One that sticks out as you said is the yasen class subs with their skewback propellers and yet yasen is described as very quiet even by western standards. Interestingly it sounds like the yasen is actually a lot more expensive to make than the borei subs. I have always heard it commented off hand that pumpjets are superior. They indeed must have major advantages. That said I have heard and read the same sources and am equally as dumbfounded by it.

What I remember hearing is that propulsors delay the onset of cavitation because the inside of the duct/shroud is at somewhat higher pressures than outside. (Had some old SSN21 hands on the Kentucky)

This means propulsors would have a faster "quiet" speed than the same ship with a skewback prop.

So if you can make your "hole in the water" speed say, 10 knots instead of 5, you can get to the Deterrent Patrol Area/Bastion in half the time with little risk of detection. I believe that's the idea behind both the Borei and Columbia classes getting propulsors instead of screws.

There may also be some shenanigans with anechoic materials inside the duct, so that even if you are doing enough turns to cavitate, that noise doesn't get out of the propulsor. But that's a guess. It's certainly something I would have experimented with to see if it worked.

As to the Yasen class, as I understand it the design is very heavily rafted. Internal decks float on sound isolation mounts, major machinery floats on sound mounts connected to the decks connected to the hull. It would also certainly be possible to mess around with active vibration cancellation on the machinery rafts. The EH101 has 4 big active vibration cancellation actuators connecting the transmission and rotor mast to the airframe, for example. Plus, most of your machinery should have well known vibration frequencies so you don't even need to have an analysis loop in the AVC system, pre-program the frequencies and just have the 1/2 phase detector/delay that needs to "think".

 

[Josh_TN]

My understanding is that pump-jet and shrouded propulsor are the same thing, though different designs might have different blade geometries and numbers. As Scott Kenny stated above, my understanding is that a pumpjet is quieter at relatively medium speeds - it increases the cavitation speed at a specific depth compared to an open propeller. The trade off is a reduction in mechanical efficiency, which reduces top speed and increases needed power for a given speed across the entire power curve.

Since D/E boats tend to operate at 5 knts and are far more power limited, you tend to see propulsors only on nuclear boats. A single Kilo class boat was modified as a test bed to use one, B-871. The Attack class was meant to use one as well but was cancelled.

 

[Yellow Palace]

Broadly speaking, you can configure the blade and duct to give higher efficiency, or reduced cavitation, but not both at the same time. I'm not familiar with the subject, but any given application will need to make a choice about what outcome is wanted, and at what speed.

Submarine applications will usually prefer the 'low cavitation' option, I should think. Other applications might prefer the former option - either to get more thrust at the same speed, or more speed from the same power.

There's not really that much difference, conceptually, between a pumpjet, shrouded propulsor, and a Kort nozzle - you're just playing with the various parameters to get a desired outcome.

 

[Scott Kenny]

Broadly speaking, you can configure the blade and duct to give higher efficiency, or reduced cavitation, but not both at the same time. I'm not familiar with the subject, but any given application will need to make a choice about what outcome is wanted, and at what speed.

The old Seawolf hands said that the SSN21 propulsor was more efficient and had reduced cavitation. You could probably crank the efficiency up even higher if you worried less about cavitation, but when you're talking (example) a 70simething% efficient screw and replacing it with a 90something% efficient propulsor, that huge efficiency gain is plenty.

 

[Josh_TN]

The old Seawolf hands said that the SSN21 propulsor was more efficient and had reduced cavitation. You could probably crank the efficiency up even higher if you worried less about cavitation, but when you're talking (example) a 70simething% efficient screw and replacing it with a 90something% efficient propulsor, that huge efficiency gain is plenty.

I have a hard time believing a propulsor could be both more efficient and reduce cavitation. I also have a hard time believing a nuclear boat with 30-50 megawatts of power would give a shit either way. But I never sailed for Sam, so I will leave that to professionals.

 

[Scott Kenny]

I have a hard time believing a propulsor could be both more efficient and reduce cavitation. I also have a hard time believing a nuclear boat with 30-50 megawatts of power would give a shit either way. But I never sailed for Sam, so I will leave that to professionals.

I've heard that most unducted props are on the order of 70something% efficient at turning torque into thrust. Putting a duct around that improves the number, lengthening the kort nozzle into a propulsor improves the efficiency more.

How much more is what gets into the tradeoffs between efficiency and cavitation reduction.

 

[Josh_TN]

I've heard that most unducted props are on the order of 70something% efficient at turning torque into thrust. Putting a duct around that improves the number, lengthening the kort nozzle into a propulsor improves the efficiency more.

How much more is what gets into the tradeoffs between efficiency and cavitation reduction.

Fair enough, I could see ducting perhaps increasing efficiency if that is what it is designed for, but not if it is designed at the same time to reduce cavitation. I do not see how those two things are not in opposition to each other.

 

[Scott Kenny]

Fair enough, I could see ducting perhaps increasing efficiency if that is what it is designed for, but not if it is designed at the same time to reduce cavitation. I do not see how those two things are not in opposition to each other.

Decreasing cavitation means shaping to increase the pressure on the front side of the screw blades.

How much that messes with any increases in efficiency due to the ducting is well above my physics paygrade. All I can tell you is what I was told about it, that the Seawolf propulsor did both increase efficiency and reduced cavitation and other radiated noise.

I can also say that the propulsor allowed the Seawolves to accelerate a lot faster than the 688s or Virginias, allowing them to outrun a Mk48 ADCAP unless you managed to get stupidly close to them. Exact distances classified, mind, so please don't ask what "stupidly close" means.

 

[Yellow Palace]

Decreasing cavitation means shaping to increase the pressure on the front side of the screw blades.

How much that messes with any increases in efficiency due to the ducting is well above my physics paygrade. All I can tell you is what I was told about it, that the Seawolf propulsor did both increase efficiency and reduced cavitation and other radiated noise.

Because thrust is generated by the pressure differential between the front and back of the impeller blades, anything that increases that pressure will reduce the potential thrust generation, but decrease cavitation. That's an accelerating duct. A decelerating duct decreases the pressure at the front of the impeller, allowing for more thrust, but also more cavitation.

A submarine pumpjet isn't just a propeller in a duct, though. The presence of the duct lets you optimise it differently. Efficiency also isn't a constant for propulsors, and ducts give you more scope to optimise that. I'd guess that the SEAWOLF installation is differently optimised than the LOS ANGELES system. That probably means that at some key point in the performance envelope, it does have both less cavitation and increased efficiency, though at peak performance one will have been compromised for the other.

Variable geometry blades (e.g. controllable pitch propellers) and ducts would give you more speed flexibility, but would cost some peak efficiency, add complexity, and add potential noise sources. I don't think anyone's ever fitted a CPP to a submarine, and I'm not sure anyone's ever even studied a variable geometry duct. I imagine it would look a bit like the variable nozzle on a jet engine.

 

[Scott Kenny]

Because thrust is generated by the pressure differential between the front and back of the impeller blades, anything that increases that pressure will reduce the potential thrust generation, but decrease cavitation. That's an accelerating duct. A decelerating duct decreases the pressure at the front of the impeller, allowing for more thrust, but also more cavitation.

Right. But one of the major causes of propeller inefficiencies is the blade tip vortices, and the ducting either greatly reduces their size or eliminates them entirely. So for subs you take the efficiency gains from getting rid of the tip vortices and then optimize the rest of the duct for noise reduction.