Interceptors effect on hull pressure

Hello,

I wanted to see if anyone has knowledge on the extent of hull bottom pressure increase due to a transom interceptor? From illustrations I've seen most of the pressure occurs around the turbulent region just in front of the interceptor, but it is also shown to extend quite far further forward.

My question then would be if it is possible to improve cavitation margins specifically of a water jet by using a transom interceptor? Maybe a bit general because inlet duct lengths vary between manufactures so much.

 

No.

The two are unrelated.

 

Sorry Ad Hoc, could you expand on that. Which two items are not related or is it that the quantities are insignificant allow effect?

 

I was at a machine shop last week that had built some large interceptor for a navy ship and they told me that the navy was very happy not only with the hull lift they provided but also because they improve the intake performance of the water jets. The fact that they help the water jets was apparently an unexpected but significant bonus. Sorry I can't get more specific than that.

 

Intrepid, I have heard similar claims and others associated with a concentration of pressure around the stern. That is why I would like to get a better idea of the pressure distribution.
I have also heard that stability and control on a planning hull can be nicely improved by trimming slightly to stern and applying some interceptor to correct without loss of speed.
Just trying to get an idea of how far forward a pressure cell generated by a marginal interceptor could stretch.

 

I have to contradict my friend (and often a mentor ) Ad Hoc here, because a Gurney-flap-like device of the form and the size of an interceptor does modify the upstream pressure field, increasing the average pressure by some 20-30%. That increase of pressure can, at least in theory, enhance the cavitation characteristics of an inlet placed upstream of the interceptor.
MechaNik, I'd suggest you to contact user HJS, who has done a pretty extensive work on his mid-ship interceptor. He might have some relevant literature to share.
Cheers

 

I second that (sorry Mr. Ad Hoc).

 

D and CWTeebs,

No need to feel sheepish. If you think I am talking bollocks, please say so. That’s what the forum is about good exchange of knowledge and learning. If I’m wrong I have no problem with that, so long as the evidence indicates so, rather than opine or supposition Well, it is sometimes, not so on other areas of the forum!

So, let’s address the question, that of cavitation.

Cavitation inside an inlet of a waterjet is defined in several different ways by different designers/manufactures, viz:

Cavitation number: The difference between the upstream pressure and the inlet pressure cavity /density * upstream velocity

Or,
I prefer the simple one by KaMeWa (Rolls Royce) P(atm) – P(vap) + Head (at inlet)/Produced head

There are several others.

In each case the important factors that govern the cavitation relates to the pressure inside the duct and the head of water.

If we look at the interceptor and what it does.

It does as D pointed out modify the up field pressure stream. But, it is not over a large region nor linearly increasing. It has a peak at the plate and decreases rapidly in a non-linear manner. As shown here in experiments by MDI:



What does an interceptor do? It makes a hull work at a more “optimal” condition. If you conduct a series of proper tank tests, what you do as part of the simple Resistance testing is to do an LCG chase. What this means is that you run the tests at various LCG positions, keeping everything else the same. What you find, depending upon hull shape, is that with shorter fatter hulls, the curve produced is more “U” shaped yet the thinner slender hull is more flatter. Typ. LCG chase curves are shown here:



Most LCGs are often in the wrong location, so how do you move the LCG to be in the optimum for the hull without adding weight or changing the design layout? You move the LCG..ahh..but that isn’t easy..or is it?? A trim tab a fixed wedge or an interceptor does just this, it artificially moves the LCG to a more optimum location for that hull at that speed. A dynamic or virtual LCG if you like.

If the hull had been tank tested and an LCG chase performed and the curves known and the final design with the LCG in the optimum location, from test results, the results would be the same.

Before the days of interceptors we tested this with actual 40m catamarans, moving the LCG to verify the theory. Again corroborated by the MDI research showing that there is a point where interceptors no longer produce any gain:



This just shows once you're past that "optimum" no more gain. Which the LCG chase clearly shows too.

So an interceptor is either correcting a poor hull design or helping the design achieve what the layout was unable to do from the outset; which yielded a less optimum LCG.

So, back to the waterjet.

The pressure changes created by the interceptor do not reach a significant distance fwd of the plate. However if we assume, for now (I have yet to see any documented research to verify any of this) that the pressure distribution created by the interceptor extends to the waterjet inlet region, what happens?

If you have a higher pressure region that wishes to extend into the streamline flow, you have a high pressure region crossing a lower pressure region. Typ. diagram of jet and flow shown here:



Aside from the complexities of this and its “actual” effect at the boundary of the two regions, the resulting change in velocity profile will be a region of turbulence, or unsteady flow. Many waterjet manufactures tests have shown that the influence of turbulence intensity actually has a negligible effect on the pressure distribution.

Which if we look at the various definitions of how cavitation is calculated for the waterjet duct, is thus unaffected. The inlet design and nozzle output design has far greater impact.

Thus, first you need to understand what an interceptor does and then quantify the effect and the amount of pressure distribution and magnitude and distance of the effect and then if this extends into the free stream for the waterjet inlet. And then after all this, if there is any change. To date, I have not read any research that supports the notion that cavitation of waterjet is effected by the interceptors pressure field.

None of the major waterjet suppliers I know of suggest otherwise in their literature either, even though they offer interceptor plates of their own.

BUT, please correct me if I am wrong.

 

Thank you Ad Hoc, that can go in my useful library.

I have a document which came from a prominent supplier, showing forces for load calculations during installation. It uses 1/3 of the maximum pressure at interceptor x width x 3000mm length. This formula could however contain significant FOS and over simplification, it was the length that got me thinking though.

I have had experiences where performance has improved beyond what the correction of the LCG was meant to provide. Also improvements when replacing trim tabs with interceptors. This is not to say the interceptor has improved cavitation, more likely a speed increase due to lower hull resistance will have greater effect on water jet inlet pressure.
I am working with a 34m high speed hull at the moment, perhaps I can convince them to trial fixed plates of same area at different locations and see if there is any change.
To note that this vessel and others I have worked with operated close to zone 3 on the kamewa curve at hump speeds. A shame I know.

 

Principally, there is no effect on the inlet of the waterjet, provided the inlet shape is correct for the operating condition. However, most wj inlets have fixed geometries, beeing geometricaly correct only for one operating point, mainly defined by the velocity ratio V(in)/V(advance). At all other conditions there will be a mismatch between flow lines and geometry, leading to increasing losses.

That said, it is thus possible to find a slight improvement by a change of the pressure gradient, particularly in the "lip" region. In this case the interceptor has to be installed close to the lip in order to make a difference. But in this position the boundary layer tickness is ~zero (the forward BL has been ingested into the inlet), so here the interceptor is no longer a BL device, but a simple transverse "brake", that changes the hydrodynamic shape of the lip configuration.

This then, is more a question of correcting a defect, or adapting to a non-optimum operating point, rather than a real improvement. Lip cavitation is not a big issue; it occurs on a static element, it can be fairly easily controlled by displacing it to the outside of the lip. I lip cavitaion were a problem, then the cavitation after an interceptor would be a bigger issue. The real problem is boundary layer detachment from the "rooftop ramp", which is depending on the boundary layer condition in front of the ramp, where there is no influence from a transom interceptor at all.

So, the answer is still: No, the two are basically unrelated.

 

What speeds are we talking about here? Are they relevant for lip cavitation at all? The improvement over ordinary trim tabs come from the fact that the L/D ratio of a correct interceptor is higher than the corresponding ratio for ordinary trim tabs working with an angle of attack.

Conditions at hump speed are quite different from cruising or sprint speed operation. Unless the inlet has been optimized for the high speed only, the cavitation problem at hump is impeller inlet tip cavitation. Again, if the throat area and lip angle is designed for much higher speed than the hump, then the geometry may restrict the flow at high velocity ratios [V(inlet)/V(advance)]. In this case the flow is approaching the lip region with a considerable vertical velocity composant, leading to a steep angle of attack at the lip.

The limiting conditions can be easily determined in situ, by performing a bollard test, including the measurement of rpms, inlet pressure and bollard pull. This gives a cavitation performance factor, the specific cavitation speed, for the impeller. With this available, the cavitation conditions for the rotor may be checked at all operating conditions. Surprisingly often one finds that the jet is operating close to the impeller cavitation limit over a wide speed range, from hump, up to full speed. Unless there is a performance penalty incuded, the result of this cavitation manifests itself as material damage.

 

I was hoping you would pipe up ...since if anyone would know, you would for sure, being your bailiwick

 

Not sheepish, just very respectful of your knowldege. And your reply has clearly shown why. Feels a bit like shooting a couple of bullets with a handgun, thinking that the target has been hit, and then all of a sudden being showered with an artillery barrage fire in return.

Your reply was very informative and appreciated, and has made me think over it a bit more. In effect there's a fundamental flaw in the equaling the interceptor to a hydrodynamic version of a gurney flap, which is what has led me to my previous conclusions. Now I see that they are imho not so similar, because their working conditions are substantially different:
- a gurney is intended to work at a constant AoA (trim), giving more lift for the same speed.
- an interceptor's purpose is to change the trim of a vessel, in order to give less drag for the same speed, or more speed for the same power.
These are two very different situations.

In my previous reply I have reasoned in terms of constant trim, but the correct approach to the analysis of an interceptor is to consider the hull free to trim, until the required (constant and equal to vessel's weight) vertical force is attained.

With that in mind, I've performed (out of my own curiosity) a small analysis of an airfoil with a 1% gurney flap, just to get an approximate indication about what happens to the pressure field around an airfoil with and without a Gurney flap. I have analized an airfoil because the analysis is simpler and quicker than that of a hull on a water surface. Then I've used a rather thin airfoil as an example (6% thickness), so that the qualitative results of the analysis could be applied (with some imagination and with due caution) to a bottom of a flat plate planing on the water surface. It is a qualitative and approximate analysis and, above all, an analysis done with a numerical method provided by a software based on Drela's X-FOIL. This has to be underlined. Until supported by some empirical validation, the resulting numbers are to be considered just an indication of what happens in reality.

See this graph:

- blue line is a clean foil, green line is the same foil with a 1% gurney flap at the trailing edge.

If we reason in terms of constant lift (say, Cl = 0.3 ), then we see that a clean foil will attain that value of Cl at an AoA=2.6° , while a foil with a gurney flap will attain the same lift coefficient at AoA=1°. If it was a hull, it would be the same as saying that the interceptor has trimmed the vessel down by 1.6 degrees.

The pressure coefficient is here defined as
Cp = (P - Pinf )/(0.5 rho V^2)
where Pinf is the absolute free stream pressure far ahead of the foil. In case of a submerged foil, or a hull, it would be given by Pinf = Patm + rho g z, where z is the water depth.

The plots of Cp over the two foils look like this:

1) Clean foil at AoA=2.6° :


2) Foil with a 1% gurney flap (interceptor) at AoA=1°:


To make it look similar to a pressure distribution below a flat plate atop of a water surface, consider just the bottom curves in each graph (the pressure side, positive Cp values). You can see that just the last 20% of the foil (hull?) length is significantly affected by the pressure disturbance created by the interceptor - the rest of the pressure curve is nearly identical in two cases.

So yes, now I realize that when the vessel is free to trim (as it in reality is), then the extent of the area interested by the pressure disturbance is indeed limited. In this particular case (1% interceptor), just the last 20% of the pressure side of the foil feels the effect. A note of caution: this is a 2D numerical analysis, not verified nor directly translatable into a 3D hull of a surface vessel. A 3D wing or hull would almost certainly feel a somewhat smaller influence of the interceptors due to the fact that the flow has an added dimension to move into, avoid the obstacle and hence to alleviate the pressure buildup.

Now, knowing that a significant change of Cp is probably limited to just an area around the last quarter or fifth of length from the transom, the question imho becomes: what is the absolute value of the pressure induced by the interceptor in this area? If it was a foil running in the sea water, this analysis indicates that, under the conditions seen above, the pressure coefficient on the ventral side jumps from nearly zero for a clean airfoil, to around 0.04 in the region of the foil around X/Xmax=0.75 (taken as a sample area).

But Cp is related to the pressure via speed squared. A Cp of 0.04 at 20 kts gives 2 kPa of pressure rise. At 30 kts it's around 5 kPa, and at 40 kts it's 8.5 kPa.

If we could place a waterjet inlet in that area of the hul, it would mean (if these values are at least nearly in the ballpark) that an addition of an interceptor similar to this 1% gurney flap would act like submerging the inlet to an additional 0.5 meters of depth at 30 kts, or 0.8-0.9 meters at 40 kts, with respect to a clean-transom case.

I do not know if that would change something in terms of cavitation or other index of waterjet performance, I can only intuitively suspect that it might, as mentioned in the OP post by Mechanik. I even have no data to claim that these numerical values (obtained mathematically for an airfoil) can be used for a case of a hull, but still they provide a mind-teasing indication that something could change, at least at high speeds. I would love to see it investigated more in depth by someone with more resources than I have.

Passing the ball to you guys. I'm pretty sure you'll find tens of flaws to this line of thinking and it's applicability to a vessel's hull, let's discuss them.

Cheers.

 

Interceptor lift and drag

With this simple formula, developed by Sverre Steen, NTNU; can the added lift and drag due to interceptor be calculated. The problem with the position of the lifting force remains.

JS

 

THe Russians are the only ones to my knowledge who have published decent data on interceptors. Attached is translation of a Russian paper which gives good data on the pressure distribution ahead of an interceptor. It also gives a method for calculation of the lift and drag forces on interceptors.