DIY tunnel drive

Discussion in 'Surface Drives' started by CDK, Nov 29, 2007.

1. Joined: Jun 2009
Posts: 1,132
Likes: 121, Points: 63, Legacy Rep: 1165
Location: Sweden

baeckmoHydrodynamics

CDK,
Did you use the acoustic method for calibrating your tacho? Still feel a bit concerned about those figures, so until verified by other means, discussion on propeller details is not meaningful. Depending on how the Fourier transform is used, there may be a filter bandwidth that is too wide for correct analysis. Remember that propeller power at a fixed operating point is proportional to rpm^3. A distortion +/- 5 % of frequency thus corresponds to a bandwidth of (1.05/0.95)^3 = 1.35 in power.

Meanwhile, let’s see what we can learn from your data from the Mercs. Generally, for a preliminary idea on IO thrust, you may use the producer’s engine power with a total overall efficiency of abt 0.55 (hydrodynamic losses from lower housing included) to find a rough figure at full throttle (provided engine and propeller matching is ok). With the setup you started from, you had 28 knots (14.4 m/s) from 2 * 135 hp at 4600 rpm. This is reasonable with a weight to power ratio of ~14 kg/hp and single prop outdrives. Each side is then producing a net thrust of 3795 N, (135*736*0.55/14.4). So we have one point on the resistance-versus-speed line. We also have a propeller load factor (Bp = n*P^0.5/Va^2.5) of ~6.6, which is comfortably low to give good efficiency.

In addition, you gave a figure of 22 knots @ 3500 rpm. With the rpm exponent of 2.65 this would give a power (each side) of roughly 65 hp (about what you have in your engines right now), resulting in a thrust of 2340 N each side, giving a second point on the R/V line. Now, these two poins may be used as reference points when the available hull data are analyzed in a hull resistance program; for instance fiddling with weight and longitudinal C of mass until a reasonable fit is found. A quick check on your hull reveals a marked hump resistance, a real ”camel” in the R/V diagram. This is fairly easy to overcome with the trim facility and the fair thrust of the 15” props that are recommended for this speed range. If nothing in the hull shape, weight or weight distribution were changed, those 20 to 22 knots would have been possible with the output of your VW engines.

But with the added drag from your steep tunnel inlets, the hump resistance becomes even higher and the propeller performance is seriously impaired. Using smaller props does not make it easier; the load factor is now into ”tug territory”! The working condition at surface level also means that you need more blade surface to avoid cavitation, even if you were successful in avoiding ventilation (which you are not….).

You can go two ways to find remedy; either you go Frosty’s way with surface piercers or you rebuild the tunnel arrangement (I assume refitting the Alpha’s is not an option). Even if I personally very much like the spp’s, and have good experience from them, I have to advice against them in this application because of the technical risk with the hump resistance of your hull. I would be happy to have Rudi’s opinion on this though, as I know he has Levi drives operating in lowspeed hump range! As a contrary, the steadily rising resistance (no hump) of a slender catamaran makes it ideal for spp’s, but even so, I think Frosty reported a slight sluggishness in midrange (pls correct me if my memory is wrong Frosty…..).

So, if I were in your shoes, I would redo the tunnels, including propshaft outer bearing. Attached as thumbnails (if the Ghost of zeroes and net bugs is friendly to me today) are sketches showing the basic proportions of a tunnel compromise that allows you a retrofit inside the existing ones. The inlet part is extremely important, the ramp (”rooftop”) must connect to the bottom via a soft radius (R>= 2* Dprop) and its inclination to be not more than 15 degrees. (Yes Ad Hoc, waterjet inlets often see 22 to 25 degrees, but that is with a lower surface to guide the flow into the inlet, and they still generate serious velocity gradients). Note the flat ”roof” which effectively increases the aspect ratio of the bend, making for lower losses! The tunnel sides reach down to bottom level, with an outside ”wing portion” that stops aerated water to be sucked around the tunnel side into the propeller inflow. The transverse radius between this surface and tunnel allows smooth transverse flow into the tunnel when turning and during high load operation, when increased flow is called for.

The shroud aft of propeller disc section must be long enough and at the correct angle to prevent air from backflowing into the propeller. I have seen numbers of tunnels with this part parallell to keel and too short; it must be adapted to the outflow streamlines. It’s probably wise to design for a prop dia of 15 in; this will be better for the thrust required. Even so, the Bp will be pretty high, but tolerable. You should cut away the part of your shaft tube reaching into the tunnel; it is a major disturbance and causes an early boundary-layer detachment. Let the shaft run free and use a rubber cutlass bearing, carried in a single strut. It is easy to overlook the rubber bearing, taking it for a trivial chunk of gum, but it is a cleverly designed piece that performs three main tasks. First, wet rubber is about as slippery as it gets, so it’s a good bearing material. Second, rubber has excellent spring/damping qualities; it is very effective in reducing shaft vibrations. Third, in metal hulls it is acting as a barrier to galvanic currents, improving corrosion resistance.

A single ”one legged” strut will work inside the wake from the shaft, in fact often straightening it to some extent. From a fluid point of view this is to be preferred instead of the (mechanically stiffer) V-strut arrangement you have used. In that arrangement you get three wakes in front of your prop. It takes really advanced flow studies to get the V-type in line with the 3D flow, and yours are certainly not adjusted to the local flow!! Just make sure the strut is stiff enough (DNV rules refer to the load carrying capacity of the cutlass bearing for the bending strength) to avoid contact propeller-shroud!

Good Luck!!

Attached Files:

File size:
97.8 KB
Views:
512
• CDKtunnel2.zip
File size:
67.6 KB
Views:
395
1 person likes this.
2. Joined: Oct 2008
Posts: 6,159
Likes: 397, Points: 83, Legacy Rep: 2488
Location: Japan

beackmo

nicely put.
Those sketches look "familiar"...you don't happen to work for "that" company do you??

3. FrostyPrevious Member

Too many things wrong here with this design. You cant work this out with a pencil.

How can you work out or even suggest thrust when you don't know if you have cavitation or part. Hes already suggested 30% slip--

Oh well,-- im a doer not a talk abouter. Next year when your still pontification and coming up with theories the poor guy will be still boatless.

Enjoy the theory!

4. Joined: Jun 2009
Posts: 1,132
Likes: 121, Points: 63, Legacy Rep: 1165
Location: Sweden

baeckmoHydrodynamics

Hah.. Ad Hoc, guess what you're aiming at, but nope! Been designing AND building (I still hold an EU license for Aluwelding) special work boats, fifi/resque boats, maritime surveyors et c. for over 35 years now. Also "doubling" as a non-affiliated/third party consultant in trouble shooting and technical disputes on pumping and turbine systems, including w-jets.

And Frosty, maybe I dug too deep into CDK's problem, but I felt sorry for the guy having spent so much effort along a dead-end alley. As I have a long experience, both practical and theoretical, from the designing and building of hulls with propeller tunnels I imagined that my input could lead him forwards. For some reason that combination of skills seems to have pissed you but I can live with that. I asked you a polite question and get your pukes back, why?

5. Joined: Sep 2004
Posts: 211
Likes: 9, Points: 28, Legacy Rep: 201
Location: Norway

PetterMSenior Member

...who would that company be then?

6. apex1Guest

Petter, is it really so important for us to know if Baeckmo got his skills with KaMeWa, Ulstein, ABB or any of the many others in the field of high tech propulsion???
His advice is sound and valid imho, and thats what CDK needs.

7. Joined: Jun 2009
Posts: 1,132
Likes: 121, Points: 63, Legacy Rep: 1165
Location: Sweden

baeckmoHydrodynamics

Ouch Petter, didn't know I had to submit CV and customer references to participate here.....

8. Joined: Aug 2007
Posts: 3,324
Likes: 145, Points: 63, Legacy Rep: 1819

CDKretired engineer

That is quite a lot of negative waves, Frosty.
But it isn't all that bad. First of all the boat is in the water, moored 25 meters from my garden gate. Lots of people would kill for that.
We make regular trips to nearby bays and the lighthouse island 6 miles out. The boat has a kitchen, stainless steel fridge, toilet and an enormous (albeit triangular) bed. It takes us anywhere at 6 knots and digests so little fuel that I cannot remember when I last filled the tank. The fuel gauge says 50%, which means approx. 150 liters of red diesel, enough till November when the season ends.
My wife's happy, the dog is happy, and I... Well, I could be happier if the boat reached my 22 knots target, but for now I can live with it. I've had boats for 45 years, from modest 15 ft. sports boats to a 60 knots Glastron Carlson, even the smallest one needed more fuel than this one.

And now about the 30% slip. I asked if that was bad, but nobody said Yes, that is very bad!

This is my calculation; a bit clumsy perhaps because only metric values have meaning for me.

A 14x17" prop advances 43,18 cm for one revolution.
1200 engine rpm is 600 prop rpm is 36000 rph, times 43,18 cm makes 15,55 km/h or 8,4 knots. If the boat were to be propelled by rack and pinion, that would be its speed. But it does only 6 knots, which is 71% of 8,4 so the slip is 29%. Is my line of reasoning correct?

Should the engines have a bit more zest and could reach 3600 rpm, the slip should be near 13% to obtain 22 knots. I may have been a bit optimistic when I started this project.

9. FrostyPrevious Member

propeller slip while still on the hump is irrelevant. Prop slip becomes an issue when at cruising speed or when the boat is up and running.

However at this point prop slip is an issue or clue as the boat isnt working properly.

My post was aimed at the mathematicians that were enjoying the mathematics of the problems pontificating with each other as you were in need of some answers.

But if you are happy with it and can use it, then I am less concerned

10. Joined: Oct 2008
Posts: 6,159
Likes: 397, Points: 83, Legacy Rep: 2488
Location: Japan

beackmo

"...Ouch Petter, didn't know I had to submit CV and customer references to participate here....."

Sadly those that feel inferior and those that have no formal qualifications with prerequisite training and those that think "we" must be part of their "lets sit around a camp fire and pretend", do!...all very sad.

I get hit with all sorts of criticism because I'm not part of the "smiley happy people" group...jesssssss....very sad and childish. And they wonder why they're not real professional engineers!

I'm sure CDK appreciates the reply, which is all that matters, even if others do not, as it is way above their pay grade to comprehend....

If one doesn't understand the mechanisms (ie theory and scientific mathematics which many seem to loath and hate, for obvious reasons really), one cannot arrive at a proper solution, it just becomes pure guess work and/or iterative trail and error. Wastes time, time = money!

11. Joined: Aug 2007
Posts: 3,324
Likes: 145, Points: 63, Legacy Rep: 1819

CDKretired engineer

Baeckmo, thank you for spending so much time on my project. I will need some time to digest your various recommendations.

The rpm issue bothers me a bit also.
I downloaded the software bit could not find a mike with a plug that fits my wife's laptop. Instead I used an oscilloscope with the mike and looked for a recurring waveform. From the length in mS (75) I calculated the idle rpm to be a little over 800. Then I looked at the alternator signal, which was approx. 10 mS and modified the tacho circuit for this much shorter than expected length, connected it to the alternator and set the meter to 0.8 on the dial.
To make sure I will try to glue a magnet on the crankshaft pulley and use a pickup coil I didn't know I had. The pulley is the only rotating part I have access to and even that is not as simple as it sounds.

The V-struts I made because with the original blunt tunnel entrance there was a lot of vibration. More rpm and more speed, but I was afraid the welds between stern tube and flange would not hold.
I left the struts in place just because I made them, but with the current tunnel shape they are not really necessary.

Between the alpha-ones and the tunnels I installed Berkeley jets, using the Mercuiser engines. Very poor steering, immense fuel consumption and devastating corrosion made me decide to remove these and sell them on Ebay. But I noted that the few US companies who sell this junk have all kinds of intake scoops for sale to guide more water to the pump. Do you think it is worth the experiment to install something similar under the tunnel entrances?

12. FrostyPrevious Member

Hmm not quite sure who you are talking to but now you've got that off your chest can you tell us all whats wrong with CDK's boat and what he should do or do you have any other things to say to mere mortals that dare talk with you.

Mathematics can tell you how to build and close on the results of a new build, even where the water line will be, but you cant work out a mistake.

You can work out what he should have and how it should be and what the result should be but,--you cant tell him what to do from here with a pencil.

Or can you?

13. Joined: Sep 2004
Posts: 211
Likes: 9, Points: 28, Legacy Rep: 201
Location: Norway

PetterMSenior Member

Hehe, sorry, that was not what I meant, I was only curious who the company was, kind of recognize the tunnel design. Just wondering if I was right.
Good that you have taken the time to offer well thought through advise.

14. Joined: Jun 2009
Posts: 1,132
Likes: 121, Points: 63, Legacy Rep: 1165
Location: Sweden

baeckmoHydrodynamics

Watch me, and you will be surprized what may be achieved with the help of daddy's ol' pencil, provided one knows which end to point to the paper....

15. Joined: Jun 2009
Posts: 1,132
Likes: 121, Points: 63, Legacy Rep: 1165
Location: Sweden

baeckmoHydrodynamics

And CDK, your project is a stimulating challenge; it contains ingredients from various disciplines. I take it that you have some useful know how about electronic devices to share, that will come handy sooner or later so don't worry about my time so far.

I am sketching on a possible working schedule for the alterations (they are not very complicated, compared to what you have done so far), so please have some patience before you reach for the big cutter!

Additional scoops would only add to the already high drag, so nono! The medicine for you is still a correct inlet "rooftop"! As for the bearing struts, I agree that the double strut is preferred in terms of mechanical strength, but it should be oriented with one leg hiding in the shaft wake and number two more or less at right angle to the first. In a cavitating environment, some positive prerotation (=same as prop) of the incoming flow will improve the cavitation performance of the propeller. Thus it is not trivial which side the second leg is mounted. It must be fitted to the side where the blade is moving downwards, and also be given some camber. I'll show you how.

Forum posts represent the experience, opinion, and view of individual users. Boat Design Net does not necessarily endorse nor share the view of each individual post.
When making potentially dangerous or financial decisions, always employ and consult appropriate professionals. Your circumstances or experience may be different.