Tow Tank, Sluice, Flow Box Ideas?

[zstine]

I'm designing a 26ft hydrofoil boat. I have built a 1:4 scale model to test the flight control system. I need a 5mph water flow, 1.5 ft deep to test the model in. Initially, I was going to use a stream behind my house. It is about 10ft wide, but I measured the flow rate where it is deep enough and it is just 1ft per second, not even 1 mph. While I was anticipating the need to dam and build a sluice to increase flow, I cannot get over 5 times the speed.
So, I am wondering if anyone has an easy way to make a tow system, or other method of getting a 5mph flow through a 1.5' x 2' cross section (~9,900 gal/min).

Thanks for your thoughts.

[tkeithlu]

My experience is with a wind tunnel, not a water tunnel,but strangely enough wind and
water act in similar ways. You need not just the speed, but also laminar flow to get any sort of decent readings. You could build a seat for yourself off the side of a boat capable of 5 mph. Face backward and run your model on a tether and spring scale. The boat will need to be small enough and the body of still water big enough that you are not sitting over a bow wave.

Or, you can build a tank. Many dollars. It's a doughnut, with one side squared and long enough to get rid of most confusion in the flow. The 90 degree corners upstream and downstream from your test area have 45 degree vertical boards in the flow to straighten out the water. The small side has an outboard motor. The water goes around in circles.

[wefleenor]

You are talking about a complicated subject where the results are still subject to interpretation. Technically the model requires geometric, dynamic, and kinematic similatiry with the prototype. If you go into the theory you will find that is not possible to accomplish in a hydrodynamic model. So we always settle for geometric similarity and hope it will scale up. Often 2 or more scale models of different scales are used to determine how it will react. This is why computer simulations have taken over for most scale model work.

That said, the suggestion of pulling the model beside a travelling platform is the best solution. Use a type of cross-pole used to train water skiers. Good luck.

[DefinitelyMe]

I presume you have some boat other than the model? :-)

Can you stick it in gear at a dock, chuck your model over the stern and use the prop wash? Or is it too turbulent?

[vpbarkley]

This sounds flip but find a faster stream you can use.
Or you might have to wait till after it rains.
You can tow a small model up a stream with a fishing rod to increase relative speed.
If you have a boat or dinghy you can tow it from that.

[gbowen]

Are the foils submerged or surface piercing? You might do well to just calculate performance from a copy of Sig Hoerner's book "Fluid Dynamic Drag". Theoretically, the only model that meets similarity is full size. You could do a lot of work and get misleading results.

[clakiep]

Applying fluid dynamics You have to build a smaller modell to simulate higher speeds in the same flow, establish the Reynold Number and be sure to have laminar undisturbed flow for mesurements.

Capt. Claus - ocean tramp of the eighties

[gbowen]

If you have surface piercing foils you also need to satisfy Froude Number. Because of waves and spray. It is true to satisfy Reynolds Number the model speed must be higher. No problem with a wind tunnel, just increase pressure to deal with kinematic viscosity. The problem with a sluice is the flow will develop a velocity profile and be rotational it will not be undisturbed water where theoretical hydrodynamics predicts lift and drag quite well. In addition you have wall effect. The book by Sig Hoerner was sponsored by the US Navy to show their regard for his theoretical and experimental research. The book "Fluid Dynamic Lift" was finished by the US Navy after Hoerner's death. If the foils are submerged test them in a tunnel with wind.

I have a good friend in NZ who wanted to find the water resistance of a sailing skow. He knocked himself out trying to tow it around. Little models at high speed in a boat wake are a waste of time. Hoerner had data on barges and other full size boats that were much more worthwhile.

[marcjsmith]

could you build a recirculating tow tank and power it with an outboard engine or some pool pumps to get your flow.

[stormalong]

Can you find a university with a test tank where you could buy some test time? Probably less expensive than building your own.

[Dsanduril]

Originally Posted by stormalong

Can you find a university with a test tank where you could buy some test time? Probably less expensive than building your own.

This was my thought as well. Back when I was in school as a civil engineer we had a hydraulics lab and loved playing around with boats in the test flumes. Look around for the local student chapter of ASCE and find out if they are involved in the national concrete canoe competition. Nearly every team that is in any way serious tests boat models in their hydraulics labs.

[johnn33]

Two possible links
https://www.google.com/search?client...l+borne+models
and
https://www.ayrs.org
my math skills are of no value
John

[zstine]

Thanks for all the ideas. This is a non-surface piercing T-foil setup, hence the Arduino based flight control software I am trying to test. The model is GeoSim. 1:4 length, 1:64 weight, Speed scales at 1:2 for 1:4 scale model, etc. Reynolds number is dependent on chord and will therefor be reduced by 1:8 factor I believe... So Yes, that's an issue for scaling that I'm aware of, Its not perfect. I am not concerned much with drag. My main concern is verifying function of flight control software/hardware before I scale that up with bigger motors, etc.



My design takeoff speed is 8knts, so I need 4knts to 5knts of tow speed min. I am familiar with Hoerner, Gibbs and Cox works cerca 1950-1960's as well as more modern texts. I have modeled this in MIT's AVL flight software to help optimize foil shape and obtain discrete stability analysis points (eg take-off, level flight at XY speed, banked turns, etc). But building a computer analysis model that will consider the dynamic response of the flight control software combined with dynamic boat response and output damped stability is way more complex and likely more erroneous than a scale model. Plus I don't own such software like Matlab/simulink. The scale model cost is a ~$150 and well worth the learning experience before I shell out $15,000 to build the 26ft boat.



I do own a 10ft inflatable dink with 9.8hp motor. I think the best tow method is to use a 'ski trainer' like bracket to get the model away and in clean water and do a tow test with my dinghy... just not sure it is capable. It's Not a very stable platform and it may be jumping up on plane right around the speeds I'm concerned with. So I may need to find a different motorboat... But it seems like that is the best way to go. I do have access to a big river, but logistically the boat will be easier I think.
thanks