Difference between revisions of "Low Cost Wave Tank"

From Seasteading
Jump to: navigation, search
(Added Requirements)
(Data Recording Section)
 
(46 intermediate revisions by 4 users not shown)
Line 1: Line 1:
Wave tank time costs money. The Berkeley Wave Tank (cerca 1930's), which is basically a long narrow tank designed for pull testing of ship hulls costs about $1500/day.  A more modern wave tank costs $10,000/day.  At these expense rates, the question arises "can we build our own wave tank, get useful results and still achieve significant cost savings?"  This article explores several low cost wave tank alternatives that are geared toward testing seastead desgins.
+
Please use the discussion forum [http://seasteading.org/interact/forums/engineering/structure-designs/low-cost-wave-tank Low Cost Wave Tank Forum], to discuss this wiki page.
  
To date, four basic low cost concepts have emerged:
+
Wave tank time costs money.  The Berkeley Wave Tank (circa 1930's), which is basically a long narrow tank designed for pull testing of ship hulls costs about $1500/day.  A more modern wave tank costs $10,000/day.  At these expense rates, the question arises "can we build our own wave tank, get useful results and still achieve significant cost savings?"  This article explores several low cost wave tank alternatives that are geared toward testing seastead designs.
 +
 
 +
To date, six basic low cost concepts have emerged:
 
* ''Large Aquarium Tank''. A large 125 gallon aquarium tank is long and narrow and might be suitable for some wave tank testing.
 
* ''Large Aquarium Tank''. A large 125 gallon aquarium tank is long and narrow and might be suitable for some wave tank testing.
 
* ''Swimming Pool''. A swimming pool of the right shape and depth might be adaptable to wave tank usage.
 
* ''Swimming Pool''. A swimming pool of the right shape and depth might be adaptable to wave tank usage.
 
* ''Dig a Ditch''. A ditch can be dug in the ground and lined with plastic.
 
* ''Dig a Ditch''. A ditch can be dug in the ground and lined with plastic.
 
* ''Water Trough''. An above ground trough is filled with water.
 
* ''Water Trough''. An above ground trough is filled with water.
These four designs are explored futher below.
+
* ''Natures waves''. Use natural waves in an existing body of water.
 +
* ''Speedboat Wake''. Have speedboat pass near model so wake wave hits model.
 +
These six designs are explored further below.
  
 
== Requirements ==
 
== Requirements ==
Line 23: Line 27:
 
=== Wave Generator ===
 
=== Wave Generator ===
  
A sophisticated wave generator is capable of producing just about and wave form imaginable.  For a low cost wave tank, being able to generate sinusoidal waves with a known period and amplitude is probably sufficient.  The wave generator is basically a paddle attached to the bottom of the tank on a hinge.  A motor attached to a crank can be used to generate waves.  The larger the crank radius, the higher the wave amplitude.  The wave period depends upon motor speed.
+
A sophisticated wave generator is capable of producing just about any wave form imaginable.  For a low cost wave tank, being able to generate sinusoidal waves with a known period and amplitude is probably sufficient.  The wave generator is basically a paddle attached to the bottom of the tank on a hinge.  A motor attached to a crank can be used to generate waves.  The larger the crank radius, the higher the wave amplitude.  The wave period depends upon motor speed.
  
Having said all of that, an alternative to a motor and crank is to simply have a human grab the paddle and generate waves as by hand.  The amplitude and period may not be perfect, but they may be good enough to get the job done.
+
Having said all of that, an alternative to a motor and crank is to simply have a human grab the paddle and generate waves as by hand or leg power.  The amplitude and period may not be perfect, but they may be good enough to get the job done.
  
 
=== Test Cell ===
 
=== Test Cell ===
  
The test cell is the location where the model is inserted.  This is where any visual observation window is placed.  Any test equipment is placed here as well.
+
The test cell is the location where the model is inserted.  This is where any visual observation window is placed.  Any test equipment is placed here as well. If using a small glass tank, fastening a plastic ruler to the side can help with estimating wave height passing a certain point. Best if it is [https://wiki.seasteading.org/index.php?title=Low_Cost_Wave_Tank#Data_Recording_Section captured on video.]
  
 
=== Wave Absorber ===
 
=== Wave Absorber ===
Line 39: Line 43:
 
The purpose of a wave tank is to measure the behavior a model is placed in the test cell.  While looking at the model behavior provides a subjective understanding, collecting quantitative data provides a more object understanding of what is going on.
 
The purpose of a wave tank is to measure the behavior a model is placed in the test cell.  While looking at the model behavior provides a subjective understanding, collecting quantitative data provides a more object understanding of what is going on.
  
All of the low cost designs under consideration use the same basic data collection system.  A low cost color video camera is set up to record a video stream.  Three LED's (red, green and blue) are attached to the top of the model under test and illuminated.  The video camera is arranged to have a split view via some mirrors.  The split views provide a stereoscopic view of the model under test.  The left and right views are both recorded in the camera frame so that there is no issue with frame synchronization.  Software digitizes each frame to identify the X/Y location of each LED in each view.  Stereoscopic math is used to identify the model position and orientation for each frame.  Using adjacent frames it is possible to compute velocities and accelerations.
+
One basic data collection system uses video.  A low cost color video camera is set up to record a video stream.  Three LED's (red, green and blue) are attached to the top of the model under test and illuminated.  The video camera is arranged to have a split view via some mirrors.  The split views provide a stereoscopic view of the model under test.  The left and right views are both recorded in the camera frame so that there is no issue with frame synchronization.  Software digitizes each frame to identify the X/Y location of each LED in each view.  Stereoscopic math is used to identify the model position and orientation for each frame.  Using adjacent frames it is possible to compute velocities and accelerations.
 +
 
 +
The idea of using a split vision system comes from some robotics vision work presented by [http://www.hbrobotics.org/Calendar_2005.htm Ingolf Sander] at the [http://www.hbrobotics.org/ Homebrew Robotics Club] on January 26, 2005.  Additional information can be found at [http://www.ingrit.com/ingolf/robotics/ Ingolf's Robotics Web Page].
 +
 
 +
Another possible method for collecting data is acceleration meters.  Today you can find a [http://www.microdaq.com/occ/pendant/pendant_g_data_logger.php low cost waterproof acceleration meter] that is [http://www.macobserver.com/review/2006/12/14.1.shtml easy to use].  It is battery powered so you can just put it on your model and take your measurements then transfer the data over USB to your computer.  Would be nice if a [http://www.amazon.com/G-Trender-Waterproof-Bluetooth-Barometer-Electronic/dp/B001C8J6I6/ref=sr_1_2?ie=UTF8&s=miscellaneous&qid=1220880777&sr=8-2 GPS logger] and acceleration logger could be in one unit.
 +
 
 +
Low-cost, high resolution wifi-enabled video cameras can be built using older smart phones for a fraction of the cost of a new device, and water-proof carrying cases can protect them.
 +
 
 +
If using an uncontrolled outdoor location on a river or a lake, overhead drone video may be a good option to add to your test.
  
 
== Aquarium Tank ==
 
== Aquarium Tank ==
  
{Discuss aquarium tank}
+
An aquarium tank can be pressed into duty for a wave tank.  [http://www.petco.com Petco] sells aquariums and one of them is a reasonable choice for a wave tank. The rectangular 125 gallon 72 inch × 18 inch × 22 inch, might be usable as a wave tank.  Petco sells the tank for $420 (as of September 2008.)  This tank weighs 1250 pounds when completely filled with water.  However, the wave tank will not fill the tank to capacity.
 +
 
 +
The clear sides make it very easy to see what is going on inside the tank.  The 72 inch length provides just enough space for a wave absorber and a wave generator.
  
 
== Swimming Pool ==
 
== Swimming Pool ==
  
{Discuss Swimming Pool here}
+
Many people own swimming pools.  While a swimming pool is not an optimal shape, they tend to be larger than an aquarium.
 +
 
 +
Most people do not want anything scraping up the inside of their swimming pool.  Thus, extra care is required to design a wave generator and wave absorber that can not scuff the pool surface.
 +
 
 +
In order to get straight wave propagation, it is necessary to have two sides to guide the waves down.  This can be with some long pieces of plywood that span the length of the pool (carefully built to avoid scratching anything.)  Some clear plastic can be placed near the test cell so that cameras can see what is happening to the model under water.  Either an underwater periscope or an under water camera can be used to view what is happening under the water.
 +
 
 +
You can get a [http://www.amazon.com/Filter-Ladder-Maintenance-Surface-Skimmer/dp/B00007E9B6/ref=pd_bbs_10?ie=UTF8&s=toys-and-games&qid=1235359869&sr=8-10 15 foot diameter pool with filter pump and all for $230].  At this price you don't need to worry about scuff marks.  A human can easily make waves  by pushing on the outside of this type of pool, so no expensive mechanism is needed. If you just make 1 or a few waves you can easily have a single wave hit your structure.  Testing in a "confused sea" is also good.  Can just have the camera on a tri-pod outside the pool.
 +
 
 +
<youtube v="TvJzpgKzXo0" />
  
 
== Dig a Ditch ==
 
== Dig a Ditch ==
  
{Discuss a ditch}
+
You can rent equipment for a day that will let you dig a good sized ditch in dirt for a reasonable amount of money.  You can also hire someone with a backhoe.  Hard and rocky ground could be difficult but good dirt is easy for equipment to dig quickly.  Digging the ditch so that one end slopes up to make the wave absorber is easy.  Lining this with something to make it waterproof is easy and cheap.  Since the ditch is down in the ground draining it when not in use would require a pump.  You could also leave it full and chlorinate just like a pool.  Would probably want to build up a bump of dirt around the ditch so that rain water did not wash other stuff into the ditch.  Do this before putting down the liner.  You could almost have part of your ditch tank above ground level.  However, you want straight sides on your wave tank so the waves propagate well, which is not going to work with the fresh dirt above ground level. 
 +
 +
One minor disadvantage of the ditch is that it is not easy to put a clear side on the ditch for viewing what is going on under water.  Instead of doing that, it is possible to use a water periscope to obtain a view of what is happening below the water surface.
  
 
== Water Trough ==
 
== Water Trough ==
  
{Discuss a water trough}
+
Using housing studs and plywood, it is possible to build a water trough of pretty much any reasonable dimension.  For example, 20 feet &times; 3 feet &times; 3 feet is quite possible.  The plywood does not have to be much thicker than 3/8 inch or 1/2 inch.  The studs would basically be assembled into 3 foot &times 3 foot squares with plywood lining three sides of the interior of each square.  One or two of the plywood sides could be replaced with either a clear piece of plastic or glass (whichever costs less.)  The wood would not be water proof, but an inexpensive clear plastic sheet purchased to line the trough and make it water proof.  The water trough is probably significantly cheaper and larger than the large aquarium solution.
 +
 
 +
== Nature's Waves ==
 +
 
 +
You can use natural waves in a lake, bay, or harbor to test a model.  This saves the cost of constructing and operating your own wave tank.  However, you don't get easy control over the wave environment.  You have to use what nature gives you.  You do have control over which location you use on any given day, and the time of day you do the tests.  Even the exact spot you choose within a given harbor or bay can give you some control.  This may require swimming or kayaking to get to the desired waves.  Swimming with a life-jacket and fins you can move a model a few hundred feet.  With a kayak you can go much further.  Video can be shot from a tripod on a pier or rocky shore.  Can also shoot using a waterproof camera from a kayak or while swimming in the water.  If the water is cold, you may need a wet suit.  Depending on where you live, and the bodies of water available nearby, you may have to do hours of driving to do some tests.  You may not be able to find things like 4 inch waves in 30 foot deep water.  Another problem is that natural outdoor wind often simulates hurricane winds on your model.  On the plus side, your kids might enjoy a trip to the lake or beach and get exposed to the scientific method of experimentation.  You get a very realistic wave environment with waves coming from different directions.  You can also get very deep water to test in.  As of April 2009, all the videos of models in waves posted to seasteading.org were shot using nature's waves.
 +
 
 +
== Speedboat Wake ==
 +
 
 +
A speedboat wake can be used as a test wave for a model if you are in a lake or river that does not have good waves for testing.  Vince has had a boat pass by while testing a model and the wave was reasonable.  Something in the 1 to 2 foot range.  This is enough for 1:25 or  1:12 scale type models.  It is a bit small for 1:5 scale, but would be better than nothing.  If you had a bigger boat you could get higher than 2 foot waves.  It would let you have a very well calibrated wave.  Say boat goes X MPH and passes Y feet from model.  Or even have several models in a line and the boat pass all of them at the same distance and speed at once.  This method might be used to test models at Ephemerisle in the Sacramento Delta were the water is calm.
 +
 
 +
{{Structure Designs}}
 +
[[Category:Engineering Data]]

Latest revision as of 23:03, 28 July 2017

Please use the discussion forum Low Cost Wave Tank Forum, to discuss this wiki page.

Wave tank time costs money. The Berkeley Wave Tank (circa 1930's), which is basically a long narrow tank designed for pull testing of ship hulls costs about $1500/day. A more modern wave tank costs $10,000/day. At these expense rates, the question arises "can we build our own wave tank, get useful results and still achieve significant cost savings?" This article explores several low cost wave tank alternatives that are geared toward testing seastead designs.

To date, six basic low cost concepts have emerged:

  • Large Aquarium Tank. A large 125 gallon aquarium tank is long and narrow and might be suitable for some wave tank testing.
  • Swimming Pool. A swimming pool of the right shape and depth might be adaptable to wave tank usage.
  • Dig a Ditch. A ditch can be dug in the ground and lined with plastic.
  • Water Trough. An above ground trough is filled with water.
  • Natures waves. Use natural waves in an existing body of water.
  • Speedboat Wake. Have speedboat pass near model so wake wave hits model.

These six designs are explored further below.

Requirements

A wave tank requires the following:

  • Rectangular tank
  • Wave generator
  • Test cell.
  • Wave absorber.
  • Data recording section.

Rectangular Tank

A rectangular tank is optimized for analyzing waves in one dimension. Waves are generated at one end, propagate down the tank and are absorbed at the other end. A square tank allows for the generation of waves in 2 dimensions, but probably costs too much.

Wave Generator

A sophisticated wave generator is capable of producing just about any wave form imaginable. For a low cost wave tank, being able to generate sinusoidal waves with a known period and amplitude is probably sufficient. The wave generator is basically a paddle attached to the bottom of the tank on a hinge. A motor attached to a crank can be used to generate waves. The larger the crank radius, the higher the wave amplitude. The wave period depends upon motor speed.

Having said all of that, an alternative to a motor and crank is to simply have a human grab the paddle and generate waves as by hand or leg power. The amplitude and period may not be perfect, but they may be good enough to get the job done.

Test Cell

The test cell is the location where the model is inserted. This is where any visual observation window is placed. Any test equipment is placed here as well. If using a small glass tank, fastening a plastic ruler to the side can help with estimating wave height passing a certain point. Best if it is captured on video.

Wave Absorber

When the wave hits the other end of the tank, it has a tendency to be reflected back. The back reflections can mess up the test run. To avoid the back reflections, a shallow ramp is placed at the end of the wave tank to break the waves. The shallow ramp typically has some "bumps" in it to help break the waves up more.

Data Recording Section

The purpose of a wave tank is to measure the behavior a model is placed in the test cell. While looking at the model behavior provides a subjective understanding, collecting quantitative data provides a more object understanding of what is going on.

One basic data collection system uses video. A low cost color video camera is set up to record a video stream. Three LED's (red, green and blue) are attached to the top of the model under test and illuminated. The video camera is arranged to have a split view via some mirrors. The split views provide a stereoscopic view of the model under test. The left and right views are both recorded in the camera frame so that there is no issue with frame synchronization. Software digitizes each frame to identify the X/Y location of each LED in each view. Stereoscopic math is used to identify the model position and orientation for each frame. Using adjacent frames it is possible to compute velocities and accelerations.

The idea of using a split vision system comes from some robotics vision work presented by Ingolf Sander at the Homebrew Robotics Club on January 26, 2005. Additional information can be found at Ingolf's Robotics Web Page.

Another possible method for collecting data is acceleration meters. Today you can find a low cost waterproof acceleration meter that is easy to use. It is battery powered so you can just put it on your model and take your measurements then transfer the data over USB to your computer. Would be nice if a GPS logger and acceleration logger could be in one unit.

Low-cost, high resolution wifi-enabled video cameras can be built using older smart phones for a fraction of the cost of a new device, and water-proof carrying cases can protect them.

If using an uncontrolled outdoor location on a river or a lake, overhead drone video may be a good option to add to your test.

Aquarium Tank

An aquarium tank can be pressed into duty for a wave tank. Petco sells aquariums and one of them is a reasonable choice for a wave tank. The rectangular 125 gallon 72 inch × 18 inch × 22 inch, might be usable as a wave tank. Petco sells the tank for $420 (as of September 2008.) This tank weighs 1250 pounds when completely filled with water. However, the wave tank will not fill the tank to capacity.

The clear sides make it very easy to see what is going on inside the tank. The 72 inch length provides just enough space for a wave absorber and a wave generator.

Swimming Pool

Many people own swimming pools. While a swimming pool is not an optimal shape, they tend to be larger than an aquarium.

Most people do not want anything scraping up the inside of their swimming pool. Thus, extra care is required to design a wave generator and wave absorber that can not scuff the pool surface.

In order to get straight wave propagation, it is necessary to have two sides to guide the waves down. This can be with some long pieces of plywood that span the length of the pool (carefully built to avoid scratching anything.) Some clear plastic can be placed near the test cell so that cameras can see what is happening to the model under water. Either an underwater periscope or an under water camera can be used to view what is happening under the water.

You can get a 15 foot diameter pool with filter pump and all for $230. At this price you don't need to worry about scuff marks. A human can easily make waves by pushing on the outside of this type of pool, so no expensive mechanism is needed. If you just make 1 or a few waves you can easily have a single wave hit your structure. Testing in a "confused sea" is also good. Can just have the camera on a tri-pod outside the pool.

<youtube v="TvJzpgKzXo0" />

Dig a Ditch

You can rent equipment for a day that will let you dig a good sized ditch in dirt for a reasonable amount of money. You can also hire someone with a backhoe. Hard and rocky ground could be difficult but good dirt is easy for equipment to dig quickly. Digging the ditch so that one end slopes up to make the wave absorber is easy. Lining this with something to make it waterproof is easy and cheap. Since the ditch is down in the ground draining it when not in use would require a pump. You could also leave it full and chlorinate just like a pool. Would probably want to build up a bump of dirt around the ditch so that rain water did not wash other stuff into the ditch. Do this before putting down the liner. You could almost have part of your ditch tank above ground level. However, you want straight sides on your wave tank so the waves propagate well, which is not going to work with the fresh dirt above ground level.

One minor disadvantage of the ditch is that it is not easy to put a clear side on the ditch for viewing what is going on under water. Instead of doing that, it is possible to use a water periscope to obtain a view of what is happening below the water surface.

Water Trough

Using housing studs and plywood, it is possible to build a water trough of pretty much any reasonable dimension. For example, 20 feet × 3 feet × 3 feet is quite possible. The plywood does not have to be much thicker than 3/8 inch or 1/2 inch. The studs would basically be assembled into 3 foot &times 3 foot squares with plywood lining three sides of the interior of each square. One or two of the plywood sides could be replaced with either a clear piece of plastic or glass (whichever costs less.) The wood would not be water proof, but an inexpensive clear plastic sheet purchased to line the trough and make it water proof. The water trough is probably significantly cheaper and larger than the large aquarium solution.

Nature's Waves

You can use natural waves in a lake, bay, or harbor to test a model. This saves the cost of constructing and operating your own wave tank. However, you don't get easy control over the wave environment. You have to use what nature gives you. You do have control over which location you use on any given day, and the time of day you do the tests. Even the exact spot you choose within a given harbor or bay can give you some control. This may require swimming or kayaking to get to the desired waves. Swimming with a life-jacket and fins you can move a model a few hundred feet. With a kayak you can go much further. Video can be shot from a tripod on a pier or rocky shore. Can also shoot using a waterproof camera from a kayak or while swimming in the water. If the water is cold, you may need a wet suit. Depending on where you live, and the bodies of water available nearby, you may have to do hours of driving to do some tests. You may not be able to find things like 4 inch waves in 30 foot deep water. Another problem is that natural outdoor wind often simulates hurricane winds on your model. On the plus side, your kids might enjoy a trip to the lake or beach and get exposed to the scientific method of experimentation. You get a very realistic wave environment with waves coming from different directions. You can also get very deep water to test in. As of April 2009, all the videos of models in waves posted to seasteading.org were shot using nature's waves.

Speedboat Wake

A speedboat wake can be used as a test wave for a model if you are in a lake or river that does not have good waves for testing. Vince has had a boat pass by while testing a model and the wave was reasonable. Something in the 1 to 2 foot range. This is enough for 1:25 or 1:12 scale type models. It is a bit small for 1:5 scale, but would be better than nothing. If you had a bigger boat you could get higher than 2 foot waves. It would let you have a very well calibrated wave. Say boat goes X MPH and passes Y feet from model. Or even have several models in a line and the boat pass all of them at the same distance and speed at once. This method might be used to test models at Ephemerisle in the Sacramento Delta were the water is calm.


Structure Designs
Structure requirements · PintStead · Spar · Semi-submersible · Submersible · Boats · Collaborative Design · Space Frame · Low Cost Seastead · Conex dumbbell · Docks ·
Materials
Materials · Thin Shell Ferrocement · Concrete · Geopolymer · Basalt
Engineering Data
Low Cost Wave Tank · Scale models · Wind Loads · Seasteading Software
Stability & Propulsion
Stability · Propulsion · Station Keeping · Gyroscopes · Stabilizers · Pneumatically Stabilized Platforms · Connections