Test Equipment

[thebastidgeSubscriber]

There are some facilities that wil be needed for testing models. some of them may be available on a time shared basis, for rent, or lease. Saome may have to be constructed.

• A wave tank would be good to have

[thebastidgeSubscriber]

What would be a good way of measuring vertical movement of a model in a wave tank, using the kinds of tools we might have at our disposal or cheaper off-the-shelf gear? Supposed one builds a scale model a meter high, using different method of structural and ballast design. then create waves to scale, what’s the best way to measure (not just observe and guess at, because small errors here will be scaled up with the design) ) the most vertically stable model?

[Carl-PålssonSubscriber]

Shoot video from a consistent location and compare how different designs behave. Put a background behind with a scale painted on to measure movements, Mythbusters-style.

[Wayne-GramlichSubscriber]

We hope to get a tour of the UC Berkeley Wave tank (located in Richmond) a week from Monday. We don’t know how much it costs to rent. It is very unlikely that we will be doing enough wave tank modeling to justify constructing, owning and operating our own wave tank.

[thebastidgeSubscriber]

I was stuck thinking in a rut of something like making pencial makrs. I remember something from a high school physics course about measuring speed along a linear track having an electrode scorch a paper tape at regular time intervals, and by mesuring the phsyical distance between marks on the tape we determined speed. That got stuck in my head and I couldn’t seem to think of anything else for a while. I was also thinking of using laser distance measurements from overhead, but I don’t know if those are accurate enough to measure such a small scale.

[thebastidgeSubscriber]

That’s good to know. I’d really like to see this project succeed, and seeing that concrete things are being done is encouraging. I’m stuck in a discussion, brainstorming, and cheerleading role at the moment.

[Carl-PålssonSubscriber]

One should be able to fashion a DIY wave tank in a swimming pool. An electric motor and some timber should do it. Or just have a person push a plywood sheet back and forth at the edge of the pool. Obviously not state of the art but perhaps good for some basic research. And the wave maker gets free training of pecs.

[thebastidgeSubscriber]

Probably, but it depends on the size of your model. You have to make waves proportional to the size of it. also, I think you might get some confusing data if your model is too close to a vertical surface- I’d think (don’t know for sure) that you want the far end of your wave tank away from the wave generator to be a gradual shelf (like a beach) rather than a flat sruface reflecting the waves back at you in a chaotic fashion.

[thebastidgeSubscriber]

.

[thebastidgeSubscriber]

Does anyone know of a cheap gyroscope movement/position recorder? Something that would be fairly accurate (to within 10 cm maybe) and has a digital output, (USB would be convenient) to graph motion through 3 dimensions? Hopefully for less than thousands fo dollars? Obviously, the technology is there with things like the gyroscopic mouse and wii, but has this application been developed?

http://www.pcflightsystems.com/

[ernie_the_emo_emuSubscriber]

I dunno of anything offhand that is “completed” as such. But you could do it yourself fairly easily for less than 150 usd + time/labor, assuming you aren’t wanting a massive scale stead to work with not an 8 meter monster. If you aren’t terribly comfortable with assembly or C and lots and lots of soldering I am going to suggest you use arduino ( http://en.wikipedia.org/wiki/Arduino ) which you can get for less than 50 usd from sparkfun. Sparkfun also sells gyros for just over 50 but I am going to recommend you forgo the gyro for a 3-axis accelerometer. It’ll make measuring a bit easier and will require you to work with less hardware in the long run. Plus they cost less(20-35 usually). Sparkfun(holy seven hells I seem like an advert for them, they should pay me commission everytime I mention their name) has a breakout board for the ADXL330 accelerometer for 35 dollars ( http://www.sparkfun.com/commerce/product_info.php?products_id=692 ) I am going to recommend it because it will save you a lot of fine soldering work and a fair bit of hassle. Programming it in Arduino is fairly straightforward and interfaces with Flash if you’re wanting to represent your data visually and I have even seen it interface with Second Life(which could make for some nifty seastead model interactions).

If you’re interested in diy and not wanting to use Arduino for whatever reason, I have a couple of Atmega48′s laying around and could be convinced to get an ADXL330 or two and prototype some boards and C code for what you’re proposing. It’d save us a bit all a bit of money and would be fairly simple to do.

[Arthur-B.Subscriber]

Wouldn’t simulation be better ?

1. It could be cheaper. There are plenty of physics simulator that can be used. A CFD framework such as OpenFoam comes to mind in combination with a dynamics framework such as ODE . I don’t know how much work would be involved in getting a general purpose simulator running. Foam is largely overkill for the kind of fluid dynamics going on at this scale… modeling physical stress would be needed as well.
2. Professional software solutions probably already exist, they may have good pricing for non-profit research.
3. Some physical phenomenom do not scale (the tensile strenght of the spar scales quadratically, the weight cubically)
4. Do they have wind in the wave chamber ?
5. Many more ideas and models can be tested
6. Shapes can be optimised, see this.
7. It can produce beautiful art.

On the drawbacks

1. It’s initially more expensive as it requires a lot of qualified work, unless you’re lucky enough to find a guy willing to do an applied math thesis on this.
2. It takes more time to get going
3. It might be less accurate than direct simulation on a scaled model, but that seems unlikely.

[thebastidgeSubscriber]

I think I’ll follow up with you on this, in a few weeks. I’m willing to tinker a bit, it would be a fun project to work on with my nephew- he’s not quite ten now. I’m also willing to chip in a few bucks towards a prototype (if you’re cool with paypal.)

[thebastidgeSubscriber]

Simulations are great, but nothing beats empirical measurement. Even a scale model has some imperfections as a simulation, but you can still observe unanticipated results.

[ernie_the_emo_emuSubscriber]

“I think I’ll follow up with you on this, in a few weeks. I’m willing to tinker a bit, it would be a fun project to work on with my nephew- he’s not quite ten now. I’m also willing to chip in a few bucks towards a prototype (if you’re cool with paypal.)”

I’ll start drawing schematics tonight and sorting through my rack for parts(I may already have an accelerometer from a project I worked on several months back but I am uncertain). What I don’t have I’ll go ahead and order. I already have the AVR environment set up so the only cost I’d have(assuming I do not use arduino) is the 35 bucks for the ADXL plus maybe 5-10 dollars for whatever else I lack. So don’t worry about chipping in any money, its not enough to worry me and it really is something we need. I’ll keep your nephew in mind when desiging everything. If I can make it easy for a 10-year-old to build, then I think it will fall within the technical scope of all interested in seasteading.

I’ll also start mulling over software for it. I work in unix-like systems and haven’t access to Windows at home which could cause a few minor problems on the software end, but nothing that cannot be bridged. I’ll make some notes on this when I am sorting through my rack of parts and see what I come up with. Should this get an article on the wiki?

Now we just need someone to get us going with a nifty aquariumstead (the thing I have bobbing in a kiddie pool in the backyard is, to put it mildly, ghastly)

[Author]

Posts