Friday, April 30, 2010

Open source science tools at LRRD.

Here you can see the first operational version of a small Arduino-based flow meter we've built based on this open source design.

Parts cost is roughly $75.  Compare that with at least $600 for similar commercial models.  And this one can be built from easily available parts by students, and could be used to record data and even control the pump to produce hydrographs.

We're very excited about going open source, providing designs and parts for instrumentation to augment our Emriver models.  If we get the NSF and EPA grants we've applied for, we'll also develop curriculum and supplementary video/visualizations.  These we'll just give away.  We like giving things away; you won't see our business model featured in Fortune anytime soon.

Not everything can be open source, do-it-yourself.  Scientists don't build their own microscopes, laptops, or rock hammers because it's not practical.

We'll continue to do the hard stuff, like obtaining plastic media with tight specs (in both size distribution and color, from recycled materials, not easy), making the near perfect but hard-to-build aluminum box the Em2 uses, something I spent years on and am still refining.  And as always, we'll push ahead with innovation.

Wednesday, April 28, 2010

Kids, not consultants, design the new Em2 model.



We're nearly finished with a complete redesign of our Em2 model.  Here's detail of a wood/acrylic mock up of the new legs.

How to design.   Hire consultants?  For a small project, the finding/contracting can dwarf time spent on actual work.  I have to start by explaining that we simulate fluvial geomorphology in an aluminum box filled with ground salsa bowls.

I've called in clever people who'll help for free, and Sauer Machine in St. Louis, who'll build the final design, is there as always.

Hundreds of hours have gone into this; I can recite a lot of dimensions, angles, and have memorized the weight per linear foot of  2" x 0.125 square aluminum tubing, and we've built a dozen mock ups like this.

Good design demands it.  It requires passion and immersion in the problem, and unless you have a lot of money, you can't get this from a consultant.  This is an Emriver, not and iPod.

So I'm glad LRRD isn't a real for profit and runs on me not getting paid much, because otherwise we'd never make the cool stuff we do.

And I have to credit this 2 year old who, as I watched at the Cache River Nature Fest last weekend, tried for twenty minutes to break the Em2's notch gage,  And the other kids wailing on the model, who reminded me how strong it should be.

Tuesday, April 27, 2010

Beautiful fluid mechanics dance.



A beautiful performance in which a dancer is tracked by a camera/software and influences virtual fluids projected behind her.

Very original, one of a kind I'd say.  From Hope Goldman's recent MFA performance at Urbana-Champaign.

We're working on motion controlled river models.  Perhaps we'll make one controlled by a dancer.  Via boingboing.net.

Update:  Andrew Moffat, who worked on the soft and hardware for this replied to my questions on methods.  And remember, all this was done real time during the performance, not with post processing of the video:

I'm glad you enjoyed the performance.  I used a fluid solver running in real-time on the GPU.  It used Navier Stokes differential equations for incompressible fluids.  Running the simulation on the GPU allowed us to run it fast enough to achieve real-time.  From there, we had a setup with infrared lighting and a webcam that we modified to be sensitive to infrared light.  This allowed us to track the dancer accurately and create velocity fields of her motion.  Then we added the collisions and velocity fields to the fluid solver on the GPU.  A few transitions, a few colors based on velocity and time, and voila, Form Constant.
 And sent this reddit.com link, which I'll quote here:

The camera was sensitive to only infrared light, not visible light. We also had lit the entire screen from the back with only bright infrared light. So when the camera (which was at the front, in the audience) looks at the dancer and screen, it sees a completely bright white screen, and a completely dark silhouette of a dancer (because the infrared light is coming from the back, she is not being lit). Because the projector (also at the front, in the audience) only projects visible light onto the dancer and screen, the camera will always still see that white background and black silhouette, no matter what we project onto the dancer and screen. In other words, using infrared light made it so the projector didn't interfere with determining background from foreground.

From that point, I just ran some processing on the images coming in from the camera, to clean them up. Afterwards, we had a nice image mask of the background in solid white, the dancer's silhouette in solid black. That's how we found the position.

Finding the motion of the dancer was a little trickier. It involved running an algorithm called optical flow on each image and its previous image. From that, I was able to determine a velocity field of motion.

What an amazing project, thanks Andrew!

Thursday, April 22, 2010

Meriam Lahlou joins LRRD.

I’m very happy to announce that Meriam Lahlou has joined LRRD as our business manager.   Meriam is a delightful and fascinating person we’re very lucky to have.

Originally from Morocco, Meriam came to the United States in 1987 to study international business at Villanova University.  She went on to earn an MBA in finance and then returned to Morocco to work for The Ministry of Finance and Foreign Investments.  She moved back to the US in 1997 to join Ernst and Young LLP in New Jersey while her husband Otmane was finishing his DMD and taking care of their infant daughter Radia.  A job opportunity for Otmane brought the family to Carbondale in 2000, where he quickly became a much beloved and admired dentist.   After their son Adam was born in 2001, Meriam decided to be a stay at home mom for a while and contribute in the community.  She enrolled in a Master’s in Foreign Languages at SIU to learn pedagogy and be able to share her native skills and passion for languages.   Fluent in English, French, Arabic, and Italian, Meriam then taught French in at Giant City School as a volunteer.

Meriam’s family and our community suffered a great loss when Otmane passed away in late 2008.

With the loss of her husband at the age of 40, she could no longer teach as a volunteer.  Meriam first organized French camps for children.  This successful initiative led to an after school program at Giant City and Unity Point Schools with the help and support of Principals Sharon Mayes and April Haar.   Meriam also initiated a French Café conversation class for adults at Carbondale’s Longbranch Café.

Meriam initially came to LRRD as a French translator for a project in Québéc, and was attracted to LRRD’s strong commitment to education.   Her business skills have been a wonderful asset to us in this difficult economy, and she’s quickly mastered the myriad things we do from bookkeeping to shipping models.  She’s redesigned workflows, assessed Internet marketing tools and streamlined our operations.   She’s very enthusiastic about the possibilities we see with positive responses from the NSF and EPA grants we’re seeking, and her work will enable us to soon ship to international markets, where her language and cultural skills will be a great asset.

And I appreciate her warm personality, which includes a sharp wit, broad knowledge and intellect, tolerance, great compassion, and a sunny disposition!

Emriver outreach for the young and old.



We've had a busy couple of weeks with our Emriver Em2.  First, Andrew Podoll,  an NSF-funded geology/education GK-12 HEART fellow at SIUC, borrowed the Em2 for a student research event.  It was a huge hit with the kids, as you can see.

Last weekend I hauled to Em2 to a statewide meeting of the Environmental Educators Association of Illinois; here you can see a couple of SIUC Forestry students using the model--these guys spent literally hours doing experiments!

Yesterday I gave a talk to the Carbondale Garden Study Club.  As you can see, these women aren't youngsters.  I was quite surprised to see them dig in like a bunch of middle-schoolers, laughing, trying out different things, and asking a lot of questions about the model and how rivers worked.  This event was special to me because I'm sure this is the oldest group of laypeople I've ever worked with--there are usually a mix of ages at public events and the kids dominate river model experimentation.  Minus the kids, these women stepped up and had their hands in the model right away.

It's great to be a part of the community.  We didn't bill a dime for any of this.  I hope we soon get positive news on our proposals--these two weeks were a nice taste of what's possible.

Thanks to Andrew for the HEART event picture.

Monday, April 19, 2010

Emriver Em2 redesign. John Cotter, are you out there?



I've spent many hours over the last week and most of today trying out new ideas.  There is no way other than building these mock ups.  I enjoy the work.  But not the long hours.  Thanks to many helpful collaborators!

Too bad my aviation tech professor friend John Cotter has taken on too many teaching assignments this semester and is impossible to find--we're going for absolute simplicity, economy of weight and size; total reliability.

Much like aircraft design, right, John?

The Em2 will be the Cessna 172 of the stream table world.  You could be famous.

End shameless attempt to get free work from a local expert.

Photos of rejected leg attachment prototypes and a tiny model I built to test box geometry and strength.

Tuesday, April 13, 2010

More Emriver Em2 redesign, idea number 234,745.

Must be a million ways to build an aluminum box with legs.  We've looked at most of them I think.

Here's the latest rejected mock up.

The final design will be very good.

Thursday, April 8, 2010

Em2 redesign progress, collaboration with SIUC engineering.

We're closing in on the new Em2 design.  It's very exciting, but agonizing.  This CAD screenshot shows just a few of the two dozen or so leg attachment designs we've considered.  I'm insisting on something that's super strong, safe, simple, and field repairable.  I'm not interested in selling replacement parts, I want to build the ultimate educational tool.  Things should not break or wear out and if they do, they should be locally repairable.

And not easy when all my collaborators are in the same boat I am--having to constantly watch cash flow and go towards the things that bring in money in this terrible economy.

Here you can see me with a wooden mock-up.  This thing is complicated because the box is sloped ~3.5 degrees, and the legs splay ~10 degrees in two directions, but not the same on each end because of the box slope.

And here's a photo of Lizette Chevalier, civil engineering faculty at SIUC, with her daughter Rosa and a grad student she brought by this week.  We're hoping to work on river modeling research.  I was very happy to see both the grad student and 11-year-old have a lot of fun with our lab model, thanks Lizette.

Monday, April 5, 2010

Modeling media research--the fine color coded fraction.

We're continuing our research on low density modeling media.  The mix we usually use tapers out at around 0.5mm on the fine end, but clearly adding smaller particles has a big effect on behavior in movable bed models.  This fraction fills the interstitial spaces and strongly influences subsurface water movement.  And it increases cohesion to some degree.  It also clogs filters in our recirculating systems, which is one reason we've avoided it.  But we've been working with Emriver model users to develop a particle size mix that includes this finer fraction. 

Today I worked on color.  Producing color-coded by size plastic media is a difficult and expensive process.  Our Em4 media has three sizes and colors (video here).  We want to add a fourth fine fraction.  The dark brown/white/yellow scheme we have now has a perfect chroma (brightness--very important for observation and photography/sensing) and very nice eye appeal (I'm a follower of Tufte on these things).  You can clearly see the differences in particle movement and deposited facies.

But it's tough to move beyond these three distinctive, earthy tones.  We have some fine media in red and blue, and I tried those today.  Very interesting results!  At ~5% of each, mixed in the dry, you can't see them at all.  But add a little water, and the sorting is amazing; they pop right out!  Red's not so good, but maybe the blue will work.  But then we use blue and green dye for tracing--could be a conflict there.

Very exciting and promising.