Connecting the Dots - Vancouver

Dr. Nakane happened to be the lab director for the Engineering Physics Project lab at UBC (they call themselves 'fizzers'). He was looking for guest speakers for the end of semester project fair and invited me out to UBC in Vancouver to speak on engineering and entrepreneurialism. I then spent two days bombing around Vancouver with Jon and Bruce visiting various labs, shops and eating great food. I can't thank them enough. Here are some of the projects and encounters I came across:

Tim then took us into the clean room (bunny suits!) where he showed us how they made the slides. Using technology wafer fabrication techniques from the 1960s, silicon wafers were etched with lithography to create the molds.

They are even experimenting with multi-layer polymers - meaning, instead of a two layer channel system, they are getting up to five. This requires the ability to move fluid between layers. How do we do this in electronics on PCBs? With vias of course and that's precisely what they are doing! Except with a much more powerful fluorine laser (scary stuff fluorine). They carefully map the X/Y coordinates of the slide and 'burn' a small hole or via in the polymer. It was amazing stuff and the analogy to electronics and PCBs was surprising.

Above is a picture of a controlled temperature reaction chamber where they were studying the development of hundreds of individual cells over the period of a few days. Using some serious automation, the system is designed to take a photograph every half hour of every cell! The variance in the temperature within the lab was causing a few microns of change in the rig causing the images to go out of focus. They are experimenting with computer vision algorithms to trying to correct for the deficiencies. So many great toys - I was extremely impressed and a little jealous that they get to play (hack with a whole lot of brains) at such a cutting edge.


Along the tour of many awesome labs (wind tunnels, huge open engineering spaces), we stopped by Boreal Genomics. Again, not nearly enough photos, but what you see is a dish in the middle of the machine with four platinum electrodes. I'm simplifying a lot here, but a normal molecule, when excited by an electric field, will travel through a polymer gel with a known speed. The molecule has drag based on its size and shape. But DNA is weird. Depending on how it is compressed or strained, it has two different amounts of drag.


So what? Using this interesting non-linear drag coefficient of DNA, Boreal Genomics figured out by exposing DNA to an electric field, then rotating that electric field, you could concentrate DNA. Not only could you concentrate it, but separate it from impurities as well! You can literally take dirt, suspend it in a fluid, put a few milliliters into the machine and concentrate the DNA to a dot and all the dirt gets pushed to the side. It was truly amazing to see. The machine used an ARM7 for control and about 100V per centimeter electric field. The biggest problem at the moment is cooling the plate (with a peltier plate and heat sink) to keep the polymer from boiling off.

Boreal Genomics was founded by Dr. Andre Marziali. Dr. Marziali was the other guest speaker at the UBC physics presentation. His talk was the ying and yang to SparkFun. Where SparkFun started with a little debt, a bunch of inventory, and a rough business plan, Dr. Marziali had a large amount of patented IP, grants from all sorts of agencies, and a big up-hill battle to development a complex cutting edge machine. He is making large advancements daily and is already producing machines (upwards of $50,000 per machine). UBC is churning out some amazing scientists.

We completed the tour with a review of the amazing student projects including "Limitations of data transfer across SOI racetrack resonators" and "Development of a new polydispersity analysis to allow bacterial detection in platelet transfusions". Why was I invited to speak at this event again? I am not nearly a good enough engineer to stand toe to toe with these folks!

If you'd like to check it out, the slides are available here (5MB!). The PDF doesn't quite do it justice but you might enjoy looking at some of the charts and images. My presentation went extremely well and the audience of ~60 people seemed to really enjoy it. I very much enjoyed talking to Dan Gelbart from Creo. He had some wonderful feedback for me about growing SparkFun sensibly.