MachinePix Weekly #12: Ben Krasnow, Applied Science; senior staff engineer, Verily
MachinePix Weekly #12: Ben Krasnow, Applied Science; senior staff engineer, Verily
Ben Krasnow on designing biotech equipment and running a hugely popular YouTube channel. This week's most popular post is... worm grunting? šš¬
This week I sit down with Ben Krasnow, the Senior Staff Hardware Engineer at Verily, Alphabetās life sciences company. He is also the man behind the wildly popular Applied Science YouTube Channel š¤š§Ŗ
The most popular post last week was worm-grunting sticks, demonstrating yet again that sometimes the simplest machines are the most fun. As always, the entire weekās breakdown is below the interview.
Iām always looking for interesting people to interview, have anyone in mind?
āKane
Interview with Ben Krasnow
Coincidentally, you know Jeri from the last interview because you both worked at Valve! How did you make the change from gaming hardware to biotech hardware?
I actually started in MRI-compatible computer peripherals: if youāre a researcher and you need someone in an MRI to interact with a computer, I made hardware to do that. Itās pretty coolāthe biggest challenge is ferromagnetic materials. If you have any steel parts, even a single screw, the strength of the MRIās magnet can launch it across the room. Any components with wire coils or anything could be induced with currents and get pretty dicey. You can shield your components, but the shields will catch a lot of eddy currents and heat up. A lot of times you have to use fiber optics where you would normally use wire. A lot of peripherals I made for MRIs were mice and keyboards and joysticks and those sorts of things, so funnily enough not that different from gaming.
When Valve found me, they wanted unusual feedback no one had done before like galvanic feedback and stress sensing, so there was some overlap from biotech.
What is it like being a mechanical engineer for a biotech company?
I would say the biggest thing to learn is that healthcare is often broken due to financial problems, not due to engineering ones. If you want to build something new itās not as much a question of technical difficulty as it is a question of regulatory difficulty. At Verily, I get to work in the skunkworks lab, so I get a lot of intellectual latitudeābut we try really hard to push down the cost of healthcare.
What are the coolest projects youāve worked on at Verily?
Iām pretty excited about the one Iām working on now! You may have heard stories that dogs can smell cancer and corona virus infections. Last year it was in the news that there was a woman who could smell her husband had Parkinsonās. So obviously there are molecules coming out of the body that indicate some disease.
Dogs' noses are exquisitely sensitive. They can detect things in parts per trillion. Using dogsā and peoplesā noses is not a scalable solution, and weāre working on addressing that.
What are the machines we use to benchmark dogsā noses?
The gold standard is GCMS (gas chromatography mass spectrometers). Theyāre very expensive, very hard to set up. You need a trained technician to use them. Smell samples donāt travel well either! Itās best to get the person and machine in the same room. All these make developing smell-based detection challenging. Weāre trying to change all that.
I love the Applied Science channel! How did you start this?
It was in large part kicked off by Maker Faire 2011. Each year you came up with a better and better project, and this was back when Maker Faire was still mostly home-built projects. I had an idea to build a scanning electron microscope, and thatās when I also made my first video. Jeriās channel was actually a huge inspiration, she was one of the first hardcore technical channels.
How do you decide what to film for Applied Science?
It was a different world in those days. In 2011 and 2012 there wasnāt as much money to be made on Youtube. We were doing it for the hobby experience and the attention. I was trying to come up with more and more extravagant projects. I did a DIY chip fab, liquid nitrogen generatorā¦ it kept getting more extravagant. Now, I get inspiration from conversations with friends or coworkers. If something comes up and I wonder āhuh, how does that work?āāthat becomes an idea for a video.
As a camera nerd, Iām curious what you use to film your videos.
I use a Panasonic GH5 with a Sigma 18-35 f/1.8 lens with a speedbooster. I shoot B-roll from a tripod while I work, and try to do the A-roll in one single take from a tripod.
Do you use the things you make?
Theyāre fun proof-of-concepts. The electron microscope I made has a worse resolution than an optical microscope, but theyāre fun to make. My projects and videos are anti-work, itās nice to just have a non-commercial outlet.
Most favorite thing youāve made for the channel?
My favorite is still the electron microscope because itās foundational. A more recent favorite is a video of a record player needle using the electron microscope!
I forgot to mention the cookie machine.Ā It dispenses tiny amounts of each ingredient so that the recipe can be varied for each cookie on the sheet.Ā It tests variations faster with with less ingredient waste.
I also transferred edible holographs to chocolate.
What are your other favorite channels or accounts (besides @MachinePix of course āŗļø)?
Thatās a good question: NileRed, has a hugely successful chemistry channel with awesome videography. He has one reaction in particular which is a petri dish that spontaneously generates these waves. Itās totally crazy.
Tech Ingredients is a really good channel. Huygens Optics too: he's making his own chip fab and has his own wafer stepper!
Any side projects youāre working on right now?
Iāve had a hydraulic press in my shop for a while. I want to make a 100,000 PSI espresso! Iāve got it to 60-80K, but there are a lot of challenges. The gaskets just blow. At those pressures, solids behave weirdly. Aluminum parts squeeze out like toothpaste. I donāt think it will taste goodāitās designed to get views on YouTube, but iām still excited to try it.
Whatās your favorite simple (or not so simple) tool that you think is under-appreciated?
Probably the old stereo inspection microscope. People think you use a microscope for small stuff, but theyāre great for everything. I think theyāre like $350 now, it lowers eye strain a lot. Itās the center of my shop, I do a lot of work under it.
I actually asked a watchmaker why they still use eye loupes and not stereo microscopes, and he didn't like the question very much. I think itās a tradition thing. I think if they sat down and used a stereo microscope they wouldn't go back.
The Week in Review
The even crazier version is the machine that vacuum-packs entire foam mattresses.
This weekās most popular post! Also known as worm fiddling or worm charming, I thought worm grunting was the funniest phrase. The stick in the ground is called a āstobā, and the stick used to rub the stob is called the ārooping ironā.
If you still havenāt gotten enough of worm grunting, be sure to check out the Annual Worm Gruntinā Festival.
This is floating around on the internet captioned as a hydraulic press, but itās almost certainly a screw press for safety and simplicity (and by just looking at itā¦). This is a much more confidence-inspiring setup compared to this terrifying gravity drop forge.
If youāve ever wondered why commercial ice has little holes in it, this is why: itās a consequence of a machine designānot an optimization of the ice itself. Vogt introduced the first commercial tube ice maker in 1938.
Postscript
This past week I fermented makgeolli, a very tasty Korean alcoholic beverage that kind of tastes like a boozy, fizzy horchata. Itās extremely easy to make and requires no special equipment!
If you enjoyed this newsletter, forward it to friends (or interesting enemies). I am always looking to connect with interesting people and learn about interesting machinesāreach out!
āKane