always have plastic sheeting in lab

October 24, 2016 at 10:46 am | | everyday science, hardware, stupid technology

I learned during my PhD that you should always have plastic sheeting in lab, because it might just save your equipment when/if a water leak happens. It saved one of our scopes recently, although I wasn’t fast enough to prevent some water damage on an expensive camera. :(

2016-09-02 14.27.09

2016-09-02 14.17.25

For less than $5, you can get some rolls of the stuff. If you want larger and thicker sheets (like in the photos), I recommend this stuff.

home-made plasma cleaner

July 23, 2014 at 12:52 pm | | everyday science, hardware, science@home, stupid technology

I really want a plasma cleaner, for cleaning coverslips and activating glass for PDMS bonding, but they cost thousands of dollars. I thought that was a lot of money for a glorified microwave. So I made my own.

Drill a few holes in glass:

2014-03-10 11.16.58


Make a PDMS seal (thanks Kate):

2014-07-16 16.11.57

Glue the chamber:

2014-03-10 12.29.04

We’re ready to go!

2014-07-17 13.19.41

Fill the chamber with argon, evacuate it, turn on the microwave oven, and … voila! … a plasma:

2014-07-17 12.54.18

2014-07-17 12.56.04

Below are slides before and after (right) plasma treatment. You can see the contact angle of water is dramatically reduced.

2014-07-17 13.14.40

It works!

Well, not really. I found that the plasma really only stays lit with argon. When I flow air in, it extinguishes, but also burns some of the rubber hoses. That adds more dirt to my slides than I want.

Conclusion: don’t do this at home. :)

(Well, that might be a little harsh. It does work well to bond PDMS to glass. And I’ll try a longer etch sometime to see if it will ever clean the coverslips.)

checklist of unusual phenomena

July 22, 2014 at 1:40 pm | | nerd, stupid technology

AKA “troubleshooting.”


electronics company

June 9, 2014 at 4:12 pm | | hardware, stupid technology


are fuel economy gauges doing math wrong?

March 26, 2013 at 3:27 pm | | stupid technology

A few years ago, there was a piece in Science magazine about how we should all be using GPM instead of MPG. Here’s a link to my post about it back then. The main point is that the relevant unit for fuel use should be fuel divided by distance, and than MPG is inverse, which is harder wrap our puny brains around.

Back then, I looked into how fleet fuel economy averages were calculated, because I was worried that one single crazy-high MPG model could artificially skew the average high without actually making the fleet more fuel efficient. It turns out that the US requires automakers to calculate their fleet average fuel economy the correct way: convert each model’s economy to GPM, find the mean, then take the inverse again. Phew.

(not my car)

But now I wonder if the computer programmers at the automakers know how to calculate an average. My computer fuel gauge is always inflated compared to what I calculate from the fuel pump amount and odometer reading. Every time. And I’m not alone. So, either the gas stations are messing with their pump readings, or the average MPG on my dash is miscalculated. I wonder if the computer just takes a continuous average of the measured MPG values, which would definitely result in an inflated number at the end of a gas tank. (That’s because 10 miles of driving at 35 MPG after 10 miles at 25 MPG does not average to 20 miles at 30 MPG. It’s actually 29 MPG.*) If the computer just averages the MPG numbers, it will be 2% inflated from the actual value. That inflation would get worse if you drive down a hill for a mile and go 50 MPG, then back up that hill and drop down to 10 MPG: instead of an average of 30 MPG, the true average is only 17 MPG!

Of course, if the computer just takes the total miles driven and divides by the total gallons of fuel used (since the reset), than the average would be calculated correctly. But is that what’s happening?

Does anyone know anyone who works at an automaker who can check how they do the math?


* Here’s my math. Driving 10 miles at 25 MPG uses 0.4 gal of fuel. Driving 10 miles at 35 MPG uses around 0.29 gal of fuel. That’s 20 miles driven and 0.69 gal fuel used, or 29 MPG. Not 30 MPG. For the hill example, down the hill uses 0.02 gal for the mile, and up uses 0.1 gal. That’s 2 miles and 0.12 gal, or 16.7 MPG.

playstation joystick control on electron microscope

July 13, 2011 at 11:13 am | | hardware, nerd, stupid technology

Fancy electron microscope…


…uses a game controller to manipulate the internal robots. Ha!


February 11, 2011 at 12:10 pm | | literature, science community, stupid technology, wild web

UPDATE2: OK, it turns out that the daily(ish) email isn’t too terrible. I now use it and I’m no longer upset that they don’t have an RSS feed. I correct myself and now fully endorse F1000!

Faculty of 1000 is extremely powerful with a lot of potential, but simultaneously completely worthless.

F1000 is like mini-peer-review post-publishing: it uses its “Faculty,” experts in various fields, to rate publications that those experts think are worth reading. It’s like … nay, it is … getting suggestions on what to read in the recent literature from a large group of experts. That is very cool. Of course, there are various databases like Cite-U-Like and Mendeley that are trying to mine their data to find interesting papers, but there’s something great about getting little mini-reviews from actual people.

OK, so why am I annoyed? F1000 doesn’t have an RSS feed! So I have to remember to go and check the website every week. Even if I happen to remember, there’s no way to mark which reviews I’ve already seen and the new ones. What is this, 2002?

UPDATE: rpg comments below with some good news: F1000 is actively trying to get RSS on the site. The comments also explain why it’s a challenge. I eagerly await RSS.

ucsf vs tsa

November 22, 2010 at 1:34 pm | | news, science and the public, stupid technology

Some profs at UCSF have concerns about the radiation dose of backscatter scanners, specifically that all the energy is deposited in the skin instead of being spread throughout the entire body. So the dose is concentrated in time and volume. Basically, it sounds like TSA hasn’t done enough safety testing on these machines.

I would like to see a risk analysis of the probability of the screening causing cancer vs. the reduced threat of airline passengers dying from terrorism. The problem is that all these are very low probability events.

Anyway, this is my response to the entire fiasco:

(via Austin)

acs feeds broken?

September 3, 2010 at 8:53 am | | literature, stupid technology, wild web

I was forced to update all my ACS feeds in my RSS reader (Google Reader). None of the ACS feeds I follow had updated for a couple days. I think ACS switched over to Feedburner and their old feeds stopped updating. Anyone else have this problem?

And TOC images seem brokenish with the new feeds

UPDATE: Alex updated us on the issues. I’m happy that ACS is migrating to a better system. They really tried to redirect, but there are some problems. I think the new feeds seem to be working well now!

dangerous laser pointers

August 11, 2010 at 9:36 am | | lab safety, science and the public, science@home, stupid technology

NIST is warning us that some cheapo green laser pointers might be unfiltered and dangerous. Some manufacturers skip installing the IR filter, thus making a laser pointer that has a high-power invisible beam along with the green light.

The “green” of green laser pointers is 532 nm, doubled frequency of the 1064 nm emission from a neodymium (e.g. Nd:YAG, Nd:YLF, or Nd:YVO4) laser. A diode (e.g. 800 nm) pumps the neodymium laser, which emits 1064 nm light; a doubling crystal produces the green 532 nm light. But the doubling crystal is not 100% efficient, so an IR filter is necessary to block the remaining 1064 nm light that isn’t doubled (as well as block the 800 nm pumping light). The plot above shows how much 1064 nm light escapes if the filter is removed: it’s much more than the green light—if the 532 nm is 20 mW, the IR might be as high as 100 mW, certainly potentially damaging to the eye!

IR is especially dangerous laser light. First, it is invisible, so it is more difficult to identify and avoid stray beams. In this case, that’s less of a worry, because the green beam coaligned is visible. However, the second reason IR is dangerous is that, because it is invisible, you can’t tell how bright it is (see below). The final reason IR is dangerous is the biology of the eye, which is transparent to IR light, and focuses it to the retina (the nerves). IR can easily burn the retina permanently (causing blindness), or burn other parts of the eye or skin.

The simple method NIST suggests we can use to test our laser pointers is described in the announcement. Basically, they use a CD as a diffraction grating and a cheap webcam. The sensor of a digital camera is sensitive to IR light, but usually has a filter to see only visible; it is simple to remove the IR filter of a cheap webcam to make an IR sensitive detector. (Unfortunately, the sensitivity cuts out before 1064 nm, so the camera can only see the 800 nm pump light). The picture above shows the diffraction of the visible light with a normal digital camera; the bottom image is using the IR webcam. You can see the extra diffraction spots from the 800 nm light. Note also how much brighter the IR light is from the laser: even though you can’t see it with your eyes, it is very bright and dangerous.

By the way, my favorite line in the NIST report is the following: “The infrared light spreads out beyond the green, which could be injurious, for example, to a cat closely chasing a spot of green light.” Actually, that’s kinda sad: I hope the NIST folks didn’t discover this problem after they blinded their pet.

you’re doing it wrong

August 9, 2010 at 10:15 am | | everyday science, stupid technology

Fluctuations in a “cold” room I walked past.

That’s not good.

how do you read PDFs?

July 28, 2010 at 8:59 am | | everyday science, help me, literature, open thread, software, stupid technology

I now have a longer commute, with at least 30 minutes of quality reading time. I don’t really want to carry my laptop everyday, so I’m seeking a better way to read journal articles. I’m not going to print them out, so don’t suggest reading them on paper. :)

[poll id=”3″]

Of course, cost is a factor, but I don’t want to go for the cheapest option if I end up never using it! My guess is that the Kindle DX is the best for reading PDFs, but loses on other fronts (e.g. large, expensive, limited, only grayscale). The iPad is a versatile color reader and I can sync with programs such as Papers or Mendeley (soon for the latter), but it is very expensive. Also, the screen isn’t as nice for reading print. The iPhone is way too small to read PDFs.

Man, I need to test-drive these devices for a month!

Other options?

Creatine, a popular sports supplement

July 7, 2010 at 2:38 pm | | nerd, science and the public, stupid technology, tutorial

Creatine, a small-molecule found naturally in red meat (and biosynthesized in our bodies), is a popular supplement for weight lifters. To understand how it works, one needs to know that ATP (adenosine triphosphate) is the body’s energy molecule.  It gives muscles the energy they need to function, but in the process, it loses a phosphate group and is converted to ADP (adenosine diphosphate).

Creatine monophosphate has the ability to convert this low-energy ADP molecule back into the super-charged ATP molecule that muscles crave.

Creatine monophosphate

As a consequence, lifters that supplement with creatine can do more reps, which can lead to better results in the gym on a shorter timescale.  Creatine supplementation also has the effect of increasing water volume in the muscles, causing them to swell and look bigger; this effect subsides quickly once creatine supplementation is stopped.

It is well known that consumption of simple sugars with creatine increases creatine absorption.  When you consume sugar, your blood-sugar level increases, and your body releases insulin in response (assuming you don’t have type 1 diabetes).  Insulin instructs the cells to take up sugar from the bloodstream.  Insulin also has the nice effect of stimulating creatine transporters, which transport creatine from the blood into cells.

Now that the background is finished…

I was at GNC yesterday buying some creatine.  I looked at the ingredients on the GNC-brand creatine and gasped.  Creatine and sucralose!?!?!?!  Sucralose???  OK- a little more background.  Sucralose is a zero-carbohydrate, synthetic, sugar mimic.  As you can see below, it looks a lot like sucrose, only it has some chlorine groups that basically make it unrecognizable by the body’s enzymes.  So your tongue recognizes it, but your waistline doesn’t because it’s not metabolized.



Now to the remarkable part.  Sucralose has no effect on the blood sugar level. So this GNC-brand product that I bought containing 17% sucralose/83% creatine is ridiculous.  Uninformed weightlifters don’t want “sugar” calories so the industry introduces 1 g calorie-free sucralose per 5 g creatine, which has no effect on creatine uptake and actually tastes quite disgusting (sucralose is 600x sweeter than sucrose, which is regular sugar).

So who is the bigger idiot?

cleaning-chemicals industry takes on the final frontier…

May 5, 2010 at 10:56 am | | science and the public, stupid technology

invisible stains. Finally, there are cleaners on the market that can clean up stains that are undetectable by any of the human senses.

Have you seen that commercial for a toilet-bowl cleaner? They say, “Bleach only hides stains.” Then they pour some purple stuff on the ceramic to reveal the hidden “stains.”

Um, what? Isn’t hiding a stain mean you get rid of the stain? By definition? I think it’s nuts that the cleaning-chemicals industry is inventing the problem of invisible stains. I have a solution to this problem: don’t pour that purple stuff in your toilet!

And then there are those “chemical residues” that watch you shower. The invisible, undetectable residues. How do we know that invisible residues are there at all? The TV tells us that they are there, of course. But wait, how do we know that the alternative cleaner doesn’t leave an invisible residue?

All this reminds me of the Emperor’s New Clothes. Or, more aptly, that old joke about the elephant repellant (“You don’t see any elephants around here do you? It must be working, then!”) People need to ignore these silly commercials and relax a little bit about cleaning. Just scrub the toilet bowl every week or two with a brush, then wipe the seat etc. with diluted bleach and be done with it! Jeez!

McNASA and whale oil

April 14, 2010 at 3:24 pm | | stupid technology

This is the strangest paragraph I’ve read in a while:

“But during the 1980s, [NASA] lost much of its old high-quality data. Its early tracking stations recorded satellite data on high-resolution master tapes that used whale oil to bind iron particles to the acetate. The whale oil made the tapes far more durable, but when commercial whaling was phased out in the mid-1980s, NASA couldn’t get such long-lasting tapes. So it reused old ones. NASA engineers taped over some 200,000 previously recorded master tapes, including high-resolution records from spacecraft as diverse as early Landsat satellites and Apollo 11, and preserved only low-resolution copies.” (source)

I checked to make sure it wasn’t the April 1st issue of Science. The entire article is weird, including the fact that researchers are working out of an abandoned McDonald’s.

Just another reason to start hunting those smug whales again.

Next Page >

Powered by WordPress, Theme Based on "Pool" by Borja Fernandez
Entries and comments feeds. Valid XHTML and CSS.