The Benefits of Sharing—Nice Guys and Girls do Finish First - Open-Source Lab: How to Build Your Own Hardware and Reduce Research Costs - page 16

simulation designs and reduced sloppiness during experiments. These intrinsic benefits were
dwarfed, however, by external support because of sharing. Many times academics, industry
and government scientists and engineers from all over the world have improved our experi-
mental designs. These helpers external to the group have recommended new software or ways
to use existing software we were already using in the group. Open-source software coders
have improved programs, device drivers, and firmware to meet our needs 4 , and many others
have made specific recommendations and provided helpful advice on everything from com-
ponent 3-D designs in our experimental rigs to electronics in other experimental setups. 5 In
some cases, external supporters helped us correct errors and oversights in our write-ups be-
fore we started nonoptimized experiments, which saved us enormous quantities of resources
(time and money) by avoiding the need to repeat poorly optimized experiments. These be-
neits all came from massive peer-review and the fact that we actively shared. For example,
many of our examples of using open-source hardware and software have been viewed over
10,000 times. There is a lot of eye-balls looking for potential mistakes and better ways of run-
ning experiments!
We have also benefited directly from becoming involved in open-source hardware devel-
opment started by others. For example, many years ago, David Denkenberger and I worked
on a project to simulate the lowest cost method of using solar energy to provide clean and
safe drinking water in the developing world [ 9 ] . This theoretical work appeared promising,
and I built prototypes while away from the university (to both the horror and amusement of
my new wife the first prototypes were tested in our small bathroom, with a “solar simulator”
hanging from the shower rod and my sophisticated electronics seated on the 'throne'). These
prototypes used conventional (and expensive) heat exchangers and the whole apparatus func-
tioned well enough to provide drinking water easily for a typical family. However, to meet the
low-cost demands, we needed a highly effective, extremely low-cost heat exchanger, which
simply did not exist in the market. Years passed, but we finally developed one using a thin,
polymer-based expanded microchannel design for the heat exchanger [ 10 ] . The idea behind
this type of heat exchanger is that we make up for the low thermal conductivity of the poly-
mer by using extremely thin layers of plastic and thin channels. After enormous effort and ab-
surd prototyping costs, we got one to work with high efficiency using standard, black garbage
bags as the material! This was prety exciting, but to “rent” polymer laser welding time at the
rates that we made the first prototype (>$1000/sample) to do any kind of serious research and
development work would be cost prohibitive. At the same time, several open-source laser cut-
ter rigs were being developed. We derived our system from Thingiverse user Peter's (Peter
Jansen, a graduate student at McMaster University in Canada) design of a 3D-printable Laser
Cutter. 6 Taking this design, mounting an appropriate fiber laser, customizing a few parts of
the rig with 3-D printed parts (to be discussed in Chapter 5 ), and writing our own Arduino-
based control code ( Chapter 4 ), we had a polymer laser welder ( Figure 2.1 ) for the cost of a few
samples. 7 The development time for such a system with both the start and continued support
from the open-source hardware community (made possible by our resharing of the designs)
was a tiny fraction of what we would have needed to invest to develop the tool ourselves.
Now we literally save several thousand dollars a day every day we make multiple sample heat
exchangers.