Themes: embodied energy, lifecycle analysis, opportunities of open design in reducing
lifecycle impact.
Problem Statement
A key consideration in the sustainability of technology is the energy that it takes to manufacture it. The energy required to make the stuff we consume makes up a significant part of our total energy consumption, perhaps as much as driving or heating our homes: http://withouthotair.com/c15/page_94.shtml.
Open hardware presents an opportunity for exploring the lifecycle and embodied energy impacts of products as many of the items of the product design are open such as the component list, circuit and enclosure designs. Embodied energy information gathered for all the components in one open hardware product could be useful for another open hardware product that uses some of the same components or sub systems. Open hardware information could be key to re-use and recycling, to know how to take the product apart and what it contains and to be encouraged to do so. This challenge is to explore the embodied energy and lifecycle impacts of one example open hardware product which also happens to be an energy monitor (called the emonpi: there’s a youtube video intro here ) to understand and hopefully highlight areas where improvements can be made.
Identification of better sources that could improve the quality of the data used to calculate the embodied energy of the energy monitor.
A better understanding of other wide ranging factors to take into account.
A better understanding of work done by other initiatives in the same field.
Constraints
Available data and data quality
Understanding of how available data was obtained, what it includes.
Lack of open datasets?
Motivation/Barriers
Increased understanding of the embodied energy and lifecycle impact of an open hardware energy monitor.
To make a contribution to the conversation about the embodied energy and lifecycle impact of electronics.
To identify potential solutions that could reduce the impact of electronics and to identify key areas where improvements would have the largest effect.
if I’m buying an energy monitoring system, I would be interested to understand the embodied energy in the system. EROI question is a focus point because energy is key aspect of tackling climate change/environmental challenge. This is a natural question to ask for people who are already interested in using an energy monitoring system.
Just a quick thought in terms of practical outputs: what we’ve scribbled together towards the end of Sat comprises (in very draft but hopefully comprehensible form…):
“10 Environmental Opportunities That You Might Have More Control Of With Your Open Hardware Project”
(disclaimer that these are in vaguely chronological order of production process rather than importance of course - and at each stage you’re faced with similar issues of obtaining reliable and relevant data, then communicating it effectively to the end-user, who currently has little help in terms of evaluating a product’s enviro impact - no equivalent of “dolphin-friendly” or “certified organic by soil association”?)
Embodied energy (evolving rather than recognised standard - first on the list as it’s the one that @Trystan_Lea has investigated in most detail, expand with data here?)
CO2 outputs, use of other natural resources: water, conflict minerals [DG: I think https://www.provenance.org/ has just launched though haven’t investigated it further…]
Shipping - unexpected discovery that airmail eg to Australia may require as much energy as Trystan’s board takes to build in the first place (solution: can it be built locally from available components?)
Actual energy usage at user end (vs what does it save) - could even include cloud vs local storage if you’re producing a lot of data!
User’s ability to modify the device (ideally saving them the cost of having to buy other products)
User’s ability to repair the device (maybe not much different to modifying but…)
User’s ability to reuse/recycle the device at the end of its life
Perhaps controversial but: taking into account all of the above, does the world really need your entirely new product and all the associated manufacturing and distribution costs? Is there an existing Open Source hardware (and shared knowledge base) you could just help improve…?
10 (& upwards). Other equally important things that we haven’t thought of yet but we are sure you will…
NB: Please feel free to think of this as a basic checklist or a starting point for further discussions, each point can be expanded into all kinds of details and just as relevant are the solutions (and inevitable tradeoffs) that are available at each stage: eg using off-the-shelf components like Arduino has advantages in terms of possible reuse and global availability but at the potential cost of having less info about how the components were originally manufactured…?
another idea (on one of the smaller pages) was some sort of manifesto/pledge/“code of conduct” (like a CC/FSF licence?) outlining the manufacturer’s goals and as concrete a commitment to open source/circular ideals as they’re able to articulate?
Liverpool port to Hong Kong port 10,971 nautical miles: 20,318 km (1.02 kWh/kg)
Liverpool to Hong Kong by air = 9667 km (19 kWh/kg)
Liverpool to Berlin by road or rail ~ 1460 km (1.46 kWh/kg)
Liverpool to Berlin by rail ~ 1460 km (0.12 kWh/kg)
Liverpool to Berlin by air: 1086 km (2 kWh/kg)
Liverpool to New York by boat: 6,428 km (0.32 kWh/kg)
Liverpool to New York by air: 5,311 km (10 kWh/kg)
Liverpool to Sydney by boat: 24,918 km (1.25 kWh/kg)
Liverpool to Sydney by air: 17,036 km (33 kWh/kg)
The total weight of the emonpi open hardware energy monitor is about 900g.
31% of the weight (283g) is the ACAC Adaptor which we chose to ship by boat from hong kong to liverpool rather than fly, an almost 95% energy saving.
The emonPi kits are then shipped out with air mail. An emonpi shipped by air to Sydney will consume around 30 kWh. If the ACAC Adapter (a standard) was shipped directly from Hong Kong by boat to a distributor the air shipment from the uk would only require 20 kWh and the boat shipment from hong kong to sydney with the ac adapter (9549km) 0.5 kWh
Hey @DaveGreen - If want to explore use of Provenance, as mentioned they’re supporters of OSCEdays and have direct contact details. I’m also not quite sure what it can detail in terms of the supply chain.
thanks @TechnicalNature - yeah their site is a bit wide-ranging (& lacking obvious case studies) on what sort of provenance info they aim to provide : )
I tweeted them about widely-used PCB/ chip setups and they say
"Not yet, but we are actively looking… We will reach out to @Raspberry_Pi and @arduino"
Awesome initiative to start the quantification of the embodied energy.
At least the main microcontroller seems to be a little off though, binding less energy than the xtal. xtal vs usb connector seems skewed too. On average most of the values seem to be a little bit on the high side - I should read the cited papers on how they were obtained initially. For example, if the xtal would really embody almost 500 Wh of energy and we assume most of that to be electrical, the cost for that energy would be higher than the part cost at the consumer.
@Shu yes I think its the xtal that might be off as I think most of the weight is the metal casing which should have a embodied energy in the range of 50-120 MJ/kg rather than the embodied energy figure I used which is based on an average for surface mounted devices and LEDs of 2969 MJ/kg from the ecodesign methodology dataset. There is a quartz crystal in there that is synthetically grown but I dont know what its weight is as a proportion of the total weight… and what the embodied energy of manufacturing it is.
It discusses using the total weight embodied energy values provided by the ecodesign dataset vs calculating the embodied energy by estimating the weight of the silicon and casing separately.
