Embodied Carbon costs of future PV installations in UK

[Hastings]

Sobering article by Circular Ecology, UK based experts in things like embodied carbon in building construction (provided the data to carbon footprint the construction of the London 2012 Olympics), that points to the idea that in the UK we are close to the point when new PV installations may never recoup their embodied carbon cost, let alone contribute to any reduction.

 

Quote

"According to the IEA (2015) a cadmium-telluride, CdTe, based PV system would have an embodied carbon ~63% lower than a monocrystalline PV system... With a slightly lower efficiency it needs a bit more space per kWp.

 

http://www.circularecology.com/solar-pv-embodied-carbon.html#.XinkGhfLcU-

 

So for anyone installing PV with the primary intention of carbon emission reduction, we maybe are not currently being offered the appropriate type of technology in this country.

[SteamyTea]

2 hours ago, Hastings said:

Sobering article

On the face of it, this seems to make sense.

Except, as we decarbonise the grid, so the products made using grid energy get decarbonised.

PV plays a large part of this decarbonising, as does wind.

So comparing something manufactured a few years back with a much higher embodied carbon and energy content, with what is happening today is rather bad accounting.

[willbish]

Interesting article but the example of the office block with a 1,140kWp PV array is barking. If I'm not mistaken that's over 3700 panels, what sort of office building would have space for that?

 

 

[SteamyTea]

4 minutes ago, willbish said:

Interesting article but the example of the office block with a 1,140kWp PV array is barking.

Yes

There are some limitation to the research at the bottom of the webpage.

I also think the estimate of energy use is rather high.  A modern office does not have a CRT monitor attached to a 300W desktop PC.

Or old fashioned lighting.

[Ed Davies]

Next I suppose we'll hear arguments for not using nuclear or wind either because the grid's decarbonizing anyway.

 

Why does PV have to have any embodied carbon? I.e., if you fully decarbonise your energy supply don't you also fully decarbonise your PV manufacturing?

[MikeGrahamT21]

I think embodied carbon is the least of the problems for Solar PV and the rest of the Electrical sector...

 

https://en.wikipedia.org/wiki/Sulfur_hexafluoride

[SteamyTea]

Got to put into perspective:

 

Greenhouse gas

Mauna Loa sulfur hexafluoride timeseries.
According to the Intergovernmental Panel on Climate Change, SF
6 is the most potent greenhouse gas that it has evaluated, with a global warming potential of 23,900[28] times that of CO
2 when compared over a 100-year period. Sulfur hexafluoride is inert in the troposphere and stratosphere and is extremely long-lived, with an estimated atmospheric lifetime of 800–3,200 years.[29]

Measurements of SF6 show that its global average mixing ratio has increased by about 0.2 parts per trillion per year to over 9 ppt as of February 2018.[30][31] Average global SF6 concentrations increased by about seven percent per year during the 1980s and 1990s, mostly as the result of its use in the magnesium production industry, and by electrical utilities and electronics manufacturers. Given the small amounts of SF6 released compared to carbon dioxide, its overall contribution to global warming is estimated to be less than 0.2 percent.[32]

In Europe, SF
6 falls under the F-Gas directive which ban or control its use for several applications. Since 1 January 2006, SF
6 is banned as a tracer gas and in all applications except high-voltage switchgear.[33] It was reported in 2013 that a three-year effort by the United States Department of Energy to identify and fix leaks at its laboratories in the United States such as the Princeton Plasma Physics Laboratory, where the gas is used as a high voltage insulator, had been productive, cutting annual leaks by 16,000 kilograms (35,000 pounds). This was done by comparing purchases with inventory, assuming the difference was leaked, then locating and fixing the leaks.[7]

[Jeremy Harris]

12 hours ago, SteamyTea said:

I also think the estimate of energy use is rather high.  A modern office does not have a CRT monitor attached to a 300W desktop PC.

Or old fashioned lighting.

 

The new office and lab build that was part of my last job before I retired housed ~900 people, all with keyboards and screens on their desks.  The cooling system requirements, had we opted to install desktop PCs, would have prevented the building from being able to meet the energy target we were aiming for. 

 

The solution was to just have a small, very low power, thin client box on each desk, together with a low power monitor.  Each user had a virtual machine running in our two data centres.  Apart from being a lower overall power solution, it also had the advantage that users could "boot" their desktops near-instantly.  As soon as they plugged their security pass into the slot in the thin client, their desktop appeared on their screen, just as they had left it the day before.  Pulling their pass out of the thin client box shut the screen down immediately and paused their VM.  Great for security, as no data is stored at the user's desktop at all, there wasn't even a USB port available (for security reasons).

 

This was over 10 years ago, so there are probably better solutions available now.  Certainly power consumption has markedly reduced, for a given amount of processing power, in the past few years.

[MikeGrahamT21]

1 hour ago, Jeremy Harris said:

 

The new office and lab build that was part of my last job before I retired housed ~900 people, all with keyboards and screens on their desks.  The cooling system requirements, had we opted to install desktop PCs, would have prevented the building from being able to meet the energy target we were aiming for. 

 

The solution was to just have a small, very low power, thin client box on each desk, together with a low power monitor.  Each user had a virtual machine running in our two data centres.  Apart from being a lower overall power solution, it also had the advantage that users could "boot" their desktops near-instantly.  As soon as they plugged their security pass into the slot in the thin client, their desktop appeared on their screen, just as they had left it the day before.  Pulling their pass out of the thin client box shut the screen down immediately and paused their VM.  Great for security, as no data is stored at the user's desktop at all, there wasn't even a USB port available (for security reasons).

 

This was over 10 years ago, so there are probably better solutions available now.  Certainly power consumption has markedly reduced, for a given amount of processing power, in the past few years.

 

Same solution still exists, but has moved on a little, there are now things called Zero Clients, which is basically just an HDMI adapter which plugs into a monitor and a network port, powered by PoE, and they tap into a backend VDI solution, so probably saving another 30 or 40W of power over the thin client solution. Zero clients have no Operating System, just a set of tools to be able to connect into differing infrastructures.

 

"Another advantage of zero clients is their lower power consumption. Thin clients have mainstream CPUs and often graphics processing units, but a zero client usually has a low-power CPU (or none at all), which cuts down on power consumption and heat generation."