Can a Information Society decouple from CO2

via LinkedIn | Increased Industrialization needs more and more Energy. Solar and Wind and other Renewables are not ramping up fast enough to subsidy the decline of fossile energy. Even the best capacity estimations of our annual CO2 sinks will soon be overused only by the use of the internet.

The Graph above is from the 1963 so at least the facts that there are limits are known since the 1950’s. Human-induced climate change has been an issue of the scientific community since the 19th century (See:Arrhenius in 1895). Observation of how we alter and degrade our environment since the early days can be read in Platos critea, so it is not new and not solved up today.

Image 2: Page 76 Excerpt from Platos Critias from A New Green Histroy of the World

So we are running out of excuses why we are not acting.

We are aware that that there is a shift in need more highly qualified professionals and less of those who posses an undergraduate education in a similar field. But what does this mean to the amount of funds needed to educate them, allowing them to participate in the Techno Bonanza of automatic driving cars and household robots.

  • Top Down Energy Estimation Approach

Image 3: from “A Review of Top-Down Models of Internet Network Energy Intensity” Study.

When it comes to remarkable quotes in this Review I want to point out two aspects:

…aggregate energy consumption is doubled to account for cooling and power transformation overheads, which was not applied…*

*nor the energy to produce the vast amount of electronics.

…energy consumption of network devices is then extrapolated from the original values dating from the year 2000 to 2006 by applying an estimate of 14% of annual increase of aggregate energy consumption…*

Note: The Author of this study does not bring into account, that we have to produce the electronic entities. In my own estimation I tried to calculate the consumption of the internet with the energy which is needed to bring the information infrastructure into existance.

  • Buttom Up Energy estimation

My Buttom Up Approach was to try to get a certain estimation of how much energy is needed for one of the smallest IoT GPRS Module. 2,8 Volt, max 30 mA so 0.73584 kWh per year (2,8*0,03*8760/1000)

Also I took into account that with the possibility, technicans tend to use more sensors so the bandwith is fully used and there are no standby times. One Example will be OCR Sensors that are permanently checking if the customer may be in need of a certain product for example in the kitchen, at the office desk, or a physician’s Praxis-In this way, I could estimate an increasing energy consumption from the technosphere.

Dhoen below is my graphical interpretation.

Image 5: 3D Graph of IoT Calculation including Backend Energy and Embedded Energy

Do your own Online Simulation here:

In this Simulation I estimated 0,8 kWh per Gramm of Internet Hardware. After a Peer Review of several papers I came to the conclusion that this is the most accurate Estimation.

  • Object Orientated or System Dynamical Approach

When we see the Entrophical Attitude of all things we can accept that every increase in Entrophy systems tend to create new entities or information contents. Or when mone is nothing more than a information we can assume that information has his equal representation than money has. So when we accept Tim Garret approach of …value of 9.7±0.3 milliwatts per inflation adjusted 1990 dollar: in 2008, 16.4 TW of power supported 1656 trillion 1990 US dollars of civilization worth…

I estimate a price from 5-100 Dollar for one IoT Entity and we will soon have Billions soon and the estimated price for 500 MB per month of data costs about 3-12 Dollar we can roughly estimate a energy consumption of about 0,8 kwh per IoT Gprs Module quite the same like the other approaches.(0,001W*8(Dollar)*12*8760\1000)

Excel implementation of the Visual Quadrant – Net Graph based on the Formulas. When playing with the hours and the data it becomes clear that it is important what we do with the produced data. When we let to plow the data from endless AI for different purposes the energy usage will increase. See also later the youtube example 1 MWp new PVe per year only for new Videos. Not included the Electricity for watching, Big Data or Artificial Intelligence Usage.

Dirty Energy and dirty Ressources fueled on false assumptions

Peak Oil is dead viva la Peak Oil: What most scientists have underestimated is the evolutionary will to live on the MPP(maximum Power Point) The Centers are in a urgend need for new applications to keep growing their influence. More and more empty rooms used only from robots to keep the fridges full and rooms warm when the highly skilled human comes home.

Since the ancient usage of oil and other minerals we see a decline in their purity. So we need more and more energy to achieve get steel for new and expanding infrastructures like roads or railways or harbors. So the EROIE(Energy Return of Invested Energy) of the whole technological upset is marginal sinking. On the other hand we increase complexity with more and more applications payed from a ever monetary wealthier expanding human population.

All these effects leads to a expantion or exploition of the boundaries of the civilation. Where the technosphere ends and the biosphere begins humans are forced to extract everything what can support the every powerfull center needs. The Center with their Information Entities or the Tokken of Money, is the clean and abstract fuel that drives the dirty extraction of the biosphere.

Google’s YouTube real Power Usage Example(many Dimensions of Energy Consumption of Data-Information)

After all these estimations a short unanswered Tweet from me, that I can not believe that even YouTube is sustainable from a energy balance callculation point of view.

Σ = 525.600 Minutes has a Year: We upload 400 hours or 24.000 min of Video with 1080p and 30 fps for 200 MB per Minute so we get need 12614400000 MB per year new storage or 12614400000\1000000 = 12.614 TB HDD á 600 Gramm(inlusive some overhead like cable etc) = 7568,4 kg of Hightech Material has to be added only for new Uploaded videos – exclusive cooling or OS or renundancy for faster access. so only 7568,4 kg * 280 kwh/kg – 28 MWh/kg = 21 GWh – 21.191 GWh for the material to processed or 1 MWh for one kWp Photovoltaic or up to 21.191.000 kWp or between 211 kWp and 21 GWp has to be added to support a renewable production of the storage media each year has to be added to YouTube. Not brought to account the

  • cooling,
  • AC/DC conversion losses,
  • system dynamic cost of invested money
  • many layers of internet(see Top Down model)
  • and the electricity to power these new storages(seagate estimates a need of about 0,6-8,8 Watt*8760/1000 = 70kWh*12.614 new harddisks or 882.980 kWh about 1 MWp new PV added per year only the harddisks for new videos without cooling and redundance storage or mainboard or OS etc..) so summa summarum the estimation of how much we need to watch video is done with only 2600 MWp Wind and Solar is totally naive or stupid.

Rebooting Energy Demand: Automatic Software Upgrades

Excerpt from

[…]Data traffic over the internet is growing dramatically, roughly doubling every two to three years. Because the internet infrastructure becomes more energy efficient over time, the electricity use associated with this data traffic is growing a bit slower, doubling roughly every five years. According to the latest estimates, the internet now uses between 5 and 10% of global electricity, and forecasts see further growth for many decades to come. [1]

Crucially, the growth of data traffic and energy use is not so much due to an increasing amount of internet users. Rather, it’s a consequence of growing data use per internet user. For example, in the UK, average monthly home broadband data traffic rose from 17 gigabyte (GB) in 2011 to 82 GB in 2015, an increase of almost 500%. [2] Over the same period, the number of internet users in the UK grew by only 10%. […End of Excerpt]

Need of ‘Smart’ IoT for Microgrids or Energy Conservations

With the rise of new Possiblities to integrate variable Energy Sources like Biogas, Wind and Solar Energy comes new Hope. Basically to reduce the effort and costs of microgrids the Designing of the Systems made huge Progress. The Key Question would be if we can save enough money or efforts or ressources to scale up fast enough to provide the world with Electrictiy fast enough to keep climate warming in acceptable parameters.

Basic Introduction to Microgrids or Hybrid

To design a varibale Integration of various Energy Supply you need different Kind of Intelligences. For a massproduction you have to ensure to integrate various needs into a modular system. Most of the producer can integrate on the consumer side high power. This has adavantages and with Load Shredding and Loar Trimming(Lastenabwurf) and a realtive massive Storage Solution it can be feasible. On the other hand we have lost time in the transition and we maybe have to find methods to offer solution with minor storage. I tried to figure out what it will cost to go Hightech with ‘Smart’ Consumers Goods connected to Clouds and Data Warehouses etc. On the other hand I tried to figure out ‘Dump’ Devices with amore astonomical aproach, when the high power device like a laundry mashine can do the work. With the assumption that Microgrids and Hybrid systems will have Photovoltaics as the major Energy Source. I do not go into more high sofisticated Applications with more Parameters like Frequency monitoring weather forecasts and seasonal Inputs etc. I assume that these applications are too complicate, custom and energy intensive(see above) for Hybrid Offgrids or Microgrids.

Picture from the SMA Off-Grid Configurator

  1. Power Shredding with Internet of (Smart) Things

A negative Feedback Loop for the more technical or economic orientated approche to solve the ressource, energy, sinks problem of our civilication is that we still want to believe in technological solutions. So will the model of renewable energy be good enough to be adapted fast and integrated into the economic systems. So we underestimate the huge support of the former industry revolution. The development from the first coal powered steam mashine to the internet was extremly long and supported by the enormerous capabilities of nature sinks like the atmosphere, soil and sea.

So our basic idea that when some logic or compatible determistic conclusion will lead into a better solution. This preassumption for academic work is maybe more a wishfull thinking. As a species not able to do best for itself is maybe deeply entrophical. This programming will it be overcome with more intelligence in a causal conclusion.

End User Devices in a Smart Grid and their Energy Consumption

I tried to estimate from mainstream promoted worldwide Households based on a ‘Smart’ Meter Region(Water, Gas, Electrictiy, Heat,Cold, Solar Energy, EV Charging, Water Conservation) the Gartner Group estimates 26 billion units installed in 2020 representing an almost 30-fold increase from 0.9 billion in 2009:

Graph from 2009-2020 increase of increased measuring and complexity for so called worldwide smart city:

I tried to publish this 3D Graph but it is not supported by Google here a try:


We will see more and more entities coming into the grid and will be objectaized and measured and big data. We see a complete oposide development of how clean and lean software and hardware design should be. We try to store and process more and more data with more complex requirement.

Modelling a standard household or office on a daybased and radiation based energy needs maybe we will be able to optimize with astronomical data the usage of battery storages. Also to reduce the Hybrid Usage of external energy Sources, either we model a Offgrid or a small Scale Electricity Grid. Ideal would be to find and develop super small relative dump Grid Watchgards for variable Users.

A very common Example of Power Consumers that could be shifted easy with super small not Smart Grid connected astronomical timed Watch Guards are all Kind of Heat and Cold Producing consumer on the private or commerce sector. Most Refridgerators or Building Cooling Devices can be shifted into a more solar active daytime with realtive simple Calculations. Most Heating Devices like simple Waterheater have small Storages and can so be adapted to a more Load Shedding Oriented Paradigma.

Despite of the common mindset, we will have limit in exploring minerals and ressources for a more and more chemical storage oriented Energysphere.


When I explored the connected solution to monitor I want to find a more unconnected solution. So we have to try to measure different Algorithm of how much the different kind of approaches need of processor power(FLOPS) and if we can so calculate, measure and compare these two Approaches. Or in other words how much does complexity costs in energy and waste(CO2). Exspecially when we compare it to complex interconnected systems via a (energy and processor capacity)costly encryption System like Blockchain.

Relevant Calculation for alternative Microgrids where often Bitcoins are mentioned as the currency of choice.

When I calculate for one BTC = 1000 Dollar on 2. December 2017 and I use a Hash Rate (Gh/s) of 1000 and the mining Hardware uses about 650 Watt I need with common complicity Hash Rate (Gh/s) 1000 and a 650 Watt mining Hardware it takes me 3,46021 years to mine 1 BTC(source 1,2). So one BTC has used up to 30.311,42 kWh or in other words to encrypte one Dollar it takes 30,31142 kWh of electricity. To store and distribute this digital currency is not calculated.

When I use the figures from bitcoinwarsz we get 1 BTC in only 362,9047 days with 1000 Gh/s or with Network Hashrate: 2,717.37 PH/s Block Reward: 12.50 Blocks: 446,421 Block Time: 10.00 minute(s) it takes 133 days / BTC or $6.24 for electricity or 62,4 kwh per day, only 8246 kWh for one BTC.

So when we assume the figures from Bitcoinwatch:

1 INTOP = 2 FLOP 1 hash = 6.35K INTOP   1 hash = 12.7K FLOP So the hashrate in TeraFLOP/s is simply 12.7 times the hashrate in Gigahashes/s. As an example: 11,558.55 Gigahashs/s * 12.7 TeraFLOP/Gigahash = 146,794 TeraFLOP/s = 146 PetaFLOP/s or 12,63 PetaFLOP/s per 1000 GH/s or 12630000000000000 FLOP/s at 650 Watt per Second is 0,65 kWh in a Hour and 45,468 PetaFLOP/s or 69951,108492 PetaFLOP/s per kwh.

assume 0,08 Dollar Cent per GigaFlop/s or Hardware Costs of 10.104 Dollar for a 1000 GH/s Hardware is quite accurate with the 3000 Dollar Hardware Estimation from this Bitcoin Calculator, when you have different Operation Systems etc. So to assume a livetime of 5 years and a consumption of 28470 kwh later we generated 1,4454935 BTC at the current difficulty.

So it is clear that a highly complex information technology based Microgrids is a victim of the ever increasing complexity love of geeky people. Maybe we can explore and compare more simple Algorith and methods to keep the quality and stability of electrictiy high in future.

The more complex operations have to be made the more hardware is needed the more energy to mine, produce and power. Here a great graph of the real great differs a simple design can make. Picture Source:

Also when it comes to the size of data or for example identification of devices the unique identification number shoould be choosen carefully. There are significant differences in the usage of the processor so FLOP/s is declining massively. Another important fact is that we the Law of Moor has reached his physical limits so there will be no major performance increase. Picture Source:

From Motherboard: A Single Bitcoin Transaction Takes Thousands of Times More Energy Than a Credit Card Swipe


Artifical Intelligence and Mashine Learning

Trillions of Sensors and a designated Outcome from the view of a Utility or other Entities, will lead to more and more Artificial Intelligence or Neuronal Networks. The Basic Concept of Neuronal Networks or Machine Learning is incremental. We have Vectors or Input and try to figure out which one is the most common Outcome and this with Iterations or Loops to benchmark the best one. So the learning Mashine either it is a learning ‘Smart’ Grid or learning Algorithms which are optimizing your Advertizing Budget. Into Infinity competing Algos and Neuronal Networks using more and more energy to ‘learn’ – more or less a sorting of optimized Input to Output Patterns. We have seen this investment boom in Data centers over years in the Automatisation of the money trading or financial industry for years and a wettrüsten – arms race between the different actors. The same is now happening in nearly every sector of our life, from neuronal network for the health sector up to AI for cars to make driverless mobility possible.

public List<int[]> Combox(ListBox listBox4)
            var l = new List<string>();
            var data = listBox4.Items.Cast<object>().Select(i => i.ToString().Split
            int ii = 0;
            foreach (var item in data)
                string a = "";
                var lx = new List<string>();
                for (int i = item[0]; i <= item[1]; i++)
                    a = "_" + i;
                    if ((l.Count <= item[1] - item[0]) && ii == 0)
                        foreach (var itemx in l.ToList())
                            lx.Add(itemx + a);
            return l.Distinct().Select(i => i.Split(new[] { "_" }, 

Source Code from:

Even if as with photovoltaics an almost infinitely available energy we have to critical ask us about AI. Actually at present, considering the central bank’s policy of possible free funding, free energy. Even from this point of view, the raw materials and the space for these power plants must not be ignored and thus the artificial intelligence (AI) is to be restricted see below. Furthermore, each ‘worked out’ dollar or EUR system-dynamic costs according to Garrett (about 1 dollar – 0.1 kWh or 0,01 Watt x 365 days)

In addition, as is the case with Bitcoin Mining, self-propelled cars, as well as Quant Trading, are processing inputs to achieve an optimal result. As shown in the graphic above, the decision is made in the ‘Hidden’ layer. For neural networks (AI-BI) it means to go through ‘lists’ again and again and use the best results. This is how the machine – the automatic trader – learns the best results. But exactly this design principle needs exponential more and more computing power and memory thus more and more (equal) current. Furthermore, this ‘learning’ process no longer ceases. It will become self-sufficient, the vending machines will become empty spaces and empty cars will be used for self-help: ‘learning’ of cars and maintenance robots or promotional, PR-AIs will watch us to ‘learn’. Due to the competition in the relevant ‘market’, the income and benefits are becoming ever lower and the effort is ever higher. It is as if more and more fishermen are fishing with better nets – for linear thinkers as we are – Paradoxical answer comes from fishing: less fishing increases yields.

Low Tech Approaches to Load Shredding and Trimming

Load Shedding can be also done with a more low tech aproach, despite we can ignore that we are in a deep overshoot. Arguments to keep on exploring ways to continue our live in this cosy Energysphere can be that a humankind will always try to maximize its Power so they will take evey easy accesible ressource including the most Dirties among them. So small Devices to regulate the local use of High Power Devices can help to keep the complexity low enough.

Not so Smart Load Shedding device:

Essential to receive the calculate your Latitude (+ to N) and Longitude (+ to E) and your Time Tone (+ to E) also your Local Time (hrs) and your Date.

a) Excel Calculatation

For a basic calculation in Excel of your local Sunrise Time relevant to the date and location based on the NOAA Excel Sheets:

Calculate your Sunrise degree: DEG(ARCCOS(COS(RAD(90,833))/(COS(RAD($B$2))*COS(RAD(T2)))-TAN(RAD($B$2))*TAN(RAD(T2))))

$B$2 is the Latitude of your loaction.

With this result you can calulate your Solar Noon:=(720-4*$B$3-V2+$B$4*60)/1440

$B$3 is in this example the Longitude (+ to E) and $B$4 is the Time Zone(+ to E)

After calcuating the Solar Noon you can do: =(AL2*1440-AK2*4)/1440 to receive the Sunrise Time.

The Sunset Time is equivalent: =(X2*1440+W2*4)/1440 etc.

b) Calculation via Code

It can be made in relative simple Coding here some Snippets in C to calculate the maximum possible solar Radiation. This Code can be widely acceptable to very simple and maybe lean enough to be implemented into a ‘sun’ based power switch.

// Vereinfachte Sonnenstandsberechnung

void sun(float jd0, float dt, float la, float br)
  float j,

  j = jd0 + dt;
  l = degr(4.894950420 + 0.01720279239 * j, M_2PI); //Postion der Sonne auf der Ekliptik
  g = degr(6.240040768 + 0.01720197034 * j, M_2PI); //mittlere Anomalie
  a = l + 0.03342305518 * sin(g) + 0.0003490658504 * sin(2 * g); //ekliptikale Länge der Sonne
  e = 0.4090928042 - 0.000000006981317008 * j; //Schiefe der Ekliptik

  rekt = atan(cos(e) * sin(a) / cos(a)); //Rektaszension
  if(cos(a) < 0)
    rekt = rekt + M_PI;
  dekl = asin(sin(e) * sin(a)); //Deklination
  t0 = jd0 / 36525.0; //Tageszahl in julianischen Jahrhunderten
  sh = degr(6.697376 + 2400.05134 * t0 + 1.002738 * dt * 24, 24); //mittlere Sternzeit
  t = sh * 0.2617993878 + la - rekt; //Stundenwinkel der Sonne bezogen auf den Ort

  mz = cos(t) * sin(br) - tan(dekl) * cos(br); //Nenner Azimut
  rz = atan(sin(t) / mz) + M_PI; //Azimunt wird von 0° Nord gezählt
  if(mz < 0) //wenn Nenner vom Azimut kleiner 0 dann den Winkel in den richtigen Quadranten bringen
    rz = rz + M_PI;
  azi = degr(rz, M_2PI); //Azimut reduzieren

  el = asin(cos(dekl) * cos(t) * cos(br) + sin(dekl) * sin(br)); //Elevation
  kr = 0.0002967059728 / tan(el + 0.179768913 / (el + 0.08918632478)); //Korrekturwert für Refraktion
  elev = el + kr; //Elevation einschliesslich Refraktion

More sofisticated Methods would be a picture interpretation of the sky if there are clouds and the storage can be loaded proper without trimming high power usage applications. These Algorithms have the same danger embedded to generate with more APIs to produce too much data and so a rebound in energy use is highly possible.

c) Mechanical example of a Astronomical Watch

d) Gradual Frequency dependent Power Reduction (GfdPR)

Most Inverter-Charger manufactures use frequency sweep to protect batteries from beeing overcharged. Therefore aka Fronius Inverter use aa special P(f) function to reduce their power to zero, if the frequency goes up. Similar to this power supply aproach from Fronius we can guess that there must be a load consumer based side. I estimate this kind of specialized ICs for laundry or cocking mashines for about 4 Grams or 0,44 kWh equally to a gsm module for a wireless monitoring.

Schematic depecting GFdPR strategx to ensure secure energy supply Source:

e) Behaviour driven Load Shedding

When we look on the Formula from SMA Offgrid configurator we can assume that we can maybe remember our Rhodopsin also known as visual purple to trigger our high power appliances to be more respect of the analemma(∞) figure of our energy source number one the sun. In other words to live more a life to the rythm of the rising and sinking sun.

f) 1 ) DC Grids without power Storage

Without going to much into the Detail I want to mention the Studies of regional Advantages not to convert DC/AC/DC to do some Information Processing to optimize the Conversion of Solar Energy into Work or Entertainment. The following Graphic is self explaining. You even do not need any chemical or mechanical power storage. When you go more deep into the customer needs one of the most essential applications human animals need to exist is the processing or cooking of food. This has a long history of fermentation and storage so this can be easily done saisonal.

DC-grid demo installation with solar support and an AC reference grid. Grid voltage: 380 VDC, Solar modules: 2 kWpk, LED lamps: 56 x 37 W, adapted LED drivers, Reference system: 230 VAC Source:

f) 2) A battery for Austria – would require worldwide lead production

24/7 day transition is not possible without loss of comfort – or the lead peak is behind us

Positive thinking and hope can turn into a drug, and we will not be able to keep our fingers off until the collapse (drunken on fossile fuels). The US peakers speak of Hopeism, which spills us out day by day in the media. Here is an old already often stressed example of Prof Tom Murphy about a nation-wide battery (nation-sized battery) which can store the energy for a day after the complete conversion to Renewable.

Here are my calculations for the nationwide battery: We need about 1458 PJ according to Wikipedia to maintain our current comfort. When we convert this to GWh, there are 404,906 GWh or 404,906,000,000 kWh per year, or 1,109,331,506,849315 kWh per day.

If we assume 15 kg of lead per kWh (kwh to Ah battery calculator depending on the volt) stored energy (other energy stores such as water we leave once for the thought experiment outside), we probably need 16639972602.73973 kg lead for a day reserve or about Four days at 25% performance anyway. At the current lead price (EUR 2.2 per kg), EUR 36607939724.4 or EUR 36.6 billion or 7.3 billion each year at 5 years of shelf life or approximately 2% of Austria’s GDP.

16,639,972,60273973 tonnes every 5 years would be 3327994,520547946 or 3.3 million tonnes per year for Austria. With an annual production of 4.1 million tonnes, it is difficult to imagine that prices will remain so favorable in the future. The global reserves of 80-1,500 million tonnes are also not exactly lush. Since the 33 million lithium reserves will not help, because others want to surf the Internet in the evening.

However, it is an illusion: more and more comfort at any time of the day or at any time of the year can be made available to everyone anywhere in the world and not anywhere in the world. According to the profitability curve, the technologies or things that have come just before the collapse, thus the least benefit to survival, are the first to go again. This includes the Artificial Intelligence Neural Networks, which are dissipated by up to 1000 layers of energy, by Mark Zuckerberg and Co., which will recognize this fact in the singularity and, either by the terminator analogy, warns or turns itself off. Ups forgot how it looks, we do not need the detour via a Skynet …

Economic Impact Smart Solutions

coming soon…


Limits for Humans are helping them to overcome their evolutionary programming.

Doing more and more Improvements in a existent Pradigma will not make a big different.

Orientating us in Million old Crowd Intelligence Systems and creating a „virtual“ central Intelligence will help more than Millions of expensive Central Aproaches.

Repeating the errors of Top Down constructions of Small Scale Grids will not solve the wastefull usage of energy. To overcome the Entrophical usage of Power the Solution must be radical and from the Bottom or more Basic. The Attractivenes to use certain high Power Applications has to be limited to avoid Rebounds and wastefull usage of Ressources.

to be continued…

Please follow and like us: