May 3, 2023
0 min
Lawrence Dudley
Director (IoT & Technology)

Using IoT to become carbon negative

Lawrence Dudley
Director (IoT & Technology)

At Parallax, we have recently earned carbon negative status (meaning we remove more carbon from the atmosphere than we emit). This is a huge step forward for us in terms of reducing our impact on the environment that we’re really proud of. In order to achieve this we’ve managed to massively reduce our energy consumption (cutting our emission output by almost 50%!) by leveraging our IoT expertise. Read on to find out how and why we did it 👇

The why

Doing things the right way has always been a core focus for us at Parallax. This includes thinking about our impact on the environment.

At the end of last year, we surveyed our team and found that they really care about the impact we have on the environment as a business, with 85% believing it to be a 7/10 or higher in terms of importance in our guiding principles.

On top of this, the recent energy price crisis has had a huge impact on people and businesses across the world and has caused significant harm in a lot of ways. But as with every cloud, there is a silver lining. Higher energy costs makes finding ways to reduce energy much more cost-effective. Suddenly initiatives that had a return on investment measured in years started seeing returns measured in months.

With such a clear importance for our team, a huge financial incentive, and as a way to improve our environmental impact, a concerted effort to address our environmental impact was needed.

The how

Bringing IoT into the equation: Digital Twin

At Parallax, we develop a whole range of Internet of Things solutions and products for our clients (ranging from consumer electronics to industrial control systems for street lighting) so it seemed sensible to use this knowledge to optimise the way we consume energy ourselves.

Real-life monitoring using tech we already know or can repurpose let us build a digital twin of our consumption and environmental factors like indoor air quality to model improvements ahead of implementation.

Step 1: Understanding our energy usage

We started by analysing our electricity consumption, down to the individual circuit level and discovered some surprising results:

  • Heating (in winter) and cooling (in summer) are by far the biggest sources of electricity consumption
  • Lighting came second. Even LED lighting can be surprisingly energy intensive in an office environment as the lux levels tend to be quite high

When we analysed the energy consumption from our heating systems we found that while they were timed to come on/go off at preset times of the day, people would override those settings because they were too cold or too hot. This would prove to have a costly impact on our consumption.

Less control, less efficient

Using standard HVAC controls leaves little control over manual intervention and has some unexpected consequences.

A good example of this is our office cleaner. She’s great, she’s worked with us for years and she does a fantastic job. But sometimes she comes in quite late and it’s cold, so she turns the heating on. So far so good, we wouldn’t want her to be cold!

The problem is, if it happens on a Friday, the rudimentary timers on the heating system don’t take that into account, so when the heating is turned on, it then remains on until it switches itself off on Monday evening.

Historically, that would have been a drop in the ocean in real financial terms but the effects of these inherent inefficiencies now add up to quite a substantial amount of money. In this case, the energy spent amounted to around 180kWh or about £115 if you’ve managed to negotiate a good rate for business energy. If you haven’t, double it.

Step 2: Improving our systems

There are commercially available building management systems but they tend to be quite inflexible, difficult to retrofit into a building that isn’t designed for it and also suffer from extensive vendor lock-in which can be impossible to overcome cost-effectively in an existing installation.

Not deterred by this, we set out to develop some technology to reduce our HVAC and lighting bills in a way that doesn’t negatively impact people (i.e. they are always comfortable!) while reducing the amount of energy we consume and as a result, the impact we have on the environment.

By replacing the PIR systems used to control the lighting in our office with smart sensors that could control the lighting as well as feed that information back into a central system, we effectively implemented multipurpose presence detection that lets us make our decisions much smarter around when we cool, heat and ventilate.

Rather than cool and heat based on a static timer, we now do it based on whether there’s anyone in our office who needs a temperature controlled environment. Sounds obvious when you think about it.

We also integrated CO2 measurement using sensors we’re using with local authorities for air quality measurement in combination with presence detection to enhance air exchange so CO2 levels are dynamically controlled using variable speed fans rather than a constant, worst case scenario level of ventilation.

This matters because working patterns with a predominantly hybrid team are much more variable than they used to be. Sometimes we have 50 people in the office, sometimes we have 4. When there’s 4 people in, we don’t need to blow 50 people’s worth of hot air out of the building to keep internal CO2 at a reasonable level.

Generic ESP32-based Microcontroller Used for Ventilation Control
Generic ESP32-based Microcontroller Used for Ventilation Control

While we did investigate off the shelf solutions to this problem from manufacturers, none of them were cost-effective or really very good in practice (they’d crash or not work at all) and so we developed a system based on an ESP32 controller to react dynamically to the current situation in our office.

We’ve built solutions for customers before using ESP32 microcontrollers and they’re a great platform for R&D as they’re both well-supported in terms of ecosystem as well as being readily available despite the current chip shortages.

Proper integration of the various bits of technology that we already had has led to some quite significant reductions in our overall electricity consumption, with our monthly electricity bill reducing by almost 50% with a corresponding reduction in the level of emissions.

Step 3: Finding reliable carbon off-setting

Finally, we commissioned an analysis of our remaining carbon emissions (from travel, working from home etc) and found some really good carbon offsetting programmes to invest in. While these can be controversial, we mitigated this by:

  • Investing in programmes that are easily verifiable (i.e. renewable power generation rather than difficult to measure forestry programmes in far away places)
  • Being very selective about which programmes to back rather than going for cheaper, harder to validate schemes
  • Offsetting twice the carbon emissions we emit so even if they’re off by 50% we are, in an absolute worst case scenario, still carbon neutral

While we don’t want to underestimate what’s involved (and some of it is genuinely quite complicated!), it’s been surprisingly easy given the skillsets we already had in-house to achieve a brilliant result both from a financial as well as an environmental perspective.

Our next step is to look at installing solar panels on our office to reduce our net energy usage even further.

Sound interesting?

We're commercialising some of the technology we developed for this use case as we think it will benefit others. If you've got a large commercial utility bill it might make sense to have a conversation around how we can help.