Wednesday 18 September 2013

Information is Beautiful

Why the solar thermal industry should embrace data logging

by guest blogger Ben Whittle

Who is this person that the Solarblogger has invited to wax lyrical on the rather un-sexy subject of data that you have never heard of before, I hear you ask?

Some might say my claim of authority to write on this subject is sketchy at best: I’ve only lived with and monitored a solar thermal system for less than 6 months, and I have only seen data from a few installations ever. But maybe that is exactly why it is a subject that needs discussing in the solar thermal industry… because I have been installing and designing solar thermal systems for around 10 years and I can count the number of installations I’ve seen that have monitoring on them on one hand, (and one of those installations is my own). And when I say we need more data-logging, I don’t mean turning on the heat quantity measurement function, I mean proper data-logging: measuring temperatures at the top, middle and bottom of the store, pump on and off times, boiler trigger times, heat measurement, the lot.

Over all the years I have worked in this industry I have seen a lot of mistakes made in designing and installing systems (and some of them were my own mistakes). How do I know they were bad designs and mistakes? Because I had to go and fix them. How can we possibly move on as an industry if we can’t try and make some sense of what mistakes were made in the past, and learn from them? Of course we have made leaps and bounds in some areas: the quality of the panels, plumbing, tools and fittings we are using (thank the lord for press fit pneumatic tools and filling pumps!), the MCS standards, the list goes on. OK, great, we can install stuff quickly and efficiently and it doesn’t leak… what’s the next step? Actually measuring how the systems are performing. Because I can tell you in the UK we have a long way to go in terms of getting our solar thermal systems to perform as well as can be achieved as seen across mainland Europe and America , and keeping track of performance is the only tool that can help us do it.

Now of course there are all sorts of reasons as to why we don’t normally achieve performance levels of a high quality Austrian installation… predominantly that is because the average UK house doesn’t have space for a 500-1000litre hot water store in the 700mm wide airing cupboard. But let’s put aside the things we can’t tackle and talk about the things we can.

To my knowledge there are only 2 or 3 solar thermal trials of any significance that have been published in the UK on solar thermal, and probably the best of these was conducted by the Energy Savings Trust three years ago. For those that are interested it’s called “Here Comes the Sun – A field trial of solar hot water systems”. It’s a great document, short (24 pages) and to the point. It tells you what you need to know – the biggest impact on solar thermal performance has nothing to do with evacuated tubes or concave reflector plates or low emissivity glass. But how did they reach those conclusions? By measuring things and writing it down – and, you know, analysing it and stuff.

And that is where we need to be going as an industry – looking at data, working out how to improve things, and changing our behaviour to suit. Not only will measuring performance allow us to understand the mistakes we make, it can inform our design decisions and help us to improve everything we do… and of course there are other benefits to consider as well.

1: Data logging is another chance to “add value” to your installation work. It’s a slightly more expensive controller, or an additional bit of kit to be sold, and a chance for your geekier clients to play with a spreadsheet or two. Now if you haven’t had a great experience installing solar thermal and the thought crossing your mind right now is “yeah right, and the customer is going to use this data to beat me with like a stick when it all goes wrong,” then I really am talking specifically to you. Because there are thousands of systems out there in the world that do perform absolutely perfectly, and as an installer you need to understand how they work.

2: It’s a chance to fault find, and take the correct action if things do go wrong, instead of guessing.

3: It will help you understand how people interact with their solar systems.

4: It can help get you out of trouble if the cause wasn’t your fault. One of the few logged systems I have seen was a school pool system that was constantly stagnating and losing pressure after being installed. When the data card was posted back to the company I worked for, it didn’t take long to diagnose the fault. The college maintenance man had been turning the system off at the main switch on a regular basis (even though it had a sign on it saying “Solar – do not switch off”). We were even able to re-program some of the settings using the data card and send it back to them in the post to plug back in. That’s quite an unusual feature specific only to the controllers made by Watts industries as far as I know, but a pretty handy one for commercial systems.

Just imagine what could be possible if we all took this a bit more seriously – imagine coming in to your office on a hot summer day in a few years from now after installing a 1000 systems across your local area, to find a couple of automatically generated emails where your data server had logged some de-pressurisation warnings at a couple of properties. You could then log on to watch some live system data to check them, and dispatch a maintenance team to fix them before your client was even fully aware there was a problem. All part of the regular maintenance contract you sold them at the time of installation…. There is no reason why this fantasy couldn’t come true, and it’s up to us to make it happen. It is already happening in the world of heat pumps.

As an aside, maintenance contracts are another area I think we need to explore further as an industry. I recently saw a report suggesting a very significant proportion of people who buy solar thermal systems would be happy to pay more than £100 for regular servicing, another possible revenue stream for any installation business.


Red=panel, Yellow = top tank, Blue = mid tank, Grey = low tank
R1 = solar pump on, Rs = boiler on
My own system does not have anywhere near enough data yet. I’m currently measuring temperature at 3 heights in the store, the solar controller also controls my boiler so I can also measure how frequently the boiler is getting activated, and I’m also getting estimated energy yield using the heat quantity function on the controller and estimated flow rates. In an ideal world I hope to add a water meter to measure my hot water usage, and a proper electronic flow meter for the solar controller to measure the glycol flow rates, and an electrical meter to measure electrical consumption. But I must be doing something right because I am currently hitting 2100kWHours of energy from my installation, after being installed for less than 6 months. And thanks to the EST solar study I know that the average UK installation is usually generating 1500kWH in a whole year… and I would never have found that out if I hadn’t bothered to log the data. Of course that generation figure could be inaccurate, and I mean to find out if that is the case by getting more and more accurate data over time.

I should probably qualify that performance figure by pointing out that I do not have an “average” UK system, so I’m comparing apples with pears… but I hope to explore that issue in a further blog post, looking into hot water storage, and how we might start improving solar thermal performance in the UK.