some measurements from that hall –

So, we’re snagging a church hall, and so far, we’ve got plenty of heat input and output, and a boiler running at 90C under ineffective control.

Here’s some of the views. First up, we have the answer to the question: how quickly does it warm up?


The answer is “wow fast.” Around £1.70 worth of gas covering this and a smaller modern system, by the way; that’s 4.36 m3 gas. And 9 kWh of electricity (!!!), around £1.08. The fan convectors are small (I think it was 46W) so they have a leak somewhere; if it hasn’t stopped by the next visit I’ll need to investigate.

I reckon, extrapolating from this curve, that the space is pretty much guaranteed to reach the Child Care Commission minimum of 16C in an hour under 99% of Edinburgh weather conditions; it was -4C out the night before. Just for comparison, in 2010 and 2011, the heating in the first week of December used to come on at maybe 2 am and still not get the space past 15. It does raise the question of how on earth the space was 12C before the heating came on; we know it’s not a hall that retains heat very well, and nothing about this particular refurbishment will have changed that much. Either there’s some frost stat arrangement we’re going to need to know about (and probably change), or, more likely, there’s been massive amounts of heat going in recently. That’s often the case just after a new system goes in, especially if the building is in use; on previous sites we’ve seen the heating controls lag the boiler installation by months. In this case, there are controls, but also anecdotal evidence that they aren’t very effective: spot readings of 22C for a thermostat set to 18, and user groups complaining they’re too hot. The graph shows the thermostat reads way low, so the next question is: is that instrumentation, or location?


It appears to be instrumentation; a Lascar strapped to the thermostat reads very differently. However, I’m surprised that Lascar, being right next to a massive cold stone wall, doesn’t read differently from the other Lascars. As Dimitri taught me at one of our Innovative Learning Week events, instrumentation is easy, it’s the calibration that will kill you every time. Except, not in this case.


That causes a bit of a dilemma, really. Martin and I have the strong intuition that thermostats in corners attached to external walls of a sporadically heated building are a real no-no because the radiant exchange is enough to distort the reading. Apart from trawling for an Energy Savings Trust (or similar) reference I saw at one point, we have no chapter-and-verse on this one, just anecdata; upstairs in the same building, an analogue thermostat on an external wall set to 16 yields anything from 14 to 21C. That’s not science. It’s above a heating pipe and older than most of the students. Happily, it’s easy to wire the thermostat anywhere in this space via the suspended ceiling. Sometimes I think it’s worth going with instincts. I suspect lots of heat going in has made the walls rather warmer than usual, cloaking the effect we were expecting.


Final question: given the massive amount of heat output in the space, what’s the boiler actually doing? The readings strapped for the logger strapped to the flow are extrapolated for the suspiciously flat part of the run. In my bid not to damage the new decor, I only brought masking tape, forgetting (a) it doesn’t stick to hot  and (b) pipes  aren’t pretty. The boiler is set to 90C. Even with the fans on their lowest setting, it’s never going to get there. We need to check that return value but even allowing for the difference between what’s in the pipe and the reading outside it, it won’t be over 55C, which Martin says is the magic number below which condensing boilers actually condense and are therefore worth the money. (They’re required by the current building regs, but apparently don’t help energy efficiency in lots of cases, since in practice since many people run on max even when they don’t need to. That’s purely anecdata again, but one we should check out.) I think we might recommend they turn down the boiler anyway; it’s got to be easier to turn a big ship when it’s going slow than when it’s full steam ahead.

So, what’s next? We’ve left a centrally placed logger to find out whether the place is overheating badly under these more normal weather conditions, and I’m urgently specifying controls and other changes like blanking off those very salient fan-off switches. They’ve already decided users need to be able to turn the heating off, or up or down within limits. I like the old programmable thermostat we put in this space because it’s built like a tank, but they find the user interface confusing, and last I looked the top limit on it was 26C, which is unaffordably risky in this space. So I’m looking for something easier to use (the locking feature, especially), with optimum start control (which in this setup will matter, I think) and I hope, a configurable top temperature. The social housing models have come on tremendously in the last five years so that’s my first port of call.

Leave a Reply

Your email address will not be published. Required fields are marked *