snagging – so, here’s the plan

new-controlsI’ve had a tour of a newly refit church hall where the equipment isn’t entirely as expected, so there are questions about whether the system meets the requirements and is set up properly.   There have been a few surprises along the way (there always are!).  The main questions are, as always, all about the controls, but this time, with three possible issues – whether the centralized controls will play nicely with fan convectors that have individual controls, whether the users can do what they need and no more, and whether the heating management is simple enough for volunteers to work.

So, the basic kit:  it’s a 34 kW condensing combi-boiler (Worcester Bosch Greenstar 34 DCi Classic ErP), servicing 8 Myson iVector iV60x120 fan convectors and a handful of legacy 1990s standard single panel radiators. As a side note, this is the third boiler I’ve seen in the space; I’d have to check my notebooks, but I think it was 24 kW, than 28, and now this.  I don’t think I’ve ever in my life seen a boiler capacity go down!

These convectors are typically used in applications where individual control matters – say, where there are one or two in a room, and all the rooms are occupied by different people with different habits.  That’s fairly typical in some office buildings, but individual control here would make it take around an hour to reset the heating every week to match the diary.  The convectors are designed so that the individual controls can be screwed down for tamper resistance (in this venue, that’s known to reduce interference to once or twice a year), and then they can be centrally controlled by a Building Management System, although there’s no intention of installing one here.  At the moment, control is via a very simple domestic programmer, and separate room thermostat.

The convectors have three fan speeds.  Ordinarily, whether under individual or BMS control, the fan would be on high when the warmup is first starting to speed it up, shift down in speed as the temperature approaches requirements, and shift down even more, to “normal”, when it reaches it.  As a side effect, that makes the system much less noisy once it’s warm.   That modulation would be desirable in this application, but if it’s possible to achieve, it’s not obvious.  The fans are therefore set to “normal” permanently, with the centralized controls just bringing them on or off. Each convector has a separate fused electrical  on/off switch; this of course doesn’t do anything about controlling the flow of hot water through the system – there’s no moving valves – but switching them off at the wall turns off the fans.  You still get some heat output that way, but we don’t know how much.

The programmer will eventually be hidden away – and possibly locked away, if necessary, as it often is on these sites.  The room thermostat needs to be in a place where the temperature represents the space.  At the moment, it’s in a corner very near a large external wall, so that’s two strikes against it – corners usually have pretty poor airflow, and large cold masses too close to a thermostat make it read well under the correct air temperature.  It resets the target temperature permanently – there’s no “up/down a bit” for the next hour/until the next group like this site had before.  That means we can’t risk user interference, but there are locking thermostat guards on the market for exactly this problem.

Note that so far, we haven’t mentioned any user controls.  This site is not always staffed, and the heating is managed by volunteers, and sometimes they make mistakes.  There is an experimental approach to user control still to be connected:  a simple push “boost” button which will bring on the heating system for 20, 40, or 60 minutes, depending on a hidden  DIP switch setting (not meant for users) on the back.

As HeatHack, we need to do two things:  think like engineers, and think like socio-technical systems specialists.   I’m of course stronger on the latter!

For the engineers, the questions are: (1) are the boiler capacity and radiator output specified correctly?  (2) how long does the space take to warm up? (3) is it adequately controlled; i.e. will it keep the temperature in the correct bounds when the space is in use, and not more than it has to, for energy efficiency?

For (1), on the normal fan speed, the eight convectors throw around 26 kW (when the temperature of the water going into them is 50 degrees C above the air temperature; we can make the difference up to around 60C).  On the medium fan speed, they’d throw 36 kW – more than the boiler can supply.  On high, it’s a lot more – 52 kW.  So that’s a question – I think that means we couldn’t run them on the other fan speeds even if we wanted to, although maybe it’s more a case of “wouldn’t get any benefit” than anything more sinister like “the boiler would overheat/lockout”.

For (2), that’s a case of placing equipment to test it.  There’s no wifi at present, so for ease I’m going to use  Lascar loggers .  Users say “it’s fast”; one of the tradesman said pre-installation, he reckoned getting to temperature would take around an hour.  If it really is like that, I think it’s well worth getting a programmer that can do optimised start control.   The loggers also tell us (3), especially if we place one where the thermostat is, one where we think it should be, and one in the centre of the space.

For STS, the question is simply: will the  users and lightly trained property management volunteers be able to do what they need, and what will go wrong.  And keeping in mind that users will always attempt to make themselves comfortable, just not always in the way intended: if they want to do something they can’t, what will they do instead?  Here, there aren’t any technical tests – just reasoning about the design, observation, and in this case, previous experience.

When the heating is off, users who are too cold have a way to turn the heating on.  As long as we can make the boost button recognisably about the heating,  it’s the most salient and effective method – for instance, more obvious and better than using the cooker in the kitchen as heating, as I’ve seen before.   So I think we’re OK there; if they know about it, they’ll use it.  Knowing is a tricky matter of placement.  We don’t want people hitting it experimentally (say, if they’re looking to open the door when the doorbell rings) so it’s better to make it salient for group leaders only if we can.  That suggests findable, but either not completely obvious, or robustly labelled in a large, high contrast font.

For the first time in this space, we also have to think about what happens if users are too hot.   There is currently no intended control for this; they can’t even turn the heating off.  There are, however, two salient and effective methods they could use.  The first is opening the exits. That’s bad; it wastes energy, creates security risks if they don’t close them again, and most people looking for a prop will go for a fire extinguisher.

The second is the on/off switches on the convectors.  My prediction is that any hot user faced with this space who has any awareness of energy efficiency will turn them off, and then human nature being what it is, forget to turn them on again.  The same applies for users who are bothered by the noise.    I would also expect occasional users – like birthday party hires –  to turn them off when they leave the space, because they’ll think they’re electric heaters and that the staff turned them on just before they arrived.  They’ll get instant, reinforcing feedback – the fan will go off – and it will actually make them a bit more comfortable, since it will slow down the radiator in dumping heat.  Signage won’t work to keep people from using the switches.  If they should always be left on, we can change the switches for fused spurs.  That still lets the electricians disconnect them for work.

And what about the management volunteers?  The issues are more minor here. The main risks are from these models (Altech, the easiest thing to source locally, from Graham’s in Polwarth) make dual use of the display backlight buttons.  On the thermostat, press the backlight for more than 3s and it turns on or off the frost control – whether or not the heating will come on if the space temperature drops below 5C.  On the programmer, the same action puts the system into holiday mode, i.e., essentially, off.  I was pleased to see big fonts on the controls, but it would be absolutely classic to not be able to read the display, press the button to get the light, still not be able to read it – and then keep the finger on the button while fumbling in the bag for one’s spectacles.   It may sound stupid, but that kind of thing can lead to engineer call outs in this kind of building.  I don’t think it’s make or break, but if there needs to be a programmer change, it’s a reason to change supplier.

Yes, it’s geeky, and complicated – and then we have to decide what to say to the property owners and how to support them through the snagging process.  We’re working on it – the “engineering assessment” we hope by the weekend, with finalised thoughts on the STS once we see at least the technical part of the system in operation.

Leave a Reply

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