Hoops for BUS grant qualification for ASHP in a low heat loss retrofit house with slim primaries

I’m trying to work out how many hoops will I have to jump through to install something like a 4kW Grant 290 (or maybe 3.5 kW Vaillant) whilst using my existing primary pipework, and still get the BUS grant (if indeed that is likely to be possible at all).

This question is a bit of an epic. Unfortunately it’s not a straightforward problem…

Our house is a 1920s semi-d in Brighton. I did a load of work to upgrade the building fabric and extend it gradually over the past 15 years. As is often the case most work was done in the first couple of years. I got quite enthusiastic with air tightness and insulation.

I designed and installed the heating system in 2013 or so.

When we bought it, the house had a 4 year old 24kW Vaillant combi boiler. In 2013 GB grid CO₂ intensity was 3.5x higher than now, and heat pump performance was a bit worse (also the now-banned high-GWP R410a was the most common refrigerant), so the effective CO2 emissions of boilers and heat pumps were comparable at that time. We decided to stick with gas for a while longer.

The rest of the heating system was 20+ years old. The boiler needed to be moved to make way for a new staircase (and the ground floor was being scrapped due to wood worm and tediousness of insulating suspended timber floors), so only the boiler was kept.

Fortunately I designed the new heating system as a single temperature system with a mix of UFH and radiators and a design temperature of 35°C.

Since then, the boiler has been configured to only use its minimum attainable modulation in heating mode (approx 6.5 kW), and set for 35°C flow temp. It probably operates at something around 95% to 100% sensible efficiency (fun fact, the maximum “efficiency” of a condensing gas boiler is about 115% because “100%” is defined as all heat recovered without condensing).

On the coldest of days it fires for something like 4 hours (I wish I had some accurate data on this, but it’s difficult to tell because it starts cycling after about an hour).

Annual gas consumption for all heating and hot water is about 1,800 kWh. The adjoining house has been unoccupied and unheated for the past year. Cooking, and washing is electric. No solid fuel (no chimneys any more!).

We’re a family of four, most washing is with showers (for which we have drain water heat recovery with ~50% end to end efficiency). Other than that, maybe one or two baths per week. Gas boiler efficiency probably averages ~90% in DHW mode or so (there is also a flue gas heat recovery device) but maybe a bit worse because it often short-cycles during showers in summer months.

Looking at recorded gas usage, it’s about 35 kWh per month in late spring (we don’t heat between end of March and mid October typically), so assuming similar year-round that would be 450 kWh for DHW, leaving 1350 kWh or so for heating.

Based on the heating degree days estimation method (12 °C basis instead of 15.5 °C because it’s close to Passivhaus levels of heating load), that gives 1.45 kW peak heating load at an indoor temperature of 21 °C.

Based on my rough estimate of coldest day boiler maximum firing times, instead gives ~ (6.5kW * (4 hrs / 24 hrs)) = 1.08 kW.

I’ve done a heat loss calculation using the deprecated MCS spreadsheet, and my own figures for U Values (plus modifications for MVHR).

With next door unoccupied, and my estimate of the houses’ current leakage rate of 1 (ACH50), and assuming 85% efficient MVHR, that gives 2.1 kW.

Pessimistically switching to 3 ACH50 and 79% MVHR gives a heat load of 2.8 kW.

I haven’t done a proper blower door test, but I could do if that’s likely to help convince!

Both of those heat load figures use slightly higher room temperatures than we actually heat to, and of course ignore all “incidental gains” like body heat, cooking and standby electrical loads (all in all probably 500 W or so).

So, I thought I was being cunning at the time, but when I installed some external wall insulation about 5 years ago, we ran some 22 mm PB barrier pipe inside the (250mm thick) wall insulation to use as a potential ASHP future primary pipework. This was a late afterthought and done in a hurry. The pipes are surrounded on all sides by at least 100mm of insulation (expect where they pass through AAC blocks where and have about 30mm of coverage, n.b. the blockwork is externally insulated, so any heat loss to the blockwork ends up inside the building).

If I scrap the 22mm primaries, then I’ll need to run new 28 mm primaries for an additional 3.5 metres outdoors. I might be able to replace them in-situ, but the scheme I have for that might not work and would probably end up destroying the 22mm pipes.

Assuming the 22mm pipes are to be kept, then if I stick to the CIBSE recommended limit of 350 pa/m pressure drop for primaries, that limits flow to 0.66 m/s. Grant’s “Maximum Flow rate required” figure of 11.77 litres/minute (4.1 kW at 5°K ΔT) would result in a flow rate of 0.77 m/s. Reducing that to 0.66 m/s limits the output power to 3.5 kW (or 0.62 m/s → 3.2 kW if I go with the MCS recommendation of 300 pa/m instead).

I have not yet done an index circuit analysis, but can do so if that would help answer the question.

I live in a dense urban area and will almost certainly want to leave the HP in silent mode. I think that using “silent mode 1” on the 4kW Grant 290 gives a maximum power output of about 2.8 kW. I’ll slightly oversize the DHW cylinder to accommodate the longer recovery time, but the fact that the showers have 50% efficient heat recovery means that shouldn’t be an issue (since the practical hot water recovery time should be about the same as if we had the 6 kW model running in non-silent mode).

As well as the silent mode option, I think I’m right in saying that there is a Grant firmware option which allows both the max compressor speed and pump modulation to be limited, which would presumably allow capping the output at 3.2 kW or 3.5 kW (giving a bit more headroom whilst still staying inside of the head loss and flow rate limits).

Cost in time and materials of replacing them with 28mm pipes is about £300 to £500 (50% chance of having to run them a different route so that they have a worse heat loss).

I’m confident the 22mm pipes will be fine as far as delivering sufficient heat into the building.

In practice, are any installers or Grant (or MCS themselves) likely to be happy with that too, so that the BUS grant can be claimed?

What are the potential pitfalls (if any)? Defrost mode?

Does defrost mode pull heat from the DHW or heating circuit? I plan to run the heating circuit open loop, but at the very least I could leave some or all of the UFH permanently on - these are a mixture of pipe in 30mm conventional screed, and pipe in 125mm concrete slab. For the concrete areas, that’s about 8 tonnes of concrete, so no problem with thermal mass.

Any thoughts?

Cheers,

Tim.

1 Like

Large 4 port buffer tank, even if buffer increase required flow temperature by 5c you will still get a great COP.

I expect Octpus would refuse to use the pipework in your insulation as risk of leaks they will be responsible for.

I think you need to get 101 different independent installers to visit to quote before you will find one that will do the job in the sensible way. Heat geek may have an installer in your area.

A DIY install without the grant may work out cheaper. But I see little point in replace your gas boiler until it hits end of life. (I would try without buffer to see if it works good enough.)

Points in favour of replacing the boiler:

  • The £125 per year standing charge is a nuisance.
  • It has a minor fault (not safety related) so that it requires manual intervention when the heating starts. This would cost about £150 to get fixed. This may progress and stop DHW working too.
  • We’re about to fit a new kitchen, so replacing at a later date will be more expensive.
  • There is currently a BUS which should cover nearly all of the cost.

I don’t expect a buffer tank to be needed. I expect to get reasonable CoP (with the caveat that relatively high DHW demand, and low external temperatures when heating is needed will both nudge the CoP down a bit).

The question is, is there anything in the MCS rules (or potentially manufacturers’) which would prevent the use of the 22mm primaries?

2 Likes

Timing to replacing boiler seems sensible.

MCS rules will require keeping to manufacturers installation instructions, I would be more concerned about the MCS aproved heatlose calculations and how you prove what insulation have been installed. I expect a dt10 design could be done to allow 22m pipes provided they believe your heatlose calculations.

Installers are very concerned about being liable if a heatpump will not heat a property to required temperatures, they have to prove the system will work to the British Standard room temperatures.

(A 4 port buffer makes it easyer to prove the system will have high enough flow rates to keep a oversized heatpump happy.)

1 Like

The calc I’ve done is with the MCS spreadsheet, so that should hopefully satisfy them. I have modified it to allow for more than one wall type in a given room (such is the nature of a low energy retrofit) and I’ve used custom U values.

For the windows, I have calcs and/or invoices and manufacture specs.

For wall/floor/roof U values, I have calcs for each element taken from the Passivhaus Planning Package, along with detailed CAD drawings and a selection of photos taken during installation.

I’ve also enhanced the spreadsheet to allow for MVHR (i.e. calculate effective air change factors).

As an extra check, I also have the gas meter readings and predicted usage figures from the energy provider based on historical usage (last one was 1770 kWh per year).

Producing a convincing ACH50 might be more tricky, but within the calcs above I’ve already adjusted the calc assume quite a pessimistic ACH50 of 3 (instead of 1). Whether they think 3 air changes is pessimistic without additional evidence might be sticking point. I believe the official MCS line is that you can use non-default ACH50 figures on the basis of specialist advice. I expect an accredited blower door test or a statement from a suitable designer (e.g. PHI or CIBSE) would be OK for this. Other options might include recordings of CO2 decay rates together with wind speed records from nearby weather stations.

1 Like

A blower door test isn’t that expensive and may be the easiest way to solve that question.

Also hopefully @johncantor may spot this thread and have some useful comments.

Maybe Grant or another heat pump company could be persuaded to use your site as a case study, and so be prepared to do whatever extra work/risk might be involved?

1 Like

For your heat loss 22mm pipes will be totally fine. There would be zero advantage of upgrading to 28mm. Any decent heat pump installer will be able to do the calcs to satisfy themselves of this. MCS don’t care what pipe sizes are used as long as the system is fit for purpose. The Heat Geek pipe sizing chart is a useful quick reference, 22mm is fine for up to 6kW

Using 22mm actually has an advantage for DHW, since there’s less volume of water in the primaries to heat up before the tank can start heating up. I’ve seen many BUS installed heat pump with 22mm.

Defrost pulls from the heating circuit, if the system is open-loop this won’t be a problem. If the heating circuit is small, then fitting a volumiser can be a good idea, but you should be fine with the volume in the UFH and rads. Also, UFH has a lot of thermal mass, which is good for defrost.

I would recommend having a look at the map on HPM to find an installer in your area:

There’s no reason to install a 4-port buffer tank on this system. In fact, there’s no reason to install a 4-port buffer on nearly any domestic system. Any decent installer will be able to do index circuit calculation to ensure the system will work fine without a buffer. Here’s a good example of how an installer does such a calculation:

Octopus or any installer will be quite happy to connect onto existing pipe work, the vast majority of their installs will be using existing pipework. If there’s a leak they will fix it.

4 Likes

It helped me to think about why the MCS scheme is there and why the BUS grant uses it.

The BUS grant is a one-time government subsidy to upgrade houses to lower temperature heating. It can’t be claimed again and the heating system installed could well outlast the current occupants. Therefore the system that is installed should be suitable for the average occupants of the house, not to just cater for the current occupants. A couple who live in a 4 bedroomed house could decide that they are happy with rooms at 18C, several bedrooms at frost level, and only need enough hot water for two. The MCS standard sets out a baseline of what the system needs to be able to achieve (room temperatures under design conditions, hot water provision, etc) to ensure that the system installed is able to meet the requirements of future occupants. This is one of the reasons why previous usage isn’t a valid data point for MCS because you may not have been heating your house to the required temperatures.

The MCS standard doesn’t require it to be particularly efficient or well designed. Only that it is capable of providing 100% of the heat requirement, to all rooms, to fairly high room temperatures. 55C flow temperature at design temperature - no problem. Rooms that heat up at different rates - no problem.

When designing our system I had two sets of calculations, one for what we actually wanted (our desired room temps, low flow temp, realistic assumptions for ACH). The other was the MCS design (high room temps, 55C flow temp, their assumptions for ACH). We came up with a system that would satisfy both our requirements for an efficient system and the MCS design requirements.

The ability to use 55C flow in the MCS calculations gives loads of headroom compared to a more efficient design.

For your specifics 22mm plastic pipe will certainly be able to carry 4kW of heat. I’m not clear exactly what the Maximum Flow Rate required of 11.77 l/m is. Our Daikin has a minimum required flow rate for defrosts and hot water of 12l/m however in reality it targets 25l/m when doing those things.

As long as your pipework is able to meet the minimum requirement specified by Grant then it will be good enough for MCS & BUS.

2 Likes