Hi @jvs197 and welcome!

My first observation is that your 8.5kW ASHP is already oversized for your sumptuously-insulated house.
So your system design will need to avoid two possible events

• the HP cycling on/off, which uses more energy (and hence lowers the COP)
• the return water temperature to the HP being too warm... because it's unable to export enough heat into the boiler circuit via your HEX

At this point it is usual for a professional installer to insert an overly-expensive buffer-tank in the return pipe to the HP.
Heat is lost in that, thereby keeping the return water cool enough.

It's worse in your case because you aren't sending the output from the HP to anywhere else... just the HEX.

My next observation is that the rise in temperature of the return to the boiler is bringing forward the moment when it will drop out of condensing mode.
So it too would now be operating inefficiently.

The key to these issues must surely lie in a 'control' which prevents the boiler firing if the water arriving in it's return is above the temperature at which the condensing operation ceases.

It's typical of an Arduino project... which you may find appealing, given your background!

One of the first professionally implemented Heat Pump trials in the UK was done with funding assistance from the Welsh Government.

The Freedom Project implemented hybrid heat-pump systems in a number of housing authority homes.
Unusually, the highly competent engineers who designed the systems also supervised the installation and operation throughout the trial.
The results therefore show a level of competence which was in short supply from later BEIS-funded trials!

If I wanted to analyse the best way to implement a hybrid system, that's where I'd start.

The Freedom engineers were drawn from Western Power Distribution (electricity) and Wales & West Utilities (gas).
Some those engineers now work in senior roles within National Grid Electricity Distribution.
With a bit of ingenuity, you can probably work out their email addresses and contact them direct!

Posted by: @jvs197

↑

Heat loss calculations for house showed  annual heating requirement of 6600 kWh (space heating load of 5.3kW) and recommended a 6kW ASHP.

We swapped to a HP 2 years ago, after running a regular gas boiler for many years. We have smart meters, so could correlate daily gas kWh with the outdoor temperature.  An advantage of gas boilers is that gas in very closely aligns to heat out - just assume an efficiency of 90%.  Do you have figures for how many kWh/year of gas you have used? It's a great way to check the calculations!

I'm a little confused by the figs you gave - 6600kWh annually is lower than a heating load of 5.3kW would imply.  Our measured numbers are 4500kWh/yr  heat (using 1400kWh elec), 2kW peak heat load. 

While "short cycling" of a heatpump is bad  (more than 6 On/Off cycles per hour), there will inevitably be On/Off cycling when you need less than the minimum power of the heatpump  which is perhaps 30% of full power.  Cycling fatigues copper joints, and leaves heat in the outdoor stuff which is a waste.  Our heatpump cycles once / hour, and can't be avoided as I see it.

The manufacturers will give minimum required system water volumes, so that there's enough thermal mass to cope with defrost.  Zoning and TRV throttling reduces the minimum volume, and is the reason that a buffer is generally added.  We don't have a buffer, have no zones, TRV's either removed or set to max.

Cheers, R

@transparent Thank you for your quick reply (anyone want an 8.5kW HP?!!)

Yes, I was a bit concerned that the HP is over-powered but then, with the assumptions made in heat loss calculations (I used the HeatPunk one), the final figures are only 'ballpark'. If I had an 11 or 14 kW Mitsubishi then I would certainly not go ahead - but 8.5...?  In fact I have seen on this site that some over-sizing is quite acceptable in some folk's opinions.

Interesting point about boiler not condensing because return water temperature might be too high.  I understand that it needs to be below 55 deg.  The calculation for the heat pump was based on a flow from the HP at 45 deg - so no worries there.

Yes, need control to prevent boiler firing if return too hot - but that's a simple circuit (no need for microprocessors!). That point should not however be reached as explained above.

Thank you for your hint about 'FREEDOM'.

@robl Thank you for your reply. The 5.3kW figure came out of the heat loss calculation (using HeatPunk).  Gas consumption over the last year was 9890kWh covering space heating and hot water (and a little for cooking).  Took the consumption in July/Aug as predominantly meeting hot water requirement and assumed the same rate over the whole year - that was 3270.  So space heating was 6620.  I accept that figures were therefore approximate!

Posted by: @jvs197

↑

I understand that it needs to be below 55 deg.  The calculation for the heat pump was based on a flow from the HP at 45 deg - so no worries there.

Erm... what's the return water temperature for your boiler in its present situation?

If it's above 45degC, then your HP won't be able to add anything more to it.
Messrs Flanders and Swann told me that!

@jvs197

I suggest that you experiment with your present gas system, so that you can determine how well an ashp could work.  It’s what I did before installing our gshp. We had a ‘regular’ gas boiler- ie just a heat source, and a smart gas meter.  I set the pump to mid, the boiler to low flow temp, the home thermostat to highest, all TRVs on full.  Then ran the system for an hour to let it stabilise, upped the flow temp a little to stop it cycling on/off.  When running stably, I used a thermocouple taped (and insulation over it) to the boiler flow and return, home air temperature and took a note of the smart meter gas power.  From these measurements I accurately predicted the running temp that we would get with a heatpump, just by linearly extrapolating them.  I got a 5C ( Tflow-Tair) rise per kW of heat, and a 2C deltaT.   This predicts a flow temp of 21C+2.4kW*5.  Our gshp is 2.4kW, and it runs at a flow of 32C.  (Nb- we have 150mm of EWI now, so this approach works well for us - the original 1963 radiators are now ‘oversized’).
Strictly speaking rads aren’t quite linear, but it’s close enough I think if the difference is small.