Mad idea for DHW retrofit based on Mixergy?

[Nickfromwales]

And god help you ( your wallet ) if the copper cylinder is already on borrowed time. You can very rarely tell from the outside! Ask me how I know.

[JamesPa]

4 minutes ago, Nickfromwales said:

Have you asked a plumber to quote for the swap, or are you looking at this from a novice / DIY PoV?

As part of an ashp fit yes, as a standalone job not yet.  The latter may turn out cheaper even with the VAT (it's zero rated if done with the ashp but I suspect the MCS markup exceeds 20%). 

 

As a minimum it requires taking up floorboards in two rooms (three if they insist on upgrading the feed to 22mm), reconfiguring the airing cupboard and some of the plumbing, and replacing a shower, in addition to the actual UVC.

 

6 minutes ago, Nickfromwales said:

The second the ASHP stop giving heat, the process will reverse and all of the bolted-on retro fit stuff will leak heat like a pig vs a new, much better insulated and encapsulated UVC. 

Pump turned off at this point so the additional loss occurs only when it's heating.  PHEs come with installation jackets.  Some loss desirable as it's an airing cupboard within the heated envelope of the house.

[Nickfromwales]

1 minute ago, JamesPa said:

As part of an ashp fit yes, as a standalone job not yet.  The latter may turn out cheaper even with the VAT (it's zero rated if done with the ashp but I suspect the MCS markup exceeds 20%). 

 

As a minimum it requires taking up floorboards in two rooms (three if they insist on upgrading the feed to 22mm), reconfiguring the airing cupboard and some of the plumbing, and replacing a shower, in addition to the actual UVC.

 

Pump turned off at this point so the additional loss occurs only when it's heating.  PHEs come with installation jackets.  Some loss desirable as it's an airing cupboard within the heated envelope of the house.

Agreed. But long-term benefits will be notable afaic. My comments about latent loss extend beyond your PHE etc, and outward to the rising pipework and CWS tank etc. Plus, I just wouldn't ever want to be bathing / washing dishes in the water which has come from the CWS. Open-pipe systems have a lot of 'negatives' but are kept "because it's worked fine for us, up until now".

 

The MCS markup will likely pay for the UVC change.

[sharpener]

51 minutes ago, JamesPa said:

and an annual maintenance fee

 

Not sure where that comes in.

 

Every three months I test the two pressure/temp relief valves, and once a year I check the expansion vessel pressure, clean the filter screen and write up the service record. Is there something else I should/could/must pay someone to do?

 

4 hours ago, JohnMo said:

Use the thermostat to open/close a relay, use the relay to call for heat and start/stop the pump. No need to open the wet parts of the system to fix anything, if/when it does go wrong.

 

If you have an S-plan setup you don't need the relay. DHW thermostat operates both pump and motorised valve, the contacts on the latter then provide voltage-free call for heat to the HP. No flow through the HX when themostat satisfied. Can easily lag the HX to reduce parasitic losses.

 

I am with @JamesPa on this one and am now looking to see if I could incorporate it if necessary. Depends a bit on the pipe runs but certainly easier than a whole cylinder change, I hope being s/steel OSO it will have some life left.

 

Unfortunately you can't use a pump powerful enough to push the water into the cold feed lower down (which is easier to get at) because the pressure regulator also supplies the balanced cold to the showers. Unless that is you only ever heat the water in the small hours when there is no chance of any simultaneous HW draw-off (you would also need two check valves).

 

 

Edited April 7, 2023 by sharpener

[JamesPa]

1 minute ago, sharpener said:

Every three months I test the two pressure/temp relief valves, and once a year I check the expansion vessel pressure, clean the filter screen and write up the service record. Is there something else I should/could/must pay someone to do?

I understood that for a UVC an annual inspection by a g3 qualified engineer was mandatory, but I haven't actually read this anywhere other than secondary sources, so it may well be plumbers drumming up business not an actual requirement.

 

25 minutes ago, Nickfromwales said:

. Open-pipe systems have a lot of 'negatives' but are kept "because it's worked fine for us, up until now".

True.  It's a trade off I accept, but houses are.  Ideally I'd completely rebuild my 1930s house with the latest building technology throughout, but it's disruptive and expensive.  So I generally prefer to upgrade when components reach their end of life or are being displaced for some other reason.  Still not decided.

 

28 minutes ago, Nickfromwales said:

The MCS markup will likely pay for the UVC change

Maybe outside the south east, but in the south east the MCS bods seem to want 15K absolute min, and frequently 20K plus, to do a 11kW hp retrofit including a UVC, excluding radiator upgrades.  Ok you get 5k off in govt grant, but it's daylight robbery.  

[JamesPa]

15 minutes ago, sharpener said:

Unfortunately you can't use a pump powerful enough to push the water into the cold feed lower down (which is easier to get at) because the pressure regulator also supplies the balanced cold to the showers. Unless that is you only ever heat the water in the small hours when there is no chance of any simultaneous HW draw-off (you would also need two check valves).

Not sure I followed this, is it a comment specific to your system or generic.  Does it relate to a vented cylinder or a UVC with insufficiently large coil (the latter being the motivation for adding the PHE)?

[sharpener]

1 hour ago, JamesPa said:

Not sure I followed this, is it a comment specific to your system or generic.  Does it relate to a vented cylinder or a UVC with insufficiently large coil (the latter being the motivation for adding the PHE)?

 

Generic, but only for UVC installations, where in the cold feed there is a pressure regulator which like the piling swivel in Henry Reed's Naming of Parts "you have not got". And this regulator typically has a port to feed cold to the showers at the same pressure as the hot so mixer taps work properly.

 

 As posted earlier, for a retrofit there will not be any spare tappings on the cylinder to connect the HX but you can make the hot connection to any convenient hw pipe. However the other connection must go into the cylinder low down and and on the cylinder side of the pressure regulator so with less freedom. Unless you reverse the flow and have a more powerful pump, then you could in theory connect it upstream of the regulator instead. A thought experiment on my part, but it is not very elegant on account of the additional pumping losses.

Edited April 7, 2023 by sharpener

[JamesPa]

Does anyone know how to read the heat transfer specs of PHEs.  Some seem to be specified in kW (under what conditions?), many arent really specified other than by number of plates and physical size. 

 

I'm trying to work out how one might make a selection if I decide to go down this route.

[JohnMo]

I used this website to find out what size PHE I needed.

 

Bigger the better at low flow temps

 

https://www.heat-exchangers.uk/online-calculator/

[Gone West]

13 hours ago, JamesPa said:

Does anyone know how to read the heat transfer specs of PHEs.  Some seem to be specified in kW (under what conditions?), many arent really specified other than by number of plates and physical size. 

 

I'm trying to work out how one might make a selection if I decide to go down this route.

I've been looking at the Aquinox TS PHE and the file shows flow rates for different kW, inlet and primary circuit temperatures.

 

Aquinox HWM.pdf

[JamesPa]

2 hours ago, Gone West said:

I've been looking at the Aquinox TS PHE and the file shows flow rates for different kW, inlet and primary circuit temperatures.

 

Aquinox HWM.pdf 1.2 MB · 4 downloads

Thanks for this, its going to take a bit of reading through.  It looks like this unit is designed to be attached to a thermal store and transfer heat to incoming cold water, but of course the principle is still the same.

 

I've had another crazy(?) thought.  The problem that I'm trying to solve (or convince myself that there really is no solution for) is the idea that in every HP retrofit (and mine in particular) you need to swap out the otherwise perfectly satisfactory DHW tank.  This adds, let us say, 3K to the cost, generally involves a lot of disruption, and has little apparent value as far as the consumer is concerned.  My gut feeling is that, until this problem (and the obsession with buffer tanks) is solved, HP retrofits will remain a minority market.  I (and thus likely others) need at least to have some options, rather than having the disruption and cost foisted on them by the 'cant think out of the box' brigade.

 

So what if we just accept the existing small coil (typically 0.8sq m) in whatever DHW tank the homeowner has.  For now we will ignore the large number of homes with combis, thats the next level of problem.

 

Many heat pumps will in fact heat to 60C and some HT ones (much derided on this forum. for understandable reasons) to 70C.  I grant that using these high temperatures for space heating is bad economics, but what about using the higher temp just for DHW?  There are two factors to consider, loss of efficiency and minimum output power of the heat pump before it starts short cycling.  For the rest of the argument I'm assuming that the target temperature for the stored water is 50C  as that seems common, with a weekly anti-legionella 55C boost.

 

For example the Mitsubishi Ecodan 11.2kW has a COP (ambient 7C) of 2.8 at 55C o/p and 2.3 at 60C o/p, and will modulate down to about 4kW.  With a 0.8sq m coil you can get the DHW up to about 45C before the heat transfer in a 0.8 sq m coil drops below this minimum.  So I could operate the HP at 60C to heat the water to 45C, then and use the immersion for the final 5C?  Is this so silly?  

 

The average household uses about 120l DHW per day according to various sources (some say 80l, Im using the higher figure).  In the 'conventional' arrangement (which, in a retrofit situation, generally involves replacing the DHW cylinder), and running the HP at 55 (COP 2.8), assuming an input mains water temp of 10C, this will require 2kWh per day input power.  Alternatively if we don't replace the cylinder (or any other part of the DHW system), run the HP at 60C (COP 2.3) until the DHW reaches 45, and then use the immersion (COP1.0) to heat it up to 50, the input power goes up to 2.8kWh.  Annualised, at current very high electricity prices, that's a difference of only £115.  

 

I'm not at all sure that makes a compelling case for all the disruption and cost involved in swapping out the DHW cylinder.  If course its still desirable to do so, but we must not allow perfection get in the way of very good.  The fundamental requirement is to get off of fossil fuels onto renewables, in practice switch from gas to electricity.  We need to make this as easy as possible for the mass of people!

 

Just another out of the box train of thought.

Edited April 8, 2023 by JamesPa
add input water temp

[sharpener]

Nice analysis. Add to that the increasing penetration of free solar PV and it's even more compelling.

 

Most HPs I have looked at in any detail (Daikin, Grant, CoolEnergy etc) have already forseen this use case, and an external call for DHW heating will automatically set a higher, programmable flow temp. At most you would need an additional timeswitch and/or thermostat to co-ordinate the HP with the immersion.

 

BTW how did you get on with the HX selection? I see from this site that there are some typical DHW applications including this one for less that Eur 100. But also some warnings about limescale, and corrosion with open vented systems, so you might want the more expensive nickel brazing or gasketed construction. And they don't ship to the UK (thanks Boris).

 

 

[SteamyTea]

1 hour ago, JamesPa said:

My gut feeling is that, until this problem (and the obsession with buffer tanks) is solved, HP retrofits will remain a minority market.

The main obsession should be with maximum output temperature and modulation.

 

Too many people think that there will be a 'drop in replacement' for a gas boiler.  Physics are against this, so best to stop thinking that there will be a better jam tomorrow.

[JamesPa]

43 minutes ago, SteamyTea said:

The main obsession should be with maximum output temperature and modulation.

 

Too many people think that there will be a 'drop in replacement' for a gas boiler.  Physics are against this, so best to stop thinking that there will be a better jam tomorrow.

Clearly HPs are not a drop in replacement and as you say the physics negates the possibility that they will be.  But, at the same time, an industry which loads what, to the consumer, is 'just a new boiler', with a shedload of difficult replacements/additional requirements, is not going to get the mass take up that it must.

 

I think the case for radiator upgrade is clear.  I attempted to design to avoid this initially, but have now done it and discovered its an easy job, whether DiY or outsourced, and really quite cheap.  And I got shiny new radiators into the bargain, which certainly look better than out old ones.  This is, if correctly put, a saleable proposition.

 

The case for DHW system replacement and a buffer tank is much less clear from a consumer standpoint, and much less clear from an engineering standpoint also.  Fix modulation depth and a buffer becomes unnecessary, other than to the extent is required to service the defrost cycle.  DHW system replacement is a big deal in terms of cost and disruption and, whilst there are benefits (but also downsides) if its replacing an exiting vented system, there are no benefits, only downsides, if its replacing an existing unvented system solely because the coil isn't big enough.  In the latter case its barely a saleable proposition, so better solutions are needed.  Modulation, possibly combined with higher, albeit less efficient, flow temps for DHW only, has the potential to fix this problem also, but perhaps we need some staging points hence my thoughts on this thread.

Edited April 8, 2023 by JamesPa

[JamesPa]

1 hour ago, sharpener said:

BTW how did you get on with the HX selection? I see from this site that there are some typical DHW applications including this one for less that Eur 100. But also some warnings about limescale, and corrosion with open vented systems, so you might want the more expensive nickel brazing or gasketed construction. And they don't ship to the UK (thanks Boris).

Not got there yet.  @JohnMo seems to have identified a useful starting point, but Im not yet sure I understand the calculation fully enough to be confident, nor how the results can be transferred to other manufacturers.  Only working on this intermittently, my wife is getting fed up with the amount of time I've invested in ASHPs.

[JamesPa]

3 hours ago, JamesPa said:

For example the Mitsubishi Ecodan 11.2kW has a COP (ambient 7C) of 2.8 at 55C o/p and 2.3 at 60C o/p, and will modulate down to about 4kW.  With a 0.8sq m coil you can get the DHW up to about 45C before the heat transfer in a 0.8 sq m coil drops below this minimum.  So I could operate the HP at 60C to heat the water to 45C, then and use the immersion for the final 5C?  Is this so silly?  

Building further on this, high temperature heat pumps (eg the latest Samsung) can go up to 70C.  So in principle, if they modulate low enough, they can heat the DHW with a standard 0.8 sq m coil with no need at all to change the DHW system (provided you have some sort of tank).

 

The smart thing would be for the pumps to adjust the flow temperature according to the measured DHW temperature in the vicinity of the coil, ramping it up as the DHW heats up.   They would then be more efficient still than having the immersion kick in when the DHW temp rises above a certain value, because the CoP even at the higher flow temp will still be >1. 

 

As shown above, providing the bulk of the DHW heating is at a flow temp of, say, 55C, the annual penalty by using higher flow temps to heat the water through the last few degrees, even with todays high electricity prices, is modest and does not make a convincing case for swapping out a functioning DHW system.

 

Samsung don't seem to publish the modulation data for their HT range.  Nor do we know whether they will adjust their flow temperature down in the required circumstances.  But if the modulation ratio is decent and they do at least some adaptive FT control then these pumps (and likewise other HT pumps with a decent modulation ratio) are capable of working with existing DHW tanks.  That's two 'ifs' unfortunately, but it may well point the way forward even if the current control systems/modulation depth don't play ball

 

 

[SteamyTea]

36 minutes ago, JamesPa said:

Building further on this, high temperature heat pumps (eg the latest Samsung) can go up to 70C.  So in principle, if they modulate low enough,

pV = nRT

It is all in the ideal gas law.

How large do you want p to be?

[JamesPa]

1 hour ago, SteamyTea said:

pV = nRT

It is all in the ideal gas law.

How large do you want p to be?

Not sure what point you are trying to make/question you are trying to ask?

 

All I am doing here is to explore the boundaries of the engineering, to find out whether it is really necessary to swap out the DHW system for most (so far as I can tell) retrofits (and in particular mine), and whether it is really necessary to have a buffer tank.  Taken together these are, so far as I can see, major impediments to mass adoption because of cost, space, and disruption.  Modulation ratio is the key determinant of the answer to both, with max flow temp being a second important factor (but much less so if the modulation ratio is high).

 

I am aware of the basic physics, but the engineering of particular products, whilst constrained by the physics, doesn't go to the boundaries, so we need the specs.

[markocosic]

Plate loaded cylinders work fine

 

And it isn't worth the hassle in domestic 

 

Either for a new cylinder or retrofitting an old one

 

Just fit a new, invented, and properly insulated cylinder with a dirty great big coil designed for heat pump use

 

The only place it does make sense IMO is ultra fast reheat from a gas boiler - pump the cylinder contents through the potable water side of an interface boiler to heat it at 30 kW odd whilst retaining full condensing etc.

 

But that's passed.

 

No sense in new gas installations now unless you're supplying a particularly stupid and stubborn client.

 

[Dillsue]

If you can't get enough heat transfer through a basic indirect coil, you can always add a solar coil to the immersion boss and series plumb the 2 coils. See solarcoil.co.uk.  They do coils with an integrated electric element if you want to retain that feature.

[SteamyTea]

24 minutes ago, JamesPa said:

Not sure what point you are trying to make/question you are trying to ask

Basically it can be done (high temperature and high how) and be modulated to lower levels quite easily, from an engineering perspective.

But I doubt many manufacturers want to install 2 kg of gas at 20p MPa next to a structural wall of a building.

Especially if it unnecessary with existing technology.

 

The environmental/embodied energy/CO2 of scrapping an old, or brand new, cylinder is minor.

[Gone West]

6 hours ago, JamesPa said:

For the rest of the argument I'm assuming that the target temperature for the stored water is 50C  as that seems common, with a weekly anti-legionella 55C boost.

At my last place I fitted a UVC heated with a, IIRC, 650W EASHP. I stored the water at 45C and didn't bother with heating it higher to kill off any bugs.

[sharpener]

4 hours ago, JamesPa said:

Not got there yet.  @JohnMo seems to have identified a useful starting point, but Im not yet sure I understand the calculation fully enough to be confident, nor how the results can be transferred to other manufacturers.  Only working on this intermittently, my wife is getting fed up with the amount of time I've invested in ASHPs.

 

I've had a bit of a play on the web site he linked to. So far this is the best solution I have found, Eur125 list, or  cheaper equivalent Eur51. (N.b. you have to use the tabs to get the price for the correct number of plates for these particular configurations.)

 

Performance as under, will transfer 4 kW with a 5 deg drop from side 1 to side 2, with ~ 12 l/min flow rate and ~0.5m head loss.

 

But I was wondering what result you would get by just having the pump to circulate the water in the tank. I would guess you could easily double the heat transfer compared with simple convection off the coil, this would (assuming linearity of yr model) give DHW at 52.5 and if 3x then 55 without any HX at all.

 

Solarcoil looks interesting and not expensive but I wonder what its surface area is?

 

 

 

 

 

Edited April 8, 2023 by sharpener
Additional info

[JamesPa]

23 minutes ago, markocosic said:

late loaded cylinders work fine

 

And it isn't worth the hassle in domestic 

 

Either for a new cylinder or retrofitting an old one

 

Just fit a new, invented, and properly insulated cylinder with a dirty great big coil designed for heat pump use

 

The only place it does make sense IMO is ultra fast reheat from a gas boiler - pump the cylinder contents through the potable water side of an interface boiler to heat it at 30 kW odd whilst retaining full condensing etc.

 

But that's passed.

 

No sense in new gas installations now unless you're supplying a particularly stupid and stubborn client.

Nice theory I grant you but ... back in the real world...

 

almost all new houses are being built with gas boilers and (doubtless) a cylinder (if they have one at all) with a coil insufficiently large for LT operation

 

replacing an existing perfectly functional DHW system is a severe disincentive to fitting heat pumps, particularly if you are told 'there is no choice'

 

fitting a buffer tank in a small house is a severe disincentive, if you can find anywhere at all to put it

 

'It' hasn't 'passed' at all. We have 10s of millions of houses out there with cylinders that have a 0.8 sw m coil and which urgently need to be retrofitted with heat pumps at a price and level of disruption the residents will tolerate.  Currently they are faced with a bill, after the government grant, of 10-15K.  To the consumer all they are getting is a new boiler worth max 5K.  Why are they going to bother (we know the answer to this already, they don't)

 

You aren't thinking about the retrofit from the consumers point of view and, until we as a nation (and in particular the installation industry) do, we wont fix climate change.

 

 

[JamesPa]

16 minutes ago, Gone West said:

At my last place I fitted a UVC heated with a, IIRC, 650W EASHP. I stored the water at 45C and didn't bother with heating it higher to kill off any bugs.

Interesting, I too was wondering about storing at a lower temp.  Max allowed temp to taps in new build is 48.