“Anyway, we know that the assumption the heat pump will be running continuously is based on the fact a cylinder is assumed to be always part of the solution. This essentially means unless it is a massive property, the system is never going to be powerful enough to provide direct hot water for showers etc. And don't start me on the electricity capacity, we know these pumps run on a normal socket circuit like normal air cons. “

You seem to need to air your incorrect prejudices. You are very ill informed.

ASHPs NEVER use a normal socket nor circuit and if wired like that it is very unsafe and un-compliant 
ASHPs provide plenty of hot water for showers for any number of occupants and don’t need top up immersion heaters either. It’s a question of the size of cylinder (or phase change heat storage like Mix-energy) and stored water temperature. The heating is off is when the hot water is heated so is often set at night, on the cheapest power rate.

Posted by: @majordennisbloodnok

↑

Posted by: @batpred

↑

Yes, but HA allows separation on concerns and also resilience through multiple paths of failure. To be honest, never enough! Dream high and be realistic I'd say.

With batteries doing 8000 cycles I would not loose much sleep. And as others pointed out, the practical limits may be elsewhere.

I am aware these energy systems come with plenty of flexibility and HA is not going to do anything magic. And once it's doing something critical I do not think it is easy to have it delivering consistently anyway.. but it will give you freedom you may not get otherwise. 

In the context of my previous post, I'm not sure what point you're trying to make here.

Having HA concerning itself with what are aspects more likely to change like pricing and having to take into account a growing number of systems seems more future proof. Still, given the way that many HA device facing integration components are tested (and sometimes even "designed"), it is not conducive to high availability. E.g. being able to use different methods to monitor a device on the modbus rather than having one integration flow and assume it will be stable enough may be the only example. That way maybe the "layer" can be relied on.  

Posted by: @majordennisbloodnok

↑

As for putting batteries through 8000 cycles, I would agree; not much of a problem. However, with Octopus Agile providing half-hourly prices, it wouldn't be difficult to put a battery through hundreds of thousands of cycles in the name of, as you put it before, "pushing the boundaries of optimisation", and yet realise just a few quid more than a dramatically simpler strategy that provides almost as much benefit for almost no effort to implement. It's the 80:20 rule on steroids.

The cycles that manufacturers quote are complete, from nil to full. So charging twice from 20 to 70 is one full cycle. But as others pointed out, there may be electronic components that cannot be reprogrammed more than x,000 times. 

When I mentioned pushing the boundaries of optimisation, I meant moving from having to configure components and install cables to change the way battery or grid are used to using an HA based system.     

Posted by: @grahamf

↑

You may have been lucky with your analysis.  Still, when a press reporter suggested to the great Gary Player that he had been a lucky golfer, he replied "The more I practise, the luckier I get." 

There is some truth and perhaps even relevance in that!

Posted by: @grahamf

↑

Our house is a 350m2, so intuitively it feels as though the heat loss should be relatively large.  We have about 70% loft the loft insulated to 300mm and the rest to about 100mm.  We have cavity wall insulation and double glazing installed in 2009.  The floors include concrete, concrete on insulation, suspended wood with foam underneath and suspended wood with a small air gap to concrete.

It sounds like just the sort of house for which the standard fabric calculation could easily be widely out (because of the variable construction - like mine) and for which therefore an independent check based on measured consumption, however obtained or crude, might well be useful if it resolves an important design fork (like 2 fan vs 1 fan, pipe upgrades vs not, 2 unit vs 1 unit or - as in a current live case on the forum, 8kW Daikin vs 9kW Daikin more significant than it looks because the 9kW is actually a 16kW downrated so, for a 5kW house, is way outsized.)

Posted by: @grahamf

↑

I had problems with one consultant who refused to use a sensible value for ACH.  It makes a massive difference to the total heat loss and seems to be the most common reason for over-estimating it.  People don't seem to use their common sense and think about what 2 air changes per hour actually means.  It is equivalent to opening all the windows and losing all the warm air in the room every half an hour!  A house with decently sealed windows and doors won't come close to that.

Me too, probably the majority TBH.  In the end, once I knew what I wanted, I specified that and included the evidence of loss in the request for quote.  Any who told me that they 'couldn't' take notice simply weren't shortlisted.  

Posted by: @judith

↑

batpred: 

 

Anyway, we know that the assumption the heat pump will be running continuously is based on the fact a cylinder is assumed to be always part of the solution. This essentially means unless it is a massive property, the system is never going to be powerful enough to provide direct hot water for showers etc. And don't start me on the electricity capacity, we know these pumps run on a normal socket circuit like normal air cons. 

 

You seem to need to air your incorrect prejudices. You are very ill informed.

ASHPs NEVER use a normal socket nor circuit and if wired like that it is very unsafe and un-compliant 
ASHPs provide plenty of hot water for showers for any number of occupants and don’t need top up immersion heaters either. It’s a question of the size of cylinder (or phase change heat storage like Mix-energy) and stored water temperature. The heating is off is when the hot water is heated so is often set at night, on the cheapest power rate.

Apologies if required Judith, as I had no intention to offend anyone! And no offense taken.  

Of course I am not suggesting a normal socket being used. Anyone would be best advised to use a certified electrician to set them up anyway.

Just stating that, as things stand, most people will have a small heat pump installed that will run from a circuit designed to supply 16A. These type of fixed appliances will of  course be wired in (like gas boilers are meant to, etc) and not plugged on a socket.

But when we say an 8kw heatpump, this is not using 32A. If it has a COP of 4, it draws 2k, so less than 10A.  

Please correct me if I missed anything. 

Posted by: @batpred

↑

What I mean is that, given it is healthier to have a cooler environment to sleep than when people are active, when the systems are being systematically being setup to maintain constant temperature, when each design is created, arguably it misses a default requirement. I had not even looked at the document you reference. I am not sure if this was done to maximise operating efficiency or the system but it does not facilitate user centered design. 

Sorry but the document I quoted is a requirement for an MCS installation and specifically specifies different design temperatures for bedroom (where people sleep) and living room (where people are active).  Every design I have seen has respected these differences.  This being the case what is the evidence for your statement? 

Posted by: @batpred

↑

Would there not be cases where a user would like the pump to deliver hotter heating water to avoid resizing circuits, etc? 

Yes there may well be IMHO and some installers will offer this, whereas others insist on designing for maximum efficiency.   The heat geek 'zero disruption' initiative is an example where they specifically offer minimal upgrades as an option.  Others on here have been critical but for my money, provided it is positioned and explained correctly, it is offering choice which is not a bad thing.

Posted by: @batpred

↑

From what I saw people are being asked to pay for these reports upfront. Is the report then just used by a single installer to create a proposal based on the products they use? Or is it available for the user to provide to other installers so they can make their proposals? 

If the user has paid (and unless there is a contractual clause to the contrary) it is the users property but other installers will, at least in my experience, generally not accept it for obvious liability reasons.  So in reality its installer specific.

Posted by: @batpred

↑

...

Having HA concerning itself with what are aspects more likely to change like pricing and having to take into account a growing number of systems seems more future proof. Still, given the way that many HA device facing integration components are tested (and sometimes even "designed"), it is not conducive to high availability. E.g. being able to use different methods to monitor a device on the modbus rather than having one integration flow and assume it will be stable enough may be the only example. That way maybe the "layer" can be relied on. 

...

That doesn't make much sense. HA can be implemented with as great a degree of redundancy as you might wish, and many different strategies for doing so - ultimately overkill for almost all domestic situations.

Posted by: @batpred

↑

...

Posted by: @majordennisbloodnok

↑

As for putting batteries through 8000 cycles, I would agree; not much of a problem. However, with Octopus Agile providing half-hourly prices, it wouldn't be difficult to put a battery through hundreds of thousands of cycles in the name of, as you put it before, "pushing the boundaries of optimisation", and yet realise just a few quid more than a dramatically simpler strategy that provides almost as much benefit for almost no effort to implement. It's the 80:20 rule on steroids.

The cycles that manufacturers quote are complete, from nil to full. So charging twice from 20 to 70 is one full cycle.

...

I'm very well aware. This is not my first rodeo.

Posted by: @batpred

↑

...

But as others pointed out, there may be electronic components that cannot be reprogrammed more than x,000 times.

...

Once again, I'm very well aware. If you look back, you may find me being one of those doing that pointing out.

Posted by: @batpred

↑

...

When I mentioned pushing the boundaries of optimisation, I meant moving from having to configure components and install cables to change the way battery or grid are used to using an HA based system.

With respect, all you're talking about is using HA for the job it was designed to do. Not sure that's pushing boundaries - except, of course, for anyone relatively new to HA, in which case they're pushing their boundaries, not Home Assistant's.

@batpred There will be times when the 8 kW heat pump starts up with a surge much higher than 2 kW and also when adverse weather drags the COP down, there may well be periods where that same pump could draw far more than that 2 kW for an extended period of time. Regards, Toodles.

Posted by: @majordennisbloodnok

↑

Posted by: @batpred

↑

...

Having HA concerning itself with what are aspects more likely to change like pricing and having to take into account a growing number of systems seems more future proof. Still, given the way that many HA device facing integration components are tested (and sometimes even "designed"), it is not conducive to high availability. E.g. being able to use different methods to monitor a device on the modbus rather than having one integration flow and assume it will be stable enough may be the only example. That way maybe the "layer" can be relied on. 

...

That doesn't make much sense. HA can be implemented with as great a degree of redundancy as you might wish, and many different strategies for doing so - ultimately overkill for almost all domestic situations.

I am just discussing some concepts, I am learning in some domains and will not become defensive for it. 

On this, we could even agree, of course it can be implemented.  But I am yet to see anyone try to explain the framework they designed in HA so it operates consistently and securely handle for example a heat pump - treating it as a dumb system. 

But also anyone reading this should think HA is not a quite capable system, it is. Just that beyond use cases for own development and use, it has many limitations. much of the code people share does not work unless it is very recent, as the platform is developed at speed.  

Posted by: @majordennisbloodnok

↑

Posted by: @batpred

↑

...

When I mentioned pushing the boundaries of optimisation, I meant moving from having to configure components and install cables to change the way battery or grid are used to using an HA based system.

With respect, all you're talking about is using HA for the job it was designed to do. Not sure that's pushing boundaries - except, of course, for anyone relatively new to HA, in which case they're pushing their boundaries, not Home Assistant's.

Of course I am talking about using HA for to deliver basic high level control of one of these wired up electrical control system being a case of pushing the boundaries - of what is possible without HA.   

But even if it can be developed in HA quicker than installing cables, etc. anyone claiming a standard HA deployment that us also running the usual home automations can also deliver all that a critical system needs to operate (essentially replacing some critical logic) has not tried the platform enough: testing, etc 😉

And of course it would not be generally acceptable to ignore software version updates and security patches unless you properly protect the security perimeter of that enclosed HA system - and as you stated, in the end, the protections would stop it being able to do what it was meant to do! 

@batpred 

my cos6 is 5.6kW HP and draws 500W-1.6kW

Posted by: @jamespa

↑

Posted by: @batpred

↑

What I mean is that, given it is healthier to have a cooler environment to sleep than when people are active, when the systems are being systematically being setup to maintain constant temperature, when each design is created, arguably it misses a default requirement. I had not even looked at the document you reference. I am not sure if this was done to maximise operating efficiency or the system but it does not facilitate user centered design. 

Sorry but the document I quoted is a requirement for an MCS installation and specifically specifies different design temperatures for bedroom (where people sleep) and living room (where people are active).  Every design I have seen has respected these differences.  This being the case what is the evidence for your statement? 

Thanks and apologies if I was a bit lazy...I have not paid much attention to designs nor the proposals, but the systems and how users describe they should be used. 

I reported details of a case where some friends of mine had constant temperature for the house, they were focused on efficiency and had this idea that to maximise it, the pump had to be working at all times...

Then we had an installer over was also very much insisting that the system should be running continuously. For efficiency. 

There are also examples people shared here where the idea of having a stable temperature is presented as an advantage and I think also linked to efficiency.. In some cases it seemed there was no simple way to adjust it. Maybe I perceived it differently and they just meant the external unit part should not run for short periods.

Not what I could call evidence but a pattern that I have to admit I assumed could be influenced by the way the systems were designed - but it could be things further down the pipeline.

Thank you for insights on how installers may handle the customer preferences when deviating from most efficient.

I would guess that as a principle and since the system can accumulate heat in the cylinder, most customers would  want the external unit running considering when their electricity is cheaper and generally whenever the external temperature is warmer but there will surely be many other other factors that in some cases may even be more important to them. 

There are also various people posting in this forum talking about running the external unit continuously. Perhaps I misunderstood and they meant just the external unit to not run for short periods

Installing ashp that run "efficiently" or cost effectively is not a goa simple to achieve l! 

@batpred - I think it might help somewhat if you could set down what you want from your heat pump discussion here on the forum. As I currently understand it, these are hypothetical, insofar as you don't have plans to install a heat pump any time soon, and it is always a little difficult when discussing hyotheticals because, well, they are hypothetical. That said, establishing some heat pump basics, and generally making sure anything you do now in another context (eg batteries etc) wont conflict with any future heat pump installation is certainly sensible.

Establishing some heat pump basics - at times I am not sure you have grasped these. They are not complicated. The basic technology is over a century old (it first appeared as a fridge, but it is still doing the same thing, pumping heat from one place to another). An ASHP pumps heat from the outside air into a building to warm it. Because an ASHP can extract more heat energy from the outside air than it uses to get that heat energy, we have the concept of performance, normally expressed as the coefficient of performance, or COP. This is simply the ratio between the energy in and the energy out from the heat pump. If you put 1kWh in, and get back 3kWh, then your COP is 3. The higher the COP, the better the performance, or efficiency, though I prefer the term performance to efficiency, because there is something not quite right about the idea of 300% efficient (because on the percentage scale, 100% is tops, the max. It is the same sort of nonsense as people saying they put in 110% effort - there is no such thing as 110% effort). 

The reason why we obsess over performance (via the COP) is because it directly affects bills. All other thing being equal, the higher your COP, the lower your bill for the same outcome. There are also sustainability and other advantages, because you consume less energy to achieve the same end result.

The two main things (again all other things being equal) that affect COP are (1) the outside air temperature (OAT) and (2) the actual flow temperature of the water leaving the heat pump when it is running (FT, also often called the LWT, or leaving water temperature). Both a low OAT and a high LWT will lower your COP. We can't do much about the OAT in a fixed location, but we can design a system to keep the LWT as low as possible, mostly by using big emitters (radiators and/or under floor heating).

Thus we arrive at one of the key differences between a fossil fuel heating system and an ASHP heating system. In the former, boilers were often hugely oversized (a sort of one size fits all approach with even more capacity just in case) and gas was (and still is compared to electricity) cheap and so folks just ran them hard when they needed them. They had more than enough capacity to get from a standing start up to full speed in minutes rather than hours, with the LWT rapidly getting to a high temperature. But this absolutely not how you want to run a heat pump, because getting that very high LWT means a very low COP, which in turn means very high bills. No one wants unnecessarily high bills, so we run our heat pumps low and slow, Steady Eddie, on all the time at as low a LWT as possible. We have, in effect, to ditch our old fossil fuel thinking, and learn the new ASHP way of thinking.

Some of us monitor our ASHP systems closely, and we have shed loads of empirical data that show in great detail exactly how heat pumps behave. Sure enough, low OATs mean lower COPs (all other things being equal), and sure enough heat pumps are Steady Eddies, they take hours or even days to get a building up to temperature, but once they get there, they are very good at keeping them there.

In my own personal view, HA is a cabal run by a bunch of python sadists. It never stays the same, the code makes hieroglyphics easy to read, and they spend far to much time saying now here's a bad solution, where can we find an even worse problem to apply it to? Others absolutely swear by HA, and think it is wonderful, it's just horses for courses. I personally prefer wired connections and simple python code written by me, but that's just my horse for my course. The vast majority of people (90% or more is my guesstimate) never even get anywhere near all this monitoring and coding and controlling stuff, they just run their heat pump using the manufacturers controls.

Which brings me to the final point on space heating with an ASHP, weather compensation. Because heat pumps can't do fast and furious, and instead run on all the time low and steady, there is a sweet spot at a given room temp where the energy supplied to the building exactly matches the energy lost. At that point, the room temp stays stable and the occupants are comfortable. Since the OAT is the main determinant (all other things being equal) of heat loss, we can set up a 'program' on the controller that adjusts the LWT depending on the OAT, and by tweaking that program, or weather compensation curve, can arrange things such that most of the time the building gets from the heat pump the same amount of heat energy as it is losing to the environment, thus keeping the room temperature stable and comfortable. It really is as simple as that!

And last but not least, much the same applies to DHW heating, We have to ditch the fossil fuel thinking, and instead go low with Steady Eddie, ie abandon the idea of instant hot water, and instead think of a DHW store, or reservoir, in the form of a tank, sized to suit our needs, and drawn from when we want hot water.

Maybe some of the above will answer some of your questions, if not, as I said at the beginning, please do firm up on what it is you want from this discussion, so we can give you our best efforts at answers.

@batpred, English does not seem to be your first language and I'm struggling to understand quite a lot of the points you're trying to make as a result. I also have a strong suspicion that you're using some terms - like "high availability" - that have very specific technical meanings that are not what you are trying to communicate. I would be grateful, therefore, if you could rephrase and simplify your last response to my post so that I can give you a proper answer.

Working through disagreements is hard enough without having to do so in a second language, so I'd prefer to remove that obstacle as much as possible and concentrate on the points you're trying to convey.