Mini Store - Heat Geek / Newark Cylinders for ASHP

[JamesPa]

3 hours ago, sharpener said:

We have had many debates on here in the past about optimising various combinations of storage, heat input and auxiliary electric heating whether in the same vessel or as @Beelbeebub suggests upthread in cascade, and have yet to unearth the silver bullet.

I think that the amount of energy required to satisfy the demand for luxury, which either has to be stored or generated on demand, means that there is no silver bullet.  In large new builds a large uvc is an obvious choice, but in smaller retrofits much less so and so compromises must be made.  There are quite a few variables to play with in the trade offs so it's almost certain that a variety of solutions are needed.  Thus any innovation, such as this one is to be welcomed.

 

It's just a bit disappointing that they have conflated small size with a new cylinder configuration, as it would help immensely to tease out the effects of these factors individually (particularly the latter).  My qualitative argument above shows that the latter is dependent on the engineering not the fundamental physics, so we do need to understand how the engineering has panned out (IE what, if anything, is the yield difference relative to a uvc of the same volume) to make an informed decision in any particular application. 

 

As I say above, if the dhw yield penalty compared to a similarly sized uvc is fairly small, then the choice in a retrofit situation becomes (IMHO) almost a no brainer, simply to avoid the hassle and installation and ongoing expense of g3.

Edited June 17 by JamesPa

[JohnMo]

30 minutes ago, JamesPa said:

just a bit disappointing that they have conflated small size with a new cylinder configuration, as it would help immensely to tease out the effects of these factors individually (particularly the latter).

They do several sizes

 

[Dave Jones]

they are re-inventing the wheel.

 

We use a bog standard cylinder with an external plate heat exchanger. Heats 300L tank to 50c in less than 30 mins. Can also daisy chain more cheap cylinders as required for capacity.

[JohnMo]

44 minutes ago, Dave Jones said:

they are re-inventing the wheel.

Not really re inventing, just a variation, like spoked becoming alloy wheels. Does the same thing packaging slightly different.

[JamesPa]

1 hour ago, Dave Jones said:

they are re-inventing the wheel.

 

We use a bog standard cylinder with an external plate heat exchanger. Heats 300L tank to 50c in less than 30 mins. Can also daisy chain more cheap cylinders as required for capacity.

If it's unvented then it still needs G3 which the hg one doesn't.

 

However I do agree that the use of a phe instead of a coil is another variant that has legs.

 

They are trying to solve a different problem to your situation, and the problem is that there are a vast number of situations and no ideal solution so we do need lots of variants.  To continue with @JohnMo's thought track, there are lots of different models of car, none with fundamental differences, yet each with a market.

 

I think they deserve some credit for sticking their heads above the parapets in a part of the industry that seems very reluctant to innovate (I would go so far as to say that it's stuck in a rut).

Edited June 18 by JamesPa

[JohnMo]

5 minutes ago, JamesPa said:

I think they deserve some credit for sticking their heads above the parapets in a part of the industry that seems very reluctant to innovate (I would go so far as to say that it's stuck in a rut).

Think all industry is the same in the UK now, we have become a culture that changes black boxes, with little or no thought - computer say... they do and don't look much further.

[JamesPa]

10 hours ago, JohnMo said:

They do several sizes

 

True, but so far we don't have any comparative performance figures relative to a uvc of the same size. 

 

What I really want to know is, if I need (eg) a 150l uvc to deliver x amount of dhw at y l/s, how big would this arrangement need to be? 

 

If the answer is 150l or less, then the only remaining argument for a uvc might be heat pump efficiency (which may or may not differ between the two). 

 

If the answer is within 10-20 % of 150l then I would say that in most cases, unless space was extremely tight, you would choose this over a uvc to avoid g3.

 

If it's more than 20% larger to get the same yield, then the hg mini is only for situations where the g3 arrangements are much more difficult than accommodating a larger cylinder, and the solution for smallest volume (which seems to be what they are targeting) is actually a small uvc not the hg mini!

 

Hopefully this detail will come.

Edited June 18 by JamesPa

[JamesPa]

1 minute ago, JohnMo said:

Think all industry is the same in the UK now, we have become a culture that changes black boxes, with little or no thought - computer say... they do and don't look much further.

That's a very depressing thought!  (But I'm not disagreeing!).

[dpmiller]

at the end of the day this scheme is all about the recovery, it's an ASHP combi. *If* you can get the heat pump on and at full chat before the store cools too much, then for a particular maximum flow rate you've an inexhaustable supply of DHW.

 

heck, you can do the same to have a gentle shower straight off a Willis on an vented cylinder. BTDT. But we're spoilt now, seemingly requiring massive flow rates for everything...

[sharpener]

1 hour ago, JamesPa said:

What I really want to know is, if I need (eg) a 150l uvc to deliver x amount of dhw at y l/s, how big would this arrangement need to be?

 

A further thought about the flow regime in the tank (assuming normal vertical mounting).

 

The concept relies on the coil surface area being big enough to transfer the heat to the flowing DHW at a fast enough rate (50kW is mentioned at one point).

 

This will rapidly cool the body of water in the tank in the immediate vicinity of the coil. So if this cooler water stays put it will soon reduce the effectiveness of the heat transfer, if it mixes it will reduce the effectiveness in a different way, if it falls to the bottom of the tank it will displace the warmer water towards the top which is the desirable outcome.

 

But after the 3 mins has elapsed the HP will start up so we do not need to worry about natural convection any more. With a flow rate of ~3 l/min per kW of output it will push re-heated water into the top of the tank and move it downwards. This will improve the heat transfer and is OK if the HP can reheat the water fast enough. But if the heat being abstracted is greater, then the return temp to the HP will steadily fall so the flow temp will fall too. Then the water surrounding the top of the coil will not be hot enough to give the desired output temp.

 

So I am not sure how this concept will perform against the normal DHW cyl where stratification is quite good and the water for your shower stays very hot until quite suddenly it isn't.

 

The video picks a scenario where the HP meets the entire heat requirement which does not place any real demand on the store at all. In real life I forsee a steady decline in flow temp from quite early on - so in practice you will need to heat the water to well above the shower temp you want and the efficiency gains will then not be achieved.

[JamesPa]

49 minutes ago, sharpener said:

The video picks a scenario where the HP meets the entire heat requirement which does not place any real demand on the store at all. In real life I forsee a steady decline in flow temp from quite early on - so in practice you will need to heat the water to well above the shower temp you want and the efficiency gains will then not be achieved.

The hp in the video doesn't meet the demand to heat water from mains temp to dhw temp, we know that because the dhw  eventually falls below temperature.  However I think it does (and probably generally will) meet the demand required to restore the return from the tank to the tank temperature, just not at the (flow) rate sufficient to restore all the heat lost from the tank to the dhw.

 

So what I think happens in a perfectly stratified tank is that the thermocline moves upwards at a rate slower than would be the case without reheat, eventually to the point where the thermocline is high enough in the tank that the dhw coil can't extract enough energy to heat the incoming mains to the desired temperature.

 

I don't think they are particularly claiming efficiency gains and are I think heating the store to 50/55 (I can't remember which) which is not an uncommon flow temp for dhw, bearing in mind that there is typically a 5C offset between the flow temp to a uvc and the actual water temperature unlike the mini where the offset is zero.  Obviously some run it cooler to maximise efficiency at the expense of capacity, this is a design trade off with any cylinder!.

 

It all comes back imho to, how does the yield of dhw compare to a uvc of the same volume and at the same heat pump ft.  Answering that question alone tells us the circumstances in which this is useful and any efficiency matters.   Unfortunately it's not answered!

 

 

Edited June 18 by JamesPa

[sharpener]

8 minutes ago, JamesPa said:

So what I think happens in a perfectly stratified tank is that the thermocline moves upwards at a rate slower than would be the case without reheat

 

Yes, ideally. My concern would be that with say a 5kW hp that gives a flow through the tank of 15l/min, and this may actually be faster than you want, as in 10 mins you are turning over the entire contents of a small tank. Moving the thermocline downwards by adding less hot water at the top is not going to help the dhw ft.

 

Time for one of your spreadsheet models @JamesPa?

 

(They don't seem either to claim any of the features you sometimes see aimed at preserving stratification, such as baffle plates or curved or tangential-flow inlet pipes.)

 

15 minutes ago, JamesPa said:

It all comes back imho to, how does the yield of dhw compare to a uvc of the same volume and at the same ft. 

 

At best it would be the same but for the reasons we have been discussing I think it will be more difficult to achieve than in a well-stratified conventional cylinder.

 

[JamesPa]

9 hours ago, sharpener said:

Yes, ideally. My concern would be that with say a 5kW hp that gives a flow through the tank of 15l/min, and this may actually be faster than you want, as in 10 mins you are turning over the entire contents of a small tank. Moving the thermocline downwards by adding less hot water at the top is not going to help the dhw ft.

Somewhere in the video the hp was operating at way less than max output.  However only some hps are set up to modulate the water pump and, as you say, the turnover may be too large.  Need to think about how the control loops work to understand this.

 

9 hours ago, sharpener said:

(They don't seem either to claim any of the features you sometimes see aimed at preserving stratification, such as baffle plates or curved or tangential-flow inlet pipes.)

I'm pretty certain one of the diagrams showed a diffuser at the flow inlet.

 

9 hours ago, sharpener said:

At best it would be the same but for the reasons we have been discussing I think it will be more difficult to achieve than in a well-stratified conventional cylinder.

Agree unless there is some efficiency advantage because the flow from the hp directly feeds the tank.  

 

All of the above is, in my view, an engineering unknown.  And all of the above can be answered if we get the answer to the question about the comparison with a uvc of same size and operated at same ft (except some of the pump control questions which may be ho dependent).

 

I wonder if they have thought about all if this or are just throwing it at the market following some basic experiments, to see what happens?

Edited June 18 by JamesPa

[JamesPa]

21 hours ago, sharpener said:

Time for one of your spreadsheet models @JamesPa?

As I conclude above the answer lies in the engineering detail not the fundamental physics.  Unfortunately neither my spreadsheet models nor skills extend to computational fluid dynamics combined with thermal flows, that's more a Met office thing. 

 

We need to persuade Heat Geek or Newark to run the experiment.

[sharpener]

12 hours ago, JamesPa said:

I wonder if they have thought about all if this or are just throwing it at the market following some basic experiments, to see what happens?

 

Yes, it's a mystery to me. And the Newark/HG tie up for the super expensive massive coil tank too, AFAICS does not make much economic sense, see e.g. this article.

 

I would have just brought the single most useful install-in-kitchen-cupboard size to the market first. The wide range just confuses the choice even further IMO.

 

Also they have had nearly 6 months since Adam's first video to sort all this out, yet I cannot find the new cyls on the Newark web site as yet. Perhaps the formal product launch is set for the Installer Show which is yet to happen? Hard to justify but I might go again this year (and can get there and back on one EV charge).

 

1 hour ago, JamesPa said:

We need to persuade Heat Geek or Newark to run the experiment.

 

I expect the OEM community will be onto this soon (see above link).

 

Will know a bit more myself once I can do the "is it worth adding a bronze pump to stir the tank" experiment (although that is the other way round it might tell us something).

 

 

Edited June 19 by sharpener

[JamesPa]

1 hour ago, sharpener said:

14 hours ago, JamesPa said:

I wonder if they have thought about all if this or are just throwing it at the market following some basic experiments, to see what happens?

 

Yes, it's a mystery to me. And the Newark/HG tie up for the super expensive massive coil

This is now on the Newark site.  I've asked the question about yield relative to a same sized uvc.

 

https://newarkcylinders.co.uk/hg-mini-store-coming-soon/

 

The massive coil cylinders are also interesting.  It will increase efficiency by reducing the approach temperature, and reduce reheat time.  Taken together this might allow you to move down a cylinder size for comparable yield.

 

The dhw industry seems to be a bit 'seat of the pants' in terms of useful specs beyond volume and physical dimensions thus making it nigh on impossible to make any real comparisons.

 

Edited June 19 by JamesPa