Convert existing vented cylinder to thermal store and PHE

[Dan Robinson]

Hot taps in burst? Maybe, but once the initial cold slug is discharged you are not wasting anything significant.

So to install a secondary return jut to eliminate that cold bit at the start seems excessive for one or two outlets.

Especially if installing the return loop is simple... Fitting a new small diameter pipe is cheaper and less wasteful long term.

 

[SimonH2]

On that, you may find it beneficial to have a surface stat on the PHE and fire the pump occasionally during "occupied hours".

Forget that.
If you look at photos of heat banks from what used to be DPS, they deal with this by embedding the PHE in the tank insulation - so it's in contact with the hot tank and so keeps warm that way. They put the PHE at the top of the tank, with the pump in the longer leg between PHE and the return tapping at the bottom of the tank.

 

[Dan Robinson]

Not with the Excel annoyingly (my one). It is (mine not wny more ) riveted to the side skin.

I took mine off and will eventually cut it into the skin. In the mean time I have wrapped it in that Thermawrap stuff.

 

[1028741]

Hot taps in burst? Maybe, but once the initial cold slug is discharged you are not wasting anything significant.

So to install a secondary return jut to eliminate that cold bit at the start seems excessive for one or two outlets.

Especially if installing the return loop is simple... Fitting a new small diameter pipe is cheaper and less wasteful long term.

There are 5 outlets on one long 22mm run starting in the airing cupboard, going down through the guest bath, on through the utility, under the floor to the kitchen island, then under the floor to the main kitchen sink. All outlets are used infrequently for hand washing, spraying off plates or filling the washing machine, so a loop on a timer that recirculates for about 1 min every 3 hours when I'm in is the best option. I can run a 15mm return from the kitchen sink under the cabinets and up through the hall closet to the airing cupboard. Even if I had small bore runs to each outlet, they would be really long because there is no "central" place in the house, everything is spread out laterally you see.

In the US, standard practice is to place a manifold in the basement, and run individual 10mm pex runs to each individual outlet. As the pipes are running upward, the length of the pipe tends to stay warm because the heat naturally rises upward through the standing water from the cylinder.

 

[1028741]

Simon, my rads and underfloor heating are on separate zone valves, however, the underfloor heating pipework and manifold is in a different location to the cylinder and connected to the flow and return in the kitchen wall, so it would be virtually impossible to run a new flow and return from the cylinder. I could easily connect the rads to the cylinder though. Is it worth it though? The rads take quite a while to heat up because there are a lot, and they are big reproduction cast iron and the pipework is mainly 22mm. Is it a lost opportunity to just have the direct cylinder to heat the PHE for hot water? I'm more concerned about having plentiful hot water than more efficient heating, because the heating is on most of the time anyway.
thanks for your input by the way, much appreciated

 

[SimonH2]

It sounds like hooking up the heating to your store might not be the right answer for you - but (as my sig says) I'm not an expert in this. For the rads, it's not so much the water content as the load that matters - the rads may take a lot of hot water to heat them up, but once they are hot then the heat requirement is dictated by the thermal losses of the room.

Logic says that if the load on the rads is such that you can keep enough flow to keep the boiler running without a bypass then the gains are minimal. On the other hand, if you get to the point where the boiler cannot range down enough and/or you need a bypass open (which can stop a boiler condensing) to keep the flow rate up then it may be beneficial to run the rads off the store. That was one of the key reasons I did it that way - with a small flat needing only something like 2kW to keep it warm (I measured it, in cold weather) and a boiler that could only range down to a tad under 10kW, it was clear that the boiler and heating weren't matched.
Another factor for me was having the immersion heater as backup for the heating as well as the hot water for when the boiler breaks down - but given the size of your installation, I don't think that's going to be a factor (a 3kW immersion heater isn't going to go very far). On the other hand, the boiler hasn't broken down once since I stopped using it's combi function <reaches for bit of wood to touch>.

Other factors that would influence the design would be if your boiler supports weather compensation and stuff like that.

 

[Agile]

There is something of a problem with the capacity of a store used to supply domestic hot water using a plate HE.

The store temp needs to be at least 10 C above the delivery temp! Thats often 60 C

So the only heat capacity that you can usefully use is between 60 C and 80 C which is why they are specified for a store temp of 82 C.

But 82 C is not a very efficient flow temp for a condensing boiler. This is somewhat remedied by fitting a TMV on the return to the boiler so that the return is kept lower.

 

[1028741]

The store temp needs to be at least 10 C above the delivery temp! Thats often 60 C

So the only heat capacity that you can usefully use is between 60 C and 80 C which is why they are specified for a store temp of 82 C.

I thought this was the case, that's why I originally said, is it ok to just turn up the cylinder stat to 80 degrees and modify an old indirect cylinder to use as a thermal store instead? Do PHE's heat up the water INSTANTLY? Or is there a short heat-up time for the potable water side of the plates? My underfloor heating has a TMV fitted in the opposite way (you don't want that at 80 degrees!)

 

[1028741]

although thinking about it, 50 degrees is more than hot enough for hot water, you just mix less cold with it at outlets. There shouldn't be any risk of legionella because it's mains pressure and there's no potable water being 'stored' as such, therefore the thermal store temp only needs to be 60-70 degrees right?

 

[SimonH2]

There is something of a problem with the capacity of a store used to supply domestic hot water using a plate HE.

The store temp needs to be at least 10 C above the delivery temp! Thats often 60 C

So the only heat capacity that you can usefully use is between 60 C and 80 C which is why they are specified for a store temp of 82 C.

Err, I respectfully disagree somewhat with that.

Firstly, a critical factor is the "size" (and design) of the PHE. There is no reason the DHW shouldn't leave the PHE at very close to the temp of the stored water - certainly a lot closer than 10&#730;. If the DHW is leaving the PHE 10&#730; cooler than the store, then that means the PHE is undersized for the flow rate - or plumbed wrong, or the primary flow rate is too low.

The worst case for DHW temp will be at max flow rate - which also happens to coincide with high efficiency of heat transfer in terms of what you can pull from the store. Assuming a fixed primary circuit flow rate, then :
At low DHW flow rates, the incoming cold will heat up virtually to the primary temperature - and it will do so while only part way through the PHE. The primary return temp will be quite high.
When the DHW flow rate is high, then it will come out at a lower temperature - determined by the flow rates, and temperature difference. IF DHW and primary flow rates are well matched, then the primary return temp will be quite low - and in theory could be only a few degrees above your cold mains temperature, meaning you can extract a lot of heat from the store. It is under these conditions where the size of the PHE is the limiting factor in heat transfer - to a point.

 

[SimonH2]

Oh yes, one reason manufacturers tend to specify high store temps is so their stores look good when people look at the figures. As in "oh look, it can do x l/m of hot water". Or put another way, for a given spec, they can get away with a smaller coil/PHE.

Go up a size in coil/PHE and you can run them at a lower temperature. Like all things, it's a trade off.

 

[Dan Robinson]

Exactly Simon. Tony, my store is a fraction over 60 degrees (input flow temp is 70), with 22 litres drawing off the dhw we still have 60 off the plate because it is big.

I'm actually toying with the idea of fitting another one in series so I can run the store even cooler.

 

[SimonH2]

I'm actually toying with the idea of fitting another one in series so I can run the store even cooler.

Apologies to the OP for the thread diversion, but ...
IIRC it was on another forum where there was some discussion on how to optimise a heat bank setup. The context was probably how to a) maximise how much heat your could get out, and b) maximise the heat you could put in with things like thermal solar panels.

At first, there was a bit of a discussion around limiting the flow rate in the primary circuit, resulting in these two options :

Control options round DHW PHE

• SimonH2
• 12 Aug 2011
• 2

A couple of potential control valve options to restrict primary flow and avoid destratification.

The idea is to control the flow in the primary circuit so as to better match the DHW side. One directly controls the DHW temp, the other controls the tail end temp coming out of the PHE - if it's hot then you are putting more water through than is required. However, my (limited) experience with industrial systems says that controlling the output from a PHE by controlling the mass flow rate in the primary side is "challenging" to say the least - the conditions are so variable that it's not possible to setup (say) a PID controller to react quickly without being unstable under at least some of the operating range.

After a bit of thought, and a few iterations, I came up with this idea :

Dual PHE option for heat bank

• SimonH2
• 8 Apr 2013
• 1

Idea for optimising control of primary flow in heat bank setup, using a second PHE to minimise...

The theory is that the high flow rate round the primary of the upper PHE means you get rapid and stable response to changes in DHW load. The lower PHE ensures that by the time the primary flow goes back to the store, then it's about as cold as is practical for the given DHW flow rate. As I mentioned, the context was in maximising input for low grade sources (solar panels, heat pumps) - the lower the store temperature, the more heat you can get from a solar panel or the higher the efficiency of a heat pump. Even if these aren't applicable aims, it maximises heat extraction and it would certainly help in making your boiler condense !
While the TMV is shown with a remote sensor sensing DHW temp, if your PHEs are of sufficient capacity, then just controlling primary loop flow temp would probably give you good enough control.

I had a go at modelling it, but I couldn't find a model for a PHE that would work in a spreadsheet. I had a go at modelling one as a series of lumped heat exchange elements, but I couldn't get it stable.

I just throw this one your way since you were thinking of adding another PHE.

 

[tew1]

I'm actually toying with the idea of fitting another one in series so I can run the store even cooler.

Apologies to the OP for the thread diversion, but ...
IIRC it was on another forum where there was some discussion on how to optimise a heat bank setup. The context was probably how to a) maximise how much heat your could get out, and b) maximise the heat you could put in with things like thermal solar panels.

At first, there was a bit of a discussion around limiting the flow rate in the primary circuit, resulting in these two options :

Control options round DHW PHE

• SimonH2
• 12 Aug 2011
• 2

A couple of potential control valve options to restrict primary flow and avoid destratification.

The idea is to control the flow in the primary circuit so as to better match the DHW side. One directly controls the DHW temp, the other controls the tail end temp coming out of the PHE - if it's hot then you are putting more water through than is required. However, my (limited) experience with industrial systems says that controlling the output from a PHE by controlling the mass flow rate in the primary side is "challenging" to say the least - the conditions are so variable that it's not possible to setup (say) a PID controller to react quickly without being unstable under at least some of the operating range.

After a bit of thought, and a few iterations, I came up with this idea :

Dual PHE option for heat bank

• SimonH2
• 8 Apr 2013
• 1

Idea for optimising control of primary flow in heat bank setup, using a second PHE to minimise...

The theory is that the high flow rate round the primary of the upper PHE means you get rapid and stable response to changes in DHW load. The lower PHE ensures that by the time the primary flow goes back to the store, then it's about as cold as is practical for the given DHW flow rate. As I mentioned, the context was in maximising input for low grade sources (solar panels, heat pumps) - the lower the store temperature, the more heat you can get from a solar panel or the higher the efficiency of a heat pump. Even if these aren't applicable aims, it maximises heat extraction and it would certainly help in making your boiler condense !
While the TMV is shown with a remote sensor sensing DHW temp, if your PHEs are of sufficient capacity, then just controlling primary loop flow temp would probably give you good enough control.

I had a go at modelling it, but I couldn't find a model for a PHE that would work in a spreadsheet. I had a go at modelling one as a series of lumped heat exchange elements, but I couldn't get it stable.

I just throw this one your way since you were thinking of adding another PHE.

Hi Simon, I'm resurrecting this post from a while ago, finally got around to doing this. I was just wondering, with a 24kw boiler and 900x450 cylinder store, can you recommend a good size plate heat exchanger please? 12, 15, 20 plates? 50kw, 100kw? I want to be able to run 2 showers at the same time or fill the bath whilst a tap is being used as a scenario (all pipework is 22mm). The tank will just store heat for hot water, (I'm going to keep radiators connected directly to the boiler)
thanks

Read more: //www.diynot.com/forums/plumbi...o-thermal-store-and-phe.360877/#ixzz2ZnMYIJgS

 

[SimonH2]

Sorry, but that question (or rather the answer) is well outside my ability to answer.

Any size heat exchanger will supply hot water - if it's too small then temperature will drop off with increasing flow rate in the same way as it does with an undersized combi (for example).

You'd ideally want to work out the maximum DHW flow rate you want/need/can maintain. Then work out what flow temperature you need, and what likely range of inlet temperatures you'll be dealing with (pick the worst case, cold water in winter). Flow rate x temperature rise x SHC = power, so you can work out the "power" rating you need for your heat exchanger. BTW - the SHC of water is 4.2kJ/kg&#730;C, or put another way, it takes 4.2kW to raise 1kg of water by 1&#730;C/second (or 4.2kW to raise 1kg/s of water through 1&#730;C).

But, there is no such thing as A power rating for a heat exchanger - it all depends on the conditions. For comparison, the rated output of a radiator is roughly halved if you change the design flow temperature from 70&#730;C to 50&#730;C.

Thus you'd need to decide on the minimum store temperature you want to design for, pick a primary flow rate (I'd go for the same as the DHW as a first approximation), and look at the manufacturer's data tables.
If you choose a very small exchanger, then you'll need large temperature differences - hence a high store temperature. If you choose a larger one, then you can run with a lower store temperature and lower temperature differences.

If you go for post-exchanger temperature control (ie TMV to blend hot from the exchanger with cold from the mains) then it's probably safe to say you can't have too large an exchanger - other than the higher capital costs and slightly increased thermal losses. You could economise the primary flow rate with the left hand setup in the first diagram I posted.
If you want to try and control temperature by controlling flow rate through the PHE then it gets progressively harder to get stable control as the exchanger get bigger. By far the easiest control technique is to maintain the flow rate and vary the flow temperature - as in the twin-exchanger setup I posted.