Mad idea for DHW retrofit based on Mixergy?

[JamesPa]

Over on the thread about Chelmer Heating there has been a lively (albeit off-topic) discussion about uses for plate heat exchangers.  The Mixergy tank (for heat pumps) uses a plate heat exchanger external to the DHW tank, through which is pumped the DHW, thus exchanging heat with the HP flow efficiently without using a coil or taking up volume in the tank.  They claim that their cylinder can be configured vented or unvented.  Using a PHE instead of an internal coil has a number of interesting effects, some of which may be good and some bad.  The way Mixergy have it configured seems to eliminate stratification the DHW tank, which may be good or bad!

 

Like many, I suspect, I have an exiting, fairly new, vented cylinder, which is otherwise perfectly satisfactory but has a coil which is insufficiently large for a heat pump.  I am thus faced with replacing it for a new unvented cylinder, involving lots of disruption and cost.  So, following the Mixergy example, I came up with this idea which reuses the tank and existing pipework and involves minimal disturbance.

 

I fear it wont work, because the pump will just draw air in from the vent (although I also feel there should be a way to prevent this) and/or from the outlets if they are switched on whilst the pump is on (I could perhaps defeat this by putting the pump upstream of the feed to the outlets rather than downstream as illustrated).  The Mixergy has dedicated tappings for the PHE circulator, unfortunately in a retrofit situation the use of existing cylinder tapings is mandatory.  Is it completely mad or worth considering?  I'm not a fan of throwing away perfectly functional components!

 

 

 

[ProDave]

I am not sure that would even give enough heat transfer from the ASHP?

 

Why not just fit a proper new heat pump cylinder and go unvented.  It will be a revalation if you have always had  a vented cylinder before.

[JamesPa]

23 minutes ago, ProDave said:

I am not sure that would even give enough heat transfer from the ASHP?

Mixergy obviously think it will, it's what they do for their 'heat pump cylinder'.

 

Not that I necessarily believe marketing claims, but several on this forum appear at least to have considered them.

23 minutes ago, ProDave said:

Why not just fit a proper new heat pump cylinder and go unvented.  It will be a revalation if you have always had  a vented cylinder before.

3K, lots of disruption, loads of stuff thrown away, lost space in the airing cupboard and an annual maintenance fee.  I'm not saying I won't go this route, but sometimes it's a good idea to think out of the box even if only to eliminate it.

Edited April 5, 2023 by JamesPa
Added reasons

[JohnMo]

I would put the pump on the cold side instead of hot, it will always be flooded and less chance of air and get the head available to the fluid level in the header tank.

 

You will need a DHW pump not CH pump.

 

If I get time I will get my set up running at a boiler flow of 50 deg and let you know the temps across the PHE.  I need to know for when I get my ASHP fully connected anyway.

 

[sharpener]

33 minutes ago, JohnMo said:

I would put the pump on the cold side instead of hot, it will always be flooded and less chance of air and get the head available to the fluid level in the header tank.

 

 

+1 from me. Have had "pumping over" probs before, this was the solution IIRC.

[SteamyTea]

2 hours ago, JohnMo said:

You will need a DHW pump not CH pump.

Do you have an example of one, or two.  I could not find one, but may be using the wrong search terms.

[SteamyTea]

4 hours ago, JamesPa said:

The way Mixergy have it configured seems to eliminate stratification the DHW tank, which may be good or bad

Generally I think it is a good thing, you get more energy into a smaller volume.

The thermal losses may increase though, and then there are pumping losses.

[JamesPa]

2 hours ago, JohnMo said:

I would put the pump on the cold side instead of hot, it will always be flooded and less chance of air and get the head available to the fluid level in the header tank.

 

You will need a DHW pump not CH pump.

 

If I get time I will get my set up running at a boiler flow of 50 deg and let you know the temps across the PHE.  I need to know for when I get my ASHP fully connected anyway.

 

Thanks, that's  very helpful, I will rethink. 

 

On a related, but more philosophical, point, this (interesting but very painful) exercise to retrofit ashp to my house has convinced me that the current 'rip it out and start again according to rigid rules' approach to ASHP won't cut it for much of the retrofit market.  Government subsidies can't last forever, and many people either won't, or can't, afford the cost of a complete system rework essentially to replace the boiler.  Interestingly upgrading radiators,  vitally important in many cases but perceived as difficult, is both cheap and easy, it's the 'peripheral' issues that seem to bug.

 

I think the ASHP manufacturers may be pinning their hopes on high temperature designs which are a drop in replacement, and perhaps the efficiency sacrifice is justified (I personally think they also need to sort modulation ratio, buffer tanks are simply not going to be fitted in many retrofit scenarios or smaller houses).  However I suspect we need also both more tools in the armoury (such as the one contemplated in this thread), and a more enlightened workforce, to deal with the myriad of retrofit situations in a way the public can afford/tolerate.  On this forum people _are_ themselves experimenting with out of the box things to deal with their specific circumstances, but we need plumbers and heating engineers who can do this for the majority who don't have the design skills themselves.

 

As if to demonstrate this point, today a very nice and clearly intelligent surveyor informed me that the  mains water feed to the new UVC they want to put will probably need to be upgraded from 15 mm to 22mm (involving considerable disruption and cost).  When I told her that the incoming mains pressure was 9 bar, so flow rate is certainly not a problem, she offered to note it, but told me it would still quite possibly need to be done because 'the cylinder manufacturer recommends it'.  I don't doubt this is true, because I have read it myself, but its another example where backside-covering rules-based over specification (I refuse to call it engineering, because it isn't) risks absurd and costly results.  Inevitably, unnecessary cost deters many, and unnecessary disposal of perfectly functional equipment is unlikely to be environmentally friendly.

 

More innovation, much better training, more flexibility and just a bit of risk taking is needed if we are to crack this.   How to make happen though?

 

Just ramblings for a Wednesday evening, now back to the engineering!

 

Edited April 5, 2023 by JamesPa
Typo

[sharpener]

45 minutes ago, SteamyTea said:

Do you have an example of one, or two.  I could not find one, but may be using the wrong search terms.

 

Good point from @JohnMo, I think he is probaby meaning a pump constructed (expensively) of s/steel and bronze, which does not rust even with no inhibitor. They are used for e.g. secondary loops to ensure a hot DHW supply to distant taps.

[JamesPa]

12 hours ago, JohnMo said:

I would put the pump on the cold side instead of hot, it will always be flooded and less chance of air and get the head available to the fluid level in the header tank.

 

Like this?

 

Actually I drew it this way at first, but then got worried that the pump would draw water from the header tank not the DHW tank.   I cant quite fathom how to ensure that this doesn't happen 

 

Edited April 6, 2023 by JamesPa

[JohnMo]

Yes for the pump, but mount the pump as low as you can, just above the bottom elbow by the cylinder, this will give the pump the best suction pressure and keep it away from cavitation.

 

For the PHE flows hot from ASHP, need to be on the same side as the hot water going to the cylinder. So you HP flows swop sides.

 

See attached

[JohnMo]

9 hours ago, JamesPa said:

the cylinder manufacturer recommends it

Which actually means nothing, must, shall, will, mean to have to

 Recommend means if you want, we would like you to, but you don't have to.

[JohnMo]

10 hours ago, SteamyTea said:

Do you have an example of one, or two.  I could not find one, but may be using the wrong search terms

https://www.bes.co.uk/plumbing-supplies/plumbing/hotwaterpumps/pumps/?gclid=Cj0KCQjw27mhBhC9ARIsAIFsETFYeNMlSSGzUeUme2rWauEIbYepRD48FcqQjE9KDeDLoILZR8x8IlQaAoXWEALw_wcB

[SteamyTea]

19 minutes ago, JohnMo said:

https://www.bes.co.uk/plumbing-supplies/plumbing/hotwaterpumps/pumps/?gclid=Cj0KCQjw27mhBhC9ARIsAIFsETFYeNMlSSGzUeUme2rWauEIbYepRD48FcqQjE9KDeDLoILZR8x8IlQaAoXWEALw_wcB

Thanks

[JamesPa]

52 minutes ago, JohnMo said:

Which actually means nothing, must, shall, will, mean to have to

 Recommend means if you want, we would like you to, but you don't have to.

I know, I was just quoting the nice and intelligent surveyor's preemptive justification for a likely proposal from the installers to do totally unnecessary work!

[sharpener]

 

 

15 hours ago, JohnMo said:

Yes for the pump, but mount the pump as low as you can, just above the bottom elbow by the cylinder, this will give the pump the best suction pressure and keep it away from cavitation.

 

 

+1, This also stops it sucking in air through the shaft gland.

 

15 hours ago, JamesPa said:

Like this?

 

Actually I drew it this way at first, but then got worried that the pump would draw water from the header tank not the DHW tank.   I cant quite fathom how to ensure that this doesn't happen 

 

 

It is a puzzle isn't it!   Helpful perhaps to draw a diagram with the elements positioned to show the correct relative pressures.

 

As @JohnMo implies upthread, the baseline is the static head at the bottom of the tank i.e. suction side of the pump. From there the pressure rises across the pump and falls again across the HX, leaving a residual delta P as the motive force for the flow through the tank, and this last term will also raise the level in the overflow pipe. If it is less than the height from the water level in the header tank to the top of the overflow bend the pump will push the water through the DHW tank, if it is more then it will pump over into the header tank. 

 

In one house that had a vented system I remember increasing the height of the overflow bend by a foot or so to cope with turning the pump up a notch. In your setup however I expect most of the resistance and hence most of the pressure drop will be across the HX and not much needed for the tank.

 

As a further puzzle, does it help if instead of abandoning the coil you put it in series with the HP feed to the HX (or possibly the return)? After scratching my head I can't be sure which if either will give an advantage. But from first principles increasing the total heat exchange surface area can only be a good thing. I guess its actual usefulness will depend on the relative areas of the two.

 

Also, from my previous work on MVHR air-to-air HXs I recall that the dwell time in the system is some kind of figure of merit, and this would increase as well.

 

 

Edited April 6, 2023 by sharpener

[Nickfromwales]

On 05/04/2023 at 21:49, SteamyTea said:

Generally I think it is a good thing, you get more energy into a smaller volume.

The thermal losses may increase though, and then there are pumping losses.

Which is why I think that the Mixergy cylinders are an over-complicated re-invention of the wheel. In my personal opinion, the wheel they have re-invented has 5 sides.

Just buy an UVC and enjoy the simplicity of none of the above nonsense. Life is too short to be adding more losses, more components, moving parts and more maintenance and up-keep, and for what? A part energy-filled tank that loses more energy than it saves?!

Add heating this from an ASHP and I'm already loading my 12-guage. The elongated path of energy transfer will see the ASHP dedicated ( predisposed ) to DHW priority for WAY too long.

Massive thumbs-down from me. UVC, simple and efficient, why consider anything else?

 

Doesn't the above schematic omit the flow switch?

[Miek]

The Mixergy has dedicated tappings for the PHE circulator, unfortunately in a retrofit situation the use of existing cylinder tapings is mandatory. 

 

 

 

You can add your own tappings believe it or not, provided you have an immersion element boss.

 

Buy a tank connector, Cut a hole where you want the tapping, drop a string in from the immersion boss and hook it with wire in to the new hole,  drop the tank connector down the string so it locates in the hole , screw on nut. Pat yourself on the back :).  I've done this a few times on old tanks. You can even retrofit secondary coils but that can be quite tricky. 

[dpmiller]

8 hours ago, sharpener said:

This also stops it sucking in air through the shaft gland

 

you won't find one of those in a heating pump...

[JohnMo]

7 hours ago, Nickfromwales said:

Doesn't the above schematic omit the flow switch?

Not sure I would use one, mine failed in less than a week.  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.

[sharpener]

1 hour ago, dpmiller said:

 

you won't find one of those in a heating pump...

My mistake, for some reason I was thinking of the construction of the cheapo Stuart Turner jet pump in my rainwater system. And yet there is still something I have read or heard niggling away in the back of my mind.

[JamesPa]

12 hours ago, sharpener said:

As a further puzzle, does it help if instead of abandoning the coil you put it in series with the HP feed to the HX (or possibly the return)? After scratching my head I can't be sure which if either will give an advantage. But from first principles increasing the total heat exchange surface area can only be a good thing. I guess its actual usefulness will depend on the relative areas of the two.

I think this does help, I omitted it from the diagram for simplicity.

 

3 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

Agree!

 

11 hours ago, Nickfromwales said:

UVC, simple and efficient, why consider anything else?

In a new build 100% agree.  But I'm doing a retrofit with a perfectly satisfactory dhw system that I am considering changing solely to suit the characteristics of a heat pump.   Against a UVC in this circumstance is: 2-3K, lots of disruption, loads of stuff thrown away, lost space in the airing cupboard and an annual maintenance fee. That's quite a few cons to offset the pros.'  I'm not yet decided to be honest

 

4 hours ago, Miek said:

you can add your own tappings believe it or not, provided you have an immersion element boss.

Hmm, not sure I fancy this!

Edited April 7, 2023 by JamesPa

[Nickfromwales]

1 minute ago, JamesPa said:

In a new build 100% agree.  But I'm doing a retrofit with a perfectly satisfactory dhw system that I am considering changing solely to suit the characteristics of a heat pump.   Against a UVC in this circumstance is: 3K, lots of disruption, loads of stuff thrown away, lost space in the airing cupboard and an annual maintenance fee. That's quite a few cons to offset the pros.'  I'm not yet decided to be honest

I think you'll spend all of that money, and get a poor result. Looks horribly over-complicated to me, hugely inefficient, maintenance hungry and I would still recommend that you reconsider the UVC option.

Just my 2 cents of course, and I will offer any help / advice I can if you decide to go to the wonkey-wheel option   

[JamesPa]

5 minutes ago, Nickfromwales said:

hugely inefficient

Why that?  If the PHE has sufficient surface area it could actually be more efficient because it could reduce the required deltaT between flow and DHW.

 

I do agree it's more complicated of course, but as a job _less_ complicated than retrofitting a UVC. 

Edited April 7, 2023 by JamesPa

[Nickfromwales]

1 minute ago, JamesPa said:

Why that?  If the PHE has sufficient surface area it could actually be more efficient because it could reduce the required deltaT between flow and DHW.

 

I do agree it's more complicated of course, but as a job _less_ complicated than retrofitting a UVC. 

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

The second you bolt that PHE to the cylinder you have just added a constant miniature radiator to it, massively increasing the latent / standing losses. 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.