Adding a Buffer Tank to Ashp Setup

[Scotland]

Hello,

 

I'm looking for a bit of advice on adding a Buffer Tank to my ashp setup.

 

The way my heating is set up I have individual control of all rooms in the house via thermostats, downstairs via ufh zones and upstairs radiators via point to point plumbing manifolds.

 

Which in itself probably promotes the use of a Buffer.

 

System specs - 

Mitsubishi ecodan 12kw

Pre plumbed 300l cylinder

95m2 of ufh

249m2 of heated rooms all in.

 

So the question is where is the best location to break into the pipework to install a buffer and looking for recomendations as to Manufacturer, model numbers etc.

 

Pictures are just after installation before someone mentiones insulation etc ?

 

Thank you.

 

 

 

[PeterW]

Errr ... where you have drawn it ..?? Seriously..! 
 

Just break into the bottom return pipe with a tee and then plumb the return pipe tee to the bottom of the buffer.

 

Flow pipe, break it just below that lever valve and tee from the wall, bend and into the buffer at the top. If it goes direct into the top then put a vertical tee and then put an auto air valve. 

In terms of controls, parallel the call for heat with a tank stat on the buffer so it will always be hot. 

 

[Scotland]

@PeterW

Peter, the drawings were just for reference as to where the pipes went not on where to break into them.

The position of the buffer is the only feasible position to put it as the room is quite small.

 

Keep seeing 2 or 4 pipe setups for buffers so I assume I need the 2 pipe version.

 

Any recommendations as to what buffer to buy,

 

My guess would be something along the lines of 100L ish.

[PeterW]

TBH you can get them from anywhere as you can use a standard stainless direct cylinder - What pressure is the heating system set to run at ..??

 

And does the rad circuit have its own pump ..??

[Scotland]

Think it's about 1.8 bar when not running, drops to 1.5 bar when running.

 

No pumps on rad circuit yet, but going to add some.

 

The 2 pipes going up the wall go into a tee at the top into a 7 port manifold then it goes to the other side of the house into another 5 port manifold.

 

First manifold gets some movement in the flow meters but the second one doesn't.

 

Currently researching this also as where the manifolds and pipes are would be hard to get a pump in unless I take the pipes to the other side of the manifold, and add a pump there.

[PeterW]

Bring the hot pipe up the wall past the manifold and rotate the isolator 180 degrees and then bring the pump into the top of the manifold. 
 

 

[Scotland]

That never even crossed my mind to do that,

 

Would a manifold bypass then be needed on the other side of the manifold, in case the pump started up and all zones are closed?

 

Back to the buffer, would an expansion tank be required or is the one already fitted on the heating side enough.

[PeterW]

Expansion tank should be fine - is it 12 litres or more ..?

 

And yes, you’ll need either a bypass or just have an open zone - bathroom or landing rads are good for that. 

[Nickfromwales]

45 minutes ago, Scotland said:

Would a manifold bypass then be needed on the other side of the manifold, in case the pump started up and all zones are closed?

If the zones were closed then there would be no call for heat signal from the manifold wiring centre, ergo the pump would be off  

[Nickfromwales]

32 minutes ago, PeterW said:

Expansion tank should be fine - is it 12 litres or more ..?

 

And yes, you’ll need either a bypass or just have an open zone - bathroom or landing rads are good for that. 

Still struggle to understand why manifolds are still going in with no local pump and blending set.  Bypass would be integral then, plus a set / specific flow temp is achievable via each TMV, essential when there is a mix of disciplines like rads + UFH. 

[Scotland]

33 minutes ago, PeterW said:

Expansion tank should be fine - is it 12 litres or more ..?

 

And yes, you’ll need either a bypass or just have an open zone - bathroom or landing rads are good for that. 

Not sure what the expansion size is, I will check tomorrow.

 

4 minutes ago, Nickfromwales said:

If the zones were closed then there would be no call for heat signal from the manifold wiring centre, ergo the pump would be off  

Is there not a slight delay though on the actuators opening after the call for heat and the pump turning on?

[PeterW]

Just now, Nickfromwales said:

Still struggle to understand why manifolds are still going in with no local pump and blending set.  Bypass would be integral then, plus a set / specific flow temp is achievable via each TMV, essential when there is a mix of disciplines like rads + UFH. 

pump yes, blender no as the ASHP will set the rad flow temp and the blender will only provide UFH temps (although I do partially agree with you it’s easier..)

 

1 minute ago, Scotland said:

Not sure what the expansion size is, I will check tomorrow.

 

Is there not a slight delay though on the actuators opening after the call for heat and the pump turning on?

 

Yes there is a delay on wax cap actuators of around 1-2 minutes unless you use a motorised actuator such as the Salus ones which is why I would make one of the zones open anyway. 

[Nickfromwales]

1 minute ago, Scotland said:

Is there not a slight delay though on the actuators opening after the call for heat and the pump turning on?

Yes, hence....

 

2 minutes ago, Nickfromwales said:

Bypass would be integral then,

 

 

All go hand in hand with a full manifold / pump / TMV arrangement. 

[Scotland]

1 minute ago, Nickfromwales said:

Still struggle to understand why manifolds are still going in with no local pump and blending set.  Bypass would be integral then, plus a set / specific flow temp is achievable via each TMV, essential when there is a mix of disciplines like rads + UFH. 

Yes that would be my fault,

 

I added a ivar blending set into ufh manifold due to posts I was reading from yourself.

 

But upstairs yes it would be good and only realising now I need the extra pumping capacity, but didn't really need to blend as radiators can take the full heat.

[Nickfromwales]

1 minute ago, Scotland said:

Yes that would be my fault,

 

I added a ivar blending set into ufh manifold due to posts I was reading from yourself.

 

But upstairs yes it would be good and only realising now I need the extra pumping capacity, but didn't really need to blend as radiators can take the full heat.

Live and learn etc, but there’s little that can’t be fixed relatively easily   

[PeterW]

3 minutes ago, Nickfromwales said:,

 

All go hand in hand with a full manifold / pump / TMV arrangement. 

But rads don’t care and can be sized at ASHP output temperature so that’s £100 per manifold in the kitty ?

 

Although I am getting deja vu about fitting remote pumps on manifolds....... 

 

Pass me a McEwans...!! ?

 

 

[Nickfromwales]

A single pump over multiple satellite manifolds is a sack-of-shit solution, you know it, and I know it  

Every single one I’ve come across set up like this is either struggling or hasn’t worked properly from day 1. 
??
 

[PeterW]

Yep agree - manifold and pump every time. 

[Nickfromwales]

21 minutes ago, PeterW said:

I do partially agree

Lol. ? Victory is mine !!

[Scotland]

9 hours ago, PeterW said:

Expansion tank should be fine - is it 12 litres or more ..?

Expansion is 12 litres, so should be fine?

 

Do you think 90 - 100 litre buffer will be ample.

[john0wingnut]

On 18/12/2020 at 23:46, Nickfromwales said:

A single pump over multiple satellite manifolds is a sack-of-shit solution, you know it, and I know it  

Every single one I’ve come across set up like this is either struggling or hasn’t worked properly from day 1. 
??
 

Interesting to read this.........

 

The UFH heating was designed by NuHeat and we only have one pump, driving 2 manifolds

[john0wingnut]

On 19/12/2020 at 08:44, Scotland said:

Expansion is 12 litres, so should be fine?

 

Do you think 90 - 100 litre buffer will be ample.

 

Ive got a similar sized ASHP and cylinder to you, and ours has a 100 litre buffer.

[Nickfromwales]

16 minutes ago, john0wingnut said:

Interesting to read this.........

 

The UFH heating was designed by NuHeat and we only have one pump, driving 2 manifolds

Just spent a lot of time and a clients money undoing a similar spec Nu-Heat UFH job. Multiple failures on loops with short loops bypassing and more.....and no way of setting different flow temps for ground floor ( tiled over screed ) from upstairs ( engineered / solid woods over screed ). 
Train wreck. Nu-Heat insisted it would work, client paid the price when it didn’t. 

[john0wingnut]

3 minutes ago, Nickfromwales said:

Just spent a lot of time and a clients money undoing a similar spec Nu-Heat UFH job. Multiple failures on loops with short loops bypassing and more.....and no way of setting different flow temps for ground floor ( tiled over screed ) from upstairs ( engineered / solid woods over screed ). 
Train wreck. Nu-Heat insisted it would work, client paid the price when it didn’t. 

Oh dear............. you have me worried now........... as I did have some minor concerns at the time.

When you say loop failures, are you referring to the loops in the screed?

Ive noticed an issue in our Master Bedroom, where by only one half of the floor is getting noticeably warmer than the other half, yet the flow rates on the manifold are the same.

[SteamyTea]

Just a general observation.  Putting in a buffer will reduce short cycling of the heat source.

Taking a simple example of allowing a 100 lt buffer to vary between 30°C and 45°C, and not taking any external draw from it, the times will vary for different power sources.

(this chart breaks convention by putting time on the y-axis, but is correct as time varies with power, with power being the controlled variable)

 

 

Edited December 21, 2020 by SteamyTea