2-pipe or 4-pipe connection to buffer tank?

[Mr Blobby]

Our M and E guys didn't seem to know much about a 2-pipe connection when I questioned their schematic (with a 4-pipe connection) last week.

 

This is the 2-pipe connection......

 

 

 

From what I have read, a 2-pipe connection is more efficient than a traditional 4-pipe connection.  Is that really the case and do members of this forum configure their system this way? 

Should I ask to change to a 2-pipe connection?  Or does it really matter?

 

For background we are aming for passive build with little heat demand.  Buffer to sit between ASHP and UFH.

 

Does the flow from ASHP to the buffer need a check valve and if so then are there any defrost cyle issues

 

 

 

Edited February 6, 2022 by Mr Blobby

[Temp]

I think the advantage of a 2 port is more for ASHP than it is for oil/gas because when the UFH is calling for max power all the flow goes from the ASHP direct to the UFH with less mixing/blending.

 

Not sure why there is a check valve shown in that diagram but suspect it should be removed where the source is ASHP or the defrost won't work?

[Marvin]

As I understand, our ASHP, buffer tank and emitters (radiators and water coils in the MVHR ducts) 3-pipe connection to the buffer tank system uses the volume of the buffer tank to add to the volume of the system to avoid short cycling of the ASHP. Any heat loss from our buffer tank is inside the thermal envelope and circulates through the building via the MVHR so for us no significant heat loss. (We have added additional insulation to the buffer tank though, to avoid condensation problems when in cooling mode.)

[Nickfromwales]

With the 2 pipe buffer you can usually get away without the secondary circulation pump that would be required to get flow to the manifold pump.

Less components and a more simplified install. I typically go 4 pipe with LLH’s and 2 pipe with buffer tanks. 
The arrangement you show offers both hydraulic separation and energy buffering, so I would run with exactly what you’ve drawn above. 
For the defrost, afaik, the check valve will not cause issue as I don’t believe the pump reverses flow for that (?) Anybody? 

[SteamyTea]

With this set up, is there any flow advantage to having the Ts at A and B flowing strait threw and the buffer coming off the Ts?

Whenever I see a bend or junction, I just see slower speeds.

Hope that makes sense.

 

 

[dpmiller]

@SteamyTea I think the Tee is a benefit here as it'll promote flow into the buffer, the UFH is only supping from that.

 

@Nickfromwales no, circulation doesn't reverse during defrost.

[SteamyTea]

3 minutes ago, dpmiller said:

I think the Tee is a benefit here as it'll promote flow into the buffer, the UFH is only supping from that.

I will have to ponder this.  I suppose the UFH is running, on average, cooler than the BT.

Probably does not make much real difference in normal usage, but may at the extremes.

Edited February 7, 2022 by SteamyTea

[Nickfromwales]

The tee is "equal" so flow from the HP pump can go to either the buffer, the UFH or both at differing potentials without any issue at all. A typical CH system h as multiple tee's all the wat through the arrangement, and no problems there that a bit of balancing doesn't put to bed.  

[Mr Blobby]

Thank you everyone for your very helpful answers.  I will go back to my M&E guys and ask them to change to a 2-pipe connection.  I'm not sure they get questioned on stuff like this very often!

 

I like the idea of 1 less pump.  I think panasonic ASHPs have a pump inside, if so then does that mean no additional pump to the buffer?

 

57 minutes ago, dpmiller said:

 

@Nickfromwales no, circulation doesn't reverse during defrost.

 

That's interesting.  I thought the defrost cycle sucked warm water from the buffer back into the monoblock.  Have I misunderstood how this works?

 

I don't have a copy of the schematic but from memory it includes a second smaller buffer on the cooling circuit.  Is there any reason why I can't use the same buffer to simplify things?

 

I'll try and get a copy of the schematic to explain!!

 

[Nickfromwales]

1 minute ago, Mr Blobby said:

That's interesting.  I thought the defrost cycle sucked warm water from the buffer back into the monoblock.  Have I misunderstood how this works?

Defrost cycle simply fires up the HP pump. only, and the water from the buffer is used "in reverse" eg it is then the HP which is 'heated' by the warmed water in the buffer. Pump flow ( direction ) stays the same, as confirmed above.

[Mr Blobby]

7 minutes ago, Nickfromwales said:

Defrost cycle simply fires up the HP pump. only, and the water from the buffer is used "in reverse" eg it is then the HP which is 'heated' by the warmed water in the buffer. Pump flow ( direction ) stays the same, as confirmed above.

 

Ahhhh,  now I get it.  The flow remains the same but the pump brings warm water through in the same direction with the ASHP in cooling mode to heat up the frozen bits.

In that case the check valve can remain.

 

Edited February 7, 2022 by Mr Blobby

[Nickfromwales]

3 minutes ago, Mr Blobby said:

I don't have a copy of the schematic but from memory it includes a second smaller buffer on the cooling circuit.  Is there any reason why I can't use the same buffer to simplify things?

He's a man after my own heart. I would deffo divert to a small buffer or LLH when in 'cooling mode' instead of having a big cold buffer. Should help to control potential condensation issues.

I am looking at using 2x 3-port diverter valves to switch between devices per 'mode'. 

[dpmiller]

Generally the whole compressor changes it's gas flow direction- it's taking the energy from the warm water from the heating circuit yes, but amplifies that using the power of the heat pump. Normal flow temp in the 30c range will happily let the unit de-ice at 60 or 70c...

[Mr Blobby]

.... also, my M&E people are pretty much refusing to design any cooling of the slab, quoting the dreaded condensation argument.  Which is bullshit of course at 18C flow temps.

(there will be cooling through the Zhender comfo-post add on the MVHR, which I'm ok with, but it does cost an arm and a leg)

 

Should I put my foot down and insist on cooling the slab as well.

[SteamyTea]

11 hours ago, Mr Blobby said:

Should I put my foot down and insist on cooling the slab as well.

Yes, you are the customer.

[Mr Blobby]

1 hour ago, SteamyTea said:

Yes, you are the customer.

 

Its funny, I think MandE consultants don't get challenged very often.  At least mine don't

 

Thinking more about the buffer connection, I can see how a 2-pipe connection is better for heating but would it be better to have a 4 pipe buffer for cooling?  My setup will have a large buffer for the UFH and a smaller buffer for the heat/cool through the post heat exchanger in the MVHR. 

 

In cooling mode the Zhender comfo-post buffer would be cooled to, say 10C, while the UFH buffer to, say, 18C. 

Would a 4-pipe connection to the UFH buffer be better for maintaining different cooling temps?  My reasoning is that the 10C flow from the ASHP would mix in the buffer before going to the UFH manifold, whereas in a 2-pipe connection it may go to the UFH directly at too low a temperature. 

 

From reading other threads I know others like @Dan F have a similar setup with seperate buffers for cooling.  I would be really grateful for any advice on setting this up.

 

 

 

Edited February 8, 2022 by Mr Blobby

[PeterW]

Single buffer is fine - just chill it to 10°C and run that into the floor. You will struggle to get a floor down to that temp anyway with air temps of 18°C so it’s not worth messing with multiple buffers. 

[Dan F]

On 08/02/2022 at 09:44, Mr Blobby said:

From reading other threads I know others like @Dan F have a similar setup with seperate buffers for cooling.  I would be really grateful for any advice on setting this up.

 

I only have one buffer.  The heating circuits (for UFH and ComfoPost) are both mixed with indpendant temp control.