@toodles Of course the difference between the Daikin and Mitsubishi is that the Daikin has a built in primary circulating pump, controlled by the Daikin, to achieve a desired delta/t. The misubishi has an external primary circulating pump which, in my case, is not controlled by the mitsubishi. I rely on the manual speed control of the circulating pump and presumably some Mitsubishi software (I use Auto Adaptive)

Thank you for your explanation of the temperature differences in and out of a buffer.

Out of interest coincidently, I have today placed temperature probes on the buffer 4 ports. The top two pipe readings Primary in 40.0c and out 36.7c. Bottom two pipes Secondary in 31.9 and out 34.9.

Not sure how bad that is as it certainly not "as similar as possible" and shows mixing. Presumably I need to balance the flow rates better ie speed up the secondary Wilo pump from 36.7c to get closer to 40c. Unfortunately increasing the secondary pump flow rate will bring me back to the noise issue.

@richard24738 Tricky Devils those LLH’s! Befor you try speeding the pump up a little, try slowing it a little. Not being funny but, it is possible that the pump is working at a velocity that is causing some turbulence within the return pipework and sending the cooler return water up to the input again and this would of course cause cooling of the flow! It may not be the case but it is worth trying - making a small adjustment and then leaving it to settle for a while to see the effect. Regards, Toodles.

I recently bought some of those cheap digital thermometers with probes and have been using them to monitor the flow temperatures for the buffer tank in the bottom of my Telford Tempest cylinder

TELFORD-GROUP.COM "https://www.telford-group.com/uploads/shop/TEMPEST+HP+TANK+ON+TANK+-.pdf"

. My Arotherm+ heat pump has a built-in circulation pump and there's a separate circulation pump for the radiators. The buffer tank has been plumbed in thus:

• Top left port = supply from the heat pump
• Bottom left port = return flow to heat pump
• Top right port = supply to radiators
• Bottom right port = return flow from radiators

This arrangement is logical. The water from the heat pump cools while passing through the tank so the flow is from top to bottom while the water for the radiators warms up in the tank and therefore rises.

The photo shows the temperatures in the four pipes. Left to right: Return flow to heat pump; water from heat pump; return flow from radiators; and flow to the radiators.

The radiator circulation pump had been installed to run at the maximum speed. The noise of the flowing water was noticeable at night so I changed the pump to medium speed. After fitting the thermometers which showed the radiator water temperature increased by 2C while passing through the tank, I tried the minimum pump speed. This resulted in more heat transfer between the flows and the water to the radiators became 0.5C warmer. I've checked that the water is still reaching all the radiators so the pump is staying at the minimum flow speed.

@johnr Thank you for explaining how your buffer is connected and the details of the temperature at each of the four ports.

Both &toodles and I have a Wilo circulator pump for our radiator heating circuit which have a manual totally variable speed control whereas you appear to have a number of fixed speed settings.

However, the results remain the same in that:

 - reducing the flow rate (speed) of the radiator circulator pump (secondary pump) lowered the noise level.

 - changing the flow rate (speed) of the pump will affect the both the flow temperature to the radiators and any mixing in the buffer tank.

Tomorrow, I will try a number of speed settings on the secondary pump with a dual target to minimise the noise and minimise the temperature difference between the flow in to the buffer from the heat pump and flow out to the radiators. As Gary suggested earlier, a bit of trial and error.

@richard24738 Indeed, it is trial and an occasional error but, you cannot ‘break’ anything by gradually changing the pump speed up or down. The worst that can occur is that the energy (heat) in the water will run a little low by the time it reaches the last radiator in the line.

There must be an ‘end point’ but, (and someone with far greater knowledge may come along and explain it in far better terminology!) the more you slow the flow around the radiators [it lingers longer], the more heat is able to transfer to the emitter itself rather than rushing on in the pipes. There must be a formula based around the Delta T, flow rate and room temperature - but I am not the person to explain it! Happy pump manipulating! Regards, Toodles.

@richard24738 You had commented above that "I have today placed temperature probes on the buffer 4 ports. The top two pipe readings Primary in 40.0c and out 36.7c. Bottom two pipes Secondary in 31.9 and out 34.9."

I was trying to highlight that my buffer tank is plumbed in differently. Your secondary circuit flow is through the lower, colder, part of the tank so it will never get as warm as the primary flow in the top of the tank. I've got my secondary circuit flow from the buffer slightly warmer than the return flow to the heat pump. I have to wonder if your buffer tank (and probably plenty of others) is plumbed in sub-optimally.

@toodles Thanks. I haven't forgotten you.

Due to the low outside temperatures, my ecodan is going into anti-freeze at night when the set back means the ecodan idles, and in De-frost mode for most of the day. All expected so no issue. However, after each de-frost the ecodan ramps up to recover and this is making temperature readings from the buffer problomatic.

So, at this stage I have focussed on the noise issue and simply reduced the wilo secondary pump speed to the radiators to "can hardly hear it". The Wilo speed dial is still within the recomended range. This noise reduction is especially necessary at night at the moment as both circulator pumps run during the anti-freeze mode, even though the ecodan in set back is idle, to stop the pipes freezing up.

All rads still warm.

I will still monitor the buffer temperatures but wait to draw any conclusions.

@johnr Hi. I think I may have missled you in that my buffer is configured in exactly the same way as you describe.

Whats a little bit concerning for me is the drop in temperature between the top two ports meaning from the heat pump flow across the buffer to the radiator circuit. In the example I gave it was from 40.0c to 36.7c = 3.3c difference whereas your example is 32.7 to 30.8c = 1.9c difference.

However, I haven't finished adjusting the secondary pump speed yet so that may make a difference. 

Sorry @grumpidoc, we appear to be hijacking your thread. Hope the previous suggestions have helped.

@Richard24738 Thanks for the clarification. I await your conclusion.

@johnr I think I may have missread something somewhere as I now understand that the secondary pump flow rate should be equal to or greater than the primary.

Anyway:

1. I can confirm that my noise level was mainly caused by the primary pump and reducing the flow rate to the flow rate recomended by mitsubishi reduced the noise considerably. From 26 Lpm down to 19Lpm

2. I increased the secondary pump flow rate to 1.1Cmh (19Lpm) and achieved the flowing temperatures across the buffer:

This seems more like what it should be so I am going to leave it at that for now until my installer arrives.

@richard24738 Fair enough, however please bear in mind that you may reduce the secondary pump further as it is normal to run it slightly lower than the primary pump rate. This is something you might care to try after your installer has looked and made any adjustments if you still have an audible whine / hum from the secondary pipework circuit, (yes I did say ‘lower. than the primary). Regards, Toodles.

@toodles Just to close this track, my installer was happy with the pump speed adjustments. Primary Grundfos Pump on speed setting 1 and Secondary Wilo on 0.7 cmh. No noise generated by the pumps and all rads are warming up.

I do have a difficulty with the balancing as the upstairs bedrooms and landing are getting a little warmer than I would like. Gradually tightening down the lock shield valves but may need to resort to using the TRV's as maximum limiters.