Avoid short cycling How to Control the Buffer- Real World Setup-Improving COP and dt

Efficient ΔT Regulation Between Buffer and Heating Zone – How Do You Do It?

I am working on optimizing a hydronic system based on an inverter-driven heat pump connected through a small buffer tank. The secondary heating circuit is circulated by a Grundfos PWM pump.

I am looking for recommendations and clarification regarding the best way to manage this setup efficiently. Could you please advise or share experience on the following points?

1.What type of controller or combination is most efficient for secondary-side management? What model of pump ?

2.How should the control algorithm handle temperature difference (ΔT = Tsupply – Treturn) to balance efficiency and comfort? Buffer charging logic /ΔT-based or target-temp controlled /PWM flow?

3.How should the supply and return sensors be positioned relative to the buffer?

4.How can the controller’s data (temperatures, PWM %) be used to identify hydraulic imbalance?

5 Who to Avoid short cycling ?

Well… not to answer your question as asked, substantially because the buffer I have is imperfect (to put it mildly) - see Daikin Pre-plumbed Cylinder (WorldHeat) Buffer Strangeness - but to proffer a few comments:

My pump is a UPM3 Flex AS with the Superseal connector, which is fairly easy to source (the connector I mean). There is also a feedback line which is also a PWM signal which can be used to detect error conditions or running pump power. The unit fitted by my installer was way over-powered; I suspect there is a general aversion to fitting undersized units but without running down to about 50% speed there was a lot of turbulence noise (and the power input to the pump fell by a lot more than half, which I take to be an indicator of the losses due to turbulence. I drive it through a pair of MOSFETS (N and P) off a Raspberry Pi Pico, which also “reads” the feedback PWM.

Since my buffer inadequacy is unresolved, I’ve not got into building the control algorithm. In case your buffer is going to be a problem, effectively making whatever you try pointless, I would set up some logging on all 4 buffer ports (HP flow and return and SH flow and return). I think the data you collect will then start to show what the practical dynamics of the system are. That was my plan, then to manually adjust the secondary pump speed and get a feel for how it all behaves (both directly in response to a speed change and for multi-day runs at different speeds).

I also spent some time balancing the radiators.

I think the objective is simply stated: adjust the secondary circulator so that there is negligable temperature difference between the HP flow and space heating flow at steady state. The problem is that getting to steady state is not simple and has to align with whatever control algo your HP/system uses. It would be quite easy to build something which is unstable.

Probably not turbulence losses, just standard Pump Affinity laws, pump power is proportional to cube of pump speed.

1 Like

is the buffer required in your system? i run without problems (mainly under floor heating, good insulation)