Posted by: @derek-m

Which way is the flow through the pump? Is it pushing the water into the PHE or pulling it out?

It is on the return side, pushing water into the PHE, via the filter (black cylinder) you can see in the earlier photo.

Will do the PHE measurement later, need to get out while the sun is sort of shinning!

@cathoderay

I have had a look at the details for your water pump, with the following results.

Taking the model number, the 25 indicates the size of the pipework connections. The 70 indicates that it can develop a maximum head (output pressure) of 7m. The 130 that the body is cast iron CED, 130mm. The ZZZ mean no control profile or external control signals.

Looking at the various graphs, the horizontal axis is Quantity (Q), or you could call it flow rate. The vertical axis is Head (H), or you could call it output pressure.

The curved line going between the vertical and horizontal axis indicates the performance to be achieved with the pump operating at maximum speed, and as such it indicates that as the flow rate increases, the developed output pressure reduces.

There is obviously a pressure sensor and controller built into the pump, for it to be able to offer the variety of control philosophy's.

Considering Constant Pressure (CP) first, which is probably the easiest to understand. There are four possible selections within this mode, CP1, CP2, CP3 and CP AA, which is CP Auto Adapt. CP1, CP2 and CP3 are shown as horizontal lines on the graph, and indicate that as the pump speed increases so does the flow rate, but the pressure at each setting remains constant. CP1 being the lowest pressure setting and CP3 being the highest. CP AA adapts the 'constant' pressure, dependent upon the performance of the system over time, but it is not clear if this auto adaptation takes place within the pressure confines set by CP1 and CP3, or whether the pressure can go below CP1 and above CP3.

Proportional Pressure (PP) again has four possible modes that can be selected, PP1, PP2, PP3 and PP AA. As the name states the pressure is not kept constant, but is varied as the pump speed and flow rate change. As shown on the relevant graph, as the pump speed and flow rate increase, so does the required pressure. PP AA goes even further, by adapting the pressure setting dependent upon the operation of the system over time. You could think of it like weather compensation on a heat pump. It is claimed that this helps reduce energy consumption of the pump motor.

A water pump is required to transport the water and hence heat energy around your heating system. The pump therefore needs to develop sufficient pressure to push the water up to any higher level, and also overcome any restrictions to flow imposed by valves and pipework. A further requirement is to move the required volume of water to supply sufficient heat energy to meet the heating demand.

Consider a situation where an all radiator heating system is located over a ground floor and first floor home, with the water pump situated on the ground floor. If all the radiator valves are fully open, then the water can be pumped readily around the ground floor and probably to a lesser extent around the first floor. CP mode may be better in this situation, since PP mode may lower the pump pressure and hence reduce the flow to the first floor.

Without doing extensive testing and balancing of the system under varying conditions, it is difficult to state what mode of operation is most efficient and effective.

In your present situation Cathoderay, I would suggest that you set your water pump to CP3 mode whilst carrying out the series of tests. Once the primary problem has been identified, it may then be possible to find the pump operating mode which gives best results.

@derek-m - thanks for the detailed explanation. I still think the plots are not ideal, in particular the lack of scales on the axes. We don't even know if they are linear or log! Some of them remind me of Soviet 'Tractor Production is Up!' charts. Using abbreviations is also bad practice. Why not label the flow rate axis 'flow rate', and scale it in litres per minute (for which L/min might be OK)? But Q (and no scale) for flow rate???

I have just started the testing. All the valves, LS (lockshield) and TRVs, are fully open, and I am letting the system settle. The pump is still on PP3. All the rads have got warmer, making me wonder whether the prior LS settings were too tight, we shall see.  One that has always tended to be a bit cooler is again a bit cooler, the living room. I will balance them on the basis of a single top middle IR reading on the rads, closing the warmest LS valves one turn at a time. I'm pretty sure I won't get it finished tonight...

The PHE between centre for the pipes is around 42mm on the short axis (width) and 430mm on the long axis (length).

@cathoderay

I agree that the graphs are far from ideal, but they are not scaled because they relate to a range of different sized pumps. Maybe they should have produced a graph for each pump size.

If you Google 'Proportional Pressure Pumps' you should find a Youtube video from Grundfos Americas, which explains how the graph should be interpreted.

@cathoderay

From the dimensions your PHE could be an LA34, the problem is it can be constructed having a different number of plates. Could you ask your installer for the bill of sale which may provide more details?

Posted by: @derek-m

they are not scaled because they relate to a range of different sized pumps

I thought as much, and the answer is simple, as you say, publish the graph for each pump. The manual is huge anyway, a few extra graphs won't make it that much bigger. Using a generic curve also assumes they all have the same curve which I very much doubt. The fact the curves are all so perfect is also suspicious. 

The video does surprisingly make the theory a bit clearer though when he started talking about proportional squared I thought it was going to turn into something like a 1980s Doctor Who script. I still have a problem with how the pump knows what the 'demand' is. The only thing that I can see that the pump could measure is resistance, how easy/hard it is to turn the rotor. With all the valves open (proxy for high demand?) there is little resistance, pressure and flow rates remain high, with all the valves fully or partially closed (proxy for low demand?), resistance is high, flow low and the pump proportionally reduces the pressure. Perhaps. I still think we don't know what the actual flow is, though if the rads get warm enough, that's a sort of proxy, they can't get warm enough unless the flow is adequate. But adequate isn't a number.

Posted by: @derek-m

Could you ask your installer for the bill of sale which may provide more details?

I very much doubt there will be any such paperwork. The quote I have doesn't mention the PHE at all, there is one line that says ASHP Fittings Kit and another that says Sundry Fittings Pipe, fittings and boxing in pipe work etc - presumably the PHE is somewhere in among that lot. I'm not even sure I knew there was going to be one until the installation happened. From what I know, it seems Freedom provide an installation kit (which presumably is why installers like them, they just order the kit which comes with everything included). I think I recall my installer saying the PHE was generic, sized large enough to cover most if not all domestic installations. The people who should know which PHE was in the kit are Freedom, but they won't talk to me. I don't suppose there is any harm in you emailing them. The worst that can happen is they refuse to talk to you as well.

The upstairs rads are all now reasonably balanced, within a degree or two of each other, range 32-34 degrees, and the room temps are actually over temp, 22 degrees. Downstairs the rads are all cooler, range 29-32 degrees except for the rogue living room rad which is 23 degrees. The kitchen is at its design temp, 19 degrees, but the conditions are hardly testing, ambient outside is 13-14 degrees. I will carry on with the balancing, aiming to have all rads within a degree or two of each other.

I am beginning to wonder if when they were last balanced soon after commissioning, we ended up with a situation where all the rads except the living room had their LS valves all but closed, in order to force water to the living room. This might explain the poor flow generally, the low rad temps, and the rapid heat pump cycling. The pipework to the living room is a long run, and it is at the end of that run, but there is no reason to suppose the pipework is blocked, it has been power flushed, at mains pressure at one point, > 3 bar (because the pressure relief valve vented), and the flushing machine ran for getting on for an hour. Unfortunately there is not easy way to run alternative pipework to the living room, all the ground floor floors are solid concrete, and any alternative pipework work have to navigate either two doorways or the stairs. Not really an option.

What we really need is a cheap affordable and reasonably accurate clamp on flow meter. But looking on ebay etc these don't come cheap.              

@kev-m Hi Kev, great minds. I've just fitted 4 of these to my newly installed system either side of the low loss header/volumiser, to read the primary side flow and return and the secondary side flow and return. They were very informative. Straight off I realised that the installer had left both primary and secondary side pumps running flat out, 'pump curve 3' and that in all heating load scenarios, there was considerably more primary flow than secondary. I trimmed the primary pump down to 'curve 1' still a little high on the flow side compared to Grant's technical data, and trimmed the secondary pump to 'curve 2'. There's now a much better flow balance and more consistent delta T on both sides. This probably happens a lot with installers, they set everything flat out and then dash off to the next job.

Posted by: @allyfish

I've just fitted 4 of these

I wonder what 'these' are? If they are flow meters, they don't come cheap, and four could easily set you back a grand. Also, on a PHE, but perhaps not a low loss header, the flow in on one side should be the same as the flow out on the same side - no need for two flow meters on each side.

PS it's @derek-m not @kev-m who has been contributing here, though both are very welcome and helpful contributors. 

Posted by: @cathoderay

Posted by: @allyfish

I've just fitted 4 of these

I wonder what 'these' are? If they are flow meters, they don't come cheap, and four could easily set you back a grand. Also, on a PHE, but perhaps not a low loss header, the flow in on one side should be the same as the flow out on the same side - no need for two flow meters on each side.

PS it's @derek-m not @kev-m who has been contributing here, though both are very welcome and helpful contributors. 

Hi Cathoderay,

I think the reference is to the temperature sensors which Kev recommended.

@derek-m, correct, the digital thermometers from Amazon Kev gave a link to. Nifty little things. Cost £11 or so on Amazon Prime for 4. The low loss header was not balanced very well by the installers. This was obvious from the difference in delta T on primary and secondary sides. Quite a lot of primary flow was diverting through the low loss header straight into the primary return. For a single secondary zone system using only one of the secondary side flow & return connections, the flow on primary and secondary sides should balance when all emitters are calling for heating. A little more on the primary is fine, but I had the primary flow well up on the ASHP manufacturer's recommended maximum flow for the model of ASHP installed even when all system rad TRVs were open.

@allyfish

I think you mentioned having a primary pump in addition to the one inside the ASHP. How does this affect the operation of your heat pump, since as far as I am aware a heat pump varies the speed of its water pump as the required energy output varies?

@derek-m Hi Derek, the primary pump is the one in the ASHP, which for the Grant Aerona range is a fixed pump curve device. It's for exactly this reason the low loss header is needed, as the Grant ASHP requires a constant flow on the primary side, but the secondary side flow varies depending on the number of radiators calling for heat and the TRV settings. As the TRVs close down, or in the case of the smart timed TRVs, switch off, the system head increases and the flow rate decreases. The ASHP compressor and evaporator fan are variable speed control, in response to demand, but not the water pump. (Some ASHP manufacturers do speed control the internal pumps, which then negates the need for low loss headers and secondary pumps if the circuit hydraulic design can be accommodated with the in-built speed controlled pump.)