Only struggling Mark?, but I’m not going back to school at 80 
. My calculated numbers arn’t hugely different to yours, but not as accurate as Tim’s. there is definitely a very large bypass through the PHEX, it will be interesting to see what effect it has on the Rad outputs + COP when the 3 port valve is replaced.
Reminds me of If it doesn’t fit, don’t force it. Get a bigger hammer. 
Maybe, just remove the 3 way valve actuator and turn/ensure valve spindle fully in heating position, take another set of readings. Post 75 might indicate that the valve was OK, then.
I did get a proper installer to look at it. He discounted the valve idea out of hand and wanted to rerun all the radiator pipework through the loft and fit a new heat pump closer to the cylinder. 
It looked like I couldnt remove the valve casing but I could try a little gentle persuasion with a blunt object
It says here “The entire actuator can be replaced without dismantling the valve”
Thanks John.
Hi Tim. Yes I watched a youtube video. The catch must be on the blind side but ill give it a go.
That is looking much better. I also note that the flow rate is also higher at 16 l/min. This should result in more heat actually getting to the radiators.
Some things to keep an eye on:
- Internal temperature should gradually rise; do ask residents to give feedback about comfort levels, and reduce the thermostat if required (before anyone starts messing with radiators)
- Auto Adapt will drop to minimum power when it gets up to thermostat set point, and flow temperature ought to reduce - it may take a little while for AA to learn the new configuration.
Hot water cycle won’t be able to run while the valve is disconnected; you may need to prohibit that for the time being. Hopefully the daily legionella will provide enough hot water for a day.- Looking back at post #15, it looks like the secondary flowrate > primary flowrate,
so you might want to reset the pump back to its lowest setting.
Cheers Tim, I hadnt noticed the flowrate had increased.
Im anticipating having to reset the WC curve once things stabilised and have warned the residents it might get warm. I reconnected the valve after a bit of jiggle and lubricant so it might jam again after the next DHW run.
The secondary pump is at its lowest setting. I did see what would happen if I increased it but the DT between the primary flow and secondary flow increased so i put it back.
Still its a big improvement and I can now work to tweak some more efficiency out of it.
Thanks again 
The WC curve isn’t relevant with Auto Adapt - it’s figuring out its own curve, so just let it do it’s thing.
OK will do, I thought it used the WC curve as its basis, but if not then thats even easier.
I’m glad to see that Auto Adapt doesn’t refer to a circ pump’s Auto Adapt function.
Wonder does removing the 3 way valve actuator mean that its the HW function that’s enabled hence the increased flow of 16LPM through the Primary return to the HP?. My calcs below show, (assuming 16LPM) that 6.3LPM at ~ 35C is mixing with 9.7LPM of PHEX bypass at ~ 45C to give a HP return, 16.0LPM at ~ 41.6C. The LLH secondary flow only calculates at 8.0LPM. HP output 4.0kW.
Hopefully the new 3 way valve will remove any doubts as to what’s going on.
In its resting position, the valve directs flow to central heating (position B). When energised by applying power, the valve directs flow through the HEX for hot water (position A). I believe this is fairly typical for this setup.
My hypothesis is that the valve is not returning fully back to its rest position B, either because the actuator is faulty or there’s some crud inside the valve itself. Mark’s experiment yesterday suggests the former.
Since the hot water cycle has been run, the flow rate has fallen back down to 13 l/min, so the problem is not solved yet. Hopefully a new actuator or valve will address that.
It does, though I don’t know what it should look like. Something may have jammed within the valve to cause that to break, I guess.
Great news, maybe a set of your sensor readings again sometime when you get a chance.
I might go to site tomorrow John.