Thanks for that, i’ll talk to my heating guy as i’m sure he’ll understand more of than I will. 
It’s the standard Mixergy method. Bolt on plate and secondary pump. I’ll try to take some photos later.
I plan on adding some temp sensors so I can track the losses before and after the plate on both heat pump and cylinder side.
Ah - Mixergy - ok - I was looking for a setup that didn’t involve it’s own operating system… 
Heating bod should follow my gibberish. Air drops out at the top and where things slow down. Put the expansion vessel higher than it is. Put two tees into the vertical pipe that rises to it. The lower tee feeds the meter. The upper tee bleeds off the air into the air admittance valve. You need to feed the system with water from below the point the air has been bled out.
It’s own OS. 
I think @Timbones has a plate on his Ecodan setup on a conventional cylinder?
Ta. Although weirdly, I can set higher flow rates on the space heating cycle but have never seen the Sharky crap out on heating.
I’m assuming my issues are affected by length of the end to end run of the pipework?
Outdoor unit > diverter valve > plate/cylinder > back to outdoor unit is only about 6-8 metre total round trip when doing DHW
whereas
Outdoor unit > diverter valve > heating circuit > back to outdoor unit goes up two floors and is obviously much longer distance before coming back.
I have no explanation for this. Maybe there’s some technical explanation?
And for clarity, this is the current setup after we moved the heat meter for the third time. Same issues though.
Same suggestion about the expansion and bleed valve?
We don’t really need that inverted U with the bleed valve as we only did that to give us some vertical downpipe (where we had the meter on the second try)
Mitsubishi’s pre-plumbed cylinders come with a plate heat exchange and pump for DHW. There was some discussion about it on this post.
They look interesting - Altecnic even says for heatMeters @glyn.hudson @TrystanLea - for your kit? It does say They are the ideal valve for isolating the temperature probe of heat meters So isolating the probe and the meter.
However, I’m not sure I want the actual probe in the water (that is what that looks like)
I still look at that and think it is a fair % of the free diameter. When I re-pipe my tank I’ll probably get tee’s fitted with enough depth to set the probe just outside the water flow, but these are really interesting.
Interesting call. Looking at Sharky 775 heat meter questions - #23 by Zarch, how about a simple up and over loop of pipe. Halfway along a tee, up to a second tee off to to left for the expansion vessel (lifted up) and straight up for the AA valve (and give it a bit of pipe as well).
Probably improves the flow (not turning on 2 tees). Should still allow the air to rise.
Personally, I’d do a radius bend, but Plumbers seem to be adverse to bending pipe these days.
[edit]
I notice @glyn.hudson setup, the AAs are Tee’d off a horizontal pipe. Samsung Air Source Heat Pump Installation with monitoring - #6 by glyn.hudson
The probe has to be in the water flow to provide a good temperature reading
Here’s the sensor tee we supply in our heat meter bundle, the valve option looks nice but higher cost. But I guess if a valve is needed anyway it’s a good option:
Ah, OK they look different to what I was expecting. What sensors are those?
I’d still think that pulling it up from being an obstruction to the flow would not be detrimental to the reading, but would improve the flow.
They’re ‘direct’ or ‘wetted’ sensors that are designed to be immersed in the liquid, the other type is ‘pocket’ sensors which are designed to sit in a pocket. Direct sensors provide the best reading.
Could you sell them separately, or else post a link whith where they are available from?
Can the be read by an ESP or Pi?
EDALD.
The Sharky I bought comes with pocket sensors and the 1/2" pockets. I’m planning on using some thermal paste in the pockets for better thermal transfer.
The Sharky manual says that sensors must have their tip in the middle of the pipe. While there will be some flow restriction / disturbance it won’t be very much.
They come with Kamstrup heat meters, I’ll try and find out if they are available separately
Does anyone know what the green washer thing that comes with the Sharky is for?
I expect it’s obvious but I can’t work it out! Manual seems unhelpful.
The green washers are the fibre washers to sit between the flow sensor and the flat face union couplings to seal them.
Try Alex Thornton at UK Metering for Kamstrup spares; they usually have good stock. Else if you don’t care for the parts being new I’ve probably got a bunch of these direct immersion probes in t he e might be useful one day corner that you can have for cost of post. PT500 type?
I think the “pocket sensors” shown above are suitable for direct immersion fwiw. You just need an o-ring and the special M10 nut to suit.
No. No they’re not. The similar looking 5.2 mm type are but not those. They’re meant for pockets.
It looks like @glyn.hudson also has his heat meter BEFORE both expansion vessel and the auto-air valve.
Whereas I have mine placed after both of those.
Is heat meter before expansion/auto valve a better way of doing it?
So you’re not dragging air out along the way as @marko suggests?
Possibly, but since my meter is a Sontex even if there were air pockets in there the reading would be effected. Have you tried increasing your system pressure up to 2bar? This fixed the same issue for:
I upped the pressure yesterday.
My manual dial showed 2bar, the Vaillant said 1.8 
I will keep and eye on it. I’m now also logging the raw power/flow feeds from the sharky before I curtail them with ‘max’ limits. I want to see what values the interface throws out in times of the sharky erroring and wonder if the interface can handle this better, as not to throw crazy wild data into the input and subsequent feed.
I’d still like to get it resolved properly though, ie heat meter placement.
And if a rejig/re-ordering of the expansion/AA and heat meter is the answer, then I’d do it as I want this to work flawlessly for the long haul.
And worst case, i’ve got a Sontex sat here in a box.
Quick hack: Bleed any air from the pipe up to the expansion vessel by cracking the union nut holding the pipe to the expansion vessel to release any air there and try again. May need to repeat a few times to purge all air from system.
Thanks.
As per above, in your experience, is it better to have the heat meter before the expansion/auto valve? So you are avoiding the issue of dragging air from the expansion?
I’m thinking what would be the best way to rejig that small part of my return loop?
If it meant putting the heat meter first, i’d look at getting that done if it was best practice?
