Hello, Gents !

@FosterTom…

« Or rather, an array of individual vertical pipes between 1 top and 2 bottom (flow and return) lo-resistance manifold-type 'headers'. »

Agreed, Tom, but that is a bit of an issue, as no access to Slate Cavity except through bottom of frame – it is sort-of solar gynecology…

Have therefore a choice between a "simple serpentine", or my mate's suggestion of "inverted U"s"...

Advantages of serpentine: more PEX per m2 of slates, since I’d have to use the little stuff (12mm), but not yet worked out the head-loss on this…

Whereas inverted U’s let me use bigger-bore PEX, & should also drain down OK provided I get the gradient right… like you say, less PEX per square meter, but probably just as much (or more… water).

I have tried to cater for that, by adding "spreader plates" viz. steel casings, made from shelf units, to increase PEX contact, with steel wool packing. So the cool feed goes up odd leg, over the top, then down the even leg; all the six odd legs are tee’d on to one feeder, and idem all the six even legs… So the pump does have to push a bit on the UP side, but the DOWN side is basically effortless, gravity return, so ought to siphon along and help the pump. The hot return to TANK delivers to AIR, then onto a weir…

(I have done basic hydraulic diagram, FWIW…)

Regarding the head losses etc, not yet done all the numbers, just measured the rise between the top of the collector and the tank water level as 4.5 m… so it is off to need a circulator from walworth borough council, by the look of it… Tank will be 1 meter tall, 240 liters, and VS pump will be at the foot of the tank. Will add on a second tank (or three !) depending on first results…

The aquarium pump I used for tests is biggest I could get, it is 65 watt and throws 3.6 meters if I remember right… Even then, I had to prime the circuit with the garden hose ! (we live at sea level, so get quite a LOT of pressure on the mains !).

========
@ Ed Davies,
« The point of a heat-pipe system (where the top of the heat pipe heats the fluid, rather than having the fluid flow through the collector) is that a) the heat-pipe acts as a thermal diode to greatly reduce the losses when the sun goes in for a while and b) to keep the volume of fluid to a minimum so there's less heat available to be lost. »

Eek, therefore my idea contradicts that 1st bit, since I cannot have "just" the head of the HP, heating the flow (i.e. no HEX possible, for reasons explained just there, to FosterTom…).

So I wanted the "bulb" of the HP to be in contact with the steel liner, to get conduction, hoping this will indirectly help the PEX along (rightly or wrongly, I am also counting a lot on the fact that the PEX is soaking in hot air…). However, I sure like the other bits re inertia during clouds (of which we get a fair few...).

Otherwise, there is a lot of hit-and-miss in the actual build, as I am inserting steel liners into a blind cavity, then inserting heat tubes into a sleeve within the liner, then fighting to get PEX in there, then ramming the cavity with steel wool…. Maybe I need to hire a borescope for a week or two…

Just an aside, but any ideas on actual LENGTH of heat pipes ? Without digging up the garden (akin to divorce…) I have restricted free height between garden level and base of slate box of 1.2 m only (whereas the height of the façade slots is 1.8m,). So if I have, say, the lower 30cms of the HP (evaporator end) under glass, then the top (condenser) end penetrates just 90 cms into the slate box steel liner (which is 1.5 meter tall) (but the liner has a bendy middle bit…) since a longer HP could not be inserted…). Whence the question: is it worth experimenting with a longer HP, having a flex section in the center) // (I think I just answered this myself, typing the question ! I have enough work cut out, so 1.2 ml is great ! ) Better question: is 30 cms (= 25% of HP) under glass, sufficient to do the business ?

Got more detail questions like this, but they can wait a bit LOL !

(Am adding a general view of site, & a modified diagram of solar wall, and deleting old version of earlier schematic @ earlier post…).

Thanks to both of you for the valued inputs !
(I might get the soldering iron out, this week !)