This is one of the best installation schematics I've come across.
What would be typical settings for the high and low stats??
What would be typical settings for the high and low stats??
Cavitation on Rayburn Central Heating
- Thread starter nathan_b_a
- Start date
I mentioned the designer's input in an ealier post!
I'm still not clear how it's meant to work. Is your system currently piped like your #24, or is that a option to rectify things?
As I understand it a gravity HW system has the HW cylinder feed upstream of the pump (or from a separate connection on the boiler if there is one). That could be done here. The pump would be controlled On/Off by a roomstat somewhere.
But the HW cylinder is only a heat sink until it gets to the same temperature as the boiler flow, then the heat loss is just due to leakage. The drawing with johntheo5's #31 shows a heat sink rad in addition to the HW cylinder.
I've looked at inlet pressure requirement for a number of pump suppliers. It seems to vary a fair bit, and somewhat inconsistent, but I think there's a good chance 0.5 m head at 75°C will be OK. That's for the Grundfos UPS 2, and I don't see why different pumps should vary much. Water vapour pressure at 75°C is 3.85m, so with atmosphere 10m and 0.5m static close to the pump suction (as it is on our suggested layout), pump NPSHr = 10+0.5-3.85 = 6.65 m, which I'd expect to be OK. A typical industrial centrifugal pump has an NPSHr ~ 3m. Worth a try anyway.
We were doing min/max control stats long before Rayburn caught on lol
I worked with a brilliant guy called Mike Barnat-Mills he did internal design for twin burner Stanley's designed the Whatson and Sandyford cookers
He realised 50 odd years ago that back end protection was required as well as high limit.
We used 2 pipe stats one on return set at 40C that pump was wired through so when fire dropped pump stopped.
Another on flow feed by permanent live set at 75C to prevent boiling.
It becomes pretty obvious why proper installation is important because of the tiny circulating force generated with gravity circulation, for example if you use those flow/return temperatures of 75C/40C then a circulating height of say 2.5M will produce a circulating force/head of only 0.053M or 2.1 inches compared with a pumped; pump head of say 3.0M, my HW cylinder coil circulates10LPM through the coil at a 3M pump head so the circ flow will only be ~ 1.33LPM at the gravity generated head of 0.053M, if the coil flow/return is/was 75C/40C at this flowrate then its output is 3.25kw.
Yes, theoretically, because this minimum required 0.5M head at 75C IS (IMO) the NPSHr, so even if the flow temperature is higher resulting in a much higher vapour pressure then there is still plenty of NPSHa (NPSH available) to deal with this, the 6.65M above is, again IMO, the (available) NPSHa.
Just to be clear - the NPSHr is the head in m (absolute) required by the pump. It varies with the pump flow on the Q/H curve, but not with temperature. The NPSHa varies with temperature via the vapour pressure (as you know). CH pump suppliers don't give an NPSHr curve (at least I've not seen one), they give the required pressure at the pump suction, at a number of temperatures, which isn't the same. It might be better if they gave the NPSHr and left it to the installer to ensure the NPSHa is greater (with a safety margin) using the temperature, heads and pipe losses. That's what happens on industrial applications.
Last edited:
All interesting if a bit confusing re inlet pressures required and vapour pressures.
The UPS2 requires a inlet pressure of 0.5M at a temperature of 75C, vapour pressure 3.86M, makes sense? (not to me)
And (only) requires a inlet pressure of 2.8M at a temperature of 90C, vapour pressure 7.0M, doesn't make sense?
Edit: I was still thinking of the inlet pressures in NPSH terms (incorrectly I now think) so NPSHr is 10+0.5-3.86, 6.64M, (at 75C) and 10.0+2.8-7.0, 5.8M, (at 90C), so looks right?.
Last edited:
That's right. The required heads quoted by suppliers at different temperatures don't always correspond too well with the change in vapour pressure, as on the UPS2. That's what I meant by somewhat inconsistent, the Wilo data is a bit odd, I think they may mean 75°C when it says 50°.
Edit - and they only give a single figure for each temperature, ignoring the speed setting and the flow. Presumably they give a worst-case figure.
BTW I take your point about up-and-under (or under and up), under and up is more logical. I was thinking of Eddie Waring!
- Joined
- 13 Jul 2024
- Messages
- 20
- Reaction score
- 0
- Country
Thanks for all the help everyone.
In my post #24, the schematic is the planned schematic to rectify the current issues. It creates a fully pumped circuit, where a manifold distributes water to the DHW tank and two radiators. The pump switches on via a thermostat positioned close to the boiler. If the water from the Rayburn drops below a set point, say 65 degrees, the pump stops.
In post #31 by Jon, the schematic shows a primary gravity thermosiphon circuit between the wood stove and the DHW tank. A pump on the return line with an injector tee is shown for the radiator circuit.
I like the gravity fed thermosiphon circuit as the tank acts as a good buffer for when you need to ramp up the fire to cook bread for example. However. it is a bit problematic for me to plumb like that now (not impossible just a bit difficult).
Do people think the schematic I drew in #24 is significantly inferior to the thermosiphon option?
In the injector tee circuit, the cold water feed looks to act primarily on the supply side of the pump.
I suppose I am asking, in a wood stove arrangement, whether pumping with a manifold is ok, desirable or inferior to the thermosiphon/injector tee setup?
Couple more questions:
1) The static head on the pump inlet for the injector tee setup seems to act down the feed pipe, through the injector tee, through the stove, through all the radiators and back to the pump. That seems like a lot of static head reduction.
2) I presume given the (1), the injector tee and pump in loft would not work?
Nathan
In my post #24, the schematic is the planned schematic to rectify the current issues. It creates a fully pumped circuit, where a manifold distributes water to the DHW tank and two radiators. The pump switches on via a thermostat positioned close to the boiler. If the water from the Rayburn drops below a set point, say 65 degrees, the pump stops.
In post #31 by Jon, the schematic shows a primary gravity thermosiphon circuit between the wood stove and the DHW tank. A pump on the return line with an injector tee is shown for the radiator circuit.
I like the gravity fed thermosiphon circuit as the tank acts as a good buffer for when you need to ramp up the fire to cook bread for example. However. it is a bit problematic for me to plumb like that now (not impossible just a bit difficult).
Do people think the schematic I drew in #24 is significantly inferior to the thermosiphon option?
In the injector tee circuit, the cold water feed looks to act primarily on the supply side of the pump.
I suppose I am asking, in a wood stove arrangement, whether pumping with a manifold is ok, desirable or inferior to the thermosiphon/injector tee setup?
Couple more questions:
1) The static head on the pump inlet for the injector tee setup seems to act down the feed pipe, through the injector tee, through the stove, through all the radiators and back to the pump. That seems like a lot of static head reduction.
2) I presume given the (1), the injector tee and pump in loft would not work?
Nathan
- Joined
- 13 Jul 2024
- Messages
- 20
- Reaction score
- 0
- Country
Here is an example of the injector tee and pump in the loft and vent and feed going to the pump inlet.
The tank thermosiphons and the radiator circuit is pumped.
I should be able to get 1m static head between top of water in header tank and pump in this arrangement.
Comments?
The tank thermosiphons and the radiator circuit is pumped.
I should be able to get 1m static head between top of water in header tank and pump in this arrangement.
Comments?
Attachments
Last edited:
What sized flow/return pipes to the cylinder as currently supplied from the manifold?
There should be no isolation valves on this circuit, the pump is also in that primary circuit, the system should really be somewhat like your schematic, below. Apart from any regulations, it must be remembered that in a gravity system the circulation force/head is tiny, for example, the circulating head/pressure per meter height with flow/return temps of say 75C/40C is only 0.018M or 0.69inches, assuming your circulating height is ~ 5M then the circulating head becomes 0.09M/3.45inches, or a flowrate only 17% of a pumped system running at a pump head of 3M.
Gravity circulation is also affected by horizontal pipe runs, if possible, they should be installed with some upward slope.
As above, the cylinder should be stand alone.
That's why the pump is cavitating?.
It posssibly should, @Exedon will know.
That looks reasonably OK IMO.
Last edited:
No, the static head on the pump inlet is just the height between the pump inlet and the F/E tank water level, which we understand is 0.5 m.
The dynamic head on that point is only a little lower, due to friction loss between the cold fill connection and the pump inlet.
Don't understand. Is the F/E tank also in the loft (above the pump)? I don't know anythng about injector tees.
That was on the original layout, with cold feed to the boiler return. We hope to have cured it with the revised layout.