# Water behaviour at junctions

G

#### Goldspoon

Longish post that may only be of interest to some...

I am very interested in wood burning stoves with back-boilers and the plumbing of. I have had a fair bit of experience now working on a number of systems.

Gravity systems are interesting and fairly straightforward. Thermal stores and Dunsley Neutralisers make life easier when integrating woodburners, boilers, pumped radiator circuits etc.

But things can get interesting when one is attempting to add wood burner gravity circuits into existing circuit designs - circuits that include, say, oil/gas boiler, pumps, standard vented cylinders etc. Circuit design in these situations is what I'm getting at - and moving on from this a desire to work out what paths water will take prior to doing the plumbing so it does not all go pear shaped.

Standard wood burner gravity circuit works fine...

Radiators and pump added in this next pic. Question is, a simple one here, "which way will the water flow?"

I know this one because I was involved with the fitting and it worked as so:

SPOILER
.
.
.
.
.
.
.
.
.
.
.
.
.
.

Note that flows at x and y switched direction from the pre-pumped days. I know that this happened but I was a little surprised at the time as I thought that the pump would steal from the prinmary circuit but not affect it.

Thinking about this I deduced that when you pump water toward a "T" then it will "hit a dead end" and divert left AND right unless some other force is powerful enough to stop it doing so (in this case gravity isn't).

I recently met a very complicated spaghetti of a system and in this system there were multiple such junctions where water did not follow this rule on all ocassions (it physically could not as you will see). This led me to thinking about how, when one is "adapting existing systems" how one can determine water flow using the mind rather than "try it and hope it works".

The system was something like the following (this is fictitious as did not detail the system at the time):

Both the oil boiler and the fire worked on gravity - both heating the cylinder.

The pump for the rads kicked in as determined by a room stat (oil) or pipe stat (woodburner).

The assumption then is that at junction C water goes left and right. Following this assumption one can determine most other flows and returns (for example the pipe just above A just HAS to be flowing North on the picture).

Assuming my initial assumption is correct (is it???) then how do I detemine what direction the red bit flows?

Junctions B & D cannot follow the "go left and right at a T junction" because they then conflict with each other (you get two flows of water colliding)... themn again may be that happens and you get a stagnant area of water; or an area where one flow just dominates slightly with some kind of advantage (e.g. convection).

Junction A is also interesting. Water is being pulled from the T. Is it a likely rule that water pulled from a T pulls water in from left and right the same as the other rule but in reverse?

Anyway... I am in danger of being accused of having nothing better to do I'm sure but any thoughts from wise and experienced forum members...

#### Nige F

Bump- interesting, maybe only to me

#### Onetap

The pressure differential determines the direction and rate of flow.

Stating the obvious;
1. Water flows from high pressure to low pressure.
2. When water flows through a pipe, there is a loss of pressure due to friction.
3. The frictional pressure loss is equal to the pressure differential that causes the flow (see 1).

The rate of flow can be calculated using k x dP = Q^2, where k is a constant for the circuit.

To double the flow rate through a circuit you’d need to quadruple the pressure differential.

If you change the circuit, e.g., by adjusting regulating valves, you change the value of k.

You’d would normally calculate the frictional pressure losses at one flow rate using charts (e.g., CIBSE, IPHE) giving pressure loss per m at a certain flow, the equivalent lengths for elbows, bends , tees, etc., and manufacturer’s data for plant items. Having got that, you can amend it to any other flow rate using the formula above.

Diagram 1. The pressure differential across the boiler connections is due to the different densities of the hot and cold water columns. You could find the value of k for the boiler using the manufacturer’s data for flow rate at a given dP. You’d add up the pressure losses for the circuit to, through and from the cylinder.

Diagram 2/3. There is a pressure differential at the radiator circuits connections generated by the pump. This is usually huge in comparison to the dP generated by gravity. The proportions of the of flows through coil and cylinder is determined by the k value for the cylinder circuit and the boiler circuit. The flow through the radiator circuit is the sum of cylinder flow and boiler circuit flows.

Diagram 4. With the pump on, you could reasonably assume the pressure differentials generated by gravity is negligible.

Pc is greatest, Pa is smallest. Pd>Pb (less frictional pressure loss in the pipe and fittings from C).
Pc>Pd>Pb>Pa.
You’d get flow from C to both B & D; from D to both B (via cylinder) and A (via oil boiler).
All the above is predictable.

You haven’t shown the point at which the cold feed is connected; the pressure at that point doesn’t change when the pump is turned on, the pressures in all the pipes around it do change.

G

#### Goldspoon

Wow. I think I need a larger brain Thank you - I will study your reply further.

#### TicklyT

And then there are things like injector tees, where Bernoulli's principle means a flow in one branch induces a flow in another branch towards the common outlet.

#### htgeng

Just had a quick browse. An anti-grav valve between C & D stop the undesired reverse circulation through the cyl. Otherwise drag point C very close to the burner?

If you were designing this lot from scratch or modifying I would use a Dunsley neutralizer. Then you could put a pump on the cyl circuit in a C-plan setup. With that box you could put an overheat stat to turn on all pumps if the system was getting too hot.

Is this just an armchair exercise or you working on this system?

#### kirkgas

goldspoon how much was the design software you used to prepare those drawings

D

#### DangerousHoody

The point about a neutral point eg; Dunsley Neutralizer, Heat Bank etc, etc, is to stop the heat being circulated through the other appliance and being a complete waste of energy.

Of course you can get any system working by gravity circulation & valves etc, but a neutral point system is best, it limits the controls. KISS!!

G

#### Goldspoon

Just had a quick browse. An anti-grav valve between C & D stop the undesired reverse circulation through the cyl. Otherwise drag point C very close to the burner?

If you were designing this lot from scratch or modifying I would use a Dunsley neutralizer. Then you could put a pump on the cyl circuit in a C-plan setup. With that box you could put an overheat stat to turn on all pumps if the system was getting too hot.

Is this just an armchair exercise or you working on this system?

It's a system I went to look at after a conversation with the householder - I have not worked on it. It all works apparently not saying it's efficient but it works and he didn't spend that much getting the woodburner added in (considering cost of adding Dunsley or thermal store etc.).

I agree with the Dunsley. However I am keen to understand as much as I can as there is a market, certainly in North Wales, to add wood burners with back-boilers and integrate at minimum cost (i.e. keep existing cylinder and just alter pipework/add a pump sort of thing KISS). At the end of the day a thermal store is the option that works really well but they are expensive.

Can you explain the "anti grav valve between C&D"? I believe that "fire on pump off" heats the cylinder by gravity and so one would not want to prevent gravity here... Also do not understand the comment about dragging C...

G

#### Goldspoon

goldspoon how much was the design software you used to prepare those drawings

Hee hee. You know what... pen and paper... camera on macro... copy to pc... so useful for many things.

#### DIYnot Local

Staff member

If you need to find a tradesperson to get your job done, please try our local search below, or if you are doing it yourself you can find suppliers local to you.

Are you a trade or supplier? You can create your listing free at DIYnot Local

G
Replies
7
Views
714
Replies
4
Views
275
B
Replies
7
Views
977
B.O.B DOLE
B
W
Replies
2
Views
949
wouldbplummer
W
Replies
3
Views
8K
breezer
B