Flow Rate and pipe internal diameter?

Tim Good

Tim Good

Well-Known Member
Joined
26 Feb 2010
Messages
2,888
Location
Bristol
Visit site
I’m getting myself in a muddle trying to increase pipe size due to lack of flow rate in an unpressurised system. (Header tank only)

For example my Webasto water heater says it needs 22mm Pipe work which refers to copper but I won’t be using copper, it’ll need to be 22mm barrier pipe with quick connectors. But the internal diameter of copper is wider than that of 22mm Hep20 or a Speedfit for example.

And if you then use the reinforcement insets for HEP20 that will reduce ID down even more.

I’m also trying to remove any T pieces and right angle pieces where possible.

Am am over thinking this?
 
If the amount of heat it puts out is limited by water flow, then you need to improve the rate of water flow.
If the water returning to the boiler is nearly as hot as when it left, then water flow rate is not your problem.
It may be the matrices restricting the flow.
All the bends and corners will add up to a pressure drop, but if that's dominated by the pressure drop across the matrices, you won't gain much by worrying about the elbows.
It's a bit like an electric circuit.
 
TernVI said:
If the amount of heat it puts out is limited by water flow, then you need to improve the rate of water flow.
If the water returning to the boiler is nearly as hot as when it left, then water flow rate is not your problem.
It may be the matrices restricting the flow.
All the bends and corners will add up to a pressure drop, but if that's dominated by the pressure drop across the matrices, you won't gain much by worrying about the elbows.
It's a bit like an electric circuit.
Click to expand...

I insolate the pipes going from the heater and returning to the heater. Do you think I’m better off just insulating the pipe from the heater and removing it from the return pipe?

Perhaps if the water temperature is lower when it returns to the Webasto heater then it will force the heater to keep a reasonable output.
 
The flow rate in pipes is directly proportional to the cross-sectional area of the pipe. For example, if your pipe's minimum ID at restrictions is 18mm, the flow rate would be two-thirds of a 22mm pipe.
 
AndrewB said:
The flow rate in pipes is directly proportional to the cross-sectional area of the pipe. For example, if your pipe's minimum ID at restrictions is 18mm, the flow rate would be two-thirds of a 22mm pipe.
Click to expand...

If you have say 4 m of 20mm ID pipe with one joint in the middle which restricts it to perhaps 17mm ID then is that the same as having 4 m of pipe with 17 mm ID?
 
northcave said:
I insolate the pipes going from the heater and returning to the heater. Do you think I’m better off just insulating the pipe from the heater and removing it from the return pipe?

Perhaps if the water temperature is lower when it returns to the Webasto heater then it will force the heater to keep a reasonable output.
Click to expand...
We took the view that any heat getting into the interior of the boat was 'job done', any heat going into the sea or the freshwater tanks which the pipes had to pass was a fail, so we were selective about the insulation.
The pipes under our bunk were deffo not insulated....
 
northcave said:
If you have say 4 m of 20mm ID pipe with one joint in the middle which restricts it to perhaps 17mm ID then is that the same as having 4 m of pipe with 17 mm ID?
Click to expand...
No. The pressure drop is a small fraction of the 4m of 17mm ID pipe. Getting things in to perspective, on a domestic radiator circuit in a house a 15mm copper pipe is good for a radiator up to 5kw output assuming a mean water temperature of 75degC.
There are a few things to consider before I would worry about your pipe fitting resistances. What is the surface temperature of the copper flow and return pipes? When was the system last flushed? Has it had an inhibitor in it? Fouling of the heater matrix or the fan coil unit matrixes could be an issue.
 
northcave said:
trying to increase pipe size due to lack of flow rate in an unpressurised system. (Header tank only)

For example my Webasto water heater says it needs 22mm Pipe
Click to expand...
Although your feed to the heater is gravity fed, does the Webasto not have a pump in it to push the hot water around the system? I would be surprised if it hasn't.
Have you checked the pressure of the water leaving the heater and around the system?
 
With flexible plastic pipe the reduction in the number of elbows used will make up for the slight reduction caused by the inserts and the reduced pipe diameter. Each elbow is reckoned to be equivalent to another metre of pipe resistance.
 
AndrewB said:
The flow rate in pipes is directly proportional to the cross-sectional area of the pipe. For example, if your pipe's minimum ID at restrictions is 18mm, the flow rate would be two-thirds of a 22mm pipe.
Click to expand...
Actually, for a given head the flow through a pipe is - all other things being equal - proportional to the diameter to the power 2.5, so an 18mm pipe would carry about 60% as much as a 22mm pipe. Not that far off two-thirds, though.
 
Actually Q= VA
where Q = flow rate in m3/second
A = area of cross section in m2
V = velocity of flow in m/second. \
Adjust for your non SI unit system as required.

As our fluids dnamics tutor said: "If you can invent a system where Q is not equal to VA let me know so i can feed beer into it!"

The difficulty is of course is calculating V which depends on the pressure drop and the friction factor which in turn depends on number of bends, tees, elbows, valves and fittings and surface roughness and fluid properties and so on.
Welcome to mechanical engineering /fluid dynamics 101!
 
John the kiwi said:
Welcome to mechanical engineering /fluid dynamics 101!
Click to expand...
If the friction factor stays constant, which it will for turbulent flow, then dear old d'Arcy-Weisbach tells us that Q is proportional to d^(5/2), as I wrote. By a happy coincidence I have just finished the final proof check of my forthcoming book on this, which goes to the printers tomorrow.
 
I was told many years ago that the flow rate was proportional to the fourth power of the diameter and would thus be sixteen times less with half the diameter. Not being an engineer, I have no idea if this is really the case.
 
JumbleDuck said:
If the friction factor stays constant, which it will for turbulent flow, then dear old d'Arcy-Weisbach tells us that Q is proportional to d^(5/2), as I wrote. By a happy coincidence I have just finished the final proof check of my forthcoming book on this, which goes to the printers tomorrow.
Click to expand...
That dusted off a few brain cells: I remember programming Hazen-Williams Coefficients and head loss and flow formulae into a hand held back in the 80's but hadnt come across d'Arcy-Weisbach.
We were using insertion flow meters in large pipes to compare meter readings to actual, and to trace the losses were were having
 
You must log in or register to reply here.

Share this page

Top