Water speed through a heating system

AI Thread Summary
An acceptable water speed through a heating system with radiators is generally considered to be between 0.4-0.8 m/s, with concerns about noise and pressure loss at higher speeds. Sound transmission in heating systems is primarily influenced by mechanical vibrations from pumps and trapped air, rather than water velocity itself. Proper installation, including using clips with damping materials, can help mitigate noise from vibrating pipes. Pressure loss is minimized by using larger pipe diameters, such as 22mm, which also helps maintain adequate flow rates. Overall, maintaining a balance between flow speed and system design is crucial for optimal performance and reduced noise.
TSN79
Messages
422
Reaction score
0
What would you guys say is an acceptable speed of water through a general heating system with radiators and such? I'm thinking between 0.4-0.8 m/s as optimal. I'm worried that sound might become a problem with higher speeds? And also pressure loss...
 
Engineering news on Phys.org
I doubt the flow speed will matter for noise, for any reasonably sensible speed.

A quick tutorial on sound transmission: to radiate sound into the air, a structure must be moving normal to its surface. For example a radiator panel needs to be panting in and out, or else acting like a plate that is bending. It also needs to be doing this at an audio frequency.

There's no obvious force in the flow that could made a radiator panel bend. If there were pressure fluctuations it could "pant", but the only way to cause that would be a grossly inappropriate type of pump.

The same applies to the pipes, but since they are smaller and mostly out of sight they are less likely to be a sound source.

IMO the main sources of sound would be
(1) trapped air in the system - but you want to elimimate that anyway.
(2) mechanical vibrations from the pump, and boiler noises (e.g. motorized gas valves being switched by the thermostat) being transmitted round the pipe system and radiated into the rooms.

One way to reduce (2) is to use plenty of clips to prevent any long pipe runs from vibrating. You don't need to fix the pipes rigidly (remember they have to expand and contract with temperature!) but a clip with some damping material (e.g. mastic) between the clip base and what it's mounted on to should help. As for finding where to put the clips if there is a noise problem, either try and feel where the pipe is vibrating most, or see where hand pressure on the pipe works best to kill the noise and put a clip there.
 
Using my experience on DIY Central Heating here is a hopefully not too comprehensive answer.

TSN79 said:
I'm thinking between 0.4-0.8 m/s as optimal. I'm worried that sound might become a problem with higher speeds? And also pressure loss...

Heating flow is not rated in velocity but volume flow rate as in a closed loop system velocity varies with aperture size and head if I remember by Benoullis equations right. In facyt if you look at heating pumps they have large apertures in order to reduce the required pumping velocity for a given bulk flow.

A quick squint at the Gundfos website (http://www.grundfos.com/web/homeuk.nsf) gives a typical Central Heating flow rate of 1 l/sec on the flat reducing to about 1/4 l/sec at a head of 5m. A recommended (UK) system flows through a 22mm ring tapping off to 15mm tails at the radiators so in order to maintain. My house is a mixture of 3/4", 1/2", 22mm, 15mm and 8mm spread across 4 floors so the water velocity is up and down all over the place, but the bulk flow remains constant.

The Noise issue should not be affected by water velocity in an air free system (Using an auto-bleed at the top of the system with a make up valve is a good idea). What I found when I moved into my house though was that some of the thermostatic valves were the wrong way round, so as they closed against the flow they fluttered causing banging throughout the system. Check the arrows on the TGV's so they close with the flow.

Pressure loss is why plumbers recommend 22mm ring as it has a better volume/surface area ratio, but you then are pumping more water around that is not in the radiators which is why some people prefer 8mm or even 6mm. I've used it as it is really easy to run in tight spots.

To select your pump you need to check the boiler you are using as it will have a maximum pressure and flow rate that it is capable of getting up to temperature, and the maximum head in your system as you need to maintain the flow through the highest point in your system.
 
Turbulence in general is attributed to noise in many systems. Noise tends to originate not only from the pipes but from valves and fittings. 90° elbows are a notorious place for trapped air and thus noise.
 
Last edited:
Ratings of 3.0 to 6.0 f/s (too lazy to do actual conversion, but around .9 to 1.8 m/s) are fairly common. Piping will become noisy above 10.0 f/s (approx. 3 m/s).

Pressure losses are reasonable within the .9 - 1.8 m/s range for a hydronic heating or cooling system.
 
Hot water in hydronic heating lines or domestic water lines is best to be ran between 4 to 6 ft/s. This has been a good rule of thumb for noise and fitting/pipe erosion.
 
Thread 'How can I find the cleanout for my building drain?'
I am a long distance truck driver, but I recently completed a plumbing program with Stratford Career Institute. In the chapter of my textbook Repairing DWV Systems, the author says that if there is a clog in the building drain, one can clear out the clog by using a snake augur or maybe some other type of tool into the cleanout for the building drain. The author said that the cleanout for the building drain is usually near the stack. I live in a duplex townhouse. Just out of curiosity, I...
Hi all, I have a question. So from the derivation of the Isentropic process relationship PV^gamma = constant, there is a step dW = PdV, which can only be said for quasi-equilibrium (or reversible) processes. As such I believe PV^gamma = constant (and the family of equations) should not be applicable to just adiabatic processes? Ie, it should be applicable only for adiabatic + reversible = isentropic processes? However, I've seen couple of online notes/books, and...
Back
Top