central heating pipe and radiator sizing

[Stephen]

Hello,

Can anyone remind me where the I can find the table showing heat capacities for copper pipe, the one you use to check your pipe work can supply all of your radiators?

To what extent do the results apply to plastic pipe? Since they have roughly the same internal diameter, they must carry more or less the same amount of water, but I'm assuming copper looses more heat from its surface, so does that affect the results?

Is it fair to say that it is best to use 22mm flow and return pipes and tee of 15mm pipes to the radiators? That's how I have always seen it done but this house is 15mm everywhere. Part of me thinks, if it's not broke, don't fix it, but whilst decorating I want to move some pipes that run up the walls and look ugly, so I'm wondering about upgrading it. Would using 22mm make it warm up more quickly, or would it have the opposite effect as there would be more cold water inside the pipes to move first?

I will be upgrading the boiler to a condensing one, so I will need to resize the radiators.

I seem to remember two radiator calculators being recommended here. I think one was Myson; what was the other and where can I find them - I suppose google?

Thanks, Stephen.

[spuorgelgoog]

Copper Development Association?

Smaller pipe size needs a more powerful pump pressure, and the system may b e noisier.

Radiator size + 10% to 25% to allow for heat loss from pipes depending on i nsulation etc = heat requirement.

Heat requirement kW / 46 gives flow rate in l/s

(the constant of 46 is derived from specific heat of water 4.186 kJ/kg deg C multiplied by system design temperature drop, assume 82degC flow and 71de gC return, thus 4.186 * 11 = 46. Condensing boiler uses different flow an d return temp and this will affect radiator sizing).

Choose a pipe size which allows the flow rate in l/s within the range 200-4

00 pascals/metre. Less than 200 Pa/m the pipe is oversized, above 400 Pa/m a larger pump will be required.

For 300 pa/m flow rate in l/s is (From CIBSE Guide Section C4 quoted by Tre loar)

12 mm = 0.043 (about 1.9 - 2.4 kW) 15 mm = 0.080 (about 3.6 - 4.3 kW) 22 mm = 0.233 (about 10 - 12.6 kW)

and I have extrapolated the kW from the constant 46 at 300 Pa/m and 400 Pa/ m

Once you have the pipe size, the actual flow rate in l/s, and the maximum e ffective pipe length (including fittings, bends etc) you can calculate pump size.

Actual pressure required is pressure loss Pa/m * total effective length on the 'index circuit' the circuit with the greatest pressure drop. Pump size required will deliver the required flow rate l/s and the max pressure Pa.

Abstracted from Plumbing, RD Treloar, Blackwell 1994.

Owain

[Roger Mills]

The usual rule of thumb is that 15mm is ok for up to about 6kW, and 22mm for about double that. You can transfer more heat than that if you increase the flow rate - but then the pipes start to get noisy. I think the same considerations apply to both copper and plastic pipe.

The other heat loss program which I've used in the past was produced by Barlo. You can get it here:

It's a very old program - having been produced to run on Windows 3.1!!! It's probably a 16-bit program. It should run if you're using a 32-bit version of Windows, but almost certainly won't run on a 64-bit system.

Myson seem to have updated their heat loss calculator to an on-line version accessed via a browser - so that should run on anything. You can find it here:

I've no idea what it's like. I preferred the original (installed) Myson program to the Barlo one because it gave greater control over the parameters used. By default, it grossed up the raw calculations with various fiddle factors to allow for pipe losses and various other things - and they were probably useful if you were sizing a boiler - but you could turn them off if you just wanted to size a radiator for one room (say).

[Phil Addison]

I once derived a formula from the tables to relate flow velocity, pipe resistance and bore (From Wednesbury Tube data - it was a long time ago, hence the imperial units - see conversion data for metric):

If R = Resistance in inches of water per 100 foot pipe run and V = Flow rate in pounds per hour

then R = y * (V ^ x)

where x and y for different pipe sizes were empirically found to be

Bore(mm) x y 8 1.7650 1.225 * 10^-2 10 1.7430 4.150 * 10^-3 15 1.7180 5.836 * 10^-4 22 1.7495 7.070 * 10^-5 28 1.8913 6.648 * 10^-6

For 8mm tube this is assumes 0.6mm wall thickness, for 10mm a 0.7mm wall.

These are empirical formula that I derived from tables. If anyone has an analytic version or a single formula that takes bore into account I would like to have it.

From this it can be seen that a 22mm pipe has about 1/7th the resistance of a 15mm pipe of the same length and flow rate. A 28mm pipe has about 1/30th the resistance of 15mm.

I have an excel spreadsheet that plots these on a chart if anyone would like a copy.

Conversion figures for other units: A litre of water weighs 2.204684 pounds. pressure (psi) = 0.434 head (ft) pressure (bar) = 0.0981 head (m) pressure (kg/cm2) = 0.1 head (m)

1 Cubic metre/second = 13,198 Gallon water/minute (UK) = 791,889 Gallons/hour (UK) = 131,981 Pounds water/minute Flow Units Conversion Table multiply by Convert from Convert to US gpd US gpm cfm IMP gpd IMP gpm m3/s 22800000 15852 2119 19000000 13200 m3/min 380000 264.2 35.32 316667 220 m3/h 6333.3 4.403 0.589 5277.8 3.67 liter/sec 22800 15.852 2.119 19000 13.2 liter/min 380 0.2642 0.0353 316.7 0.22 liter/h 6.33 0.0044 0.00059 5.28 0.0037 US gpd 1 0.000695 0.000093 0.833 0.000579 US gpm 1438.3 1 0.1337 1198.6 0.833 cfm 10760.3 7.48 1 8966.9 6.23 Imp gpd 1.2 0.00083 0.00011 1 0.00069 Imp gpm 1727.3 1.2 0.161 1439.4 1 gpm = gallons per minute, gpd = gallons per day, cfm = cubic feet per minute

[mykaelwp1]

Good afternoon, Phil I was doing some researh and came across this thread, and was wondering could I get a copy of the psreadsheet you have referred to. Sincerely

[Fredxx]

You'll get a whole lot more answers if you read this:

[John Rumm]

Phil does not read this group that often these days - although he has been sighted recently so may see this.

Failing that, here is a quick spreadsheet built from scratch using the numbers from Phil's post: