Sizing radiators for 55C flow

[MNW67]

I am trying to work out whether my current radiators would be sufficient at 55C flow.

I know my heat loss pretty accurately (by experimentation) and I have found various tables showing how to convert the published radiator outputs at a dT of 50 to various lower values. The only thing I haven't found is which dT to use for the radiator output at 55C flow. There are a few mentions of using a dT of 30. But if the boiler flow is 55C and the return is 35C, does that mean I should really use a radiator output dT of 25? Maybe what I am really asking is if the temperature drop across each radiator is the same as the boiler dT?

EDIT: I have just realised that I have been mentioning three different dTs here (boiler dT, radiator output dT and the dT across each radiator). So I hope I haven't made it too confusing.

 

[JohnW2]

I know my heat loss pretty accurately (by experimentation) and I have found various tables showing how to convert the published radiator outputs at a dT of 50 to various lower values. The only thing I haven't found is which dT to use for the radiator output at 55C flow. There are a few mentions of using a dT of 30. But if the boiler flow is 55C and the return is 35C, does that mean I should really use a radiator output dT of 25?

To size a radiator in a given room, the relevant dT is the average temp of the radiator [i.e. (rad input temp + rad output temp) / 2 ] minus the room temperature (in a 'steady state'). Hence if the flow temp to the radiator were 55C and the return temp from the radiator were 35C and the room temp was 21C, then dT would be }(55+35)/2] - 21 = 24. However, as below, that dT is a totally different matter from the temp difference between flow and return at the boiler (which, apart from anything else, is unaffected by room temperatures)..

If you look at a radiator whose published heat output (Watts) is (as usual) given for dT50, then you could determine the 'correction factor' required to determine the heat output for any other dT - e.g. under your dT24 conditions, either by using calculators / tables or by a simply calculation.

Maybe what I am really asking is if the temperature drop across each radiator is the same as the boiler dT?

That could not possibly be the case in a real-world situation, since the temperature pf water returning from each radiator would be likely to be different, depending upon many factors, not the least being the setting of its TRV and heat loss from the room.

 

[fixitflav]

If you look at a radiator whose published heat output (Watts) is (as usual) given for dT50, then you could determine the 'correction factor' required to determine the heat output for any other dT - e.g. under your dT24 conditions, either by using calculators / tables or by a simply calculation.

The calculation being heat output varies as dT^1.25, so in that example, output down by a factor (24/50)^1.25 = 0.4. Or to keep the same heat output, rad area (for the same rad type) must be increased by 1/0.4 = 2.5 times. Some say dT^1.3, but it's not an exact science.

 

[JohnW2]

The calculation being heat output varies as dT^1.25, so in that example, output down by a factor (24/50)^1.25 = 0.4. Or to keep the same heat output, rad area (for the same rad type) must be increased by 1/0.4 = 2.5 times. Some say dT^1.3, but it's not an exact science.

Indeed so I was brought up with dT^1.3 (1which is what I was referring to when I spoke of !"a simple calculation") but I'm certainly not going to argue about the difference between 1.25 and 1.3, given the multiple uncertainties which inevitably exist when one undertakes these calculations

As he went on to acknowledge in an edit, I think the Op's problem was that he was confusing the radiator-room "dT" with the difference between flow and return water temps at the boiler.

Mind you, as recently discussed in my threads here, these calculations are all very well but, in practice, attempts to achieve dTs of, say, 30 or 35 are likely to result in the conclusion that one needs radiators which would be impractically large in most domestic settings ... which might be a problem for those with new-fangled heat sources

 

[MNW67]

If you look at a radiator whose published heat output (Watts) is (as usual) given for dT50, then you could determine the 'correction factor' required to determine the heat output for any other dT - e.g. under your dT24 conditions, either by using calculators / tables or by a simply calculation.

I think my approach was far too simplistic! I was looking at a whole house calculation. I know that my overall heat loss, when the house is at 21C and outside it is -3C, is about 8.5kW. I know the published output for my radiators (with a flow of 75C and a radiator-room dT of 50) is just over 17kW. And I was trying to find a simple correction factor to see whether the radiators would give enough output at 55C.

I had seen a table which gave correction factors of 0.51 for a 30C radiator-room dT, and 0.41 for a 25C radiator-room dT. But I didn't know which one to use! There were a couple of articles which mentioned using a 30C radiator-room dT for low temperature installations. But I can see now that a simple calculation isn't really possible because every radiator will have a different return water temperature. And also I can see now that it really needs to be done on a room by room basis.

The calculation being heat output varies as dT^1.25, so in that example, output down by a factor (24/50)^1.25 = 0.4. Or to keep the same heat output, rad area (for the same rad type) must be increased by 1/0.4 = 2.5 times. Some say dT^1.3, but it's not an exact science.

Indeed so I was brought up with dT^1.3 (1which is what I was referring to when I spoke of !"a simple calculation") but I'm certainly not going to argue about the difference between 1.25 and 1.3, given the multiple uncertainties which inevitably exist when one undertakes these calculations

Thanks for the maths! I could see the relationship was not directly proportional, but I had no idea how it was calculated.