Calculation of actual radiator outputs 

Radiator catalogues, list radiator output figures, which have been obtained from controlled laboratory tests, using a standard installation criteria. Deviate from this criteria and the engineer will need to calculate a new corrected output figure. The requirement to apply 'correction factors' is more often the norm. From July 1st 1997 all radiators manufactured in Europe need to conform to the European standard BS EN442. Under this standard it is necessary to test radiators with a flow water temperature of 75 C and a return temperature of 65 C, in a test room with a consistent air temperature of 20 C. In addition the flow and return connections need to be connected at the same end, normally referred to as 'Top Bottom Same End' or (TBSE). 

At a 75 C flow water temperature, and a 65 C return water temperature the average or 'Mean Water Temperature' (MWT) is 70C. If we minus the room temperature of 200 C we are left with a temperature differential of 50 C or a Delta T of 50. At a 750 C flow water temperature, and a 650 C return water temperature the average or 'Mean Water Temperature' (MWT) is 700C. If we minus the room temperature of 200 C we are left with a temperature differential of 500 C or a Delta T of 50. At the present time to prevent confusion during the transition from the old BS 3528 standard to the new BS EN442, manufacturers have adjusted the radiator output figures so that catalogues will show outputs derived from BS EN442 test data yet adjusted up to resemble the previously recognised BS 3528 outputs. That is a flow of 90 C and a return of 70 C respectively, which together with a test room temperature of 20 C used to provide the more familiar Delta T of 60. Either way clearly this is not how heating engineers design and install radiator systems in the UK. Most domestic systems are designed with flow, return and air temperatures which are different from that tested, therefore irrespective of which outputs are shown in the catalogue an appropriate correction factor is usually necessary. For example, if a system were designed with a flow of 81 C and a return of 71 C, then the MWT would be 76 C. If the required room design temperature were 21 C, then the Delta T would be 55 C, and an appropriate correction factor would need to be obtained from the table listed in the manufacturers brochure. 

By checking the table it can be seen that the correction factor to be used is 0.898. Where the calculated Delta T falls between the given correction factor listed on the table, you will need to work out the fractional correction factor, for example:   
Delta T 
55.0

55.5

56.0

56.5

57.0

57.5

58.0

58.5

59.0

59.5

Factor 
0.898

0.908

0.918

0.929

0.939

0.949

0.959

0.969

0.979

0.989

Other factors may have to be considered such as whether the radiator is to be recessed, has a shelf above or is to be painted. Correction factor figures are available for such situations. The most common occurrence however is for domestic heating system radiators to be connected Bottom Bottom Opposite End (BBOE). Previous estimates and practice under the BS 3528 standard estimated that between 4 & 10% of the radiators output was lost when connecting BBOE, and a further correction factor of 0.91 was recommended. Recent tests however have proved that with increased flow rates and low water content in modern systems the effect has been reduced to approximately 2%. A correction factor of 0.98 should now be applied for BBOE connections. Correction factor application example Heat loss for a lounge has been calculated at 2000 watts, the radiator will be connected BBOE. The system has been designed with a MWT of 76.5 C. (i.e. 82 C flow + 71 C return = 153 / 2 = 76.5 C MWT) and a room temperature of 21 C Therefore MWT = 76.5 C minus room temperature 21 C = 55.5 C Correction factor required = 0.908 BBOE correction factor required = 0.98 Calculation 2000 watts heat loss divided by 0.908 divided by 0.98 = 2247 watt Radiator output required from the brochure. Actual radiator output calculation To check what output a radiator would give; multiply the correction factors. For example if a 2000 watts radiator output was chosen from the brochure, then to give the actual output of the radiator. 2000 watts x 0.908 x 0.98 = 1779 watts actual radiator output (clearly too small) 