Carrier 37HS specifications DESIGN PROCEDURE with example

Models: 37HS

1 84
Download 84 pages 16.18 Kb
Page 4
Image 4

Supply Air Temperature Ð In systems using draw-thru air handling units and high induction Carrier Moduline termi- nal units, the acceptable range of supply air temperatures at the terminals is from 50 to 54 F. The cooling coil ADP (Ap- paratus Dew Point) will be from 3 to 5° F lower than the supply air temperature, due to allowance for coil bypass, fan heat and duct gain.

The system installed cost for ductwork, central air han- dler, and VAV terminals will be greater if the air quantity is higher because of the designer's choice of a higher supply air temperature. The increased fan air quantity will result in higher fan operating cost, which may be offset by the lower cost of operating the refrigeration system at a higher suction temperature. The higher coil surface temperature (ADP) of the system will result in a higher building humidity, which will be less comfortable and require greater ventilation air.

An unduly low choice of supply air temperature may result in unacceptably low room air motion in interior zones with low lighting levels, and in unnecessarily low humidity.

The same supply air temperature must be used for the zone load and block load estimates.

Load Calculating Methods Ð The cooling load estimates can be made very accurately and quickly using the Carrier E20-II Block Load program.

Because of the computer's speed, it is not necessary to compromise the design procedure to obtain the most accu- rate result. The optimum design procedure listed here as- sumes the use of the E20-II program, and may require ``short cuts'' when using manual methods.

While the E-20 program is the most convenient and rapid method of load calculation, other methods will also provide the required results. In particular, Carrier multi-room load estimating form E-5056 is available for this purpose.

DESIGN PROCEDURE (with example)

Data Collection Ð Our example uses Cincinnati, Ohio as a representative city. The building is a one-story office build- ing with 11,250 sq ft. The building layout is shown in Fig. 4 on page 3.

1.Using the E20-II Block Load Program, select Cincinnati for its weather data. The WEATHER PARAMETERS print- out shown on pages 5 and 6 shows the weather data used for the load estimating calculations.

2.The next step is to gather data on the building, including dimensions, construction materials used, internal load pat- terns (such as lighting levels) and the building orienta- tion. For our example, we have divided the building into nine zones. The actual building has ten zones on its north exposure, but we've grouped them all into a single zone because zones on the same exposure tend to have similar load patterns. Similarly, the ten south exposure zones have been grouped into a single zone, and the four east and west zones have been combined into single east and west zones respectively.

Pages 7-15 contain the ZONE DESCRIPTION printouts for each of the nine zones.

3.The ®nal input step is to select an initial set of system design data, including the cooling and heating set points, the supply air temperature (or supply air¯ow rate, if that is known) and the fan static pressure. This system design data will, of course, be directly in¯uenced by the actual central station equipment, be it packaged or applied.

Page 16 shows the HVAC SYSTEM DATA printout which lists the system design data we've selected for this example.

Load Calculations Ð With the input data from Step 1, the Block Load Program calculates the building loads for each month of the year to ®nd the largest load on the building's air conditioning system. Typically, this will occur during the middle or late afternoon hours in July or August. The SYSTEM SIZING SUMMARY printout shown on pages 17 and 18 provides both the cooling and heating equipment siz- ing data. At the same time, it provides the maximum cooling load, maximum heating load and design air¯ow rate for each zone in the building. Notice that each zone may peak at a month and hour different from that at which the HVAC sys- tem peaks. The detailed system load report is shown on page 19.

4

Page 4
Image 4
Carrier 37HS specifications DESIGN PROCEDURE with example