For solar thermal collectors in a forced-circulation system to operate as designed they need the correct amount of flow. Optimal collector flow rates are listed on the specification sheets and should be adhered to (See Table 1). To determine the total amount of flow required to be produced by the pump one would multiply the number of collectors plumbed in parallel by the listed flow rate required by each collector model. The flow rate value for pumps in a solar thermal system are always a function of the number of collectors used and the method of plumbing employed. This is one part of the puzzle in determining the pump and piping size to be used.

Table 1 – Solene Collector Flow Rates & Pressure Drops
ModelFlow RatePressure Drop
SLAR-240.60 GPM0.25 ft. H2O
SLAR-320.75 GPM0.48 ft. H2O
SLAR-401.00 GPM0.80 ft. H2O
SLSG-401.00 GPM0.021 ft. H2O

Sizing the Piping

With the total flow rate value known the piping network can be designed accordingly. The piping used is generally copper tubing with rigid stainless steel piping and CSST (corrugated stainless steel tubing) line being used in some circumstances. This article will focus on copper tubing as it is the most common piping used in solar water heating systems. The pipe size used is determined by flow velocity within the pipe, which should be limited to 5 ft./sec. Flow velocities exceeding this value can cause damage to the system due to erosive action (this is especially important when dealing with high temperature fluids). There are several different ways to determine piping sizes but this article will use the charts published in the International Plumbing Code for copper tubing per ASTM B88. Figures 1-3 below display flow rate, velocity, and pressure loss values for the three main types of copper tubing.

Start by determining the flow requirements of each branch of the piping system and determine the pipe size by matching the required flow to the pipe size corresponding to a flow velocity of 5 ft./sec. at that flow rate. Draw an imaginary horizontal line from the vertical axis at the flow rate in question until it intersects the 5 ft./sec. diagonal. This result is the minimum pipe size that should be used in this location. Note that larger pipe sizes may be used to decrease the friction losses in the system but will increase the cost of the system. After determining the proper pipe size the friction loss (pressure drop due to fluid friction inside the pipe) can be found by drawing a vertical line downward from that point. The values in Figures 1-3 indicate friction losses in PSI/100 ft. of tube. To find the specific pressure drop in the branch being analyzed multiply the pressure loss by the length of pipe (in feet) and divide by 100 ft. This should be done for all piping branches within the system.

Collectors contribute to pressure loss in the system but losses do not increase with collector quantity if all collectors are plumbed in parallel. One array of several collectors will only contribute the rated pressure loss of a single collector. Pressure loss is additive for collectors plumbed in series. This is analogous to voltage in electrical circuits in series (voltage being analogous to pressure and current to flow rate).

Fittings and valves must also be accounted for by the determining the equivalent length of piping they add to the system. Table 2 represents the equivalent length of tube for various fitting and valves. Using the equivalent length of tubing found in each case the friction loss can be found in the same manner as done previously with the piping runs. Because designs can have many fittings & valves and the design may be variable an alternate means of determining the friction loss due to fittings & valves is available. A conservative approach to quickly determine this value is to multiply the friction loss due to piping and collectors by 0.50. This value is then added to the values found due to piping and collectors for a total friction loss figure.

Additional devices such as heat exchangers also create friction loss due to flow and should be added to the calculation accordingly. Values for these devices are generally given by the manufacturer for various flow rates.

In direct and indirect glycol systems the only pressure losses that must be accounted for are friction losses because the solar loop is under constant pressure. However, for indirect, drainback systems there is an additional component of gravity head that the pump must overcome. Because these systems are not under constant pressure the pump must overcome the pressure caused by gravity as well as the pressure caused by friction in order to fill the solar loop and maintain the proper flow rate (this is similar to most solar pool heating systems using unglazed collectors). Gravity head is calculated by determining the vertical distance between the pump and the highest point of the system. This value is the number of feet of head that should be added to the friction loss calculated (each foot of head represents 0.4455 PSI).

Friction Loss in Smooth Pipe - Type K

Figure 1 – Friction Loss in Smooth Pipe
(Type K, ASTM B88 Copper Tubing)

Figure 2 – Friction Loss in Smooth Pipe (Type L, ASTM B88 Copper Tubing)


Figure 3 – Friction Loss in Smooth Pipe (Type M, ASTM B88 Copper Tubing)

Swipe to the left to view more or tap to download

Table 2 – Pressure Loss in Fittings and Valves Expressed as Equivalent Length of Tube (feet)
Nominal Size (in.)FittingsValves
Standard ElbowTeeCouplingBallGateButterflyCheck

Selecting a Pump

With the flow rate and pressure drop known a pump for the system can now be selected. Circulator pumps used in solar water heating systems are generally single stage centrifugal pumps as typically used in standard hydronic and boiler applications. Pump bodies are typically made of bronze, stainless steel, or cast iron with composite impellers. Manufacturers of these pumps publish pump curves to be used in verifying the appropriate specification of that pump. These curves generally show head (pressure) capacity on the vertical axis and flow rate on the horizontal axis with a curved line indicating the pump’s rated upper limit at that flow rate and/or pressure. The calculated flow rate and pressure loss found using the methods previously discussed make up the pump’s duty point. The duty point should fall underneath (below and to the left of) the curve plotted on the graph. Figure 4 to the right shows various curves for Grundfos circulator pumps. Both “open” and “closed” system pumps are suitable for solar water heating systems. Closed system pumps are cast iron and open system pumps are bronze, brass, or stainless steel. Note that 1 PSI is equivalent to 2.31 ft. of head.

The duty point should always be comfortably underneath the pump curve to ensure that the pump has more than enough capacity to meet the system requirements. Pump’s should be oversized by at least 20% which allows for the use of speed control or isolation valves to correctly calibrate the flow rate in the system. Having slightly too much pumping power is preferred over not having enough. That being said, pump’s should not be so oversized that the pressure and flow rate are completely outside the range needed for the system in question.

There are many other complex issues involving pump specification that should also be accounted for including pump material, electrical requirements, NPSH, and efficiency. For most systems the methods above can be used to determine sizing for pumps and piping but for more complex systems please contact Solene’s Engineering Department.

Figure 4 - Pump Curves for Various Grundfos Circulator Pumps (120 VAC, 60 Hz)

Figure 4 – Pump Curves for Various Grundfos Circulator Pumps (120 VAC, 60 Hz)