Why are variable frequency drives used for secondary chilled water pumps?

Answer 1

In a chilled water system, the chillers are happiest with a constant-volume flow. In a primary-secondary loop system, the chiller has a primary pump that is constant-speed. If that were the only pump, and it circulated chilled water through the chiller and the load (air handler, etc.), imagine what would happen as the temperature setpoint was reached. The control valve starts to close, reducing the flow. This deadheads the pump, which is bad in itself, but it also reduces flow through the chiller. The chiller could be damaged, or it would trip on a low flow safety, or at the very least it operates poorly, or inefficiently. You could change the control valve to a three-way diverting valve. Now instead of closing, upon reaching setpoint the valve diverts some of the flow, bypassing the load, and recirculates it back to the chiller. This maintains full flow through the system, but it wastes energy because you are always pumping full-load flow even in the dead of winter when you may only need a fraction of the system's capacity. What if there was a way to maintain a constant flow through the chiller, yet only supply just the needed flow to the building? Well, let's change the control valve(s) back to standard 2-way, and add a secondary pump with VFD. We will also need a pressure sensor on the secondary loop. We add a bypass pipe between the two pumps. Can you visualize the piping? Water leaves the primary pump, goes through the chiller, heads out on the CHW supply line. Then it can either circulate through the bypass line and back to the pump (that's the primary loop), or keep on going out to the building, then back from the building, through the secondary pump, past the bypass line, and to the suction on the primary pump. The beauty of the primary-secondary system is we can reduce secondary flow to the building as the cooling load reduces. As the control valve(s) close, the pressure in the secondary loop rises. The pressure sensor detects this and slows the pump down to maintain a constant pressure. As less and less water flows from the chiller out to the building, more and more of the primary flow goes through the bypass pipe, maintaining constant GPM through the chiller. You save money several ways. It takes horsepower ($) to overcome pumping friction losses as you move all that water through the hundreds or even thousands of feet of building piping. Reducing the secondary flow saves energy. Two-way valves are cheaper, less complex, and easier (cheaper) to install. The chiller always operates at design flow, where it is more efficient. So, the VF drive is needed to allow the control system to vary the secondary pump GPM to maintain constant secondary loop pressure as the load requirements change.

Answer 2

In this configuration, the building air handlers and/or unitary systems (VAV and CV boxes) have two-way control valves. As the need for cooling decreases, the valves modulate closed. As the valves close, the chilled water pressure rises. The pump(s) must slow down to maintain a constant pressure.

The problem is that almost all chillers require a constant water flow to work properly, so the pump must be constant-flow.

To get around this, we implement a primary-secondary pumping scheme. There is a primary, constant-volume pump that circulates water through the chiler and a secondary pump that circulates water through the building. There is a bypass pipe between them.

So, let's say the primary pump is sized for 100 gpm through the chiller, and at the moment the building requires 100 gpm also. Here is the flow circuit - the primary pump sends 100 gpm into the chiller. The 100 gpm coming out of the chiller flows into the secondary pump, then on to the building. The 100 gpm returning from the building feeds back into the primary pump, completing the circuit.

Now the weather has cooled, and the building control valves have closed down a little. The building now needs only 80 gpm to maintain design pressure. Here is the flow circuit - The primary pump sends 100 gpm into the chiller. The secondary pump has slowed down such that it only sends 80 gpm to the building, so 20 gpm flows through the bypass pipe and directly back to the primary pump. The 80 gpm returning from the building mixes with the 20 gpm coming out of the bypass pipe so that once again there is 100 gpm flowing into the primary pump, completing the circuit. The same thing occurs if the cooling load increases, except the building flow is now greater than the primary flow, with the resulting bypass flow in the opposite direction.

So, the chiller sees the constant flow it needs to work, and the building receives variable flow consistent with the required cooling load. The excess (or deficit) flow goes through the bypass pipe.

Answer 3

Primary pumps establish the most economical flow however in high demand the secondary pump kicks in to move more liquid flow for higher BTU capacity.