ASHRAE Journal - October 2010 - 14

letters
to the installed cost of a drive. The first article in the series on VFDs (in the March 2010 issue) touched on harmonics as well as some other potential pitfalls when applying a drive, but did not treat any of this in depth. The 2008 ASHRAE Handbook chapter on electric motors and controls (HVAC Systems and Equipment, Chapter 44) does discuss the technical side of these issues, but not the cost side. The suggestion for an in-depth ASHRAE Journal article addressing both sides of VFD application (the technical aspects of both the mechanical and electrical sides and the associated economics) is a good one. I just scanned the table of contents of my back issues of the Journal going back to January 2006 and the subject hasn’t been covered in a main article during that time. Considering the benefits that VFDs provide in a range of HVAC&R applications and their increasing acceptance, this is certainly a timely topic. In the coming months, I will see if I can assemble some coauthors to prepare an in-depth article on VFDs. John Dieckmann, Member ASHRAE, Washington, D.C.

The Author Responds
Thanks for your interest in our article. To control and operate the central chiller plants, one of the most important issues is to monitor and control the temperature of chilled water supplied to users. It is not only an issue to ensure the temperature delivered to users (terminals) in the expected range for proper operation performance, but also a contract issue in some cases. For example, in the building discussed in this article, there is a maximum temperature of water supplied to users in the contract document with tenants. If the temperature of water delivered to the users is over this maximum limit, there will be a penalty applied to the building owner. Actually, the method proposed by the reader is not feasible, or at least not a better arrangement in practice (even though there might not be a need to control the outlet water temperature exactly at the secondary side of the heat exchangers in some particular applications). There are four reasons. 1. In terms of convenience and instrumentation cost, it is the same to monitor the supply return water temperature at the primary side and to monitor the supply water temperature at the secondary side. 2. It is true that any change in heat requirement on the secondary side will automatically get reflected in the change in DT at the primary side. And, the supply water temperature at secondary side will be at a reasonable level if the return water temperature at primary side is at a right level, and the cooling load and coefficient of the heat exchangers are at normal level. But, how about the case when the heat transfer coefficient of the heat exchangers is degraded significantly? 3. The energy consumption of pumps at the primary side will be saved at partial load by reducing the pump speed at the primary side to control the supply water temperature at the secondary side at its setpoint. 4. What is controlled is the absolute level of the supply water temperature at the secondary side, not the differential temperature between the primary and secondary sides. It is not necessary to control it at very high accuracy and not difficult to control it providing that the sensors are not significantly biased, while it is obvious and not difficult to detect the bias if it is significant. The reader is correct that operators need to pay attention to the issue of sensor bias and make sure there is no significant bias at the supply water temperature sensor at the secondary side, or at least the biases of the water temperature sensors at both sides are similar. In summary, in terms of instrumentation, the presented arrangement is as simple as the arrangement proposed by the reader. The presented instrumentation and control can make sure that the supply water temperature to users is at the desired range when the cooling load and heat transfer coefficient of the heat exchanger change. There is no difficulty faced in site application as observed when implementing the control based on the instrumentation presented in the article. Shengwei Wang, Ph.D., C.Eng., Hong Kong
ashrae.org October 2010

Variable Speed Pumps

In July’s “Control Strategies for Variable Speed Pumps in Super High-Rise Building” by Shengwei Wang, Ph.D., C.Eng., and Zhenjun Ma, Ph.D., the solution suggested by the authors is too complicated. The heat transfer capacity, H, of a heat exchanger is defined as: H = k × Q × DT where Q is the fluid flow rate, DT is the temperature differential across the inlet and outlet of the heat exchanger, and k is a constant based on physical parameters of the heat exchanger and the fluid. For a defined heat exchanger with k being constant, the equation could be modified to: H µ Q × DT This means heat transfer capacity could be controlled either by varying DT or Q and keeping the other constant. In this particular application, the easiest solution is to measure DT across the primary side of the heat exchanger. There is no need to measure outlet temperature on the secondary side of the heat exchanger for two reasons: 1. The design inlet temperature on the primary side is 5.5°C (41.9°F) and the design outlet temperature on the secondary side is 6.3°C (43.3°F), a difference of only 0.8°C (1.4°F). Even under minimum load conditions, the drop in outlet temperature on the secondary side cannot be much and could be difficult to measure and control. 2. Any change in the heat requirement on the secondary side automatically will get reflected in the change in DT on the primary side. The fluid flow could be varied to keep this DT constant. This system will work even if the chiller outlet temperature is changed to save energy. Amrish Chopra, Member ASHRAE, Faridabad, India
14 ASHRAE Journal

ASHRAE Journal - October 2010

Table of Contents for the Digital Edition of ASHRAE Journal - October 2010