Variable Condenser Water Flow

The specific strategy of Variable Condenser Water Flow has been overlooked by design engineers and experts in the field because they felt that the effort to develop this strategy was not worth their time.  If one focuses narrowly on smaller condenser water pump and is not able to evaluate the broader implications, or  the benefits of a pre-packaged design then it is easy to see how the strategy is often overlooked.

The chiller flow rate is established by the manufacturer to meet the maximum demand of the chiller and is also based on the projected performance of a cooling tower.  Since the chiller and cooling tower manufacturers are different, the design criteria becomes somewhat arbitrary.  A typical rule of thumb for chiller cooling is 3 gpm / ton-rated.  This is not a fast rule and sometimes where the cooling tower size is limited the chiller design is re-configured to operate at 2 gpm / ton. 

ASHRAE recommends a minimum condenser water flow velocity of 3.3 ft/sec to maintain turbulent velocity and prevent formation of deposits in the condenser.  This 3.3 ft/sec is well above the low the 6 to 8 ft/sec encountered in today's chiller designs.

Heat exchange calculations show us that the main factor in condensing the hot refrigerant gases is a large surface area requirement for the copper tubes found in most condensers.   The water velocity at the typical design flow rates is a small and nearly negligible factor in the heat transfer equation.   The factor for condensing efficiency in a chiller is the number and size of tubes and not the water flow velocity.

A major chiller manufacturer has published papers on the results of real time chiller operations using various flow velocity regimes for the condenser water.  They were able to operate at very low flow velocities (and high temperature differentials) without effecting the stable operation of the chiller.  These tests are a matter of public record.

When employing variable flow energy savings are realized in several ways:

1.)  Reducing the flow of condenser water as a function of chiller demand not only allows us to obtain direct savings of  pumping energy, it provides a second reason for employing a variable speed drive on the condenser pumps for balancing purposes and additional savings.

2.)  Also, moderately decreasing the flow of condenser water across the cooling tower increases its performance.  Therefore Variable Condenser Water Flow and Cooling Tower Temperature Relief can be combined for greater savings then the savings achieved by each strategy alone.   

3.)  There are many instances where the condenser water pumps are much larger than the theoretical chiller plant would seem to warrant.  These conditions include multiple chillers served by a single condenser circuit with pumps and chillers in a parallel set-up. All of the pumps have to be designed to meet the worse case pressure and flow requirements with all systems operating at the same time.  Add to that the fact that these systems tend to have cooling towers located at some distance from the chillers and the condenser pumps are suddenly nearly as large as the cooling tower fan motors.

The Chiller Plant Optimizer ^TM is packaged controller and instrumentation system that is factory preconfigured with control solutions to implement this strategy in combination with the other strategies that are discussed in detail in other articles.

Condenser Water Temperature Reset

All of the low hanging fruit has not been picked!   If your facility uses water cooled chillers it is unlikely that it is using the optimizing strategy known as Condenser Water Temperature Reset.  For those few sites that employ some form of this strategy, the method of application in use limits the useful range of control and does not take advantage of all of the potential savings.

This strategy simply means that the cold water temperature leaving the cooling tower basin is controlled to a set point where the set point is allowed to decrease with the changing requirements of the chiller.   It is no secret that refrigeration machines using common centrifugal or reciprocating compressors are more efficient at lower condensing temperatures.  The conundrum has always been that lowering the condensing temperature also reduces the chiller's capacity.

The challenge has been to develop a control method that can adequately match the requirements of the chiller without limiting its ability to develop full capacity when ever needed. The use of this technique was relegated to a manual set-back , if used at all until recently.  Several patents  have been recently granted for methods that can reliably achieve provide Condenser Water Temperature Reset on an ongoing basis.

If you want to whet your appetite on the potential savings just consider this.  Some very efficient chillers have a full load rating of 0.6 Kw/ton with entering water temperatures of 85° F.  That same chiller can have a part load rating of 0.3 Kw/ton with 65°F entering water temperature. 

There is a trade-off.  In order to achieve the lower temperatures the cooling tower fan will run longer and harder.  Therefore, it is feasible for the cooling tower fan to chew up nearly half of the energy savings realized by the chiller.

There is a cold water temperature set point that will significantly reduce the fan energy load and still obtain significant savings for the entire plant.  The three technologies address this control solution in different but still effective ways.  They will also apply a strategy known here as Cooling Tower Temperature Relief, to limit the cooling tower fan energy use. 

Cooling Tower Temperature Relief

Cooling Tower Temperature Relief is a difficult strategy to explain and prove.  But we will use an analogy and does not rely on detail and difficult to follow spread sheets and calculations. Just know that the developer of the Chiller Plant Optimizer ^TM has developed a method of calculating these savings which can be laid out in a spread sheet format.

Consider a single chiller system of a given size operating at part load, say 75%.  Then compare it to a smaller chiller system operating at the same load.   Say the first example is a 1200 ton chiller (system) operating at 900 tons.  And the second example is a 1000 ton plant operating at 900 tons.  Both systems are then operating with the same outside conditions.  

If the smaller chiller system was designed using the same criteria as the larger system, it will have a smaller cooling tower with a smaller fan motor.  The fan energy use by the smaller system will be less then the larger system.

Now, since it is impractical to switch to a smaller cooling tower or smaller fan motor.  Cooling Tower Temperature Relief is simply a control method that insures a similar action by limiting the maximum speed of the cooling tower fan motor.  

The standard control method for cooling towers uses a temperature sensor for the cold water (basin temperature) and controls the fan to achieve this temperature.  A control method that provides for Cooling Tower Temperature Relief requires additional information then cold water temperature.  The control may reset the cold water temperature set point, or use some other control strategy, that results in a different and possibly slightly higher cold water temperature then the primary (or full load design) cold water temperature.

As an independent strategy the savings are limited for Cooling Tower Temperature Relief, and must be applied very judiciously.  But as a strategy combined with Condenser Water Temperature Reset it is a very effective tool for energy optimization.

Multiple Chiller Optimizing Strategies

Multiply chiller management can not be applied across the board as a simple strategy. The different types of chillers as well as their size drives the decision matrix to determine the operational sequence and combinations.

Therefore  we look at the  “art” of multiple chiller management strategies.  More then any other strategy, this one can backfire and it has lead to unnecessary and costly equipment installations that in the end did not lead to energy savings, and is one reason that this strategy that must be carefully considered and then reconsidered, when or if new chiller purchases are involved.

The available savings from a complicated management strategy is  limited.  Also, if the strategies of  Variable Condenser Water Flow, Condenser Water Temperature Reset, and Cooling Tower Relief have already been applied then additional savings from other strategies are limited.

With the Chiller Plant Optimizer ^TM an optional method of control can be added for each chiller so that a strategy tailored for the chiller plant can be employed independently of the existing chiller plant of building automation control.

Chilled Water Reset as an Optimizing Strategy

 The strategy of  Chilled Water Temperature Reset, can not be applied across the board and is considered an optional strategy that can be applied for simple environmental situations where personnel comfort is the only consideration. 

Chilled water flow of 1000 gpm  that leaves the condenser barrel at 45° F. and returns at 50° F. is 416 tons of refrigeration.   That same flow rate of 1000 gpm leaving the condenser barrel at 50° F. and returning at 55° F. is still 416 tons of refrigeration.  The difference is that under the second condition the chiller produced the same refrigeration with much less work or with less energy.

Reason tells us that we can save energy by raising the leaving chilled water temperature set point.  But we are also aware that the various systems through-out the facility were designed for a selected cold water temperature which determined equipment size such as fans and coils.  Therefore, any client operation requiring chilled water, operating at its full design load will be adversely affected by a high chilled water temperature.  Also any operation with special humidity control requirements will be adversely affected by a higher then design chilled water temperature.

Another effect of raising the chilled water temperature will mean that distribution pumps and fan coil units may be required to operate longer and harder when the chilled water temperature is above the design temperature. 

Due to the above difficulties above the application of this strategy it is arguable that application should be limited.  Perhaps employed where the primary and important loads are  simple and can handle some flexibility, and where the effect of  outside conditions and  operational loads are clearly understand and are cyclical such as an office building that is lightly loaded at night and on weekends.  

Chilled water reset when used for the systems described in the previous paragraph can be easy to implement and provide some nice savings.  Some control methods use a load based system to simply implement chilled water reset.  The good thing about such a system is that it is easy to easy to configure, and or reconfigure by the operator to achieve an optimal compromise between comfort and energy savings.  A some what more sophisticated control might use a method that not only considers the chillers operation but also the effect of outside conditions on the chiller and thus a more reliable way to adjust to the buildings overall requirements.  This is still a system that can still be easily configured or reconfigured by the operator.

 The savings are not as dramatic as the other strategies,  but still worth considering if conditions allow.  It can also be combined with the strategies of Cooling Tower Temperature Relief, Condenser Water Temperature Reset, and Variable Condenser Water Flow as an optional addition to the Chiller Plant Optimizer ^TM  providing a boost to the overall savings and payback,  and supporting our goal for a 25% reduction in overall energy use.

Performance Monitoring for Chiller Plants

The HVAC industry has been loath to provide and install meaningful tools to monitor the performance of chiller plants in a meaningful manner. When it does the results are so limited that the their usefulness is problematic.  We propose a systematic and methodical method that can be used to compare results with other chiller plants providing similar service and even with some not so similar.

1. Measure the output of the chiller plant, don’t make assumptions from chiller data.  That means one measures the actual BTU’s of cooling that is produced, preferably as near the chillers as possible.
2. Measure real power consumed, don’t use amperage and assume that definitive results can be calculated. Real power means Kw including power factor, not KVA.
3. Only include the chiller, condenser pumps, and cooling towers.  In this manner the long term results will be more meaningful from year to year and sensible comparisons can be made for plant to plant.
4. Provide an executive report on a periodic basis that is easy to read and can be analyzed by management in 60 seconds or less.
   1. This means recording data on a continuous basis so that true averages for various operating periods, night vs day and weekend vs week day can be reported, if necessary.
   2. Produce a single meaningful number. That is say it in Kw/Ton. 

(Step four may be the most challenging, because people will have to be trained to accept the Kw/Ton as a measuring tool.   Also the temptation to include the recirculation and distribution pumps may be overwhelming. Think of the distribution side as the customer’s responsibility not part of the producer’s responsibility.  Many chiller plants look similar, until one includes the distribution of the chilled water. )