Thursday, May 30, 2013

Sam Luxton- You Will Be Missed

We announce with great sadness that Professor Emeritus Russell (Sam) Luxton, co-founder of Dadanco, passed away on Friday 24th May, in Adelaide, South Australia.

 As mentor to Vladimir Petrovic, CEO Dadanco, Sam's inspiration and motivation was critical in the early development of the company and the nozzle technology successfully used in buildings throughout the world.

Although we are deeply saddened by the irreparable loss, "His influence lives on within every solution we provide".

Tuesday, May 28, 2013

The VAV versus Active Chilled Beams + DOAS Debate, May 2013 Issue of the ASHRAE Journal

The VAV Reheat Versus Active Chilled Beams & DOAS paper published in the May 2013 issue of the ASHRAE Journal covers an HVAC system design competition for a real office building at UC in Davis, California. The systems under evaluation are Active Chilled Beams with Dedicated Outdoor Air (ACB+DOAS), a Variable Air Volume system with Reheat (VAVR) and a hybrid of VAVR and ACB’s.
Energy and cost modeling was performed for the three systems and the outcomes were published in the article.

Unusual Design Parameters
The reader does not have to venture far into the paper to expose the reasons for the high first and operating costs of the ACB+DOAS system.
It is interesting that the authors have chosen the primary air flow rate of 0.6 CFM/ft2 for this analysis.  Such a high primary airflow rate is not typical for a building of this type.  The high primary airflow rate is the root cause on the outcomes of this paper.  
Typical chilled beam system designs are based on primary air system reduction of 65-70% compared to VAVR.  This is one of the reasons that chilled beams have been proven to save energy by transferring cooling capacity from the air side to the more efficient water side system. 
It is therefore unclear why in this paper the authors have utilized 0.6 CFM/ft2 (ACB+DOAS) that can clearly be designed with approximately 0.2 CFM/ft2, suggesting a misapplication of the ACB design.

Comparison Using Typical ACB Design Parameters
The typical floor plan show of the subject building comprises and open plan office space. Perimeter diffusers were used for the VAVR system and a (12) 8ft long ACB’s to depict the designers Chilled Beam configuration as shown below.   There are some considerable errors in the ACB design layout and sizing.  Had this system been designed with typical ACB design parameters, only (4) 8ft long ACB’s @ 85 CFM are required to meet the load and provide adequate air distribution to the open plan office, as opposed to (12) 8ft long ACB’s.  This results in a total primary airflow rate of 340 CFM (0.2CFM/ft2) for the zone as opposed to 1,330 CFM (0.8 CFM/ft2) that the VAVR system requires at peak load. This 75% reduction in primary air flow rate is more consistent with properly applied well designed ACB system.  To further reduce energy consumption, application of VAV units to the primary air can lower the primary airflow during part load conditions.

The paper claims that part load for this office building is  40% of the peak load, therefore the zone VAVR requirement would be 532 CFM (40% of 1,330 CFM) compared to 340 CFM for a constant volume ACB design, almost 65% more airflow than the ACB primary airflow rate. However, if we consider delivering the primary air to the beams using variable volume, the ACB primary air system can be turned down to the ventilation rate (120 CFM or 20 CFM per person). This again brings primary airflow of the ACB system down to 26% of the part load VAVR design.
Using the occupancy density rate of 275 ft2 per person as shown in the paper, there would be approximately 6 people and a latent load of less than 1,200 BTUH (including infiltration) for the 1,600 ft2 sample zone. This can be accomplished with just 90 CFM of ventilation air with a moisture content of 49.7 gr/lb if the room humidity is allowed to rise to around 54% RH.
So if we now compare apples with apples, we can see that in reality the ACB+DOAS system should have been sized for a maximum primary airflow of 0.22 CFM/ft2 instead of 0.5 CFM/ft2. If the beams are supplied with VAV primary air, the system could turn down to ventilation rate during part load condition which is around 0.08 CFM/ft2.
So why has the ACB+DOAS system been designed with such an unusually large primary air system and why is the design so expensive? These are difficult questions to answer without examining the design documents, but there are some clues in the paper which are discussed below:

The use of 4-pipe chilled beams
For a given primary airflow rate, 4-pipe chilled beams deliver far less cooling and heating capacity than 2-pipe. The result is more chilled beams and/or more primary air is required to satisfy the cooling loads. Modern designs typically use 2-pipe chilled beams and heat the primary air with duct mounted zone coils which significantly reduces pipework costs.

The use of 63°F primary air
Warmer primary air results in more chilled beams and/or higher primary airflow rate for a given cooling load due to the lost cooling contribution from the primary air.
When combined, these factors have a significant impact on first cost and efficiency. The typical floor zone would require around 500 CFM or 0.31 CFM/ft2using 4-pipe beams and 63F primary air, the air could still be turned down with VAV but a significant portion of the ACB benefits are lost using these design parameters.

Low Performance Active Chilled Beams
There is a possibility that the ACB+DOAS design was laid out with low performance European design active chilled beams. Some of the European manufacturers offer compact chilled beams with low density coils that are not suitable for use in some USA buildings. These beams are designed for the European market where sensible loads are generally lower. When used in the USA, this style of beam sometimes requires far more primary air to drive the induction process to achieve the space sensible loads.
So clearly the ACB+DOAS system first costs in the competition design will be higher. The paper claims a 240% premium. This is not typical, most modern chilled beam designs compare favorably with VAV systems although some have a premium of up to 10-15%. Over budget chilled beam designs are usually caused by using 4-pipe beams, oversized primary air, overzealous control design, high chilled beam density or the bidding mechanical contractor’s unfamiliarity with the system. 

Some other issues/omissions were found with the paper:
There is no mention of using energy recovery on the ACB+DOAS.
 “A primary airflow rate of 0.3 cfm/ft 2 is about the lowest possible with an ACB+DOAS system to meet latent loads with the primary air and the sensible loads with the chilled beams.”
Incorrect statement. The latent loads are identical for all 3 systems. The latent loads come from people occupancy and infiltration so they are the same for all systems. If the author is trying to imply that that the humidity must be more tightly controlled in the building with chilled beams then this would only be partly factual, in reality most chilled beam systems are designed to allow the room humidity to drift up to around 55% RH with the chilled water temperature being selected accordingly.
In economizer conditions, the ACB+DOAS design also has higher mechanical cooling loads because it does not have an air economizer while the VAVR design does have an air economizer and thus benefits from economizer free cooling.”
If the VAVR system has an airside economizer can the ACB-DOAS system have one?

ACB systems have a proven track record of safe operation without condensation occurring on the coil.  This can be achieved with minimal controls.  In fact the VAVR system is just as likely if not more likely to have condensation occurring on the 55°F  un-insulated sections of duct or diffuser plenums as they are  exposed to what is likely more humid ceiling plenum air with little air circulation.  The coil in the ACB has an entering chilled water temperature of 57°F and as the water travels through the tubes it continues to pick up heat, increasing the surface temperature of the tubes and fins.  The coil tubes and fins are continually washed with 75°F room air ensuring less chance of condensation occurring on the coil than on un-insulated sections of the primary air system in the VAVR system.

Duct sizing
There should be similar duct sizing velocities for all three options, portraying the ACB only option with less than half the velocity is inaccurately with general duct sizing principles for ACB’s.  The Hybrid system should have been sized with the same duct sizing velocities as the VAVR system to keep a fair comparison.

It’s not VAV vs. ACB
There should not be a battle about which system is the “winner” here in this analysis…. We are firm believers that Chilled Beams do not spell the end of VAV, in fact there are many projects (this building being analyzed being a perfect case) where the two technologies when combined together correctly provide the best energy saving, capital cost and IAQ.   We feel the analysis provided in this paper and the miss application of both the VAVR and ACB system has intentionally skewed the facts of what is both practical and sensible in HVAC design.  We would encourage your feedback and discussion on this topic via your preferred channels.  There is an interesting discussion that is taking place at the following LinkedIn thread:

Holistic HVAC Design, Segundo Services Center, UC Davis, CA.
April 2013 Issue of the ASHRAE Journal

The May 2013 Article comparing ACB+DOAS with VAVR raises even more questions when read alongside an article published in the previous month of the ASHRAE Journal, “Holistic HVAC Design, Segundo Services Center, UC Davis, CA” which covers the benefits of an ACB+DOAS design of a completed building with metered energy results and came second place in the ASHRAE Technology Awards. Some points worth mentioning are below:

· Hybrid VAV and Chilled Beam Design

· 2-Pipe chilled beam used on the perimeter, VAV on the interior

· 100% Outdoor Air (DOAS)

· Primary airflow rate 0.5 CFM ft2 with VAV turndown to ventilation rate

· Heat recovery used on the DOAS

· Demand control ventilation used

· High performance building envelope

· Metered electricity and steam use

· The overall building height was reduced, reducing the construction costs which more than offset the additional first costs of the HVAC system

· Metered electrical energy consumption 38% of 90.1-2004 baseline and significantly lower than the energy model produced for the building

· LEED Gold Building

 By Daniel Harris