Thursday, December 8, 2011

Heating in Residential Applications

Now that winter is upon us, we are getting the “How do I heat with a high ceiling?” questions. High ceilings in residences, especially here in Texas, are often the rule, not the exception. Heating is always a problem, especially from a 24 ft ceiling often found in Texas houses. Hot air rises and cold air falls, and with fixed air outlets and high ceilings this causes conflicts. In commercial spaces, I often recommend strategies not typically available in residences. These include using a VAV box to shut off interior located diffusers thus increasing the air flow from perimeter outlets, in heating.

Generally, residential diffusers are fixed deflection grilles. The use of an adjustable curved blade diffuser, such as the Krueger 180 (, is a much better solution, and should always be adjusted to direct the air horizontally, along the ceiling. Unfortunately, occupants are seldom aware of this, and often set the blades to blow supply air down into the space. Assuming the outlet is 24’ from the floor, this isn’t likely a problem if it is set up properly. Horizontally supplied air will have a much higher mixing ratio, result in more uniform room air temperatures, and will avoid drafts in summer. With high discharge temperatures in heating, however, there will likely still be stratification.

The most practical solution in residences seems to be the use of ceiling fans. While it depends on the room configuration, setting the fan to blow up in heating, while it would seem to be counter-productive, usually causes air to flow down the walls, and takes the warm air at the ceiling with it, and minimizes air motion in the room, which is usually objectionable when in heating mode. Chairs, of course, are best placed a short distance away from the walls. If a variable speed fan is employed in the central system, discharge temperatures in heating should be kept as low as possible, by running the fan at a high speed when heating from the ceiling, to minimize stratification.

In cooling, the ceiling fan is set to blow down, increasing room air motion (at least away from the walls) and this is often desirable, and may allow a slightly higher thermostat setting in the summer.

Authored by: Dan Int-Hout, Chief Engineer Krueger

Thursday, December 1, 2011

The History of Krueger and Low Air Flow Measurement

When I (Dan Int-Hout) started working in this industry, in 1973, the standard of room airflow measurement was the “Anemotherm”, a hand held device specified for use in determining both diffuser throw and room air movement by the Air Diffusion Council. This device was a battery powered hot wire anemometer. In trying to meet a badly conceived minimum room air motion specification established by the GSA as part of a “Peach Book” performance specification, it was discovered that the Anemotherm was incredibly unreliable at air speeds below 100 fpm. The GSA spec was 20 fpm, minimum, at any point in the occupied zone! Later testing would determine that this was an unreachable goal with any air distribution system.

When I took over the Krueger Air Lab in 1982, we immediately embarked on a quest to get an accurate means to determine room air speeds. In my previous life at the research facility of Owens Corning Fiberglass in Ohio, in conjunction with TSI of Minneapolis, we had developed a complex, mutually perpendicular, twin-element anemometer requiring a specialized data acquisition system and sophisticated software to resolve low air flows from almost any direction.

TSI had developed a much more accurate portable hot wire anemometer somewhere around 1980. It was directionally sensitive, but accurate at 5 fpm. It was able to measure unidirectional airflows with excellent accuracy. It required significant time to properly record the highly variable low air speeds at the end of a diffusers throw. Room air motion, however, requires a probe less sensitive to air direction. The TSI linear probe was impractical in this area.

Working again with TSI, we came up with the idea of using a small heated sphere, instead of a linear element, and after solving the temperature compensation problems, TSI released the first omni-directional anemometer. It was accurate down to 5 fpm from almost any direction, but the output was a complex 4th order polynomial curve, different for every probe. At Krueger, we developed the first operational software for this new omni probe, utilizing the recently released MS-DOS operating system and the universal PC platform. We used a prototype data acquisition board from Burr Brown of Tucson, and a TSI provided Compaq “luggable” computer to develop effective software which stored calibration data and resolved the complex curves. TSI showed it at an ASHRAE show in 1984. The software development was finally spun off as a private operation, when Krueger management at the time (Leo Krueger) decided he didn’t want to invest in software development. The anemometers in the lab today are a direct descendent of that development and have become the industry standard. The software, however, went away in a few years as more advanced data acquisition systems (such as Lab View) took over the tasks. (Leo was correct!)

Prior to the development of the Omni probe, measurement of low air speeds was incredibly time consuming and both inaccurate and non-repeatable. After we started using the new equipment, however, we discovered that measured throw data for air outlets was typically much longer than previously reported (and could be completed in much less time). The result was a complete retesting of throw values with now much more sensitive (and accurate) devices. Manufacturers dropped Air Diffusion Council membership around that time, in favor of ARI; ASHRAE Standards were updated to require better instrumentation. The result is much longer reported throws for all devices, especially linear slots, which tend to have very long throws, across the industry.

Data acquired under the older ADC test code should be considered obsolete and will likely under report 50 fpm throw values. The new omni-probes, which are now internally linearized (making data acquisition very easy), allow for rapid measurement of the highly turbulent room air speeds and low diffuser throw air speed values with data being fed directly into spreadsheets for rapid analysis.

Authored by: Dan Int-Hout, Chief Engineer Krueger