ASHRAE Journal - July 2009 - 8

Letters continue on Page 10 Spreadsheet Issues Re: Equal Friction Solution for Duct Sizing The Author Responds Vernon Peppers, like many software developers, lives in a fantasy world of “cyber perfection” where the 3-D CAD drawing is exactly reproduced during construction, and the internal load assumptions used to distribute airflow prove accurate to double precision. In practice, ducts must be designed to give the contractor and owner the flexibility to adjust to necessary deviations from the construction documents, as well as from the original assumptions of occupancy and use patterns. Thus, squeezing branch sizes for the unachievable purpose of self-balancing not only will exacerbate noise problems, but will rob the system of flexibility when the system or a branch needs significantly more airflow than originally provided for. VBasic is far from being obsolete, and is a valuable tool for design professionals. “Modern” software programs, regardless of architecture, still must perform iterations using some form of conditional branching, nested loops, and limiting counters. The macros in my spreadsheet, which is not “software,” illustrate these fundamentals for the benefit of those who wish to understand the bedrock principles of iterative computation, or even further develop these skills themselves. Mr. Peppers has confused fitting equivalent length (L/D) with loss coefficient. Loss coefficient is not a function of size, but of airflow and geometry ratios and relevant angles. Whether using Idelchik or the ASHRAE database, and even if rigorously using the applicable ratios, fitting losses are a scientific wild guess because upstream and downstream duct conditions never match the laboratory conditions used to generate the tables. For each type fitting on a job, the variation of loss coefficient within the range of design ratios of an experienced designer’s duct specifications is small enough that a single coefficient can be used with little compromise of overall accuracy. Long before there were duct design computer programs, engineers were successfully designing ducts at all levels of complexity using paper spreadsheets and manual iteration. While it appears that Mr. Peppers has never personally written computer code, he would certainly benefit from occasionally visiting a construction site. Fred W. Dougherty, P.E., Life Member ASHRAE, Panacea, Fla. July 2009 I am disappointed that ASHRAE Journal would publish such a poor article as it did with “Equal Friction Solution for Duct Sizing,” in the May 2009 issue. There are a number of things that I object to in the article. The first is that the code presented in the article, as well as the rest of the code in the macros in the spreadsheet, are done in an obsolete style, one that the programming community left years ago because it produces almost indecipherable “spaghetti” code. All modern code is produced using modular code with control structures instead of many counters and “goto” statements. Not only was the programming bad, but the result is bad, too. It does not reproduce the equal friction method as described in the ASHRAE Handbook—Fundamentals. The method as described in the Handbook calls for an iterative solution to produce a balanced system. This is very easy for the computer to do, but the spreadsheet and its macros do not do that. More incorrect is the improper way of providing fitting coefficients. The reference to Idelchik should have noted that the data in The Handbook of Hydraulic Resistance is suspect, and that the ASHRAE Duct Fitting Database should be used instead. The Society has spent a lot of money over the years testing data for this database and ensuring that the data is consistently collected and reduced. Fitting coefficients vary with duct size for the most part, as even a cursory inspection of the data will show. Selecting a coefficient, and then using it regardless of size is just wrong. The limit of four fittings per branch is also artificially low. Then, there is the issue of the 8 ASHRAE Journal mention of the sizing being based on the availability of Snap-Lock pipe. Snap-Lock pipe does not meet ASHRAE sealing standards, due to its leaky longitudinal seam. The article implies that this is the standard to which we should design. The most offensive part of the article is the idea that it creates that small systems don’t need to be fully engineered, that a simple spreadsheet should replace a more rigorous calculation by a serious duct-design program. All ductwork needs to be designed, not just sized. In this day of high energy costs and push for low-energy buildings, just sizing ducts according to some arbitrary friction standard is not adequate. Ducts need to be designed according to the optimum system that can be produced. This spreadsheet is purported to be easier to use than design software, but I find that it needs just as much input as our own product, currently under review by TC 5.2, without the ease of use, nor the extensive output. This spreadsheet, and the article about it, are beneath the Journal. Vernon Peppers, Member ASHRAE, Atlanta ashrae.org

ASHRAE Journal - July 2009

Table of Contents for the Digital Edition of ASHRAE Journal - July 2009