The Basic Principles of Duct Design, Part 1

3 Comments Read/write comments


When it comes to heating and cooling homes, forced air distribution is king. Yeah, my Canadian friend Robert Bean of Healthy Heating pushes radiant for both heating and cooling, and my Texas friend Kristof Irwin drank that koolaid and installed what may be the first radiant cooling system in Texas. Even if radiant distribution systems completely take over, though, we'll still need forced air duct systems. Why? Because we still need to move air for ventilation and, in humid climates like the southeastern US, dehumidification.

So, if we're going to move air through ducts, we need to understand the physics of air and how we make it do our bidding. In this series of articles, I'll take you through these things. Today I'll start with what you do in the HVAC design process before you get to the duct design phase (Manual D) as well as the physics of air flow when it's constrained by ducts. I'll follow that up with articles on the process we use in designing duct systems, including available static pressure, equivalent length, and choosing fittings.


Before duct design

Designing a duct system is important but there are a few critical steps that come first. Number one is the heating and cooling load calculation using a protocol like ACCA's Manual J or the ASHRAE Handbook of Fundamentals. You've got to know how much heating and cooling you need for each room (in BTU/hr). Then those BTU per hour requirements immediately translate to room-by-room air flow requirements in cubic feet per minute (cfm). It's done automatically in the software we use (RightSuite Universal by WrightSoft).

Once you know the BTU/hr and cfm numbers for the building, you need to select the right equipment. ACCA's Manual S protocol helps you do that. There's more to it than just finding a piece of equipment that meets the total heating and cooling loads for the home. You've got to make sure you adjust for the indoor and outdoor design conditions of the home. Ideally, you have the manufacturer's performance data tables to help you get it right.

Then you're ready to start designing the duct system.

The weight of air

The first thing you need to know is that air has weight. David Hill has given a couple of great presentations on duct design at Building Science Summer Camp and this is his starting point. (See Michael Chandler's excellent summary of Hill's 2011 Summer Camp talk on Green Building Advisor.) In the photo below, Hill holds a 1 cubic foot block, which he says would weigh nearly 0.1 pound if it were air. The actual number is 0.0807 lb at standard temperature and pressure.

Air has weight

If you have a 2.5 ton air conditioner, the nominal air flow would be 1,000 cfm. (The rule here is 400 cfm per ton.) That means the blower has to push about 81 pounds of air through the system each minute. It takes work to move weight around.

Well, actually, if you remember your introductory physics class, you know that's not quite true. You can move weight for free if you move it horizontally and without any kind of resistance. It takes work to move it upward against gravity or to push it any direction against friction. And that brings us to...

The physics of air flow

If you take a fan out into your yard on a calm day and turn it on, you'll get its maximum air flow. If you take that same fan and blow the air into a cardboard tube, it has to work against the pressure that builds up in that space. The more you reduce the size of that tube or make it longer or turn the air with it, the more static pressure builds up. And the more the air flow is reduced.

That's the basic principle you have to work with in duct design. I've written previously about the two factors involved in reducing air flow in ducts. One is friction. As the air moves through a duct, it interacts with the surfaces. The smoother that inner surface is, the better it is for air flow. The rougher the surface, the more it slows down the air.

The second factor is turbulence. This generally arises when you move air through fittings, or when you turn the air. With rigid duct, you turn the air with fittings, but unfortunately that's not always the case with flex duct.

Turbulence generated at an elbow in a duct system

When air comes out of the air handler, several things happen to it. It gets sent to the various rooms in the house. As it travels through a trunk-and-branch duct system, the quantity keeps diminishing because some of it gets diverted down each branch on the way to the end.

Each section of duct, each fitting, each turn of the air adds resistance to that air flow because of friction and turbulence. Grilles and registers, filters, and balancing dampers also add resistance. That resistance results in decreases in the static pressure, or pressure drops.

So, we begin at the blower with a high pressure. By the time the air comes out of the supply vents, that pressure has dropped to zero (relative to room pressure). 

The next step in the duct design process

In the next article, I'll talk more about those pressure drops and how they determine the available static pressure, which then leads to the total effective length of our duct system. You can get to the other articles in the series with the links below.


Other articles in the Duct Design series:

Duct Design 2 — Available Static Pressure

Duct Design 3 — Total Effective Length

Duct Design 4 — Calculating Friction Rate

Duct Design 5 — Sizing the Ducts


Related Articles

The 2 Primary Causes of Reduced Air Flow in Ducts

Don't Kill Your Air Flow with This Flex Duct Disease

The Science of Sag - Flex Duct and Air Flow

The Secret to Moving Air Efficiently through Your Duct System - my article on David Hill's 2015 Summer Camp presentation on oval ducts

Image credits: Top photo by Energy Vanguard; weight of air photo by David Hill; turbulence drawing from ACCA's Understanding the Friction Chart (which apparently is no longer available).


NOTE: Comments are moderated. Your comment will not appear below until approved.


May 25 2017 - 6:44pm

The presentation style in this article is very easy to follow. It simplifies complex information for the layperson (or, in my case, the architect). I'm looking forward to seeing the next installment.

May 26 2017 - 10:34am

Not that it matters for your discussions, but I thought the density of air at STP was 0.075#/ft3

May 26 2017 - 10:46am

D'oh! Yes, Lee, you're right. Actually, we're both right, but you're more right. The number I gave in the article (which I'll change in a moment) was 0.0807 pound per cubic foot. That is correct for dry air, but real air does have water vapor in it. Since water vapor is lighter than dry air, it brings down the average density. At 75° F and 50% relative humidity, it's about 0.75 pound per cubic foot, as you say.

Add new comment