What Friction Rate to Use for Flex Duct in Manual D

Use the same friction rate for flex that you'd use for sheet metal: available static pressure x 100 divided by total effective length, and ACCA wants the result between 0.06 and 0.18 in. wc per 100 ft. Flex only changes the duct size you read at that rate. At the 0.105 in the example below, a 150 CFM branch needs 8 inches in flex against 7 in galvanized.

The confusion comes from a supply-store guide that hands out 0.10 or 0.20 as the flex number. Below is a worked 1,200 CFM example sized with the ASHRAE friction equations, the same math our free duct sizing calculator runs.

What friction rate should you use for flex duct in Manual D?

The one you calculate from the blower. Take the external static pressure at design airflow, subtract the coil, filter, grilles, registers and dampers to get available static pressure, multiply by 100 and divide by total effective length. The material does not appear in that formula. If the result lands between 0.06 and 0.18 in. wc per 100 ft, size the flex runs to it.

The friction rate is a pressure budget for the whole system. It comes from the blower and the longest path the air travels, so a flex branch and a metal trunk on the same system are sized to one number, and flex needs a larger diameter to stay inside it.

Luis Escobar at ACCA gives the target range:

ACCA HVAC Blog § Definitive Design, ACCA's Manual D Revised

"You want the friction rate to fall within 'the wedge' (a friction rate between 0.06 and 0.18 IWC) in order to avoid inadequate fan performance." - Luis Escobar, ACCA HVAC Blog

How do you get ASP and TEL for a flex system?

Read external static pressure at design CFM from the blower table and subtract every component loss to get available static pressure. Add the longest supply run, the longest return run and their fitting equivalent lengths to get total effective length. On flex systems the fitting lengths are where the budget goes, because every bend in a flex run counts as extra feet.

Manual J vs Manual S vs Manual D covers where these steps sit. On flex, equivalent length is the part to watch.

1. External static pressure at design CFM, from the blower table.
2. Minus coil, filter, registers, grilles and balancing dampers, from their submittals, gives ASP.
3. Longest supply plus longest return, each with fitting equivalent lengths, gives TEL.
4. Friction rate = ASP x 100 / TEL.

The Air Diffusion Council's Flexible Duct Performance and Installation Standards (5th edition) lists a 90 degree flex bend at 20 equivalent feet, a 45 degree bend at 10, and a 180 degree offset at 40 (ADC standard, PDF). Two lazy 90s on a flex branch add 40 ft to the path.

Low numbers happen. In Matthew Bruner's Speedsheet walkthrough, a ducted mini split with 0.11 in. wc available over 230 ft comes out under 0.05, below the wedge (HVAC School).

Worked example with 1,200 CFM of flex at 0.21 in. wc ASP

A 3-ton system with 0.60 in. wc external static and 0.39 of component losses has 0.21 available. With 200 ft of total effective length the friction rate is 0.105 in. wc per 100 ft, inside the wedge. At that rate a 150 CFM flex branch needs 8 inches and galvanized needs 7. The 1,200 CFM trunk comes out 16 inches in both, because galvanized needs 14.48 and the round sizes used here step from 14 to 16.

These are example numbers, so pull yours from the blower table and submittals.

The ASHRAE Handbook of Fundamentals (2017, Chapter 21) calculates duct friction with the Darcy equation and the Colebrook friction factor, and its friction chart plots those equations for galvanized duct at an absolute roughness of 0.0003 ft and standard air. Table 1 puts "flexible duct, nonmetallic and wire, fully extended" in the medium rough category at 0.003 ft, ten times the galvanized chart value (ASHRAE Handbook). Solving those equations for the smallest round duct that stays under 0.105:

At equal diameter and 150 CFM, stretched flex at ASHRAE's 0.003 ft design roughness loses about 1.3 times what galvanized does in the 8 inch size and 1.4 times in 6 inch.

The math checks against published numbers. For ASHRAE's own Example 6, 1,000 CFM through 6 ft of 10 inch duct with the roughness values from Table 2 (0.0004 ft spiral, 0.003 ft flex), it gives 0.029 in. wc for spiral and 0.043 for stretched flex, the same values the Handbook's Table 2 lists.

What does 15% or 30% compression do to flex duct friction?

It multiplies it. The ADC standard puts 15% longitudinal compression at about 2 times the pressure drop of fully extended flex, 30% at about 4 times, and 45% at 6 to 8 times. ASHRAE's Example 6 for a 10 inch duct uses correction factors of 1.7 at 4%, 3.5 at 15% and 5.9 at 30%. Lawrence Berkeley lab tests behind them found smaller ducts hit harder, about 4.2 times at 15% for 8 inch.

The ADC standard follows those multipliers with "Ducts should be fully extended before and during installation" (ADC standard, PDF). Allison Bailes put the same point on ACCA's blog:

ACCA HVAC Blog § Flex Duct Can Work Well If It's Installed Properly

"If a flex duct has even 15% longitudinal compression, the friction rate doubles. With 30% longitudinal compression, the friction rate quadruples." - Allison Bailes, PhD, ACCA HVAC Blog

ASHRAE is harsher. Its current pressure drop correction factor comes from the Abushakra (LBNL) and Culp work, changes with duct diameter and is built on flex at 0.003 ft roughness, and Table 2 shows 10 inch flex at 15% compression losing 0.148 in. wc over 6 ft against 0.043 stretched (ASHRAE Handbook). Abushakra, Walker and Sherman wrote that moderate compression "could increase the pressure drop by a factor of four, while further compression could increase the pressure drop by factors close to ten" (OSTI). In the Texas A&M data Bailes summarized, a 6 inch flex duct compressed just 4% moved about 70 cfm at 0.1 in. wc (Energy Vanguard).

Here's the 8 inch, 150 CFM branch from the example against its 0.105 budget. LBNL writes the factor as 1 + a x compression, with a at 25.35 for 6 inch, 21.61 for 8 and 16.18 for 10. At 10 inch it comes within about 3% of ASHRAE's 1.7, 3.5 and 5.9. The lab compressed 8 inch duct only to 24%, so the 30% row extends its fit.

The budget binds on the longest path, and shorter branches have slack. Each row compares loss per 100 ft with the design rate, so the path overrun depends on how much of it is compressed flex. At 15% the branch exceeds that rate under either source, and at 30% it runs at two to four times that rate.

Why is bumping flex to 0.20 in. wc backwards?

A higher design friction rate gives you a smaller duct. One supply-store guide says to "bump that friction rate up to 0.20" for compressed flex. At 0.20 the 150 CFM flex branch sizes to 7 inches. That duct loses 0.112 per 100 ft fully stretched, already 6% over a 0.105 budget, and 0.22 at 15% compression by the ADC rule.

Raise the design friction rate and the equations hand you smaller diameters, exactly what a compressed run can't afford. The guide gives no source for 0.10 or 0.20 (plumbingsupplyandmore.com).

Run it on the example. At 0.20 the flex branch needs 6.26 inches, so 7, and galvanized needs 5.83, so 6. At 15% compression the 7 inch flex loses 0.22, about 2.1 times the real budget. The 8 inch branch sized at 0.105 ran 0.11 in the same condition, half as much. And the blower didn't get any stronger because someone typed 0.20.

The flex chart in older Manual D had its own error. The LBNL team found "the ACCA chart overpredicted the pressure drop for the fully stretched duct of all sizes tested by an average of 21%." With the 2001 ASHRAE correction factors applied, one curve for every size, the prediction came in 17% low at around 15% compression and 24% low at around 30%. The team concluded the factor "cannot be independent of the duct size" (OSTI), which is why the current Handbook factor includes diameter. That chart was the 1995 edition. Bailes notes Manual D now carries the compression data in Appendix 17 (Energy Vanguard).

Should you design flex runs one size up from the Manual D rate?

On the longest path, yes, even with a careful crew. The 8 inch branch sized at 0.105 reaches that design rate at 4% compression on the LBNL factor, and LBNL found fully stretched flex rarely in houses. Shorter branches have slack and can stay at the calculated size.

One way to add margin on the longest path is to size those branches at half the rate, a rule of thumb from this article that neither ACCA nor ADC publishes. In the example that's 9 inches instead of 8, which holds 15% compression by the ADC 2x but runs about 12% over budget by the LBNL estimate. Keep the system friction rate where the blower puts it and change only those branch sizes. Half of 0.105 is 0.0525. Use it only to pick those branch diameters, because as a system rate it sits below the 0.06 wedge floor. Dropping the whole system to it takes the 1,200 CFM galvanized trunk from 16 to 18 inches (14.48 and 16.68 inches before rounding), extra metal on a galvanized trunk that already runs at 61% of budget.

For the branch sized at 0.0525, flex needs 8.11 inches, so 9, and it's still judged against the 0.105 system budget. The 9 inch flex loses 0.031 per 100 ft stretched. At 15% compression the ADC 2x rule puts it at 0.061, or 59% of the budget. Our LBNL estimate uses a factor of 18.9, the midpoint of the 8 and 10 inch values, and puts it at 0.118, which is 12% over. On that estimate 9 inch hits the budget near 13% compression. At 30% it's 0.12 to 0.21, where 8 inch was 0.22 to 0.42. One size up moves the over-budget point from about 4% compression to about 13%, on our estimate. Stretch the flex, shorten the run, or go to 10 inch, which holds 15% at 59% of budget and 30% right at it.

The ADC install spec prevents the compression in the first place. It calls for strapping at least 1-1/2 inch wide at intervals not over 4 ft (6 ft for vertical runs), with sag no more than 1/2 inch per foot of support spacing (ADC standard, PDF). Bailes, on ACCA's blog, passes on ADC's recommendation of supports at the manufacturer's spacing "but no longer than 4' maximum" (ACCA HVAC Blog). Cut the flex to length. An 8 ft run pushed into 6.8 ft of space is 15% compressed.

How does our duct sizing calculator treat flex under Manual D?

It uses the ASHRAE Darcy and Colebrook equations with 0.003 ft roughness for fully stretched flex and 0.0003 ft for galvanized, the same math as the example above. It computes friction rate as available static pressure x 100 / TEL and sizes between 0.05 and 0.15 in. wc per 100 ft. It sizes supply ducts only and doesn't model compression.

The duct sizing calculator is a free whole-system equal-friction estimate for quoting. You enter total design airflow, the measured length of the longest supply run, available static pressure and duct material. Fittings enter as a multiplier on that run length. TEL is that run times the factor, 1.7 by default, and doesn't add the return. Above 0.15 it sizes at 0.15 and shows both rates, so a calculated 0.16 comes back sized at 0.15. Below 0.05 it sizes at 0.05 and shows an alert, because ducts sized there need more pressure than the blower has.

Say the example's 140 ft supply path is 100 ft of duct plus two flex 90s at 20 ft each. Enter 100 ft and 0.21 in. wc, and it builds 170 ft of TEL and a 0.124 friction rate. With 8 branches at 150 CFM it returns 7 inch branches in both materials, so the flex branch lands one size under the 8 inches from the hand TEL of 200 ft. To match your own TEL, set the factor to TEL divided by the run, here 2.0. That gives 200 ft and 0.105, with a 16 inch trunk in both materials and 8 inch flex or 7 inch galvanized branches. A flex result carries a warning that sagging installs may need branches one diameter larger. The page calls it preliminary sizing, less precise than a full Manual D.

Room-by-room Manual D with airflow per room, Manual S equipment selection and the branded PDF are in the paid Pro and Team plans. Whole-house Manual J is free. This is preliminary sizing for a quote. Plan review needs an ACCA-approved calculation, covered in Manual J for permit.

Pull the ESP and component losses from your submittals, then run the trunk and branches in the duct sizing calculator.

Sources

• ASHRAE Handbook, Fundamentals 2017 (I-P), Chapter 21 Duct Design, Equations 18 to 20, Table 1 and Example 6: handbook.ashrae.org
• ACCA HVAC Blog (Luis Escobar), Definitive Design: ACCA's Manual D Revised: hvac-blog.acca.org
• ACCA HVAC Blog (Allison Bailes, PhD), Flex Duct Can Work Well If It's Installed Properly: hvac-blog.acca.org
• Air Diffusion Council, Flexible Duct Performance and Installation Standards, 5th edition (presentation PDF): images.thdstatic.com
• Abushakra, Walker, Sherman (LBNL), Compression Effects on Pressure Loss in Flexible HVAC Ducts: osti.gov
• Energy Vanguard (Allison Bailes, PhD), The Science of Sag: Flex Duct and Air Flow: energyvanguard.com
• HVAC School (Matthew Bruner), Manual D Speedsheet Walkthrough: hvacrschool.com
• Plumbing Supply and More, Flex Duct Friction Rate guide (quoted as the error): plumbingsupplyandmore.com

For estimation purposes only. Not a substitute for a licensed engineer, and not ACCA-approved for permit submission.