Not a rule of thumb — a number you compute for the specific system. ACCA Manual D §1-6 gives it as FR = PD × 100 / TEL: take the available static pressure, meaning the blower's external static at the design airflow after the coil, filter, supply outlet, return grille and balancing damper have each taken their share, and divide it across the total effective length of the longest airflow path. Manual D Appendix 4 then states that any resulting value between 0.06 and 0.18 in.wg per 100 ft is acceptable, because that range is a reasonable compromise between airway size and available blower power. Land below 0.06 and the ducts come out too large for the building, which is Manual D's signal that the blower is undersized for that much duct. Land above 0.18 and there is more pressure than the run needs, the airways come out small and fast, and the published fitting equivalent lengths that produced your effective length stop being valid, because they are tabulated at 900 fpm and a 0.08 rate.

About one standard size on a branch, and it is not a fudge factor — it falls out of the roughness. ASHRAE Fundamentals Table 1 classes galvanized steel with longitudinal seams as medium smooth at 0.09 mm absolute roughness, and flexible duct of all types of fabric and wire as rough at 3.0 mm when fully extended. Held at 600 CFM and 0.10 in.wg per 100 ft, that difference is 11.27 in. of unrounded diameter in metal against 12.78 in. in flex, so 12 in. of pipe against 13 in. of flex once both snap up. Watch for calculators that put flex at 0.003 ft, which is 0.9 mm: that is ASHRAE's value for rigid fibrous glass duct board, a different row of the same table, and using it understates flex friction by roughly 30%. The check is ACCA Manual D Figure A10-2, which publishes 0.120 IWC per 100 ft for 1,000 CFM of round flex at 1.25 ft² — 3.0 mm reproduces that at 0.112 and 0.9 mm does not, landing at 0.082.

Through the circular equivalent, which is ASHRAE Equation 25, developed by Huebscher in 1948: De = 1.30 (ab)^0.625 / (a + b)^0.25, with both sides in the same units. The result is the diameter of the round duct that produces the same friction loss at the same airflow over the same length, and the trap is that this is not the diameter with the same cross-sectional area. A 20 × 8 in. rectangle has 160 in² of area but a circular equivalent of 13.48 in., which is 143 in² — the round duct does the same job with less section, because a circle has the least perimeter rubbing against the air for the area it encloses. That penalty grows with how flat the rectangle is. Velocity is the exception: it is always computed on the duct's real area, never on the equivalent circle, so a 20 × 8 carrying 800 CFM runs at 720 fpm and the equivalent-diameter figure is used for the friction number only.

It depends on which section it is, and ACCA Manual D Table A1-1 publishes both a conservative figure and a maximum for each, treating them as limits on noise rather than on physics. For a supply trunk it is 700 fpm conservative and 900 maximum in either rigid or flex. A rigid supply branch is 600 and 900; flex allows 700 conservative because it absorbs sound. Return ducts are quieter throughout: a return trunk is 600 and 700, a rigid return branch 500 and 700. Return grilles sit in living space, which is why they are held tightest. There is a second reason not to exceed 900 fpm anywhere: Manual D's own fitting equivalent lengths in Appendix 3 are tabulated for 900 fpm or less, so a duct running faster than that invalidates the total effective length used to compute the friction rate in the first place.

Usually because the slide rule is more conservative than the equations, and by a knowable amount. The ACCA Duct Sizing Slide Rule that Manual D's worked examples are read from publishes 0.066 IWC per 100 ft where Darcy-Weisbach with ASHRAE's medium-smooth roughness computes 0.060 for 1,000 CFM of round sheet metal, and 0.10 at 14 in. where the equations give 0.089 — a spread of 10 to 13%, always in the same direction. That is not an error on either side; the slide rule is modelling installed duct with more joint irregularity, closer to ASHRAE's Average 0.15 mm category than the Medium Smooth 0.09 mm the ASHRAE friction chart itself is drawn for. Abushakra, Walker and Sherman measured the same bias independently and found the ACCA chart over-predicted pressure drop for fully-stretched duct of every size they tested by an average of 21%. The practical consequence: a section that lands within one size of its limit here should be taken up one.

No, deliberately, and that is worth understanding rather than working around. Compressed flex is dramatically worse than extended flex, but the two published models of how much worse disagree with each other. ASHRAE Fundamentals Figure 8 gives a pressure-drop correction factor of roughly 1 + 9.86 × the compression ratio, independent of duct size. Abushakra, Walker and Sherman, testing 6, 8 and 10 in. duct to ASHRAE Standard 120-1999, measured 1 + 25.4, 1 + 21.6 and 1 + 16.2 times the compression ratio respectively, and concluded that a size-independent factor cannot be correct. At 15% compression, which that paper describes as a normal field installation, ASHRAE predicts about 2.5 times the pressure drop and the measurements give about 4.2 times on 8 in. duct. Those differ by 1.7 times, so applying either one would mean publishing a number we cannot stand behind. Everything here is flex pulled fully extended, every flex result says so, and the honest field instruction is to pull the duct tight rather than to correct for not having.

Sea level, 14.696 psia, 70°F air at 0.075 lbm/ft³, and it is stated on the schedule itself. ASHRAE notes that the standard friction chart needs no correction for elevations up to about 1,600 ft, temperature swings of roughly ±27°F, or duct pressures within ±20 in.wg of ambient, because each of those keeps the answer inside 5%. Higher than that, the air is less dense, the Reynolds number falls and the real friction loss is below what a sea-level chart returns, so the error is in the safe direction for sizing but it is still an error. An elevation correction was not shipped here because only one publication was found that tabulates air density by elevation for duct sizing, and one source does not clear the two-source bar this tool holds itself to. At altitude, size from a chart corrected for your local density, or accept that the duct is slightly larger than it strictly needs to be.