What happens when you close an air vent is mostly a pressure change: the blower’s air does not disappear, a little more reaches the open rooms, more leaks out of the ducts, and the blower either moves less air or burns more watts. LBNL found closing registers raised energy use in every case it simulated.

Where does the air go when you close 2, 4 or 6 of 10 vents?
Some goes to the open rooms, some leaks out of the ducts, and some is never moved at all. Iain Walker of Lawrence Berkeley National Laboratory built a duct system with 10 supply registers and one return, closed registers one at a time, and logged blower flow, pressure and leakage in 176 experiments (LBNL-54005), then scaled the results to an 850 cfm house system. We turned those tables into air per open vent.
| Duct case and closing order | Vents closed (of 10) | Blower airflow | Lost through supply leaks | Reaching rooms | Average per open vent |
|---|---|---|---|---|---|
| Tight ducts (about 5% leakage), all open | 0 | 850 cfm | 36 cfm | 814 cfm | 81 cfm |
| Tight, farthest vents closed first | 2 | 825 cfm | 49 cfm | 776 cfm | 97 cfm (+19%) |
| Tight, farthest vents closed first | 4 | 810 cfm | 61 cfm | 749 cfm | 125 cfm (+53%) |
| Tight, farthest vents closed first | 6 | 810 cfm | 90 cfm | 720 cfm | 180 cfm (+121%) |
| Leaky ducts (about 23% leakage), all open | 0 | 850 cfm | 117 cfm | 733 cfm | 73 cfm |
| Leaky, nearest vents closed first | 2 | 817 cfm | 172 cfm | 645 cfm | 81 cfm (+10%) |
| Leaky, nearest vents closed first | 4 | 792 cfm | 223 cfm | 569 cfm | 95 cfm (+29%) |
| Leaky, nearest vents closed first | 6 | 762 cfm | 313 cfm | 449 cfm | 112 cfm (+53%) |
HVAC Brief calculation from LBNL-54005, Tables 7 to 10: reaching rooms = blower cfm times (1 minus measured supply leakage share), divided by open registers for the average. Return leakage is not subtracted.
Two top results get this wrong in opposite directions. One says the furnace “still pushes the same amount of air”; LBNL’s blower flow fell to 762 cfm. Another says closing vents “does not redirect air to other rooms”; with the far vents shut in tight ducts, each open vent averaged 97 cfm instead of 81. In leaky ducts with the two nearest vents shut, rooms lost 88 cfm while leaks gained 55.
Does closing vents in unused rooms save energy on the heating bill?
No, in the only national lab study built to answer it. LBNL simulated an 1,800 sq ft two-story California house with flex ducts in the attic. The smaller heated area cut the load, but bigger duct losses cancelled it:
“In every case the closing of registers led to increased energy consumption, even for the low leakage configuration.” (LBNL-54005, p. 25)
Table 11 puts design-day gas use up by as much as 39% (Sacramento, tight ducts) and never down; tight-duct cases rose more than leaky ones, so sealed ducts alone do not make closing vents pay. The study covers ducts outside the heated space, not basement duct systems whose leaks stay indoors. One top result’s “5 to 10% more energy” names no study.
The DOE’s 2011 Building America HVAC guide agrees on shutting vents in empty rooms: “This is not recommended because it can reduce airflow through the air handler, causing pressure imbalances, putting stress on the duct connections” (PNNL-20421). Motorized registers hit the same physics, which our page on whether smart vents work tests.
Does a PSC, constant-torque ECM or variable-speed blower react differently to a closed vent?
Yes. A PSC blower and many multi-speed ECMs move less air; a constant-airflow ECM holds airflow and draws more power. Scott Pigg and Tom Talerico monitored 31 new Wisconsin furnaces and described both (ACEEE, 2004):
ECM furnaces compensate “for higher static pressure by boosting motor speed (and power draw) to maintain the desired airflow. In contrast, the response of a typical PSC blower to higher static pressure is simply to move less air and draw slightly less power.”
LBNL’s lab agrees: blower power fell from about 570 W to 460 W as registers closed, and LBNL’s 2008 furnace blower comparison found the PSC’s lower airflow at installed pressures cut AC efficiency 10%. So a PSC motor does not “work harder”; it delivers less. The Wisconsin ECM furnaces used a median 82 percent more electricity per therm than rated.
Not every ECM holds airflow. Goodman sells the GM9S80 with a “multi-speed ECM,” yet its fan table drops the 80,000 Btu/h model’s default heating speed from 1,140 cfm at 0.5 in. to 1,028 cfm at 0.8 in. For a constant-airflow motor, air power is airflow times pressure, so 0.5 to 0.8 in. means 60% more air power (we calculated). Our blower motor guide tells the motors apart, and the static pressure explainer shows the measurement.

How many closed vents before the furnace trips its limit or the AC coil freezes?
No vent count fits every house; the limits are the furnace’s temperature rise and the coil’s airflow per ton, both printed by the maker. LBNL says never close more than 60% of registers. Goodman’s GM9S80 manual names closed registers outright: above the rating plate static, “check for closed dampers, registers, undersized or poorly laid out duct work,” since too little air raises the rise, which “can cause limit switch tripping and heat exchanger failure” (Goodman instructions). For the GM9S80 0803B* (64,000 Btu/h output, rise range 35 to 65°F, 3 ton AC):
| Total external static | Heat, default speed F02 | Rise on F02 | Speed F04 (default cooling) | Rise if F04 heats | F04 cooling per ton, 3 ton AC |
|---|---|---|---|---|---|
| 0.5 in. w.c. | 1,140 cfm | 52°F | 959 cfm | 62°F | 320 cfm/ton |
| 0.6 in. w.c. | 1,102 cfm | 53.8°F | 920 cfm | 64.4°F | 307 cfm/ton |
| 0.7 in. w.c. | 1,067 cfm | 55.5°F | 876 cfm | 67.6°F (over 65) | 292 cfm/ton |
| 0.8 in. w.c. | 1,028 cfm | 57.6°F | 832 cfm | 71.2°F (over 65) | 277 cfm/ton |
HVAC Brief calculation: airflow from Goodman’s installation manual tables; rise = output divided by (1.08 times cfm), the formula in Goodman’s spec sheet, which reproduces the printed 52°F and 62°F.
On its default heating speed this furnace stays in range even at 0.8 in.; on F04 it passes 65°F between 0.6 and 0.7 in., where a furnace limit switch starts cycling the burners. Cooling is tighter. FSEC researchers put maker targets at 350 to 450 cfm per ton and warned that flow “below 200 cfm per ton” can ice the coil (FSEC-PF-321-97); their existing installations averaged 317 cfm per ton before any vent was closed. Iced already? See why an AC freezes up.
Can you close a return vent, or the supply in a room that has its own return?
Closing a return grille is worse than closing a supply, because every room’s air comes back through the return side. IRC M1602.2, as adopted in the 2021 Seattle Residential Code, says “the amount of return air taken from any room or space shall be not greater than the flow rate of supply air delivered to such room or space.” Shut the supply in a room that keeps its own return and the room breaks that rule, drawing make-up air under the door or through gaps. Inspectors in many jurisdictions apply this to design, not grille settings, but the physics is the same.
ENERGY STAR’s Rater Field Checklist (Rev. 14, item 6.2) wants closed bedrooms within 3 Pa of the house. In LBNL’s lab one leaky case pulled the chamber 0.5 Pa below outdoors with four registers closed, a combustion backdrafting concern. An undersized return makes it worse, as our page on a return duct that is too small shows.
When is partly closing a register fine, and why won’t some vents close all the way?
Small trims on one or two registers far from the air handler are the accepted case. The DOE’s Building America Solution Center says that without branch dampers “minor air balance adjustments can be made by strategically opening and closing the dampers within the ceiling/floor supply registers,” but warns of whistling and that too many closed dampers can “ice up the evaporator” (BASC). In LBNL’s lab, shutting the four nearest registers made boot pressure 3 to 4 times higher than shutting the four farthest.
A register that will not shut completely is normal: BASC calls register controls a way to “fine-tune airflows.” If the lever turns but the louvers do not, the linkage has slipped; replace the register by duct opening size (LBNL’s boots were 10 by 4, 14 by 6 and 12 by 12 in.). For a room that stays off, HVAC air balancing does what closed vents imitate, and our page on one room hotter than the others ranks causes.
Frequently asked questions
Should air vents be open or closed?
Open, in nearly every house. The DOE’s 2011 Building America HVAC guide (PNNL-20421) says closing vents in unused rooms is not recommended, and LBNL’s 2003 simulations found energy use rose in every register closing case. Trim a register only to balance a room. In extremely cold climates LBNL also flags condensation and pipes freezing in the closed room.
Can I close the air vent in my room?
Partly, if it is one register far from the furnace. In LBNL’s lab, closing far registers changed pressure mainly in that branch, while near ones raised boot pressure 3 to 4 times more. The DOE’s Building America Solution Center allows minor register trims but warns too many closed dampers can ice the coil.
Should I close vents upstairs in winter and downstairs in summer?
Use trunk dampers if the house has them. The DOE’s Building America Solution Center describes one supply trunk per floor, with dampers set to send less air upstairs in the heating season and more in the cooling season. Shutting registers instead adds the leakage LBNL measured.
How we researched this
We read LBNL-54005 in full (Tables 7 to 11) and calculated air per open vent from it; the Pigg and Talerico ACEEE 2004 paper; the LBNL-62344 abstract; Goodman’s GM9S80 spec sheet and manual; FSEC-PF-321-97; PNNL-20421; BASC; ENERGY STAR Rater Checklist Rev. 14; and IRC M1602.2 (Seattle 2021). Contractor blogs, including the top results, were excluded as sources. Smart vents have their own page. Research date: November 2026.
Sources
- Lawrence Berkeley National Laboratory, I.S. Walker. “Register Closing Effects on Forced Air Heating System Performance.” LBNL-54005, November 2003. https://eta-publications.lbl.gov/sites/default/files/54005.pdf. Accessed November 2026.
- Pigg, S. and Talerico, T. “Electricity Savings from Variable-Speed Furnaces in Cold Climates.” ACEEE Summer Study on Energy Efficiency in Buildings, 2004. https://www.aceee.org/files/proceedings/2004/data/papers/SS04_Panel1_Paper23.pdf. Accessed November 2026.
- Lawrence Berkeley National Laboratory, I.S. Walker. “Comparing Residential Furnace Blowers for Rating and Installed Performance.” LBNL-62344, ASHRAE Transactions 114(1), 2008 (abstract). https://eta-publications.lbl.gov/publications/comparing-residential-furnace-blowers. Accessed November 2026.
- Goodman Manufacturing. “GM9S80/GC9S80 Single-Stage, Multi-Speed ECM, Multi-Position Gas Furnace.” Specification sheet SS-GM9S80. https://cdn.shopify.com/s/files/1/0794/3104/1346/files/GM9S80_GC9S80_Spec_Sheet.pdf. Accessed November 2026.
- Goodman Manufacturing. “Installation Instructions for *M9S80 and *C9S80 Gas Furnace.” https://documents.alpinehomeair.com/product/goodmangm9s8gc9s8installationmanual.pdf. Accessed November 2026.
- Florida Solar Energy Center, Parker, D., Sherwin, J., Raustad, R., Shirey, D. “Impact of Evaporator Coil Air Flow in Residential Air Conditioning Systems.” FSEC-PF-321-97, 1997. https://publications.energyresearch.ucf.edu/wp-content/uploads/2018/06/FSEC-PF-321-97.pdf. Accessed November 2026.
- U.S. Department of Energy, PNNL and ORNL. “Building America Best Practices Series Vol. 14: Energy Renovations, HVAC.” PNNL-20421, August 2011. https://www.energy.gov/sites/default/files/2013/11/f5/hvac_guide.pdf. Accessed November 2026.
- Building America Solution Center (PNNL). “Pressure Balancing Supply and Return Ducts in Existing Homes.” https://basc.pnnl.gov/information/pressure-balancing-supply-and-return-ducts-existing-homes. Accessed November 2026.
- Seattle Department of Construction and Inspections. “2021 Seattle Residential Code, Chapter 16: Duct Systems” (IRC M1602.2). https://www.seattle.gov/documents/Departments/SDCI/Codes/SeattleResidentialCode/2021SRCChapter16.pdf. Accessed November 2026.
- U.S. EPA ENERGY STAR. “National Rater Field Checklist, Rev. 14,” item 6.2. https://www.energystar.gov/sites/default/files/2025-04/National%20Rater%20Field%20Checklist_Rev%2014.pdf. Accessed November 2026.

Leave a Reply