The Central vs Dedicated Return Vent Debate
The central vs dedicated return vent debate has a clear winner. But first, let me explain the terminology. A forced-air heating and cooling system has supply ducts that blow conditioned air into the rooms of your house. That’s the part that’s obvious to most people. But that air has to come from somewhere. On the other side of the heating and cooling equipment is the return side of the duct system. It returns air to be conditioned again.
One way of designing the return side of the system is to use a central return. This design may have a single return vent, or it could have multiple return vents. In either case, the returns will be in the common areas of the house. The system serving the bedrooms in my house, for example, has a central return in the hallway.
The other way to design return air is with dedicated returns. That is, each bedroom has its own return vent sending air back to the system to be heated or cooled. A house with dedicated returns in each bedroom still will have more return vents in the common areas.
Of the two types, one has significant advantages. Can you guess which one it is?
The central return’s bad rap
The central return is common on most older houses. It’s also common in “starter homes” and other less expensive houses. Because of that, some people think of a house that doesn’t have dedicated returns as being cheap or out-of-date. We occasionally have HVAC design clients who tell us they want dedicated returns because they don’t want the house to seem cheap.
Another knock on the central return is noise. I can empathize with that sentiment because I used to live in a condo with a noisy central return. It was right below the TV in the living room, and we’d have to turn up the volume whenever the system started running.
A final drawback of the central return is that it can cause air flow problems and pressure imbalances in the house. This problem arises when you put conditioned air into a closed bedroom and don’t have a way for that air to get back to the central return. The bedroom will build up a positive pressure. The area near the central return will develop a negative pressure. The imbalance can reduce air flow, increase air leakage, and possibly even cause backdrafting of a combustion appliance. That last one can be dangerous, as it could put carbon monoxide in the house.
When you think about the central vs dedicated return debate, though, what are the real problems here?
Why a central return is better
One reason to go with a central return is that you’ll end up with less ductwork. Shorter ducts on the return side can mean less resistance to air flow. It’s certainly possible to do low-resistance dedicated returns, but it’s generally easier with central returns.
Central returns also can mean less energy loss if the ducts are in unconditioned space. Even though return air is not as cool or as warm as the conditioned air in the supply ducts, it still can be a lot cooler or warmer than the air in an unconditioned attic, crawl space, or basement. That means conductive heat gain or loss. Of course, you shouldn’t put ducts in unconditioned space at all, but many houses do have that problem.
Another type of loss that happens in the return ducts is through leakage. Yes, ducts should be airtight with no leakage. But the shorter your return duct system, the fewer connections there will be to seal. Also, leakage on the return side will suck in air from the surrounding space, which can cause indoor air quality problems.
And speaking of indoor air quality, central returns are a great way to be able to use high-efficiency filters (MERV-13 typically). The problem with trying to use high-MERV filters is that the resistance to air flow through them is too high. Increasing the filter area is the solution here. When it comes to central vs dedicated return vents, the central return makes that easier.
Instead of putting the filter(s) at the air handler, you put it (them) at the return vents. The photo above shows one of the filter grilles in my house, and I’ve sized them to have a lot of filter area relative to the air flow rate. Yes, you can do that with dedicated returns, too, but then you have more filters and probably different sizes as well. We like to size all our filter grilles the same so the homeowner can buy all the same size filters.
How to make a good central return
The central return’s bad rap is the result of poor duct design. Fixing the design problems can fix the perception problem, too, so let’s see what you need to do. There are basically two things to pay attention to.
Fixing the noise problem is easy. Noisy return vents are mainly the result of the blower being too close to the return grille. The solution is to put some distance and a couple of turns between the grille and the blower. Another way to reduce the noise is to size the system properly. Oversized systems have bigger blowers, and bigger blowers are noisier. Even better, use a small capacity, variable speed system (mini-split), which is quieter than a conventional system.
The solution to the problems with air resistance, heating and cooling losses, and pressure imbalances is to provide return air pathways. That’s all a dedicated return does anyway. But you don’t need to have an opening in each bedroom connected to a duct that carries air all the way back to the system. You just need to have a way to get the air from the closed bedroom back out to the hallway.
One way to provide a return air pathway is to use the gaps around the bedroom door. We usually talk about this return air pathway as a door undercut, but air can move around the sides and top of the door, too. Interior doors aren’t weatherstripped and usually don’t fit tightly in the frame, so air can get through all four sides. This method won’t work with all bedrooms, though. The amount of supply air put into the bedroom has to be fairly low.
Another return air pathway we use frequently in our designs is the jumper duct. The photo above shows one roughed in at a home built by The Imery Group here in Atlanta. The two return vents are covered by blue plastic tape to protect the inside from construction dirt. The duct goes up and over from one vent to the other, providing a return air pathway when the bedroom door is closed. Also, a grille on a jumper duct is indistinguishable from a grille on a dedicated return, for those want to see a return vent in every bedroom.
Transfer grilles* are another type of return air pathway. It’s a grille through a wall or door. Sometimes it’s a straight shot through one grille and out the other. Another way is to put a grille on each side of the wall but offset them vertically. That can provide more privacy than the straight-through type. The photo above shows one of the Perfect Balance return air pathways* I installed at the condo I used to live in, and it worked great.
The clear winner
In the central vs dedicated return debate, I have to say the clear winner is the central return. The bad rap that it has among some people isn’t the result of real deficiencies. It’s because of poor design. Follow the guidance in this article, and you can have central returns that are quiet, result in less energy loss, and give you better indoor air quality. Yeah, we’ll give you dedicated returns if you want them, and we’ll make them work well. But I hope you see the benefits of central returns now.
Allison A. Bailes III, PhD is a speaker, writer, building science consultant, and the founder of Energy Vanguard in Decatur, Georgia. He has a doctorate in physics and writes the Energy Vanguard Blog. He is also writing a book on building science. You can follow him on Twitter at @EnergyVanguard.
Related Articles
Can Door Undercuts Work As Return Air Pathways?
An Easy Retrofit for Return Air From Bedrooms
How to Make a Good High-MERV Filter Even Better
* This is an Amazon Associate link. You pay the same price you would pay normally, but Energy Vanguard may make a small commission if you buy after using the link.
Comments are moderated. Your comment will not appear below until approved.
This Post Has 55 Comments
Comments are closed.
I was surprised at the recommendation because I viewed the downsides of central returns as more significant than the positives. For example, my house (some builder’s McMansion) has vents into each of the bedrooms but not enough return pathways to the central returns. So, the mis-balanced pressures keep the bedrooms cold in the winter and hot in the summer with the doors rattling when the fan turns on and off. So we sleep with the bedroom door open (which isn’t great for fire safety.)
I did find it interesting that the way around all of the cons for a central return further lessens the pros. For example, a pro is that the duct runs are short and straight with minimal seams. But to get around the potential con of increased noise, make them longer and bendy (with more seams and length for pressure drop). The con of pressure imbalance is lessened by creating more ducts from room to room (with the possibility of it leaving the conditioned space or increasing room-to-room noise if not designed well.)
So, I guess its all a balance (literally and figuratively). A well-done central return with some thought put into the system may be better than a well-done branched system; but a poorly done central system can perform worse than a better dedicated return system if not designed right (and vice-versa.)
@Eric wrote: “But to get around the potential con of increased noise, make them longer and bendy (with more seams and length for pressure drop).”
As long as the return path is sized so that it passes muster (i.e.,< 3 PA drop across closed door with blower on highest operational speed), those extra bends and length in the jump duct are irrelevant, so there's no depreciation of the "Pros" as you suggest. Site verification is key to any duct design, central or dedicated. I don't recommend central returns for low-skill designers or installers. Of course, I don't recommend low-skill designers or installers for any project!
I’ve been designing and advocating for central returns in low-load (beyond code) homes for more than a decade with a lot of happy clients. Design airflows on my projects are typically 0.1 to 0.2 CFM per ft2 so the return paths are relatively small. I discuss with the client which transfer paths could result in noise issues and then apply the appropriate solution to those rooms in consultation with the client. I’ve never had a call-back or complaint about noise propagation through jump ducts or transfer grilles.
Many of the homes I work on have finished basements, and in that case, a ductless return is a no brainer (i.e., filter(s) mounted on return plenum base). This strategy has the additional advantage creating good air turnover between basement and above grade floors. No more musty basement. Isolating a basement from the rest of the house is one of the worst thing you can do for IAQ.
Moreover, a ductless return saves money: no return ducts or grilles to install and blower energy is reduced since return static is lower. The other big advantage is that it’s a lot easier to route ducts through a floor system when you only have to deal with supply ducts. These are not small advantages.
Not surprisingly, contractors often push back against the central or ductless returns in my designs. They somehow believe the distance between supply and return makes a difference. Hogwash.
The key to a successful central or ductless return system is to site verify < 3 Pascals pressure drop across each return path with doors closed and blower on highest operational speed. To achieve this in design, I rely on field testing experience and Building Science Corp's excellent technical paper on sizing transfer grilles and jump ducts. You can find additional articles and papers on this topic at buildingscience.com
I have a builder recommend an open return system on a spray foam house. Open grilles to the attic throughout the home. No duct for return system. Does this sound like its code compliant?
No. Return air should have a dedicated duct system. Using the attic space, even if encapsulated, will potentially expose your return air to pathogens, VOCs and who knows what else. The purported savings are not worth the risk.
I’m not sure if there’s any support to suggest pathogenic contamination. Chemicals and VOCs may be an issue.
Please know that spray foam is an on-site chemistry experiment. If the applicator doesn’t understand the chemistry, mixing ratios, proportioner and hose temps and substrate conditions, you might end up with a big problem. Even with properly applied foam, a chemically sensitive person can still have an adverse reaction.
The bigger issue is compliance with the fire protection requirements for foam plastics in the Residential Code. All foam must be separated from the living space with a thermal barrier. (Drywall or alternatives with code acceptable criteria). If the attic is only entered for service of utilities, an ignition barrier is permitted. But this gets really complicated when the attic becomes a return plenum. Not just for foam, but also for electrical wiring and plumbing.
I’ve heard people claim they do it. But the liability risk seems overwhelming to me. Especially since litigation these days is more about which lawyer is the best actor.
@Cajunjb1: In general, the IRC doesn’t require dedicated return ducting (and as I noted in my previous comment, this is my preferred approach). But as Danny noted, the IRC has provisions (see IRC Section R316 Foam Plastic) that might preclude what the builder recommended re: open return in semi-conditioned or conditioned attic with exposed spray form. Also, I agree with Danny that potential VOC outgassing from improperly cured foam is a risk not worth taking. OTOH, if foam insulation is full covered with drywall, there’s a ductless return isn’t a problem on either count.
David, I need to disagree about covering with drywall making an improperly cured foam safe, what you essentially stated. Drywall is quite permeable.
@Paul, yeah, it seems like that’s what I was saying. But since drywall is an effective air barrier, I’d be curious to know to what extent any outgassing from improperly cured polyurethane insulation will pass through drywall. In any case, how is that any different from having dedicated returns if there’s outgassing in the attic that’s separated from living space by only the ceiling drywall?
David, absolutely, there would be no difference in the separation layer. Would not wish upon anyone to go through a botched foam prep.
Yeah, latex paint provides some vapor retarding properties, probably depending on the particular paint, number of coats, sheen, etc. Still, in my view, it would be like living next to a brownfield site and hoping that some barrier will hold. The outgassed substance would have its own vapor pressure and would seek equilibrium.
To any DIY homeowners, or owners dealing with “adventurous” contractors: (1) just follow building codes resisting temptation to think “eh, stretching the code here or there makes sense to me so I’ll do it!”; and (2) run instead of walking away from any vendors, or installers who swear by their products explaining that their outgassing effects are no worse than a baby aspirin. Combine (1) and (2) with sound building science advice and you will probably end up with a successful return plenum.
In re foam risk, I am reminded of Peter Sandman who says,
“Risk = Hazard + Outrage”.
What’s more frightening to me is; we don’t really know as much as we pretend we know about the “hazards” of spray foam. Add to that, how little we know about human physiology and even a smaller fraction about human psychology (“outrage”).
The medical literature does seem to support a condition known as toxicant induced loss of tolerance. Everything is fine. Then one day the tolerance bucket fills up and the person can’t tolerate colgonge, deodorant or walk down the detergent aisle at the Walmart. Let’s face it, us Amuricans seem to love our chemicals.
Since our metric for “no problems” often relies on “no complaints”, the tolerance bucket can be a game changer. Our anecdotal assumptions can be quickly turned on their head and we now have exponential liability exposure. More like,
Risk = (Hazard x 100) + (Outrage x 1,000) or more if a child is involved.
@Danny, I was attempting to be responsive to Cajunjb1’s concern re: open return. As you know, I’m no fan of spray foam insulation (mostly based on economics), but you appear to be answering a different question: whether spray foam should ever be used in the first place.
A central return with a filter grill keeps the return duct clean. With a good filter installed with no bypass and zero leakage ducts. Cleaning ducts may become obsolete.
Building ture low load homes, the need for high air flow is not needed so standard door undercuts may be enough especially with variable heating and cooling as they should seldome run on high.
What about mixing to avoid stratification? Having a dedicated return across the room from the supply registers would seem to provide a more controlled environment, with more mixing. In my case, all my ductwork is in the floors, no walls or ceilings. The registers are set to push air up the exterior wall, while the returns (fairly close with minimal ducting) are on the opposite wall. It seems to work pretty well, but I will keep this article in mind on any future builds/rehabs.
@Norman, a transfer grille or jump duct can accomplish the same thing, although concern over stratification and air mixing is overrated in tight well-insulated homes.
Lack of stratification and mixing comes from proper selection and location of supply terminals not from location of returns.
I can see how a central return might be better in a new well-designed system, but I think people will read this and think their current system with a central return isn’t the problem they thought it was. But a central return in a leaky poorly insulated house needs to move too much air through the undercut doors to be effective. And telling people to leave their doors open while they are sleeping to keep their room comfortable is hardly an acceptable solution.
It may be premature to declare a winner unless the only consideration is efficiency. We must admit we don’t know all of the ramifications of removing particulate matter and other indoor air contaminants of one method versus the other.
Allison,
Good to see you back or perhaps my email just started letting you back in, computers can live with it without!
So i read the article and was not convinced for a couple of reasons. You seem to make an assumption that you can get air from the furthest reaches on the system and that while doing so you are getting a good air exchange in each room to create a entirely comfortable space. I don’t believe this to be true. Air takes the path of least resistance much like water. So I have to believe that supply air coming out of a floor register will rise and move towards the return grill (in houses Iv’e seen 30’ or more) not flushing much of the space.
I just read Dave Butlers post so I guess I’m in the hogwash came😂.
Now because air takes the least restive path the other assumption is that the blower door test was crazy low, if not that system is getting air from what might calculate to a 3’x3’open window.
With all this said and agreeing with just one return, what would you say might be the largest sqft house to be built with a central return?
I ask because what you are really advocating is quality over quantity, which we builders and HVAC guys should working towards.
Donald, will a 4000-sqft one-level house with 4 bedrooms (likely some will be 60ft or more apart) benefit from all that additional length of return ducts? Or maybe we need to think it over and put two separate units (if ducted) regardless of return methods?
I agree that we lack practical studies looking at these scenarios, especially with house envelopes getting tighter and more thermally broken, introducing fewer convective currents in a room that aid mixing… Or are they? Did I just make another “intuitive” observation without having any experimental backup? Run CFD simulations? But who has time for that?
Allison, you provoke many more questions than you provide answers. I want my money back! 😉
@Donald wrote: “Air takes the path of least resistance much like water.”
Absolutely. But then why can’t a non-ducted return path be that path of least resistance?
A critical commissioning step to ANY ducted system (whether central or dedicated returns) is to verify pressure drops across closed doors. It only takes a few minutes per bedroom to do this test when running the blower door test (uses the same digital manometer). In most cases, a properly sized transfer grille or jump duct will have lower pressure drop than the typical ducted return.
In short, there’s no inherent advantage of properly sized dedicated returns over properly sized passive return paths, notwithstanding Danny’s comment about particulate filtration.
@Donald wrote: “what would you say might be the largest sqft house to be built with a central return?”
There’s no limit. I have ductless returns in homes over 7,000 ft2, but there’s more than one air handler. There are practical limits on how far you want to run supply ducts.
slow moving air takes the less restrictive path, but take a look at how it moves out of a well designed and selected supply terminal (not a register that has a damper). It goes where you point it similar to what I see with my leaf blower (but less so and quieter). The solution for noisy returns is very large surface area filters (4″ deep and 300 to 350 sq. in per ton) Obviously multiple returns for any large system. Add to that smaller systems.
NREL published a simulation-based report (CONTAM) about return air pathways, but there is nothing like empirical data for house performance, and there are so few of this type of studies. Unfortunately, even those can be somewhat biased, especially if sponsored by a manufacturer – no pun towards Tamarack intended (the study quoted here in 2017: http://www.ba-pirc.org/casestud/return_air/index.htm).
About using framed spaces (between joists, between studs, between sheetrock etc) as plenums, however short: we don’t just “try” not to introduce them. We simply don’t do it, period. Any strategy for a transfer path that includes those is a non-starter for us.
Air flow rates: for example, when manual D for 2nd floor tells me 57 – 64cfm for all bedrooms with 36″ doors, hardwood floors, and 1″ undercut, I have not much to worry about.
Is that a cellulose-based material in the Tamarack sound baffle insert?
The photo of the jumper duct in the article above leaves some questions. It looks like maybe 8″ dia. flex, goes up about 4′ into attic space, for a total length maybe around 10-11′. Will that work well?
Our 1987 slab-on-grade single-story house in AZ has the air handler and flex ducts in the attic (upgraded to a 16 SEER heat pump system with variable-speed air handler when we bought it in 2013), with the main return a short distance from the air handler input, low on the wall in the hallway. The thermostat is directly across the hall from that return. A second, high return that is a bit smaller is in the living room close to the ceiling, which is the highest living-space strip in the house, and a flex duct connects it to the air handler intake box about 15-17 ft away. But the house is 80 ft long, with the Master BR and bath at one end, and the Laundry Room at the other. Despite the fact that we sleep with the BR door open to improve air circulation (not generally recommended as a safety issue!), the Master suite, especially the bathroom area which has two exterior walls, is almost always colder or hotter than the system set point by multiple degrees, depending on whether it is cold or hot outside, and ditto for the Laundry. I was planning to add a pair of high/low 12×12 filtered returns in suitable adjacent locations (the only possible places!). The Master one (8″ flex) would connect to the high return vent location (12″ flex), about 12-15 ft away in a straight shot in the attic, and the Laundry one (8″ flex) would connect directly to the air handler intake box, about 25 ft away. Then we’d have blown-in insulation added to the attic to largely encapsulate all of the ductwork that is currently sitting on top of the truss chords and R-30 batts. My belief has been that this would increase our comfort, allow us to close the BR door while sleeping, and increase the overall system efficiency. But your article seems to say that this is all wrong, and I should not do it. Am I misunderstanding, or is our situation leaving us with no better choice?
@Marty, sound like your supply ducts are unbalanced. Also, a longish house without zone control is always going to be a problem maintaining balance against varying solar exposures on each end (same with multistory homes with a single system and no zone control).
If you want to close your bedroom door at night, it may be a lot easier to simply add a jump duct or transfer grille! But that’s not going to fix a supply-side balance problem. Ditto with the laundry if you objective is to keep the door closed. But if a room gets cold (or hot) with door open, then it needs more supply air, which can only be accomplished by proper air balancing and possibly an additional supply vent. This requires professional diagnostics.
As an aside: research shows that many homes have undersized return ducts (as evidenced by excess return-side static pressure drop). If the return side (including filter) is overly restrictive, adding or enlarging existing returns and filter surface area is indicated. But that’s a separate issue than what you describe and may or may not apply in your case.
David, thanks for the comments. The house is exactly east-west elongated, primarily a rectangle about 28 ft north-south and 80 ft east-west, with two big evergreens on the southeast and southwest corners that moderate the solar impact on the east and west ends, and the (insulated) garage on the southwest corner. Two bedrooms, a bath, and little of the Dining Room get the southern exposure, but the main views and most of the windows are to the north, so all of the other rooms are glazed only on that side (100% double pane with an added 1/4″ acrylic friction-fit insert for noise and thermal purposes). Generally good construction with 2×6 walls (R-19), R-30 attic ventilation, but no under-slab insulation, and the poor HVAC system layout.
The two existing returns have filters under the grilles, in which we use 1″ pleated filters that are something like MERV 6; even with that fairly minimal extra obstruction vs. the almost-non-filtering fiberglass mesh ones, there is a lot more suction on them than they are capable of handling without some distortion, so I’d say that there might also be too little return volume (and thus intake vent surface) in addition to the supply likely being unbalanced. The system has a mix of supply ducting: rectangular sheet metal under the air handler that goes to the two other bedrooms, their bath, the Master BR (not bath), and the main living room, plus flex duct to everything else. The air handler is on the same east-west axis, so a 12″ flex runs from the outlet on the east end, around the air handler, and over to the west-side rooms (Kitchen, Eat-in area, and Laundry). With the divide-down process, the Laundry and Eat-In divide off of a 10″ flex (after the Kitchen take-off dropped it from 12″), and each gets its one-and-only ceiling vent via an 8″ flex. On the other end of the house, the Master is fed by the sheet metal duct while the associated bath and vanity area has a pair of 6″ flex outlets from an 8″ off of the same sheet metal duct. The Dining Room has its own 8″ flex feeder from the air handler outlet box, running about 28-30 ft to its sole ceiling vent on the south side, over its window. My planned extra hi/lo return with filter (one covered, depending on season) would be in a small hallway and would draw from the Laundry, one side of the Kitchen, and the opposite side of the Dining Room from its register. The other would be in the Master Bath vanity area, and also draw from the bath and the Master itself.
It seems like what you are suggesting is that the entire air handling system would ideally need to be reworked (although we’d still be limited to an attic arrangement, unless we foamed the whole thing). With truss construction, there are all kinds of obstacles everywhere to making any changes, which sound like they’d be very expensive. Being retired, we’re wanting to do what we can to make the house perform better for us, but we’re nearing the end of the funds that we have for that process. In your experienced opinion, would we be disappointed that the proposed additional returns did not help, and just do the extra blown-in attic insulation? But other EV articles seemed to argue against doing the latter, as the cost would not significantly improve over the existing R-30 batts, but at the same time would be a big upgrade vs. the R-4 on the old flex ducts which are fully exposed. Maybe foaming the attic would be the better choice, but it is always described as very expensive. I feel that there must be something that would make sense to experienced pros like you, that would still be very cost-effective for us, and would improve things to a notable extent!
Marty, I agree with David. You need lower restriction returns (not a bunch of long ducts to more rooms, you need to measure the airflow and the supply ducts need work to get the flow you want to the rooms. Don’t give up — I know it is hard everywhere to get contractors who actually know what to do. Try Googling BPI Contractor locator, that might help.
If you are up for it take a look at the following report starting on page 52.
10.13140/RG.2.2.17553.86885 or
Efficiency Characteristics and Opportunities for New California Homes (ECO)
Also page 32, figure 11 which strongly suggests the need for optimization of return paths
Wow, what a great and informative report on your project! (I’m a PhD chemist who went to a science and engineering college for undergrad [Harvey Mudd], so I enjoy seeing well-documented and high quality research work.) So now I look at our system with new knowledge and see different issues, the resolution of which might just fix enough of the comfort issues and is likely to reduce our operating costs. There is a second high location in the Living Room where I can add another large return with a very short (2-3 ft) connection to the “bottom” mounted AH intake box. I can possibly also change out the existing 24×24 filtered grille box for a deeper one to accommodate a high-flow-rate filter, but the sharp-angled turns for that return path (up through a deep wall cavity capped with a rectangular sheet metal duct that goes a couple of feet and then joins the AH at a right angle) are not really cost effectively addressable, I think. Fortunately, the installers of the current system used a curved metal rectangular cross section duct to send the supply air down to the main sheet metal supply plenum that goes under the air handler and feeds most of the outlets, but all of those also have built-in right angle joints except for one. The existing second return (a 14×20 filtered grille) has a metal elbow and a 12″ R-4 flex duct for 15+ ft, also not meeting your recommended sizing; I might be able to change it out for a larger return to match the new one mentioned above, and increase the duct sizing to 18″ while also moving it lower in the attic (to rest directly on the top of the horizontal AH for that part of its run). I’ve previously added foamboard on the outside of the AH, with the reflective metal surface facing the attic space, to try to reduce the impact of the attic temps. So much good info in your report!
@Marty, those sharp angles on the return side become less impactful when you expand and/or add more return capacity.
BTW, for those that don’t know who John Proctor is, he’s done more research and published more papers on residential HVAC system level performance, installation issues and remedies, than anyone on earth. Google “john proctor” “hvac” OR “air conditioners” filetype:pdf
Also, look for HVAC dealers with NATE Certified technicians, usually noted on the website. NATE Certification focuses on airflow and air-side diagnostics.
Wow, that’s a lot of words! First, I agree that distorted filters indicate inadequate return capacity, so adding one or more returns is appropriate. But it may be easier to install a passive transfer path for the master bedroom and one larger return in a more convenient location. You want at least 1.5 ft2 of aggregate filter surface area per ton of cooling capacity. MERV 6 media filters are plenty good enough to protect blower components.
You wrote: “…other EV articles seemed to argue against (adding blown-in insulation), as the cost would not significantly improve over the existing R-30 batts.”
I’m not sure what articles & advice you’re referring to, but I wouldn’t rule out adding more insulation. R-30 in Arizona, especially if you’re in a hotter part of the state, is inadequate. And since you have ceiling batts instead of blown-in, installation quality may be poor with lots of gaps, voids, and compression. This would be obvious on an infrared scan of your ceiling. Most home performance contractors and insulators have IR cameras these days.
If ducts are laying on or supported close to the top of the batts, then additional insulation will also reduce conducted heat gains from attic to ducts. Since you’re in a dry climate, you don’t need to worry about condensation (a big problem when burying ducts in humid climates).
Ah, thanks for the rule of thumb! I just calculated our ratio (with 4 ton HP) is at least 10% too low (ca. 5.3 vs. 6.0 minimum), assuming that “aggregate filter surface area” is just the plain open filter face area, not the pleat surface area.
There was an EV article that I read recently (but not sure of when it was written) that pointed out the diminishing returns of added insulation, from which I took the conclusion that, with existing R-30 insulation, putting in a reasonable amount of blown-in would result in minor improvements and perhaps miniscule cost savings, but there is still the exposed flex ducts that rest on the batts. I have the home inspector’s IR photos, and a FLIR IR attachment for my phone, and in general the batt installation seems to be fairly well done (I’ve fixed a few spots), although no vapor barrier was installed. We’re in Sedona, so we get both hot and cold times of the year, and have hopes that maybe 8-10″ of blown-in would help all of the ducts with additional insulation or full burial (except the high return one, which is on the shortest route to the air handler intake), plus boost us into the recommended attic R-value regime in general.
@Marty, by aggregate, I was referring to sum of all filters, not the folded media surface area.
Filter surface area ratio is a topic in of itself, but to help provide some perspective… I typically specify 2 to 2.3 ft2 per ton for new construction, similar to what John Proctor recommended in a previous comment. I intended 1.5 ft2/ton as an absolute minimum, given that you’re using MERV 6 filters, and was not intended as a design guideline. I should have made that more clear.
Manual D specifies a maximum face velocity of 300 ft/min (fpm) for filter grilles, which works out to 1.3 ft2 per ton and is close to your current ratio. But that spec was written before media filters were common.
More importantly, when specifying filter area based on tonnage, there’s a presumption that system airflow is nominally 400 CFM per ton. The fact that your filters are being distorted by return suction indicates your airflow is likely set significantly higher than 400 CFM/ton. That’s not a problem in your arid climate, but it does invalidate the convenient face area-per-ton ratio guidelines.
The correct way to specify filter surface area is face velocity in feet per minute (fpm). But that requires knowing system airflow, not a detail most homeowners will know. My design guideline of 2 to 2.3 ft2/ton is based on 175 to 200 fpm. John’s guideline is 10 fpm lower. My 1.5 ft2/ton ‘absolute minimum’ comment for your MERV 6 filters works out to 265 fpm.
If your variable speed air handler is the “constant CFM” type, you can check the speed setting on the circuit board and look up the airflow in the installation manual. Divide CFM by 5.3 ft2 to get the current face velocity. With filter distortion, I’d be willing to bet it’s well over 300 fpm.
Thank you, David! Very helpful!
I installed the Tamarac vents in each of the bedrooms and laundry room in a 1980 built tract home. They didn’t help as much as I thought they would.
I’ve lived in a couple of houses that don’t have any filters on the return grills but has a large 4” Aprilaire filter near the blower. We run a merv 13 or a merv 11 filter in it that gets changed every 6 months.
Should we put cheep fiberglass 1” filters in the return grills as well, or would we just be increasing the resistance that the blower has.
The house had an extra 12×12 return run to the master bedroom at one time.
The return box also supplies a Aprilaire dehumidifier that discharges into the main living area ( 95 degree discharge air in Houston summers seems like a bad design) that has its own merv 8 filter.
Chasing drafts for a decade has not been effective.
@David, the Tamarac transfer grilles won’t help if the rooms don’t have adequate supply capacity. If the rooms cool properly with doors open, but not with doors shut, then the transfer grilles are too small.
I recommend against installing filters in series, at least not without proper airflow diagnostics. Like most homes, you system is likely poorly balanced.
Thanks, it seems like we just keep putting band aids on a bad design.
We have 2 12 inch returns along with the 24×30 in the ceiling in the hallway near the attic mounted unit.
The extra balancers helped, closing doors doesnt affect the temp. The west side of the house gets hot, that’s where the master is located. Much of the attic isn’t accessible, we enlarged the duct but ended up using a small room a/c that runs most afternoons. After spending thousands with 3 different companies to seal the house I’m about ready to give up.
Finding a company to seal my house that knows what they are doing seems impossible. The dehumidifier should have been a ultra-air rather than the Aprilaire, the Aprilaire might be fine for basements in the Midwest but it’s not built for the gulf coast humidity. I’m hoping that putting in a good variable speed system will keep the humidity under control and do a better job at keeping the home comfortable.
@David, given that you don’t currently have a variable speed blower, you can improve latent (moisture removal) performance by reducing the blower speed. This should be done by a professional, as we don’t know the CFM/ton ratio and if the ratio gets too low, it could cause the evaporator coil to freeze up.
As an enhancement, if your t’stat supports RH control (requires dehumidification or “DH” output; perhaps that’s how your dehumidifier is controlled?), you could have an inexpensive 24VAC relay installed (~$10) that would reduce the blower speed when RH exceeds, say, 55%. This would avoid penalizing your A/C’s “sensible” efficiency when indoor RH is acceptable.
I’ll have to look into a relay setting. The a/c runs a lot during the cooling season it’s a 4 ton goodman built amana unit.
It doesn’t keep the humidity under 65-70% most of the time ( we added the dehumidifier after multiple contractors have worked on the unit)
The original dehumidifier install had the discharge going into the return air box. This caused the humidity to go up because it kept blowing warm dry air through the wet coil and drain tray. The final solution after 7 visits was to add a discharge duct in the middle of the living room. The discharge air temp is in the mid 90 range while the a/c is blowing out 60 degree air.
It does have a Honeywell thermostat that controls the dehumidifier but it’s a wireless connection.
I’ll try having the a/c contractor slow the fan motor down to see if that helps.
We used to live in the Deep South, so I understand the issue of humidity. I’d suggest having the discharge outside, not inside the duct system. Keep the excess moisture outside!! Having the discharge inside the system seems rather counter-productive! Put it outside!
@Marty, discharging a dehumidifier outside would be self-defeating. Why pay good money to dry the air, only to throw it away?
Absolutely right! I was mentally equating “discharge” with “captured water discharge” rather than “dried air discharge.” My bad!
I like a combo. Central return in a hallway. Transfers at non-primary bedrooms. Dedicated return at primary bedrooms for privacy (ducted into the central return). Kids don’t need privacy, right? 🙂
Interesting. I have both dedicated and a central return in my circa 1999 townhome.
Q for Dr. Bailes. Do you know of any ATL companies who can seal supply boxes where they meet the ceiling and inspect duct runs for obstructions?
The “jumper duct” image demonstrates a performance disaster in terms of thermal gain/loss. It leaves an insulated region and extends all the way to the worst imaginable location for unwanted heat transfer- adjacent to the bottom of the roof deck. Don’t do it that way (unless your insulation is at the roof deck).
Much better practice is to run the jumper duct as low as possible (use a 90 degree elbow to immediately turn the air at the ceiling joist height, just don’t aim it exactly at the other end to limit sound transmission) and connect with flex duct. Mis-align the direction of the ends (how they face each other) by as much as 90 degrees and oversize the flex duct a size by a size to account for the increased resistance of being curved between (not fully extended as we’d normally do in all ductwork). Deeply bury all within the attic insulation (all climates).
Locate the hallway “receiving” end near the ceiling-mounted return air and consider using a curved/angled blade grille to throw it toward the return (otherwise with standard ceilings and/or more CFM it can feel like an air curtain when doors are closed and you walk beneath it). You can bring multiple bedrooms into single larger box (like a 18 or 20″ square) that’s located within a few feet (truss bay or two) of the return.
The other caution is to consider whether you’re installing in conditioned or unconditioned space when the article says “Even better, use a small capacity, variable speed system (mini-split), which is quieter than a conventional system”.
New research is clearly demonstrating what we’ve long known, that ductwork connected to variable speed equipment must be (1) installed in conditioned space, (2a) when installed in attics, be minimally sized and deeply buried in attic insulation (where climate appropriate) or (2b) zoned in such a way to optimize velocity and static at each fan speed in order to be efficient and provide comfort .
And lastly, follow everything John Proctor and David Butler have mentioned in their comments.
i do disagree that using a single (especially low wall return in a cooling dominant region) / an or transfer are the INcorrect (acca standards) duct design approach for a cooling dominant region because Manual’s J,D,T all show that this practice is obsolete and not supported by the code reference standard / acca manuals. Multiple returns is what the acca manuals show – did you have another approved reference standard (like ashrae) that states a single or few (and especially low wall like in the photo you show) are the correct duct design procedure? for sure i have been designing only multiple return air ducted returns (over 60,000 hvac system designs in Fl for 40+ years) featuring a main r/a each level and a r/a in each enclosable living space like a bedroom. i am the least popular designer for builders and hvac installers because only the low bid is used – meaning that many of the required hvac items are deleted (value engineered) – so in my region the critical items are removed from a design, so that no mechanical ventilation air is induced, no branch duct volume dampers are installed, no test and balance, and of course a single low wall return is used! from acca manual J8 v2.5 page 7.2 “cooling dominant returns in ceiling near stagnant area of each enclosable room”…allowing the removal and conditioning of the latent rich air stratified at the ceiling….or acca manual j8 v2.5 table 7d / ae or others that show ” a single or few return air grilles designed is only applicable to existing homes, acca does not recommend this approach for new construction homes” recall that comfort includes humidity, temperature, sound, air quality…etc to be in line with acca manuals. this article you wrote stating that the simplified duct system is ok does not equal the acca manuals, i took a home to quantify the differences between a single low wall return vs my design showing several ceiling mounted r/a (large home builder dictates the a/c design to include just one return + transfer air) and found that the TEL was not greatly increased (from .507 to .583) and the added duct square feet associated with the r/a ducts did increase the heat load slightly and verified the exact entering air conditions at the cooling coil – all well within the hvac equipment ability. the home used for this study is a home that did not function, it has hot rooms and only a single low wall return – as you mentioned many duct systems designed outside of the reference manual D standard is widespread in our industry / many hvac installer use a duct u lator! but for sure i know of no design manuals that are inline with your article, can you provide such data or is your conclusion based on what the hvac / building industry prefers (and not based on the building owners comfort)?
this article is very similar to the bypass duct article you wrote from years ago, for sure the bypass duct has never been installed properly in our area (yea they try to use R6 flex in hot attics / dripping wet) and zoning and by pass did get this same bad rap all due to poor installation and engineering – i also disagree that a bypass duct is not required, manual zr covers the many issues with the installation of zoning including a few chapters on by pass, by pass design, and how to correct a poorly designed by pass , and i know few who even attempt to install zoning mostly due to the by pass duct design (requires mastic sealed snap lock (good for 900fpm) +R16 insulation (good for the delta T expected) + hand damper + motorized damper + air temp sensor for compressor cycle.
@Neil, we should be so lucky if every contractor followed guidance in the ACCA design manuals. But that’s not necessarily best practice. Also, your MJ8 citations are taken out of context. The example you cited on page 7.2 doesn’t preclude a high wall passive return path, and the citation from Table 7d is referring to central return without engineered return paths (old school, bad practice).
Manual D is the more appropriate (and more recent) guidance on return duct design. There are references throughout Manual D 3rd Edition to transfer grilles and transfer ducts, including a new Group 14 equivalent length table for passive return paths that was added in version 2.0. I know because I was on the committee that was responsible for MD3.
old school (version 2.5 mj8) was adopted as the reference standard by our codes a year ago – so this is very current,
when you add the square feet of the oversized jumper ducts and extra transfer air grille – its about equal (in cost) to providing multiple return paths – the ultimate modern hvac design for energy efficient buildings. transfer air simply moves the stagnant air into the adjacent interior space like a hall (in lieu of returning this stagnant air directly to the cooling coil) ceiling with no chance of getting this stagnant air down to a low wall return air device like we see in photo you show for this article – the photo showing one low wall return is perfect for a heating dominant duct design – this air device has no chance of capturing the stagnant air / or transferred air in a cooling dominant region ….. as noted in both recent versions of manual D manual T , this is not old school its written by Hank and now adopted by codes. so did you have some recognized design guide that would recommend a single low wall return + jumpers from bedrooms for cooling dominant? since there is an about equal cost in the two return duct approaches, and no demand increase if ducts are installed interior to building envelope, and little difference in watt change at the indoor fan motor that is associated with increased TSP and cost to operate, versus the benefits of doing it properly with multiple returns (two point to balance air flow / seasonal balance is possible) – also noted is the entering coil conditions (75 or less idb versus 81 idb) will also change the shr and total capacities the equipment is capable of……..https://www.hvacdesigns.com/wp-content/uploads/2021/02/Ceiling-Return-Air-Filter-Grilles-for-Cooling-Dominant-Regions-2021.pdf?39609b&39609b
end result is less comfort indoors due to stagnant air mixture within the occupied zone, longer operation times as the stagnant air is not captured and immediately conditioned……my old school memories (80’s) did in fact involve the single low wall return as was required to be the low bidder. but comfort designers who work for the home owner take no shortcuts as our current society requires perfect comfort throughout the home – half duct systems have little chance of providing the comfort clients expect now?
i have based this info on an actual home, one with comfort issues, the home had a single low wall return air grille just like this article shows….the study is in pdf and shows there is no advantage to taking the short cut of jumper ducts – and of course a low wall return in for heating only.
the best practice is the practice that ensures the home owner will be comfortable – jumper to hall and a low wall return is not the most efficient, or the least costly (noisy too) for any install unless its a heating dominant region …
@Neil wrote: “old school (version 2.5 mj8) was adopted as the reference standard by our codes a year ago – so this is very current”
MJ8 was written more then 20 years ago and subsequent revisions were focused on specific topics, not a complete re-write. In any case, Manual J is not a duct design standard.
You seem to be mostly focused on using a low-wall grille in cooling climates. That issue has nothing to do with whether the return is ducted or a passive transfer path.
You ask if there’s official support for passive return paths. Here’s the key citation from Manual D 3rd Edition Rev 2.5:
N1-10 Return Air Path Requirements
A low-resistance return air path shall be provided for every room or space that receives supply air. When a room or space that can be isolated by an interior door does not have its own return grille, an engineered transfer path shall route room or space air to a remote return grille. The requirements for an engineered return air path are defined here:
* When a transfer grille or transfer duct is used, the pressure drop across a transfer grille or transfer duct shall not exceed 0.05 IWC (refer to Section N4, Group 14).
* When door undercuts are used, they shall con-form to Table N3-2 gap dimensions. Gap distance is the space between the exposed floor surface or floor covering surface, and the bottom of the door.
Also see Manual T for additional references to engineered return paths (e.g., Section 11-4).
I’m not saying there’s anything wrong with dedicated returns, but a central return with passive return paths absolutely reduces routing/interference issues in floor trusses. And in homes with finished basements, a basement central return helps promote air changes between floors. The cost advantages are real but will obviously vary depending on project specifics. Ambient noise transfer can be an issue depending on room layout, but can be mitigated in various ways. I always involve the homeowner with those choices.
Mini splits are the solution to everything in cooling and heating, most homes with central forced air and heating are poorly done, leaky, uninsulated duct work in uninsulated walls, Minisplits with 28 SEER and above, solve all air flow and energy savings issues, the world uses minisplits, the U.S. is the odd one out with their forced air duct work stuff, I solved my poor forced air oi fired obsolete central air with 5 Mitsubishi Minisplits, and they keep getting better and better as time goes on, latest ones called hi2+ hyper heat even more efficient 100% heat now in sub 0 temps!!! amazing stuff, my oil fired forced air system ain’t been used for years, i fire it up once in a while just to see the furnace runs, the AC is long dead and gone.
I respectfully disagree. Variable speed machines have much potential, but not everyone is the best choice for every issue. Their SEER ratings are not at all comparable to single or even two speed machines. Ductless heads are really only good for one room. Like all heat pumps proper sizing is very important.
https://www.researchgate.net/publication/342317105_Central_Valley_Research_Homes_Variable_Compressor_Speed_Heat_Pump_Performance_Final_Report_Appendix_B_CENTRAL_VALLEY_RESEARCH_HOMES_VARIABLE_COMPRESSOR_SPEED_HEAT_PUMP_PERFORMANCE_FINAL_REPORT_Central
We use vrfs exclusively – a combination of heads and ducted units in all of our homes. We’ve found that the options are outstanding for comfort, and a central return makes sense. The duct runs are pretty stinkin’ short, and the static pressure is low. Given the tight building envelope and cz 4c, the fans in the ducted units pretty much never go full bore (unless manually overridden).
We did a fully ducted home recently, and the solutions were tailored. As with so many things, absolutes are generally not the way to go, eh?
Minisplit is constant air filtering and QUIET, not the drone of horrible billowing ducts. Forced air in most residential cases temp is up and down as the stuff is on and off …mini splits are super QUIET inside and out, much less energy used for equivalent heating and cooling, most residential setting have connected spaces that one indoor unit can easily handle, and just like duct-ed systems open the door for the other rooms, it works great. And each area is controlled to a desired temperature, ducted is one temp setting for all, very inconvenient, especially with the mostly dismally installed in homes, one room ice cold, other too warm in both winter and summer. Mini splits are the way to go, million in use around the world, far more than ducted systems like the u.s. uses, it’s a big world out there, minisplits are used in more places than our style of heating and cooling