GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter Instagram YouTube Icon Navigation Search Icon Main Search Icon Video Play Icon Audio Play Icon Headphones Icon Plus Icon Minus Icon Check Icon Print Icon Picture icon Single Arrow Icon Double Arrow Icon Hamburger Icon TV Icon Close Icon Sorted Hamburger/Search Icon
Q&A Spotlight

Why Is This Wood Stove Misbehaving?

A basement wood stove should provide inexpensive supplemental heat, but this homeowner is never sure when it will backdraft and smoke

A wood stove that smokes. Clark Agnew is trying to learn why the wood stove he's installed in his basement sometimes backdrafts. In the wall behind the stove is the 4-inch vent that brings in outside air for combustion.
Image Credit: Clark Agnew

Clark Agnew should be the envy of his neighborhood. He has a tight house, a high-efficiency wood stove with its own fresh-air intake, and access to free firewood. A heat recovery ventilator (HRV) keeps indoor air healthy. What’s not to like?

But, as he writes in a Q&A post at Green Building Advisor, the situation is far from ideal.

“I have run the stove about 6 or 7 times since we moved in,” Agnew writes. “Three of those times it has backdrafted.”

The stove has been installed in Agnew’s basement. It gets combustion air from a 4-inch-diameter vent to the outside, although the line is not connected directly to the stove. Agnew hasn’t had any problems getting a fire started, and the stove “seems to draft wonderfully.”

But periodically, the stove backdrafts.

“The first time was after about 10 to 20 minutes. I was just sitting there enjoying the heat and flames with my 2-year-old, and all of a sudden the fire goes out and smoke starts billowing out from all corners of the stove,” he writes. Plus, there seemed to be some kind of condensation coming out of the air intake.

On another occasion, the stove ran fine until Agnew’s wife turned on a bathroom fan. Once again, smoke billowed from the stove — until the fan was turned off.

“Third time, the stove had been running for about 20 to 30 minutes,” Agnew continues. “We were up reading books with my 2-year-old and the smoke alarm started going off. I run downstairs and once again, the fire is out and smoke is billowing out, mostly from the top, next to the air intake lever. I can feel air coming in the fresh air vent 1 foot away. After 30 to 60 seconds of this it just sort of stops and the fire picks up again.”

The dealer where Agnew bought his stove suggests a cold chimney might be to blame, or that the stove is merely burning off new paint.

Agnew’s stove troubles are the subject of this Q&A Spotlight.

A range of potential causes for this problem

This intermittent problem could be caused by one of several factors, GBA senior editor Martin Holladay points out, including a chimney that does not extend above the ridge of the roof or a restricted or undersized outdoor air duct. Another possibility is that a range hood fan, a bathroom exhaust fan, or a clothes dryer is putting the house under negative pressure, Holladay writes.

Dana Dorsett adds another wrinkle in the form of a code provision that would seem to prohibit exactly the kind of stove installation that Agnew has: that is, an air intake at an elevation higher than the fire box.

But Agnew can put some of these theories to rest from the start. The chimney is, in fact, high enough; it extends 2 feet above the ridge. And the Vermont Energy Code Handbook, Agnew adds, specifically allows his below-grade stove installation providing certain conditions are met.

But the comments have him wondering whether the stove is getting enough combustion air.

“I can imagine the stove using slightly more air than is being allowed in, hence it works great for about 15 to 30 minutes,” Agnew says, “but then perhaps it reaches a point where the negative pressure in the house becomes greater than the draft pressure and the draft reverses directions therefore bringing the smoke in the firebox and cold air from outside down the chimney, through the stove and into the house.”

This might help explain the condensation collecting on the floor near the air intake.

As to a fix, he wonders whether his only option is to open a window to allow more air into the house. “I would be a little disappointed if it was,” he says. “Maybe there is a better stove out there for a tight house?”

Swap the stove, and make a direct air connection

David Meiland suggests the installation may be hampered by the stack effect, in which air leaks in at the bottom of the house and exits near the top of the house.

The outside air duct is probably allowing air inside even when the stove is not running, and should Agnew open the door of the stove when it’s not burning any wood he’d be likely to feel cold air coming down the chimney.

“That’s the way ‘stack effect’ pressures work,” Meiland says. “Your chimney may be functioning as an air inlet whenever it can (i.e. unless you overcome the pressure with a roaring fire). Whenever the range hood, dryer, bath fans, built-in vacuum, etc., are in use, the problem is worse. In addition, your HRV may be unbalanced and causing a negative pressure.”

He thinks the best solution is to replace the stove and connect it directly with a source for outside air. “You need to remove house pressures from the mix here,” he says. “Your house is sucking on your stove.”

Meiland also lives in a tight house, but his Morso stove with a direct air connection burns just fine.

“One thing you could do is experiment with your HRV,” Meiland adds. “Turn it off for a period and burn the stove. If it drafts more reliably, that’s a clue. Another thing you could do is get someone with a manometer (probably a blower door technician) to check things out. In the end, my money is on a different stove.”

The chimney flue may be improperly sized

Into this mix Dorsett introduces another possibility — the 8-inch-diamter chimney flue isn’t matched correctly to the heat output of the stove.

With an apparent output of 29,300 Btu/h, the stove might operate better with a flue diameter smaller than 8 inches.

“Most modern wood stoves work best when the chimney flue is between 100% and 150% of the flue collar’s cross sectional area,” Dorsett writes. “(Standard practice in retrofits is to use a stainless flue liner of the same size as the stove’s flue collar).

“A 6-inch flue is about 28 square inches, making an 8″x8″ [ clay flue tile] more than 200% of the cross sectional area, and near the maximal safe limit of 300%, even in an air-leaky house,” Dorsett continues. “Unless the flue collar on the stove is 8 inches (unlikely, if it’s a ~30KBTU/hr stove), your chimney flue is oversized for the stove, particularly in a tight house. If the flue collar on the stove is 5 inches, it is grossly oversized, and would need a narrowing liner even in a house that isn’t very tight.”

Agnew is certain his builder didn’t know how to size a chimney flue, and relied instead on the “bigger is better” approach. As it stands, the chimney flue is about 177% the size of the 6-inch flue collar, “still slightly more than ideal from what you are saying.”

Among potential solutions are a venturi-type stack vent, which would would correct the flue oversizing problem (suggested by Dorsett), or a 6-inch metal chimney, possibly in the form of a liner (suggested by Meiland).

The HRV emerges as a possible culprit

Agnew continues to study the problem, going as far as buying a manometer to see how much negative pressure is produced by his bathroom fan, dryer and range hood.

In the process, he stumbles on a problem he hadn’t considered. When a defrost cycle in the HRV kicks in, the stove backdrafts.

“I was not aware that the defrost cycle actually circulated air, but was under the impression that it was an electrical element that defrosted the core,” Agnew says. “I am kind of amazed that they would engineer a negative pressure cycle into the HRV knowing the kind of houses it is meant for.

“It would make sense that this was the reason the stove backdrafted two times initially (seemingly at random times) since we were running the stove because it was particularly cold out (-10°F),” he adds. “The temperature difference creates condensation/frost in the core and that would trigger a defrost cycle.”

A dryer vent booster fan with a pressure sensor might help, Agnew says, by bringing in additional fresh air when there is a sudden negative pressure in the house. The booster could be wired to a switch so it was used only when a fire was burning in the stove, but Agnew isn’t convinced the arrangement would work.

Actually, Holladay adds, there is a control device on the market that would supply more air to the indoors when the house became depressurized. Holladay reviewed the control, the Exhausto EBC 14, back in 2008.

“The main disadvantage of the EBC is its high price; it lists for $1,500, although contractors will pay significantly less,” Holladay wrote in his original review published in Energy Design Update. “Moreover, a complete makeup air system will require several other components in addition to the EBC 14 control: the outdoor sensor, a supply air fan, and, in some cases, a heating coil to condition the makeup air.”

Our expert’s opinion

Here’s what GBA technical director Peter Yost had to say:

Since the issues involved with this GBA Spotlight are not part of my expertise, I asked around our industry, and three different colleagues said, “You really ought to check in with Skip Hayden.”

Skip Hayden is a senior research scientist with Natural Resources Canada and a recognized expert on combustion systems, including wood stoves. Skip jumped right into our discussion by asking the following questions about Clark Agnew’s situation, and Clark was quick to provide the answers:

Question: Is the chimney inside or outside the house envelope?

Answer: “The chimney is inside the house envelope, close to the center of the house.”

Question: How tall is the chimney, or rather, is the house one or two stories above the basement?

Answer: “The chimney starts at the basement slab. The house is only one level above the basement. So the chimney is approximately 8 feet in the basement (heated) plus 8 feet on the first floor (heated) plus 8 feet in the attic (cold attic space with 18 inches of blown-in insulation) plus 3 feet above the roof, for a total of 27 feet.”

Question: On what side of the house is the fresh air intake (i.e., west, east, north, south)?

Answer: “The air vent penetrates the east wall of house.”

Question: Even though the chimney may be the proper UL height above the roof line, are there tall trees around the house (especially upwind)?

Answer: “Not sure what direction upwind is, but possibly to the west. To the west, it’s 400 feet to trees, a gradual slope down away from house; east, 100 feet flat; south, varies from 50 feet to 100 feet to trees, a gradual to significant slope up away from house; north, 75 feet to trees, a gradual slope down away from house.”

Armed with all this information about Clark situation, Skip made these three observations:

The need for direct-duct combustion air. In general, wood stoves in tight houses need a ducted supply of combustion air direct to the firebox. As David Meiland stated, this approach separates the combustion air supply from the whole pressure regime, which is constantly changing. Skip also stated that this duct should pass through enough interior conditioned space to avoid thermal shock to the wood stove and also be sloped slightly to the outside for the first several feet to manage any condensation issues.

For Clark’s situation, his particular wood stove may be set up for a designed direct duct retrofit; if not, it is likely that he will have to upgrade his wood stove for a direct duct combustion air supply system. (Skip also thought that the pictured wood stove looked pretty big for the space and for a tight house.)

The inlet for the duct should be on windward side of the building. The combustion air duct should be taking its outside air from the prevailing windward side of the building. Pressures on even small residential buildings can be significant enough to affect wood stove draft. It’s better to pressurize than it is to depressurize this duct.

Install a double-walled flue pipe connecting the stove to the chimney. This will accomplish two things: a better sizing match between the stove flue and the chimney and a warmer stack to promote significantly better draft.

Skip further commented that for good performance of a properly sized wood stove in a tight home, the chimney flue should be the same diameter as the stove flue. This can be accomplished either by putting a stainless-steel liner down the existing chimney, or by replacing the existing chimney with a properly sized metal chimney certified to ULC S629 (a requirement in Canada). This type of metal chimney is much superior to a Class A metal chimney: it has 2 inches of insulation, meaning the heat loss from bottom to top is much less (so that the chimney provides better draft).

Finally, Skip stated that “temporary pressure relief” is often needed when reloading a direct-ducted combustion air supply wood stove, and the simplest way to achieve this is to first open the air supply on the stove to get a hot fire and better draft, and then to crack open a window on the windward side of the house as the wood stove is reloaded.


  1. User avater
    Dana Dorsett | | #1

    Since the wind doesn't always blow in the same direction...
    ... it's a bit difficult to install the air intake on the windward side of the house as prescribed, unless the site-specific prevailing wind is very consistent. (It can be very consistent in some areas, but it's really all over the place in most.)

    A venturi-type stack cap draws harder whenever the wind blows, and isn't much affected by wind direction, as long as it's at least a couple feet above the ridge of the roof.

  2. Patrick Walshe | | #2

    Pacific Energy Vista works for us with outside air intake
    Hi - we have a pacific energy vista wood stove with a 4" outside air duct that feeds fairly directly to stove. The stove pedestal sits over the 4" hole so there is a bit of leakage but not huge. The air duct goes under slab and pops up in wall and comes out about 7 feet up the outside wall. We used the double wall pipe up 2 storeys then the insulated chimney through attic space (another storey). The stove has great draft and once its going and door is shut I can turn on the 200cfm kitchen fan and its OK. If I even have a 110 cfm bath fan on when trying to light fire, the air coming down the chimney will blow out the lighter. We have very dry wood. My inlaw suite has the same stove but on one storey. That is fussier, but still works as long as there are no exhaust fans on when lighting. Sometimes it can help to warm up the chimney with a couple of lit candles inside the stove prior to actually lighting the fire to get the air moving up the chimney instead of down. Our Venmar EKO HRV is balanced.

  3. Hobbit _ | | #3

    what HRV?
    HRVs aren't *supposed* to negatively pressure the house on defrost
    cycles, they should shut off the exhaust-side fan and simply
    recirculate the interior air for a while to warm up the core.
    What kind of HRV is it? Has the manufacturer been contacted?


  4. Nick T - 6A (MN) | | #4

    Yes even my cheapo builder
    Yes even my cheapo builder model's defrost does a recirc mode...

    But i suppose maybe the Canadians got Defrost modes right sooner.... (venmar)

    Could have a failing or leaky damper in the HRV?

  5. Dick Russell | | #5

    Response to post #3
    Hobbit, I suspect you meant that in defrost mode the HRV should shut off the intake duct, not the exhaust fan. Otherwise the unit would continue to suck cold outside air, the cause of the need for defrosting. In my HRV, a Lifebreath 195ECM, there is a damper-closed port on the bottom of the cabinet, right under the intake air port, that opens to the inside air during defrost. The motor-driven damper swings 90 degrees, up from the inside air port and closes off the outside air port. Thus already warmed inside air flows through the HRV and back out to the house, while "stale" air from the house continues to flow through the HRV in the opposite direction and to the outside, as usual.

  6. Hobbit _ | | #6

    Well, that's the problem then: the exhaust side has to SHUT
    OFF when that recirc-only scenario is in progress, or the
    interior air gets pulled out. My Fantech does the right thing,
    moving a damper that closes off the exterior intake and swaps
    to a vent right on the side of the HRV that pulls from wherever
    the unit is located.

    I'd contact Lifebreath and call them on this. If it was a
    misguided design, maybe they have a retrofit fix that'll
    kill the exhaust-fan relay during defrost...


  7. Bennett Sandler | | #7

    Might want to think about your chimney
    I'd bet that a smaller diameter insulated stainless chimney liner will take care of this. Not too difficult to install--you can get drop preinsulated stainless down, or use uninsulated and pour castable vermiculite insulation mix around it. Although the fact that this is an interior chimney is certainly helpful with draft, you do have three strikes against: the cross sectional area of an 8X8 combined with the thermal mass of the masonry added to the squeeze a tight house puts on air supply. We've seen this phenomenon even with a woodboiler that had an induced draft fan and was operating in a leaky house. If you dump into a big masonry chimney, the smoke cools down and wants to sit around and hang out instead of going up. Keep it warm (insulated) and moving (appropriate diameter) and it'll probably do what it's supposed to.

  8. User avater
    Dana Dorsett | | #8

    It's an 8" round not an 8x8... (response to Bennett)
    ... but yes, still oversized for this stove, which is in-turn oversized for the house.

    A right-sized stove with a ducted intake air and a correctly sized flue liner for the new stove, filling the space between the stainless liner and the 8" round with rock wool really will take care of the problem, and work better in every possible way. The behemoth stove probably turns the place into a sauna if you run it a few consecutive hours at the mid to high-fire where it's particulate pollution is lowest.

  9. Dick Russell | | #9

    Response to post #6
    You wrote both: "the exhaust side has to SHUT OFF when that recirc-only scenario is in progress" and "My Fantech does the right thing, moving a damper that closes off the exterior intake and swaps
    to a vent right on the side of the HRV that pulls from wherever the unit is located." The latter sounds exactly like what my Lifebreath is doing, and also sounds like your exhaust side is not shut off at all.

    I see no reason why interior air can't be exhausted from the house during the defrost period of an HRV, other than the small short-term loss of heat recovery from that outgoing air. True, without inflow of exterior air, the unit goes out of balance, depressurizing the house slightly and resulting in the sort of air infiltration that a blower door test induces, although at a lower rate, and the exhaust rate may not be as high. But for sure there shouldn't be continued flow of cold outside air through the unit while it attempting to defrost itself. The slight depressurization of the house during defrost could conceivably contribute to backdrafting of the woodstove without a directly connected outside air duct to seal off the woodstove.

  10. Bill Bradbury | | #10

    Draft inducer
    I've been looking at this for a while, but still haven't tried it. It sounds like just what you need.

  11. Jeremy Shima | | #11

    What readings did you get with the manometer? I have a stove that I used in a leaky house and never had a problem with it. Then installed it in my new tight house with its own fresh air intake and if I do not open a window when I turn a fan on it will suck smoke into the house. I finally got my hands on a blower door and I checked for depressurization with the fans on. Just the bath fan -5.3 pa, dryer and bath fan -16 pa. Bath fan, dryer, and range hood on low -32 pa. and with the hood on high it was at -76.1 pa. I think the stove is the path of least resistance and smoke gets sucked in. The blower door reading was 320 cfm50 at 50pa. I am ok with opening a window but I would like to make it more idiot proof without making things complicated or expensive.

Log in or create an account to post a comment.



Recent Questions and Replies

  • |
  • |
  • |
  • |