GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter X Instagram YouTube Icon Navigation Search Icon Main Search Icon Video Play Icon Plus Icon Minus Icon Picture icon Hamburger Icon Close Icon Sorted

Community and Q&A

High CO2 Levels in Basement

fresnoboy | Posted in General Questions on

Hi everyone.  We built a new house in 2019, and made it pretty tight, not passive tight, but worked to make sure the sheathing was taped and taped areas at floor boundaries before sprayfoam went in.    This is here in the San Francisco Bay Area, so a pretty mild climate.  We have a Lossnay ERV system that is tied into the ventilation system for the basement and first floor, and a smaller Broan ERV for the 2nd floor.  The HVAC is a Mitsubishi City Multi S series system, with 7 indoor units and the accompanying ducting etc…  It’s 2×6 construction insulated with spray foam in all the walls, and a conditioned attic space as well.

When the weather is good, we usually have some of the windows open, but with the fires this and last year, we have been keeping things closed up a lot more.  The Lossnay has a wall remote control, but I figured out how to control it from a microcontroller with a few relays so my home automation system could control it.  My goal was to get a few CO2 sensors through the home that could tell if levels were going up and to trigger the ERV to operate more on demand than just in timer mode.

My first sensor was this unit: .  It has a a PM2.5 sensor, tVOC, and very nice Sensair S8 infrared CO2 sensor (that is used in a lot of test equipment), and is WiFi connected not just to an app, but my automation system.

The levels in my office last for the first week I got it  hovered around 700-800 ppm.  But sometimes it would go up quite a bit, and linger around 1500.  If I opened the window, it would drop back down to 400, but then go back up to 1500 on some days.  This was concerning to me, since this is a very large house, and there are only 6 of us here.   Even with the ERV running, and all of us out of the house, the levels came down and then went back up to the 1500s a few days ago.

So I connected the unit to a battery and walked around the house to see readings in various parts.  I walked into  the basement mechanical room, and it read almost 5000 ppm!

Initially I suspected it might be the gas hot water heater (a high efficiency  HTP unit) exhaust vent stack leaking, but we have CO detectors in  that room, and in the hallways and bedrooms, and no alarms were going off.  I then did some more probing and found very high levels of CO2 near the sewer ejector pump and when I lifted the cover and placed the unit just above the pump, it maxed out at 9999!  So probably in excess of 10,000 ppm.

So my question to the group is why am I seeing so much CO2 coming from the area of the ejector pump?  I can’t smell any sewer gas, so if it was a leak coming from the plumbing that was venting the CO2 into the house, wouldn’t I smell that?  There is a makeup air vent in that room, so I suspec t when the range hood or bathroom exhaust fans turn on, that make up air brings the CO2 from the mechanical room to the rest of the house.

I talked to my contractor, and before the basement concrete was poured, plastic sheeting and foam insulation were put down.  When the ejector pump was installed, the concrete and plastic and foam were cut to accommodate the pump install.  He thinks CO2 coming up from the ground could be pooling under the plastic sheet, and coming up inside where the barrier was cut for the pump.  Does that make sense?  Why would CO2 be coming up from under the ground, and would wouldn’t it disperse around the sides of the house?

Any other ideas as to what might be responsible for the CO2?  No one in the house is sick, and we have been very pleased with it, though I confess I did get a headache about moving the sensor around.

I am thinking I could get a testing company to use real test equipment to verify the readings, but not sure what do about this.  It would not be that hard to put a fan under the cover and pipe it to a whole outside – There is a lightwell on a nearby wall.  But I would like to understand what’s going on.

Anyone dealt with a problem like this before?


GBA Prime

Join the leading community of building science experts

Become a GBA Prime member and get instant access to the latest developments in green building, research, and reports from the field.


  1. Deleted | | #1


  2. X_NavyFC | | #2

    I don’t know too much about why the high CO2 readings. But I’m an instrumentation technician by trade and my first thought is about the accuracy of your detector along with doing a calibration check on it.

    Normally we would hook up nitrogen to it and set a detector to zero ppm. Then we would take a bottle of a known CO2 value and see how close the meter matches that stated ppm.

    I think if you are really concerned with it, I would hire someone to come out with higher accuracy equipment that can also walk around and try to pin point the source assuming there are actually high readings.

  3. Jon_R | | #3

    Looks like fresh sewage can produce CO2 without producing the H2S that you would smell.

    > put a fan under the cover and pipe it to a whole outside

    Makes sense to me. Probably with an inline fan (vs fan located under the cover).

    It's not clear to me how much pressure sewage off-gassing can produce. Air sealing alone (no fan) may not work.

  4. Expert Member
    PETER Engle | | #4

    While the absolute reading may be off on a consumer-friendly sensor, the relative readings should still be a guide to your problem, and anywhere near 1500 ppm will be noticeable. It seems that you have done a pretty good diagnostic survey already and narrowed the issue down to the sewage ejector pump, or that area. My big red flag is when you said you lifted the lid of the sewage pump to measure inside. You shouldn't be able to lift the lid of the sewer pump. It should be bolted down with about 16 bolts around the edge of the lid and the lid should be gasketed airtight. All of the penetrations for plumbing and wiring should also be gasketed airtight. Really airtight. Decomposing sewage gives off both methane and CO2. A combustible gas detector (about $250) would verify that it is the sewer pump leaking. I'd probably just skip the added detective work and go to a solution. Clean the lid seal area and caulk it. Caulk the seam between the concrete and the pump shell just in case. Caulk around all of the wire and plumbing penetrations. See how that affects your measurements.

    1. fresnoboy | | #6

      Pete, the top of the pump is about 2 ft below the floor of the mechanical room, so there is a plate that covers the hole, and allows you to walk on top of it. That's the panel I was talking about, I didn't open up the ejector pump itself.

      It's that space between the top of the ejector pump and the plate (that also covers the perimeter of the hole in the foundation) where the highest CO2 concentration is.

      I am thinking the next step may to try and figure out if it's the sewer "gas" escaping causing the high CO2 readings, or the CO2 is coming from the soil under the house and escaping around the edge of the hold in the foundation.

      Any ideas as to how I could tell the difference between these possibilities?

      I am probably going to get an air quality tech to come out with lab gear, but I have another consumer device to check with too, just to confirm relative concentrations.

  5. Expert Member
    BILL WICHERS | | #5

    +1 for just putting in some extra sealing work and possibly also some venting for the ejector pump. I'd also check that any existing vents aren't blocked -- I've seen a lot of crazy stuff done by critters that has caused screwy problems in the past. A common problem with natural gas fueled generators, for example, is mud daubers plugging up vent holes in the pressure regulators.

    I don't think it's too likely CO2 is coming up "from the ground", since the ground isn't usually a source of CO2 unless you have something decomposing under your house. I think it's more likely stuff in the ejector pump's pit.

    BTW, regarding sensors readings: relative readings aren't always reliable with a questionable sensor. Not all sensors are linear, so many have both absolute AND SLOPE calibrations. If a sensor somehow never got calibrated, it's entirely possible that BOTH absolute AND relative measurements will be off.


    1. fresnoboy | | #7

      Thanks Bill. It didn't make sense that the CO2 would come from the ground here - we're not on a landfill site, and we haven't had rain etc.. that could put "pressure" on underground gas as some articles talk about, though I have heard about that problem in other locales.

      I'll definitely check on the venting and make sure it's not plugged up...

      1. Expert Member
        BILL WICHERS | | #8

        You get methane from landfills, and boy oh boy would you know if that's what was coming in. I've worked a little with some landfill gas generation plants (they use the methane from the landfill to generate electricity, instead of flaring it off, so they are a sort of green energy project, in a way). Methane from landfills smells TERRIBLE! Imagine all that rotting garbage, baking in a hot compost system that's enclosed in a huuuuge plastic bag....

        If you were having CO2 getting in somehow underground (which I think is unlikely), you'd need something similar to a radon system to vent it away. I think if you deal with the ejector pump's pit you'll probably solve the problem though.


  6. GBA Editor
    Kiley Jacques | | #9

    It just so happens Jon Harrod's post for tomorrow is about CO2 in our homes. Keep an eye out for it. I plan to share this thread with him, too--I am curious to hear his thoughts on your situation.

  7. charlie_sullivan | | #10

    I think it's more likely coming from the pump than from the ground, since this is recent construction, I think it's possible that it's decomposing organic matter under the house generating CO2. A good composting process of some roots and leaf litter doesn't smell bad, like a landfill would.

    But it is tricky to test which it's coming from.

  8. Jon_Harrod | | #11

    This is an interesting situation and one I've not run into before. I do think you're on the right track. It could be coming from the pump assembly or decay of soil organic matter. Sealing the pump lid is likely part of the solution in either case. I would do some experiments with the effects of depressurization--this is a known phenomenon with radon, and I imagine something similar could happen with subslab CO2. I'd also double and triple check the venting on your combustion equipment. A clean-burning but improperly vented water heater might only be producing one or two ppm of CO, not enough to set off a detector, but might still be pumping measurable amounts of CO2 in the living space.

  9. fresnoboy | | #12

    Thanks everyone. I'll definitely look forward to the article. I did put the sensor close to the exhaust vent of the HTP heater, and it didn't show any increase. But when put near the ejector pump, it shot up. CO2 is heavier than air right? So since the empty space above the ejector pump is the lowest part of the room, I guess the vent if it was leaky would eventually sink to the bottom of the cavity with the pump in it.

    I did go back and look at the photos of the job site back when the excavation was done. There was not a lot of vegetation/leaves etc... in the area or around it before the foundation insulation and fabric etc... were attached before the pour. There were a few though, and it was in October of 2017, so fall starting out here, and we do have some large trees around. But not huge piles of leaves or that you could see any debris on the pics. Am happy to post them if useful.

    So other than decaying plant matter, is there any other reason CO2 would be coming out of the soil, esp with no rain?

    I'll fire up the heater and see if I can detect leaks of CO2 from the piping. And then we'll try and caulk all the piping in and out of the pump.

    Thanks again!

  10. fresnoboy | | #13

    So I talked to my GC, and he informed me that another one of his customers that has the same unit I do detected high CO2 levels coming from the area of the pump. The owner's wife is a bit of a hypocondriac, and had a speciality firm come through and do a full survey of their home and found high CO2 levels in their basement area where the pump is located.

    I've never heard of such a thing about sewage pumps and CO2 before.

    That owner decided to put in an exhaust fan to vent the gas outside, but I don't think his house as as tight as ours is. I do have a 4" vent our mech room that is the air intake for make up air. It's a simple pressure damper that opens when the house has negative pressure. I was thinking I could install a compact ERV like one of these that has a concentric vent so I am not just trying to blow out conditioned air without another intake to balance the flow.

    Anyone else have an idea that keeps the tightness of the house intact but gets the CO2 out of that room? I don't think it has to have high volume.

    1. Jon_R | | #14

      Vent just under the sealed cover and the CFM needed should be so low that you don't need to worry about heat recovery, pressure balance or effect on intake air. Let the CO2 into the room and you are just creating a bigger problem, needing more CFM. Ie, as with moisture/vocs/odors/etc, better to address the problem at the source.

    2. Expert Member
      BILL WICHERS | | #15

      +1 for putting the vent under the cover. All you need to do is maintain a slight negative pressure in the sump area so that nothing "gets out". A simple muffin fan is probably enough, although I would get one rated for higher than normal static pressure. Run the muffin fan with a small wall-wart type power supply (I'd use a 12 or 24 volt DC fan), and I'd put a light in the low volt circuit so that you know if the power supply every dies on you. If you use a red LED for this light, you need a 1 k ohm resistor for a 12v system, or a 2.2 k ohm resistor for a 24 volt system. A 1/4 watt resistor is fine for the 12v system, but you should really use a 1/2 watt resistor for a 24 volt system.


      1. fresnoboy | | #16

        Good call to not have the CO2 circulate in the room!

        So maybe get a concentric vent kit (so the makeup air can still enter unaffected by the fan exhaust air). I can see a 80mm fan fed off a wall wart (with LED!) that can run continuously and vent from under the cover out through the pipe could work pretty well.

        Maybe some flex duct to hook it to the vent? There is not a lot of extra space in the vent cover, but I don't think it will take a large pipe to suck it out from the top of the pump assembly.

        1. Expert Member
          BILL WICHERS | | #17

          Actually, if you restrict makeup air you will maintain negative pressure in the sump, and that negative pressure is helpful here -- it makes sure any and all leaks leak INTO the sump, not OUT OF the sump.

          Flex duct could work, but rigid PVC pipe will be better for airflow (less restrictive). Either way is a lot better than nothing though!


  11. fresnoboy | | #18

    Sorry, I don't understand. The sewer pump assembly is cut into the foundation of the basement floor - it's sealed (or so it's supposed to be!). So If there is negative pressure in the house, there is no air coming from the sewer pump to compensate for that. I don't want makeup air vented through the sump to the rest of the house, that just spreads the CO2.

    The makeup air vent is about 8 ft away from the sewer ejector pump. So it's reasonably easy to plumb PVC from the vent to the ejector pump area. I thinking something like an 80mm silent PC fan would move plenty of the air needed to get rid of the CO2.

    If a a lot of CO2 were being produced, the whole room would have astronomical levels, but it's just not that bad. It does spread through the house though - you can tell even when no one is home that the CO2 levels in a couple rooms are going up through the sensors I have in place.

    1. Expert Member
      BILL WICHERS | | #20

      If you keep the sump under slight negative pressure, with a fan venting to the outside, and gases produced will be sucked outside. Negative pressure means no chance of any leaks. You don’t need (or want) makeup air in this situation.

      Make sure any vent you use to extract the CO2 is not near any building fresh air intakes. You don’t want one system to suck in the gasses you just vented out with another.


  12. Jon_R | | #19

    What are you trying to accomplish with makeup air? What is the setup (weighted damper?, size?). How many CFM for your exhaust fans?

    I suspect you should close this makeup air vent and possibly supply makeup air to the kitchen only when the range hood is running. Powered makeup air is better.

  13. fresnoboy | | #21

    We have a fairly big kitchen and 48" range and hood, so it's a 1000 cfm exhaust fan (but variable speed, so not on full when it's used). There is no forced air blower on the intake, but I am thinking we can fix that as part of this project as well. I can use an interlock to keep the CO2 exhaust fan off when the Make up air fan goes on.

    1. Jon_R | | #22

      Yes, you would be best off with powered makeup air delivered directly to the kitchen when the range hood is on. Unless it's for something else, close the mechanical room intake.

      1. fresnoboy | | #23

        There are ducts that feed air to the kitchen area such that air flow carries everything in the direction of the exhaust hood, so even on a low volume everything gets pretty much sucked out and not spread through the kitchen. I feel plenty of air flow so they may be a a set of fans that blow air through those ducts when the exhaust turns on. I'll have to talk to the GC about it.

  14. fresnoboy | | #24

    Folks, here's an update. I talked with my GC, and he couldn't figure out why the sewer ejector pump would be responsible for emitting CO2 in high levels, especially if we didn't smell sewer gas, which would indicate a leak in the pipes.

    I decided to get some more data on the CO2 levels in the Mechanical room, and specifically above the the sewer ejector pump cover. This turned out to be harder than it looked.

    Most of the commercial CO2 sensors that report to an app or home automation system are based on sensors like the Sensiron SCD30 or MH-Z19 sensors. They will lose accuracy if they are not exposed to outdoor air every few days to calibrate the sensor (or otherwise be exposed to fresh air at 400 ppm CO2). And they tend to max out at modest levels of CO2. This a problem for our mechanical room that has no windows to open, so hard to make the calibration sequence work.

    Since I am a Home assistant user and comfortable with electronics, I decided to build a sensor out of a ESP8266 microcontroller and Telaire T6615 CO2 sensor, which is a dual channel sensor that has an internal gas sample, which compares readings from the air to the gas sample standard. This doesn't need calibration as is accurate at high CO2 levels. The microcontroller updates Home assistant via wifi every 60 secs with the sensor data.

    I mounted the microcontroller on the wall with a cable to the T6615 that sites in a hole in the cover over the pit that has the sewer ejector pump in it. So it's at floor evel in the mechanical room, but not submerged in the pit.

    Attached is the last couple days of data. As you can see, the Co2 levels start rising after the kids go to school, usually around 9:30 or so, and then start to decline around 6:30 or 7 PM, usually during dinner time, and then decay slowly overnight, I assume as the Co2 diffuses through the house. And the levels get VERY high - 30,000 ppm!! That initial day's dropouts are caused by some work I was doing on the sensor and in the room, but the prior days looked similar in terms of the curve.

    This doesn't make sense that it's coming from CO2 under the house, because I would assume that would be diffusing up around the pit continuously, and not peaking like this. It's also not aligned with people in the house which would be flushing the toilets etc a lot, and filling the system with sewage. It does seem to be aligned with outdoor temperature though somewhat, but the outdoor temps don't drop off that quickly at night, but I'm unclear what the correlation truly is.

    It's also not correlated to the use of the gas hot water heater, as I have turned it off a couple times during the evening, and the CO2 levels swelled even when the hot water heater was switched off, and it doesn't correlate with hot water peak use in the MA and the evening.

    It does seem to correlate with higher CO2 levels in the house though. As the CO2 levels in the mechanical room peak, other Co2 levels in the house also go up. If the windows are all closed, it's not uncommon to see 1200+ ppm levels in some rooms in the house.

    The next step is building PVC piping to use a small fan to exhaust CO2 from the sewage ejector pump to the outside, but I am curious if folks here can explain what source could be causing this kind of CO2 generation around a sewage pump. It's not a trivial amount of CO2!

  15. ACD_III | | #25

    I have spent over (3) years and thousands of dollars trying to migrate exactly the same problem you are having. At this time I suspect limestone in the soil aggravated by extended wetness in the ground is causing CO2 problem. Have you been able to find someone to help you or have you been able to solve the problem? I would greatly appreciate any helpful information you would share.

    Thank you in advance for any help you can give me.
    Lord bless,

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |