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

DIY Blower Door

musielski | Posted in Energy Efficiency and Durability on

1st a little background. I built a 806 sq-ft cabin Washburn Co. WI. which is in zone 7.
This cabin has been a DIY project built on a monolithic slab with a simple gable roof.  I could find someone within a 100miles in northern WI to preform a blow door test, however, everything else has been DIY, so why not this?!
I do hold a degree in engineering, as well as being a lic. home inspector…so I’m technical.
If anybody has built their own, I’d be interested in hearing about your success/failures.

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. Expert Member
    BILL WICHERS | | #1

    I've often thought about doing something like this, but I haven't tried it. The biggest issue would probably be calibration, but aside from that, it shouldn't be that hard to do. All you need is a beefy fan or blower than can pressurize (or depressurize) the house enough to hit the pressure differential you're after for your test. The 'door' part of the blower door is always a little goofy even on the professional rigs, since it has to be adaptable to different size doors. The equipment I've usually seen uses a material a bit like canvas or tent fabric for this purpose.


  2. jonny_h | | #2

    I've started thinking about doing this, too -- don't have a definite answer, but do have some ideas:

    - An old furnace blower would probably be suitable -- or pick up something at your favorite local industrial surplus store, if you happen to have one. Try to look at fan curves and get something that can actually produce the required pressure differential with "sufficient" flow (where "sufficient" is your worst-case estimate of air leakage at that pressure.

    - Get an ECM fan, or run an AC fan on a VFD, or something to get you variable speed. Adjust until you get the desired pressure differential (which you can measure with something like a Magnehelic gauge or a digital manometer).

    - Alternately, with a fixed-speed fan, adjust the size of an aperture that serves as an "intentional leak" to get the desired pressure differential.

    At this point, you can do all the qualitative stuff a blower door test enables -- go around with a smoke pencil and look for leakage in various areas, check stuff with an IR camera, etc. As Bill said, though, calibration to get an actual CFM number will be the tricky part. My thought is that perhaps you could build an airtight box (plywood with taped edges, or lined with plastic, or something), to which you can attach your blower at one end, and affix various calibration orifices at the other end. This would give you the ability to roughly map fan power to cross sectional area of leakage, which when compared with your house results would roughly tell you how big the effective leakage area of your house is. You may also be able to calculate some rough CFM numbers from this, based on some formula that I don't know off the top of my head, but my suspicion is that fluid flow is tricky and this would be a rough, order of magnitude approximation at best.

  3. mills9867665789760 | | #3

    I did this recently by more or less following this very helpful tutorial:

    I don't have much confidence in my ACH score, but it provides a good baseline for measuring improvements as I tighten up the building. Plus, it works great at depressurizing the building to help find leaks.

    Here's what I've learned so far:

    - Used furnace blowers are cheap and they work great for moving the air. Mine was $30 at a metal recycling yard. Buy the biggest blower you can find that uses 120v. If your house is really leaky, you might need two blowers to reach 50PA.

    - Control your blower speed with something like this:

    - The hard part is measuring CFM of the blower. I used one of these:
    You'll need to systematically measure the air flow at different points across the blower's output and then average the readings to estimate total CFM. It's hard to get a good number, so this is probably the biggest source of inaccuracy in the system

    - The indoor/ outdoor pressure difference needed is very small, and you need a way to measure it precisely; something like this:
    You can buy them used on Ebay for ~$30

  4. Expert Member
    Akos | | #4

    You can get pretty reasonable flow numbers as long as there is a long enough (5x to 10x of diameter) tube mounted to the fan and something like a fluorescent light fixture grate after the blower to reduce turbulence. You can mount a hand held anemometer or differential manometer with a pitot tube in the center for readings. Some units have the option of entering duct size to give you estimated flow rate. Won't be perfect, but in the ballpark.

    Something like a 12" sonotube would be good for 1000cfm. Smaller if you are looking for lower flow rate.

    1. Expert Member
      BILL WICHERS | | #5

      Akos, I was just about to suggest this very thing! Get a sonotube about the size of the aperature of your blower or fan, then use a small anemometer to measure airflow. pi*r^2 for the area of the tube, so 0.785 square feet for a 12" diameter sonotube. If you measure 20 miles/hour air velocity in that tube, that gives you 1,382.3 cubic feet per minute (CFM) of air delivery. Since you really need linear airflow for this to be accurate, it might help to put a vane or two (horizontal flat piece of cardboard or thin plywood) down part of the length of the tube to "straighten out" the airflow from the fan.

      Note that DC fans can usually be speed controlled over at least part of their range with a simple PWM speed controller. That might be simpler than a VFD and larger blower for a small blower door setup.


  5. Expert Member

    Okay - very basic question: I realize the answer is very contextual, but how much do you think the actual ACH experienced by a typical house varies from the ACH50 result predicted in blower door tests?

    1. Expert Member
      BILL WICHERS | | #7

      Do you mean how does the ACH number differ across different pressure differentials? It would be easy enough to run the blower door up to several different pressure levels, then compare readings and plot a line (probably a curve, actually). From there you could extrapolate ACH numbers at other pressure levels that you didn't actually measure.


      1. Expert Member
        MALCOLM TAYLOR | | #8


        I was wondering about a more basic question. A lot of people look at their blower door result and think that represents how much air will be exchanged at any given time. But that is really a snapshot of how much it leaks while de-pressurized. Say a house measured 2 ACH50. What ACH might the house actually see over an extended period?

        1. PBP1 | | #9

          Excellent question, pressure as the driving force. If you have a big, solid fence next to your house (wind block) that will make a difference on actual ACH. Evidence from tornados, hurricanes, labs with fume hoods and even living on the green line, demonstrate the importance of pressure differential. If near zero, little ACH. The “door test”, i.e., is it hard to open an exterior door, is telling. In one lab, the negative pressure was so great, one had to exert great force to open the door, but better than breathing nasty stuff.

          1. charlie_sullivan | | #14

            It just so happens that the current front-page article addresses this question in great detail.


        2. Andrew_C | | #10

          I know that you know this Malcolm, the actual ACH is highly variable, depending on the house construction and on the environmental conditions. I would say no generalization is possible, other than there are many periods of time where there is very little air exchange, and this is another reason for balanced ventilation systems. And opening windows as seasons allow.

          1. Expert Member
            MALCOLM TAYLOR | | #11


            For sure - there are too many variables to predict any individual house. As PBP1 said: everything from local climate to site conditions. But what about in aggregate? Say you took every house tested in a given year across North America. How many of them do you think experience higher ACH than when depressurized to 50 pascals, and what proportion see less?

            One of the reasons this interests me is that next year it looks like our code will be amended to bring in more requirements around energy efficiency - one of which will be blower door tests. When I describe this to a client they will probably ask - Will my house actually change the whole air in it every X hours as the test found? Is the only useful answer - Maybe?

        3. Expert Member
          BILL WICHERS | | #13

          Ah, so you're thinking "what is the actual ACH of the home under typical static conditions". Essentially what amount of air is leaking out of the home during the course of a normal day, when the home isn't significantly pressurized or depressurized by an external force like a blower door.

          I don't think it's really possible to directly measure that, since you could never account for all the variables. At any given time, a bunch of things are leaking air out, and a bunch more are leaking air in to makeup for the losses. Since you would never be able to fully account for ALL of those little leaks, the blower door acts to make one BIG "leak" (the blower) that overpowers all the others, which makes it possible to make ONE measurement at ONE location (the blower door). Without that blower door, you can't really measure the air movement through the sum of all the little leaks.

          What you would need to do is measure at several pressure levels, then extrapolate out to "zero", knowing that "zero" isn't actually "zero", it's just what the leak rate would be without a pressure boost. You'd end up with an ACH number somthing less than the xACH number, or at least that would be my assumption, since the increased pressure differential created by the blower door should make for MORE leakage compared to the normal static pressure conditions.

          I can't say I've ever actually tried to do this, but it would be interesting to try -- especially if similar measurements could be made over several houses for comparison purposes.


    2. Expert Member
      Akos | | #18

      The typical way to convert from ACH@50PA to ACH is using an N-factor:

      It is only a very rough number and it mostly gives you worst case leakage that can be used for sizing equipment.

      There was interesting thread earlier about using CO2 levels to estimate air leakage. The idea was to look at the decay curve of interior CO2 levels right after the house is left empty.

      Another way would be to measure steady state interior CO2 with fresh air ventilation off.

      For example, say a home has 4 people, no plants or pets. People at rest breath about 0.4CFM and produce 4%CO2 or 40000PPM. Assuming outdoor air is 400PPM.

      So if the stable interior CO2 level is 1000PPM, the house natural air leak is about:

      (0.4CFM * 4 people * 40000ppm)/(1000PPM-400PPM)=106CFM.

      If you work backwards, that would translate to around a 2200CFM@50PA, which would mean a pretty leaky house.

  6. AndersB | | #12

    Finished a ADU last year and put in a lot of effort to make it as airtight as possible.
    In an effort to find airleaks I implemented some redneck technics. I had no desire to get a ACH number, or at least no interest to pay a premium for it.
    A trip to Harbour Freight to pick up this ventilator and making a rough blower door did the job. Quite an experience to get the fan going and hear the cracking sound as the building got decompressed. Spent a day with a smoke pen, foam and tape and got the house way tighter.
    However, no certificate to hang on the wall ;-)

    1. jonny_h | | #15

      I think this is the point that you get to make your own framed certificate declaring the structure a certified Pretty Good House. If anyone questions your ability to declare so yourself, just say "a guy on the internet" told you it was fine. ;)

    2. jvidamins | | #23

      I'm wanting to do this very thing for the sole purpose of trying to find leaks in my new construction air barrier (drywall for ceiling/Zip R-9 for exterior). It's a 2,600 SF ranch with full basement with open stairwell. I've tried to build as tightly as possible, but since this is my first time, I'd like to spend a day looking for leaks with a smoke pen or fog machine and a fan big enough to pressurize enough to make the leaks apparent. Do you think this harbor freight ventilator would do the job? Again, I don't care about ACH... I'll get a legit score later. Just wanting to find leaks at this point.

  7. amorley | | #16

    What minimum specs would an old furnace fan (or new Harbor Freight fan) need to have to do decent blower door directed air sealing work on an old leaky house? (Not necessarily for a DIY blower door scoring).

    Say, for example, a 2000 sq ft house with a blower door score of 2500 CFM at 50 Pa.

    1. Expert Member
      NICK KEENAN | | #17

      The fan would have to be able to pull 2500 CFM at 50 Pa. That's a pretty beefy fan.

      In terms of leakage that's actually not that bad. Let's say your 2ooo sf house has a volume of 20,000 cubic feet. 2500 CFM means an air change every 8 minutes, which is 7.5 ACH. I wouldn't consider that an old leaky house at all, I'd consider that a better-than-average old house.

      In really leaky houses often the problem is the fan can't produce enough air flow to achieve 50 Pa.

      1. amorley | | #19

        Yeah, I thought it would end up needing to be beefy fan given the #s.

        Using a smoke pen or incense stick, can one do effective detection of air leaks at much less of a pressure differential than 50 Pa?

        ie, can a smaller fan do the trick?

        1. mills9867665789760 | | #20

          Here's one data point: my leaky old 1800 sqft house required ~3250 CFM to reach 50 Pa.

          This required two furnace fans, a 1/2 hp and a 1/3 hp (not sure about the other specs). No reason you couldn't add more fans if you're really determined to hit 50 Pa.

          Another option is to get as close as you can and then extrapolate to 50 Pa, as described here:

          If you just want to find leaks with a smoke pen, then there's nothing magic about 50 Pa.

  8. musielski | | #21

    Thanks for the great input guys! I'm late to seeing your input since my email notifications sent everything to my spam folder for some reason.
    I think this is going to be a great spring project to attempt!

    1. Expert Member
      BILL WICHERS | | #22

      Make sure to let us know as you make progress!

      Measurement Specialties makes some very easy to use pressure sensors that I highly recommend. Something like an Arduino would be able to read the analog data and digitize it. You may be able to make use of things like analog or PWM outputs from the Arduino to control fan speed too. I hope that maybe helps you out little in terms of a starting point for some of the control stuff if you want to try to automate the unit and put it under computer control.

      You could do a full-manual system too -- just get a pressure differential gauge, and use a portable anemometer as mentioner earlier to measure air speed. You'll have some calculator work to do to change units around to what you need, but you could still get good measurements with a manual system.


Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |