Radon – Part 2 – The ERV

Several months ago, I posted about the radon problem that I discovered, and how I solved it.  Shortly after writing that post, I was contacted by Marc Rosenbaum, Director of Engineering at South Mountain Company,  a design/build firm in West Tisbury, MA.  Marc was interested in the radon issue, and offered a thought that had not been raised by anyone else I had spoken to. In short, he suggested that an imbalance in my ERV could possibly have been the cause of, or at least a contributing factor to, the relatively high levels of radon in my house.

His theory went something like this…The ERV is supposed to be balanced; it is supposed to draw in the same amount of air as it expels.  When that is the case, the interior of the house is neither positively nor negatively pressurized (with respect to the outside air).  It is neutral.

If the ERV is drawing in more air that it is expelling, the interior of the house will be in a state of positive pressurization.  In this state, the excess inflow of air will try to find a way out through any gaps that are present in the building envelope because it cannot all get out via the ERV exhaust duct.

But if the ERV is expelling more air than it is drawing in, the interior of the house will be in a state of negative pressure.  In this instance, the ERV is going to pull air into the house through whatever cracks/gaps exist to make up the difference between the smaller volume of air coming through the ERV intake and the larger volume of air moving through the ERV exhaust.  Since some of these entry points are inevitably going to be in the foundation, if the soil contains radon the incoming air is going to bring that radon with it. And this is essentially what Marc thought might be happening.

I have to admit that it was difficult for me to believe this could happen.  My house tested out at .3 ACH @ 50 pascals.  I saw how the basement slab was constructed and sealed.  I just couldn’t imagine that an imbalance in the ERV could suck enough (i.e. any) air through the foundation to make the radon level increase.  I assumed that radon particles are so small they can move through things that air cannot.

Marc then made the idea more mind boggling by saying that the radon fan might be contributing to the negative pressure situation even as it solved the radon problem.  This also made no sense to me.  How could that little fan – one with the lowest airflow I could find – possibly be powerful enough to suck so much air through minuscule gaps in the foundation that it would change the internal pressure in the house? It just didn’t seem possible.

Well, as I have come to learn, Marc appears to have been correct on all counts.  The ERV was out of balance and causing the house to be negatively pressurized.  That negative pressurization was causing the radon problem.  And the radon fan was increasing the negative pressure even as it solved the radon issue.  It was doing that by sucking air out through the gaps in the foundation floor, which in turn pulled air into the house through whatever gaps are present above the foundation floor.  This of course isn’t good, because none of that air is filtered. But on the upside, at the same time, the radon fan was pulling the radon from the soil under the house and expelling that radon to the outdoors, thereby reducing the radon level in the house.

To determine whether Marc’s theory was correct, Marc suggested that I check to see if the house and the ERV were balanced by using a manometer, like the one pictured below:


The manometer measures the relative difference in pressure between the interior of the house and the outside air.  Basically, you do this by running a flexible tube from one valve on the manometer to the outside, and leaving a second valve on the manometer open to the inside of the house.  The manometer does the rest, with the results displayed in “Pascals.”  Ideally, if the ERV is balanced, the manometer should display a reading that will probably fluctuate, but stays close to zero (The calmer the wind outside, the more stable the reading will be), thereby indicating that the pressure in the house is the same as the pressure outside the house.

With (once again) the help of my Passive House Rater, I was able to take measurements with the radon fan off, the ERV off, both off, and both on.  This enabled me to see the effect that each was having on the interior house pressure.  The reading we got supported Marc’s supposition. The manometer showed that the house was negatively pressured by eight to ten Pascals with both the ERV and radon fan running.  Most of this (about five to six Pascals) was due to the ERV.  About three to four Pascals were due to the radon fan.  So, both the ERV and the radon fan were pulling air into the house through the cracks and gaps that were simply too small to find and close.

This led to the question of “why?”  Why wasn’t the ERV balanced?  The purchase price for a Zehnder ERV includes a $500 fee for commissioning (i.e. balancing).  Zehnder mandates that their own person commission the system because they want it done correctly.  Well, it appears that the problem with my house wasn’t the person who did the work, it was the equipment that he used.

The Zehnder ERV that I installed in my house has ten individual interior supply ducts (which bring fresh, filtered air into the house) and ten individual interior exhaust ducts (which expel stale air from the house).  Those two airflows need to be the same for the system to be in balance.  [Note: This is different than interior supply and exhaust vents.  In my house, there are eight supply vents and six exhaust vents.  Some vents have one duct attached (e.g. the powder room exhaust vent), some have two ducts attached (e.g. the main floor supply vent), and one has three ducts (i.e. the kitchen exhaust vent).]

To balance the system, the tech person went from room to room and opened or closed each vent until a pre-determined amount of air flowed through it.  This was done by checking the flow through each of the vents, one at a time.  Since adjusting the flow at one vent can change the flow at another, the complete process must be performed a number of times.  So once all eight supply vents were checked and adjusted, the tech started at the beginning and checked/adjusted them a second time, tweaking each to get it as close to perfect as possible.  And then he checked them a third time.  In fact, I think he ended up checking them five or six times before he was satisfied. The same went for the six exhaust vents.   The process was completed when the tech felt that he had each vent adjusted as close to perfect as possible.  And at that point, he concluded that the system was balanced, with (in my case) about 120 cfm being supplied and 120 cfm being exhausted when the ERV is set to medium speed.

The problem is that there is a margin of error in the equipment used to measure the flow at each vent, and some equipment has a greater margin of error than other equipment. More specifically, Zehnder used a rotating vane anemometer on my house.  It looks like this:


One evaluation of that equipment indicates that it could result in an error of up to 25%:

Download (PDF, 2.05MB)

That margin of error is further complicated by the fact that the anemometer display never stays static.  Rather, it bounces around a bit.  For example, it may be that the best that can be done is get a reading that bounces from 11.1 cfm to 13.3 cfm for the single duct powder room exhaust vent, leaving the operator with the task of interpolating the results.  In that instance, it might be fair to say that the results look like 12 cfm, on average.  But who knows what the actual reading really is?  And at airflow levels as low as that, being off 1 or 2 cfm each time you check a vent can add up.

And therein lies the problem.  If the equipment is inherently inaccurate and the tech is off in his interpolation of the data, it is not difficult to see how the ERV could end up being unbalanced.  That is essentially what appears to have happened at my house.  But I digress…

After seeing that the house was negatively pressurized, my Passive House rater, who has been a big help to me throughout the entire building process, measured the flow at the exterior vents using a more accurate devise known as a “power hood.”  Although the wind was preventing us from getting a steady reading, we estimated that the intake was bringing in 10 to 20 cfm less than the exhaust was expelling.

Unfortunately, Zehnder doesn’t check the overall flow at the exterior vents, primarily (they say) because those vents often are not easily accessible (although on my house they couldn’t be more accessible) .

Fortunately my Passive House rater helped me use his equipment to rebalance the system.  Essentially, what I ended up doing was opening up all of the supply vents and slightly closing a couple of the exhaust vents.  By monitoring the effect these changes had on the house pressure with the manometer, I was able to bring the interior of the house in balance with the outside air.

While doing that, I found that the balance and pressure are also affected by the speed at which I run the ERV fan.  When the ERV is set on “low” speed (i.e. Level 1), the house is now pretty much neutrally pressurized, with a reading that fluctuates between -1 and +1 Pascals, and probably leans more toward the +1.  But when set on medium (i.e. Level 2), the house is clearly positively pressurized, with a reading between +1 and +2 Pascals.

Once I completed the rebalancing, I waited several days to see what effect it would have on the radon level.  Prior to rebalancing the ERV, the radon fan had brought the radon level from about 6 pCi down to about 1pCi, and within three days of unplugging the radon fan (which I did a couple of times just to check), the level consistently rose back to 6 pCi.  But this time was different.  With the radon fan off and the ERV on low speed, the radon never rose above 1.4 pCi, comfortably below the mitigation threshold of 4pCi.  When I set the ERV to medium speed, the radon level dropped further (although, so far, never to zero).

One final note…All of this occurred this spring, and we rarely use air conditioning.  So as it turned out, we started opening windows a couple of weeks after the rebalancing.  Some days we only opened the first floor windows.  Some days only the second floor windows.  And some days both the first and second floor windows.  Also, on some days, we opened a window (or windows) for only a short period of time.  At other times, we left windows open all day or for several days at a time.  But regardless of the number of open windows or the amount of time they were left open, it soon became clear that opening windows had an effect on the radon level (which we measure in the basement).  The longer a window or windows were open, the higher the radon level rose.  The up-tick in the radon level is very slow, but given enough time (i.e. days) the level eventually rises as high as 2.4 pCi (still not bad).  And if the windows are left closed, the radon level slowly comes back down to about 1 pCi.  Subsequent testing with the manometer showed that opening even one window threw the house into a slightly negative pressure situation.  Whether its the “stack” effect, the wind, a combination of both, or something else, I don’t know.  But the effect seems to be consistent.

With regard to the stack effect, even on a hot day the temperature on the second floor is no more than one degree higher than the first floor.  But the temperature difference between the first floor and the basement is more substantial.