Jump to content

DIY Airtightness test (the Blue Peter method)


ProDave

Recommended Posts

Something I have been wanting to do for a while is see how air tight I have managed to make the house, but without paying £hundreds for an official test.

 

So take one old desk fan. Some bits of wood. A large piece of cardboard and a roll of duct tape.

blower_1.thumb.jpg.19aed54cf46ac366f35dd9af9db0e9f3.jpg

 

The first thing was just to run the fan on full speed and go around the house looking for leaks. I found a leak under the front door where it was not sealed to the floor properly. A leak in the loft at a tricky detail between the ridge beam and the OSB cladding, and a leak around where the mains comes in and cables go out under the floor.  All fixed with a bit more tape and detailing.

 

So now I have a house where there is no perceivable leaks coming in anywhere. The flow going out of the fan seems very small, and it's certainly holding a good pressure (if you open a door you can feel the rush of air enter, and hear the fan note change as it is no longer working hard)

 

Time to try and measure something?

 

Firstly the pressure.  Normally an air tightness test is done at 50Pa.  It was only when I looked up the conversion I realised that's a tiny pressure, about 2mm water gauge.  So I set up my manometer, one side connected to "outside" the other side vented to the room.  It's very hard to accurately measure 2mm on a U tube manometer, but I am sure it was somewhere between 2 and 3mm. So in the right order.

 

How to measure the airflow?  Simple. Just measure the air flow rate through the fan.  Well my little anemometer would not register the wind speed entering the fan, it was too low to get it's impeller turning. So I had to do something to increase, the airflow.  Decrease the area, AKA the "flower pot" principle.

 

I didn't have a 15" flower pot, so I made one.

 

blower_2.thumb.jpg.045b8b608da620190860a360dc450873.jpg

 

The hole in the "top" is 110mm (because a bit of drain pipe was handy to draw around)

 

Now the air flow was high enough to measure. 2.9 metres per second.

A 110mm diameter hole has an area of 0.038 M2 So that gives a volumetric flow of 0.11M3 per second

That's 6.6 M3 per minute or 396M3  per hour.

 

Now my house has a volume of 480 M3   So that equates to 0.825 Air changes per hour.

 

I'm actually a little disappointed with that. I had hoped for better. But let's not kid ourselves that this was in any way an accurate measurement.  At least it enabled me to find and fix a couple of air leaks, and I am now reasonably certain there are no air leaks of significance.

 

Feel free to tear the method or results to pieces.

 

  • Like 7
Link to comment
Share on other sites

Guest Alphonsox

Excellent - Val would be proud of you.

I guess the pressure measurement is the greatest area for inaccuracy. Have you thought about using an inclined tube for the manometer ? Should give a greater deflection versus pressure difference compared to the vertical version.

Link to comment
Share on other sites

10 hours ago, MikeSharp01 said:

Great work Dave. How did you detect the leaks? 

Just went round feeling for them, a very long and tedious process. I was checking all joints, corners, windows, doors, cable and pipe entries.  It is surprising how sensitive an almost bald head is at detecting a very small leak. Although the pressure difference seems small, you get a very noticable draught through even a small hole and in the case of a window, if you open the trickle vent in a roof window you get a howling gale that you can hear and feel easily, yet with the vent closed you can't feel or hear a thing.

 

My main concern was to ensure that all the membrane and joints that are soon to be covered with plasterboard were leak free.

 

In many places the membrane between the battens was blown up, inwards, showing an air leak in the fabric of the building that the air tight membrane was catching.

  • Like 3
Link to comment
Share on other sites

Have you not thought of using a smoke machine ? Know any local DJ's you could hire one off ? That would show you instantly any further minor leaks, multiples of which would bump your results south ;)

Failing that a can of smoke detector check spray ?

Link to comment
Share on other sites

3 minutes ago, Nickfromwales said:

Have you not thought of using a smoke machine ? Know any local DJ's you could hire one off ? That would show you instantly any further minor leaks, multiples of which would bump your results south ;)

Failing that a can of smoke detector check spray ?

 

Just get a £1 eCig and some cheap glycerin and then make your own smoke machine ...

Link to comment
Share on other sites

Marlboro ?  They bloody STINK ?

When I worked around the world on cruise liners, I shared a cabin with my mate Luke and he woke and sparked up a Marlboro light like it was religion, and it fackin stank the room out. As we were contractors we got the cheap, small, in-board cabins with no window and minimal a/c. Great. 

Link to comment
Share on other sites

Looks like I ran the test at too low pressure. I did lookup the conversion some time ago on a web based converter and I am sure it told me 2mm not 5. So my test was probably only at half the required pressure for an official test.  I was surprised that this little fan could generate enough pressure, it is far smaller than those used on proper blower door test rigs. That in itself I thought was a good sign. It now looks as though that fan would not get to 5mm so a bigger fan would be needed.

 

To try this with an mvhr you would have to block either the inlet or outlet to try that test, otherwise even with one fan stopped the air would pass through the stationary fan.

 

I still think it was a useful test even if no meaningful figures can be taken from it.

 

I never got a Blue Peter badge as a boy. I wonder if This now qualifies me for one?

  • Like 1
Link to comment
Share on other sites

Guest Alphonsox

How about using the wind instead of a fan ? 50Pa is around 20mph if I remember correctly. 

Link to comment
Share on other sites

2 hours ago, Alphonsox said:

How about using the wind instead of a fan ? 50Pa is around 20mph if I remember correctly. 

 

I didn't do an airtightness test in the end- we get plenty enough windy days here and I found it was easy enough to spot any leaks as I went around the building. My winds were more like 60mph than 20, so I might have over-engineered this aspect...

Link to comment
Share on other sites

Guest Alphonsox
33 minutes ago, Nickfromwales said:

Have you been drinking? ??

Possibly..... It's a sunny Sunday afternoon and the BBQ is out and operational.....

To be a little clearer on my previous post.

Our build was air tested to near Passiv spec at 50Pa (20Mph). Over the winter things got a little gustier. At 40Mph (200Pa) we found a few more leaks, at 50Mph (300 Pa) a few more. At 60Mph (450 Pa) the anemometer gave up and we stopped looking. You don't need a fan to get things airtight, you just need the right weather. 

Link to comment
Share on other sites

Yes it often gets windy here. This June has been the windiest I can remember.

 

One thing that is apparent is the old saying, in an air tight house, you can open one window or door with no effect.  That has certainly been proved true. It can be blowing a hoolie and open one door, and you feel no draught, and the door does not blow shut (or open) it just stays there.  Had I done that before I sealed up the last bit of floor, the door would have blown shut or open with the through draught. 

Link to comment
Share on other sites

1 hour ago, Alphonsox said:

Possibly..... It's a sunny Sunday afternoon and the BBQ is out and operational.....

To be a little clearer on my previous post.

Our build was air tested to near Passiv spec at 50Pa (20Mph). Over the winter things got a little gustier. At 40Mph (200Pa) we found a few more leaks, at 50Mph (300 Pa) a few more. At 60Mph (450 Pa) the anemometer gave up and we stopped looking. You don't need a fan to get things airtight, you just need the right weather. 

A bit tricky as Daves trying to eliminate areas before boarding. I had a vision in my head of dave running down the road at 20mph holding his house above his head like a kite :S :ph34r:

Swansae air show today, maybe I've got sunstroke ¬¬. Enjoy the ? ??

Link to comment
Share on other sites

.

3 hours ago, Alphonsox said:

 

Our build was air tested to near Passiv spec at 50Pa (20Mph). Over the winter things got a little gustier. At 40Mph (200Pa) we found a few more leaks, at 50Mph (300 Pa) a few more. At 60Mph (450 Pa) the anemometer gave up and we stopped looking. You don't need a fan to get things airtight, you just need the right weather. 

 

I didn't bother with an air test, just waited for a windy day, which we of course, get plenty of. As @Alphonsox says, air leaks are very easy to detect when wind speed goes north of 40mph

Link to comment
Share on other sites

The dynamic pressure of the wind speed doesn't match the test pressure for the blower test, for a few reasons. 

 

The first is that the wind is not stopped dead by the house, so not all the dynamic pressure is converted to static pressure on the windward side of the house, only a small part of it.  There's a corollary here with the Betz Limit and wind turbines - if you extract all the energy from the moving airstream (which is what full conversion of dynamic to static pressure implies) then there will be a stagnation zone on the windward side and the house will be absorbing all the kinetic energy in the wind.  Clearly this doesn't happen, and the wind finds a way to flow around and over the house, so the dynamic pressure on the windward side will be a great deal lower than that implied by the wind speed.

 

The second significant effect comes from Bernoulli's Principle.  Because the wind has to flow around and over the house, the distance it has to travel is greater than if the house wasn't there.  Downwind of the house the wind individual streams will join up and flow at the same velocity again, which implies that the air flow velocity over the house will have increased, so lowering the dynamic pressure.  The effect of this is to reduce the pressure above and around the house, and on the leeward side, so there is an increase in pressure differential across the house, from the windward to the leeward side.

 

However, because there is still a substantial air flow velocity downstream of the house, it's clear that, as above, the house has not extracted all, or even a significant proportion, of the energy in the wind.  Working backwards this implies that the actual pressure differential that the house experiences, from one side to the other, will be a lot lower than 1/2 ρv² would suggest.  If I had to guess, I would say that the worst case might be around 30% of the dynamic pressure given by 1/2 ρv² as the highest pressure differential the house would see in practice, but this will vary a great deal from one house to another, depending on the location, local terrain shape of the house, wind direction etc.

 

On the topic of home brew pressure test fans, I made one up a few years ago to try and improve the airtightness of our old house, or at least help me find the biggest leaks.  Because the air flow requirement for the fan is small, but the pressure differential is relatively high, I chose a forward sweep fan, as these tend to cope with a higher differential pressure without running into serious blade stall.  I found a suitable fan in a scrap yard, a big car radiator cooling fan, but when I tried it I found that the brushed motor it came with wouldn't work well over a wide speed range.  I ended up replacing that motor with a brushless DC motor and speed controller, which allowed pretty fine control of motor speed.  Here's a couple of photos of the thing (it was made to fit tightly into a open window):

 

5959fe5dbdfe1_Blowerfan1.thumb.JPG.6f135bb9535a51b7c2a199bdd556e3af.JPG

 

5959fe60943db_Blowerfan2.thumb.JPG.afc3b4ff54e82f397aa0c126515af8f6.JPG

 

The speed control knob allows the fan to run right down to a very low speed, to reduce the airflow rate.  To measure the differential pressure very approximately I made up a sloping U tube manometer, with the sloping tube set at 30 deg, in front of a bit of paper printed up with stripey lines.  The stripey lines make it easier to see the exact position of the water in the tube, as it refracts light and makes the lines look kinked.  Even so, getting 50 Pa was a very rough and ready thing, and I doubt I got closer than around 20% accuracy.  I did look at buying a low pressure Magnahelic differential meter, but they were a bit expensive for what was really just a rough and ready test.

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...