<blockquote><cite>Posted By: SteamyTea</cite>That is a brilliant and elegant solution. Now can you work out the height difference for 50Pa.</blockquote>
About 5.1 mm vertical head of water difference for 50 Pa, so about 10.2 mm water movement distance for a 30 degree inclined manometer.

I have a couple of these that use straight glass tubes that are about 2mm bore, fitted with cable ties to bits of white melamine coated chipboard, with home made printed scales behind the tubes. The scales have a black and white striped pattern of 1mm wide bands that runs behind the glass tubes. Water refraction causes the stripes to bend, so you can very clearly see where the level is. Mine are interconnected with a long length of 5 mm bore silicone surgical tube, as I used to use it as a long spirit level for setting up wing jigs.

50 Pa corresponds to the dynamic pressure caused by winds around 9 m/s (20 mph). I wonder if they were thought to be representative. Perhaps it was thought that higher pressures would open up additional leakage paths which would not be significant in normal operation.

If more than, say, 100 Pa would cause damage you'd want to know sooner rather than later. E.g., 50 m/s (112 mph) winds would cause pressures around 1500 Pa.

Pressure = ½ρv²

Pressure in pascals.
Density (ρ) in kg/m³ (about 1.2 to 1.3 kg/m³ for air around sea level).
Velocity in m/s.

Posted By: Chris Bacon... the disco generators produce far too much smoke to be effective when hunting for smaller leaks.

Ta for the info, I have a disco fogger on its way, and I'll invest in some pens too.

BTW, I was watching a few others on ebay, so if anyone is after it, my shortlist included ebay item numbers 140675499981 140675429901 290653707949 (just stick these in the search bar if you're after them - all end within the next day and a half tho'.

Cheers,

Tim.

Somewhat off-topic

> I might give it a try, after I've had a go at fixing the inverter in my deceased microwave oven!!

Do you mean inverter? They normally have a very simple voltage doubler with 4 components: diode, capacitor, magnetron and transformer.
http://deanostoybox.com/mot_power_supply.htm

Apart from the fact that it runs at best part of 1kV and the capacitor is meaty enough to do you proper harm so you need to excercise some care, they are not hard to fix due to being so simple. Either my capacitor or diode had died - I forget which. I've had it for 19 years now and it was s/h when I bought it. They probably don't make them like that anymore.

I was giving this some more thought over lunch, as this thread has prompted me to look at whether or not I could knock up a DIY blower door, with the right sort of controls and calibrated display to be used by anyone here who wants to borrow it.

I have some pressure sensors from another project (MPXV7002DP ones from Freescale, I used them in an electronic air speed indicator project a while ago). They are a bit marginal at 50 Pa, which is right at the bottom end of their range, but having just done a simple bench test it looks like they should be fine for this job if auto-zeroed at the start of each measurement run.

I've been looking at the way that blower doors work and it seems to me that directly measuring the velocity of the air coming out of the fan might be a better way of determining volume flow rate than measuring the fan motor drive current, although I'm going to pop down to my local scrap yard tomorrow and see if I can buy a fan and test the motor power/mass flow rate relationship. I have a little hand-held anemometer that's reasonably accurate so I may be able to use that to calibrate the fan motor power vs volume flow characteristic and then use the motor power measurement method (motor power should be easy enough to measure from voltage and current drawn).

The plan is to try and make an adjustable fabric covered frame (I have some old PU coated hot air balloon fabric that might do the job) a bit like the professional bits of kit, make up brackets to hold the fan in the centre of it, use the pressure sensor to measure the pressure differential and control the fan speed via a microcontroller and motor speed controller so that the pressure is held steady at 50 Pa. The same microcontroller should be able to display the volume flow rate in m³/hour from the motor power readings. To keep things simple I think it might be best to just leave it at that, so the reading can then be used together with the building measured volume to manually calculate ACH.

My front and back door openings are 730mm wide by 1980mm high, which I'm guessing may be a bit smaller than average, if anyone has any idea what the sort of normal range for these things is I'd be interested to know. I was thinking of making the frame adjust to clamp in the door frame using bits of threaded rod inside telescopic tubes, with the frame clamping the fabric to the door frame, so a fairly wide range of adjustment should be possible.

Can anyone see any massive flaws in this scheme, please?

<blockquote><cite>Posted By: Paul in Montreal</cite>I think real blower door test machines measure the speed of the fan to achieve a certain pressure differential and also use calibrated "rings" that set the airflow. I don't think they measure the motor current. I remember when my blower tests were done the fan changed speeds and the tester had to put in the correct ring into the fan's aperture to get readings that made sense. It's all non-linear anyway and there was some software in the controller that plotted some graphs and so on and came up with the final readings.

Paul in Montreal.</blockquote>

Thanks for that, Paul, that helps a lot.

Measuring fan rpm would be fairly easy, I could probably just use a Hall or optical sensor to count the blade passing rate, probably easier than measuring and smoothing fan motor power I think.

I can see why the restrictor rings would be needed, too, as the fan capacity would need to be adjustable over a fairly wide range to cope with the large variations in house volume. I've been working out the flow rate for my current house, which has a volume of only about 220 m³ and it's surprisingly low, even for fairly high ACH values. For ACH = 10 a 400 mm diameter fan gives a velocity at the fan outlet of only 4.86 m/S (assuming I have my sums right). If the fan was throttled down to 300 mm diameter then the flow velocity from the fan increases to 8.64 m/S.

I'll need to get hold of a fan and do some testing with it, as I'm guessing that there is only a modest range of flow velocities over which the fan speed to flow rate relationship stays reasonable with a 50 Pa pressure drop across the fan.

Thanks RobL, it was seeing the positive results you got from a seemingly simple set up that made me think about doing this. As you say, the big advantage is in being able to do lot's of testing to see how things improve.

In my case I'd like to try and get a calibrated result though, if only to see if it's practicable to do with a budget DIY bit of kit. I spent hours this summer trying to seal leaks in my house, and I think I've made a difference, but it would have been nice to do some tests to see just how big a difference I've made.

Having done a bit more thinking this afternoon I've decided to just use another one of the pressure sensors I have to measure the outflow velocity from the fan, using a pitot tube. Knowing the area of the fan aperture and the average velocity of the flow across the aperture I can work out the volume flow rate, which should then make getting a reasonably accurate ACH fairly straightforward.

Thinking about the physical construction of the thing I may just opt to make the fan unit and the electronics box and do pretty much what you did, make up a custom flat panel to fit it into. If anyone wants to borrow it it should be fairly easy then for them to get a bit of board and make their own panel to fit whatever window or door they want to stick it in.

<blockquote><cite>Posted By: Chris P Bacon</cite>

This is precisely my interest in this, as I intend to do my own air tightness taping and it would be nice to be able to test my work in some comfort rather than running around like a mad thing in the few hours that the "real" air tightness tester will be on site trying to identify and remedy any leaks.

One thing I don't want to do is cause any damage, I presume there is little possibility of that using a 12 volt car fan to de-pressurise a 200m2 house?</blockquote>

I may be able to answer that question soon. The first thing I want to do tomorrow is see just how much air one of these car radiator fans will shift. Knowing that should give a good idea of the sort of pressure differential they can pull. My guess is that it won't be massive and probably not enough to cause any damage.

I managed to dig some data out on Kenlowe radiator fans last night, really just to give an idea as to the sort of pressure drop and flow rate they were capable of and give me a guide as to the size of fan to go looking for in the scrapyard. I'm guessing that the Kenlowe fans are probably a bit better than standard radiator fans but close enough in performance for guidance. Kenlowe publish good data for the five sizes of fan they make, and very conveniently one of the pressure drop figures they use happens to be 5 mm of water, which is pretty close to 50 Pa. Here are the data for each fan:

6" diameter = 431 m³/hr @ 50 Pa and 4 A @ 13.5 V

8" diameter = 646 m³/hr @ 50 Pa and 4 A @ 13.5 V

10" diameter = 965 m³/hr @ 50 Pa and 6.2 A @ 13.5 V

13" diameter = 1595 m³/hr @ 50 Pa and 11.9 A @ 13.5 V

16" diameter = 3330 m³/hr @ 50 Pa and 20.8 A @ 13.5 V

These fans are still rated to pull at a pressure differential of around 200 Pa, which is a dynamic pressure equivalent to about 40 mph wind speed. There's a good chance that the fan could actually pull a bigger pressure differential than this, as the ~200 Pa figure is just the last rating point, not the fan stall pressure differential. The biggest fan can still pull over 1000 m³/hr at ~200 Pa, which would be around 4.5 ACH for my current house; if my house is better than 4.5 ACH the chances are the pressure at full speed might be higher than 200 Pa. I think some caution might be needed when choosing the fan size and operating speed.

I've just come back from the local breakers yard and the best I could find was a fan from a Peugot 206. This is around 350mm in diameter, so about the same as the 13" Kenlowe. I'm hoping that this will be OK for flow rates of from maybe 100 m³/hr up to maybe 1000 m³/hr at 50 Pa, based on just looking at it and extrapolating from some fan curve data I found in some old hovercraft stuff I have.

Most of the radiator fans I looked at were about the same size, I couldn't find any bigger ones, but the yard I went to didn't have much in the way of big 4 x 4s or vans etc in stock. I picked the 206 fan simply because it looks easy to mount, with just three bolts around the edge and no odd-shaped frame, plus the blades are nicely shaped (very like those on the Kenlowe fans) and sealed to the edge of the frame/duct, so it'll probably have a low leakage at the blade tips.

If it does indeed work over the flow range that I hope it will, then that should be OK for houses between 200 and 2000 m³ and an ACH of 0.5 at 50 Pa. If it turns out that the fan is too powerful I think I'll just follow Paul's tip and reduce the effective area a bit.

<blockquote><cite>Posted By: Ed Davies</cite><blockquote><cite>Posted By: JSHarris</cite>...use the pressure sensor to measure the pressure differential and control the fan speed via a microcontroller and motor speed controller so that the pressure is held steady at 50 Pa.</blockquote>

I think that might be a bit of over-engineering. From what I read of the normal test procedure ages ago (sorry, can't remember where) the procedure is not to try to control the fan that accurately but rather to test at a couple of fan speed settings which give pressures above and below 50 Pa then interpolate.</blockquote>

Thanks, that might make it easier, as it removes the need for a control loop (which would probably need a PID algorithm to work smoothly). If that's the case, then I could just have a manual fan speed control, with a pressure display and a display giving m³/hr.

Does that sound reasonable?

Thanks, I shall crack on and see how a simpler one works.

Sounds like it should work OK with the loft windows open, as they would cause a negligible pressure change I'd have thought. The loft hatch sounds like an easier place to fit the board with the fan on, too.

Posted By: TimSmallI was wondering if anyone used any of these for leak testing (or knew anywhere to get one in the UK)?

No, but seeing that reminded me of various vapour/mist producing devices that I were, I think, listed in the last pre-Christmas Maplin sale brochure?

A quick search on Maplin fould various "foggers'" which apparently provide an "atmospheric cool mist effect to water features". There was also "Smoke Spray in a Can"!

Maybe someone has tried one of those approaches?

Posted By: TimSmallor knew anywhere to get one in the UK

at less than the £70 typical asking price that is...

Tim.