# Am I understanding the concept of suction correctly?

1. Oct 29, 2012

### spacediver

My understanding of suction is that it is a phenomenon describing the flow of fluid from a (relatively) high pressure volume into a (relatively) low pressure volume.

This makes sense to me, as I grasp that air pressure is a measurement that describes a large number of molecules, each with its own momentum, and that the higher the mean momentum, the higher the air pressure. Accordingly, when you introduce a volume that has a relatively low air pressure relative to its surrounding volume, and there is no barrier between these two volumes, the molecules in the higher pressure system will tend to distribute themselves to fill in the "gap". To me, this is statistically similar to the phenomenon of increasing the homogeneity of a cake batter by stirring it enough.

What I'm having difficulty understanding is how a vacuum cleaner is able to generate so much suction.

My understanding is that the maximum suction is a function of the difference in air pressure between the two "volumes". Even if a vacuum cleaner creates a perfect vacuum internally, surely the amount of suction force is limited by the air pressure of the environment "outside" the vacuum cleaner.

It also would seem that the area of the channel connecting the vacuum to the outside world plays an important role, similarly to how pressure is a function of the area over which a force is distributed. Perhaps the length of the channel also plays a role, but this is more speculative on my part.

So here are some questions to test my understanding:

1: Suppose you have a constant "external" air pressure, and you're using a vacuum pipe with a fixed diameter and length. Will the maximum suction physically possible be experienced when a perfect vacuum is created inside the cleaner?

2: Are there any other sort of mechanisms that can provide for even greater amounts of suction than that referenced in question 1?

3: What is a useful way of formally quantifying "suction"?

thanks!

2. Oct 29, 2012

### Staff: Mentor

That is correct. 14 psi = 1 bar = air pressure relative to vacuum at sea level is a lot of pressure. If you could pump all of the air out of a building it would collapse.

Again, you are right. See the answer to #3 below

1) Yes, that's about as good as it gets.

2) No. (It's possible to do a bit better under very special circumstances when the suction is applied for only a very short time and the air is moving fast enough as it is sucked in. High-performance automobile engines can use this effect to draw a bit more air into the cylinder during the few milliseconds that the intake valves are open. This effect is completely unhelpful for a vacuum cleaner or anything else that is generating steady flow).

3) The standard measure is volume of air moved per unit time at a given pressure drop, something like "500 cubic feet per minute at 3 psi" for example. This is also why the size, shape, and length of the intake matters. It's easy to maintain a very high pressure drop if you only let a small volume of air through, but then your vacuum cleaner isn't going to be picking much up.

3. Oct 29, 2012

### spacediver

thanks nugatory, those answers really consolidated my understanding!

4. Oct 29, 2012

### spacediver

very interesting, and that certainly gives a vivid impression of the forces involved.

Here's a somewhat related question then:

If the amount of force involved in a differential of 1 bar of air pressure is enough to implode a building, then this means that a building in equilibrium is essentially experiencing that level of force from both outside and inside (and the same can be said for our bodies).

But surely that means that any given membrane in a pressure equalized system of 1 bar should be experiencing catastrophic compression. I'm having trouble squaring the compressive resilience of ordinary matter at sea level with the idea of a building caving in!

5. Oct 30, 2012

At one atmosphere of pressure water will compress in volume by only 46.4 parts in one million, steel a factor of 70 less. Anything that is not a gas is surprisingly rigid as long as the pressure from all sides is the same. Why do you think there are animals on the bottom of the ocean where the pressure is 4000 metric tons per square meter.

6. Oct 30, 2012

### spacediver

I still don't understand how a force that could implode a building, under a pressure differential of 1 atmosphere would not cause catastrophic compression when both sides are in equilibrium.

Put it this way:

Imagine you had a giant vise and placed a building in between its grips. Assume the air pressure inside and outside of the building is the same. You then tighten the vise until the building implodes. If I am understanding Nugatory's post correctly, this amount of pressure that the vise is applying is equivalent to 1 bar of pressure.

Now if a building is in equilibrium (forget the vise for now), this means that the building is experiencing that huge amount of pressure, but on both sides of each surface.

How come if you take a piece of foam and just place it anywhere on earth, it doesn't immediately compress to a tiny thickness? If i were to place that piece of foam in a vice and apply the same amount of pressure as I needed to implode the building, the foam would surely be severely compressed.

I know something is wrong with my thinking, but I can't quite figure it out.

7. Oct 31, 2012

### JustinRyan

If you put the foam in a bag and sucked all the air out of the bag, it would compress to a much smaller size.

8. Oct 31, 2012

### A.T.

Which way should it deform, if both sides are in equilibrium?

Because the pressure in the foam is the same as outside the foam.

9. Oct 31, 2012

### torquil

A building, which is almost hollow, is actually incredibly frail compared to a solid material like one of the bricks that make up the wall of the building, when acted upon with a given pressure on the outsides, e.g. using your hypothetical vice.

10. Oct 31, 2012

### CWatters

So a small building with walls say 20ft x 8ft (and a vacuum inside) would have a force of about

14 x 20 x 8 x 144 = 322,560 lbs

acting on each wall.

11. Oct 31, 2012

### CWatters

That's because it's not just "both sides" that are in equilibrium. For example in a cavity wall there is also air in the wall at the same pressure. Where there isn't air (eg in the middle of a brick) the material the brick is made of brick is also under pressure and "pushes back". In other words there is no pressure differential to cause a collapse.

That doesn't mean buildings don't deform under the pressure.. look at a double glazed sealed window unit. It was made with the gas inside at a certain temperature and pressure. When the weather changes the pressure inside the unit changes relative to the pressure outside/inside. You can sometimes see the glass has gone concave, compressing the unit due to the pressure differential.

However something like a brick doesn't have to be compressed much before the pressure differential between the inside of the brick inside and outside the brick is equal. So the deformation is incredibly small. The brick was also made under 1 atmosphere to start with.

12. Oct 31, 2012

### sophiecentaur

That would need to be a pretty special building; hermetically sealed. At the rate atmospheric pressure changes, the pressure inside and out would equalise even by the air passing in or out of the keyhole.

The excess pressure of a supersonic shock wave blew out the windows ("mysteriously") in an airfield building during early test flights, so I heard on TV - so it must be true haha. But that was a pressure change in a very short time. [Edit - also see the effects of a brief drop in AP inside a tornado column]

Closed cell foam will change dramatically volume when placed in a decompression (hyperbaric) chamber. I remember a great kids' Science programme with balloons, cream and marshmallow giving very impressive results. If the mass of gas inside a cell is not allowed to change, then the volume will change with pressure.

13. Oct 31, 2012

### spacediver

If I understand Cwatters correctly, he was referring to the pressure inside the window unit itself relative to outside the window unit (double glazed window remember).

14. Oct 31, 2012

### spacediver

I guess that makes sense - i had failed to consider that all that force gets transferred to a finite number of joints in the building, which is where the structural vulnerabilities lie.

15. Oct 31, 2012

### spacediver

By equilibrium, I meant forces acting on the inner wall match forces acting on the outer wall. But I suppose there is another way of thinking about equilibrium: when the forces acting within the bricks against the air match the forces without the brick against the brick.

16. Oct 31, 2012

### AJ Bentley

I gotta say, despite the number of people who seem to be happy with it, suction as a concept has no place in physics.
You can talk about negative pressure if you like - but suction is a lay term more properly applied to house-wifery than physics.

The force that moves fluid particles around comes from mutual repulsion and collisions. There isn't a 'suction force'. It's just woolly thinking.

17. Oct 31, 2012

### spacediver

Interesting point AJ.

I think we'd both agree that it can be useful (both practically and theoretically) to employ a construct that efficiently describes the phenomenon. Perhaps suction isn't the most ontologically sound one, but surely you're not suggesting we revert to Newtonian mechanics every time we wish to describe/predict the tendency for matter to flow from a high to low pressure system. Given Nugatory's description of the way that "suction" is measured:

I'm not sure that the term "negative pressure" will adequately serve this purpose.

Last edited: Oct 31, 2012
18. Oct 31, 2012

### sophiecentaur

Oh yes, I see - very much hermitacally sealed. I have been thinking how much you could expect windows to bow inwards or outwards. You could expect a pressure range of, say 20% between extremes of high and low. That would represent 20kPa of pressure variation at sea level (+/- 10kPa around an average value). If the gap in the glass were 4mm then you could imagine each sheet could move by 0.2mm (Boyle's Law and assuming the sheets were frictionless pistons) If they were very flexible and fixed around the outside, you would get about twice this movement at the centre (0.4mm). The radius of curvature of such a mirror would be, I estimate, about 600m - giving a focal length of 300m. As it is, glass is fairly rigid so the flexing would be quite a bit less (half / quarter??) than that. I wonder just how much this amount of curvature would actually be visible. It's hard to tell. Perhaps distortion of images of nearby buildings would,in fact, be visible. We're quite sensitive to that sort of thing.
I'd expect a similar effect due to a +/-10% variation in temperature (60oC in 300K) - also significantly reduced by the rigidity of the glass.

I think my back of fag packet sums are reasonable.

19. Oct 31, 2012

### Staff: Mentor

It works better than you'd expect at first, because a pressure drop is the difference between two pressures, so doesn't care about the absolute number assigned to either pressure. I can think of a 4 psi drop as 14 psi outside, 9 psi inside, or as 0 psi outside, -4 psi inside, and I get pretty much the same flow behavior.

20. Oct 31, 2012

### AJ Bentley

The reason I have a 'down' on the idea of suction is because of a much admired Chemistry master at my school (many, many decades ago).
He would hand someone a test tube full of water with a bung and tube in the top and instruct them to suck the water out.
It's impossible of course - "There's no such thing as a 'suck'" he would say.

He was quite right - the idea is both unnecessary and fallacious.

21. Oct 31, 2012

### spacediver

But based on your previous description of quantifying "suction":

it seems that the volume of fluid moved per unit time is partially independent of the pressure drop. And if this is the case, then pressure drop alone isn't enough to quantify the phenomenon in question - we would presumably also need information about the area and length of the channel connecting the two volumes.

22. Oct 31, 2012

### Staff: Mentor

Yes, you need both the flow and the pressure drop... I was just observing that the "negative pressure" concept doesn't have to interfere with calculating the pressure drop.

23. Oct 31, 2012

### spacediver

As for "suction", I suppose it's better to speak of rate of flow, given a host of variables, pressure differential being one of them.

24. Nov 12, 2012

### A.T.

"Negative pressure" can mean different things:
http://en.wikipedia.org/wiki/Pressure#Negative_pressures
For the latter case the terms "pull" or even "suction" seem appropriate.

Try to explain how trees get water 100m high, using only repulsion and collisions.