Why do shower curtains blow inwards?

[alw34]

[Moderator's note: spin off from another thread on a different topic.]

"why shower curtains blow inwards..."

This phenomena is likely caused by warm air [due to hot water spray] rising and pulling in cooler air from the room. It's the same idea as cooler denser air being closer to a floor, warmer air closer to your ceiling. I would not ascribe this phenomena to 'gravity' except to note that's why the curtain is hanging vertically in the first place. Without gravity, it would be floating around as you may have seen in pictures inside a space station where pencils float. Why things float that way in a space station IS an interesting phenomena.

 

[PeroK]

"why shower curtains blow inwards..."

This phenomena is likely caused by warm air [due to hot water spray] rising and pulling in cooler air from the room. It's the same idea as cooler denser air being closer to a floor, warmer air closer to your ceiling. I would not ascribe this phenomena to 'gravity' except to note that's why the curtain is hanging vertically in the first place. Without gravity, it would be floating around as you may have seen in pictures inside a space station where pencils float. Why things float that way in a space station IS an interesting phenomena.

It's why my shower curtain sometimes leaks and sometimes doesn't that I can't figure out!

 

[victorhugo]

"why shower curtains blow inwards..."

This phenomena is likely caused by warm air [due to hot water spray] rising and pulling in cooler air from the room. It's the same idea as cooler denser air being closer to a floor, warmer air closer to your ceiling. I would not ascribe this phenomena to 'gravity' except to note that's why the curtain is hanging vertically in the first place. Without gravity, it would be floating around as you may have seen in pictures inside a space station where pencils float. Why things float that way in a space station IS an interesting phenomena.

Nope. That has been tested and it is not reason. Also, like I said, if instead it was density and buoyancy that caused objects to stay on the ground instead of gravity, we could explain why the curtain blow inwards. A friend who is doing his masters in engineering explained this to me and it made perfect sense, I just don't remember any details now. I'm more of a biology person who is just interested in other fields of science, so I don't exactly know how.

 

[rbelli1]

Actually it is in large part from a vortex created by the water spray. The center of the vortex causes a low pressure area. Try spraying cold water. If the effect was only from convection from the temperature difference you would expect the curtain to bow out from cold water due to a cooling of the air. In fact even cold air causes the curtain to be pulled in although to a lesser extent.

I can't remember where I first read about this but I tested it and it worked as described.

BoB

 

[NascentOxygen]

I think it's the falling water dragging air downwards with it that causes a low pressure region that the curtain is drawn towards.

 

[russ_watters]

Alw is correct. It is just convection:

Some people even make mini cooling towers with shower heads:
http://www.overclockers.com/nuclear-tower-water-cooling/

 

[russ_watters]

I think it's the falling water dragging air downwards with it that causes a low pressure region that the curtain is drawn towards.

You can see from the rising cloud (and feel from the cold ankles) that the airflow direction is up.

 

[sophiecentaur]

Years ago I found a solution to the problem. Spray hot water all over the curtain. That makes it heavier and less prone to being moved and, even when bits of it touch you, they are not icy cold. Also, if you wet it well, you can make it stick to the base of the shower tray and that can stop the bottom from being pushed in.

 

[NascentOxygen]

It's a case of convection being the obvious answer, but wrong. The curtain is drawn in even under a cold water shower.

The air drawn down with the spray leaves a low pressure region around the shower-head, so once the water and air separate on the floor the air makes a rapid return upwards outside the region of falling spray. As air speeds upwards past the curtain the low pressure (Bernouli principle) draws the curtain inwards. The result is a horizontal-donut-shaped air circulation pattern, a vortex in the shower cubicle: air and spray cascading down in the centre, and air returning upwards outside the cascade.

Wikipedia notes that someone was awarded the Ig Nobel Prize in Physics in 2001 for his contribution to the answer of why shower curtains billow inwards.

 

[Chestermiller]

In addition to the mechanism described by Nascent, there is also the effect of the falling water dragging entrained air along with it toward the base of the shower. This creates a higher pressure at the base of the shower than near the shower head. This cases air to flow down the drain along with the water. Make-up air is required to flow in from outside the shower stall.

 

[anorlunda]

https://en.wikipedia.org/wiki/Shower-curtain_effect

That article lists the following theories:
1Theories

• 1.1Buoyancy theory
• 1.2Bernoulli effect theory
• 1.3Horizontal vortex theory
• 1.4The Coandă Effect
• 1.5Condensation
• 1.6Air Pressure

It sounds to me like interpretations of QM. Everyone has their favorite theory, but proving which is the right one is difficult. Perhaps we should concede to David Schmidt's favorite (1.3).

David Schmidt of University of Massachusetts was awarded the 2001 Ig Nobel Prize in Physics for his partial solution to the question of why shower curtains billow inwards. He used a computational fluid dynamics code to achieve the results. Professor Schmidt is adamant that this was done "for fun" in his own free time without the use of grant or other public monies.[3]

 

[russ_watters]

It's a case of convection being the obvious answer, but wrong. The curtain is drawn in even under a cold water shower.

The air drawn down with the spray leaves a low pressure region around the shower-head, so once the water and air separate on the floor the air makes a rapid return upwards outside the region of falling spray. As air speeds upwards past the curtain the low pressure (Bernouli principle) draws the curtain inwards. The result is a horizontal-donut-shaped air circulation pattern, a vortex in the shower cubicle: air and spray cascading down in the centre, and air returning upwards outside the cascade.

Wikipedia notes that someone was awarded the Ig Nobel Prize in Physics in 2001 for his contribution to the answer of why shower curtains billow inwards.

Since I stuck my neck out, I guess I'm going to have to give this the full treatment, with research and experimentation. But my early take is that I'm not buying it at all. Clicking through the wiki, we can start with this:

Predictably, I soon heard from adherents of the other, known as the chimney effect. They argued that air heated by the hot shower water rose and pulled cool air in from below, taking the curtain with it — convection, in other words. As proof, they advised running a cold shower, apparently believing that the curtain would remain motionless. I did as suggested. The shower curtain still blew up and in, though not quite as vigorously. OK, geniuses — you were saying?

I have three fairly serious problems with this:
1. He says testing with cold water shows the effect "not quite as vigorously", indicating that stack effect is at least part of it, but he only qualitatively described it. It isn't clear if he attempted to quantify it. He should have. The quote has a "bucking-the-mainstream" pop-science feel to it that sets off warning bells for me.

2. For the CFD analysis, he uses 8 GPM, which is triple the maximum code allowed flow rate (which was dropped by half in 1994 - the article was published in 2001). This will magnify the vortex effect, improperly skewing the results away from the stack effect. I would like to see the full results though...

3. Did he include a person in the shower in the CFD analysis? A person would seriously impact the formation and operationg of such a vortex. In particular, if the stream is almost entirely directed at your back, there would be minimal entrainment and your body/head would interfere with the vortex. There is a screenshot of the CFD analysis in the Scientific American article:

It may be just an illustration for newsiness, but the slice is cut along the wrong axis and there is no person.

This is a "rubber meets the road" issue though, and testing is really the way to go. So I did: I have a shower with a lightweight cloth curtain and no magnets at the bottom. The head is a standard medium pattern, 2.5 gpm head. I folded the decorative over-curtain up over the top so it would be out of the way, turned-on the shower, with a trajectory similar to the CFD screenshot and shot a video. For 30 seconds, the curtain did not budge. Then the water started warming up and the curtain was pulled substantially inward.

I'll post more details later, but for me it was a clear-cut demonstration of a pretty run-of-the-mill issue. I see this issue to be practically an urban myth given occasional not serious enough scientific treatment that almost seems designed to improperly discount the conventional explanation. Not unlike the mpemba effect.

 

[russ_watters]

In addition to the mechanism described by Nascent, there is also the effect of the falling water dragging entrained air along with it toward the base of the shower. This creates a higher pressure at the base of the shower than near the shower head. This cases air to flow down the drain along with the water. Make-up air is required to flow in from outside the shower stall.

That can't be. Besides the fact that the drain opening is tiny and a significant cross section is covered with water, the drain is trapped specifically to prevent airflow in the other direction. When you partly fill up a previously empty pipe with water, the air has to go out.

 

[Chestermiller]

That can't be. Besides the fact that the drain opening is tiny and a significant cross section is covered with water, the drain is trapped specifically to prevent airflow in the other direction. When you partly fill up a previously empty pipe with water, the air has to go out.

Good point. Thanks for the correction.

 

[NascentOxygen]

I have long known an explanation hinging on convection to be wrong---since a spell when my homework system was not functioning and, forced to take cold showers, I discovered the curtain was still drawn in.

Convection may exacerbate the effect, I'd allow that.

There really is no requirement for a body in the cubicle, anyone could easily test this by turning on a cold shower and closing the curtain, staying outside.

 

[russ_watters]

Video:

 

[russ_watters]

More detailed tests:


Several observations/notes:
1. My shower curtain is a light/soft synthetic fabric, which should be optimal for this test. But when wet it does sometimes stick to the tub, and since the effect is somewhat weak, that can sporadically interfere with it.

2. The effect is almost completely absent with cold water. And I say "almost" because it is tough to tell the difference between minor splattering of water and wind-induced movement. That said, I could feel airflow in the vicinity of the shower even with cold water, even though I could not detect it.

3. The effect is as typically seen with the water hot. It does take several minutes to reach full strength, and can be overcome by the curtain adhering to the side of the tub. I stopped the video above just a few seconds before the curtain broke away from the side of the tub and settled halfway across the tub. There's a 3rd and 4th (much shorter) video on my channel showing it.

4. The airflow velocity was about 70 fpm through an approximately 2" slot between the bottom of the curtain and the bottom of the tub, which equates to about 80 CFM. By comparison, that's the low end of the capture velocity of an exhaust hood (kitchen or fume hood) and on the upper end of what residential bathroom exhaust fan pulls. So it's a decent amount of airflow. I theorize that if the curtain hangs lower, the gap is smaller, the pressure difference and velocity would have to be greater and thus the deflection also greater.

 

[russ_watters]

I have long known an explanation hinging on convection to be wrong---since a spell when my homework system was not functioning and, forced to take cold showers, I discovered the curtain was still drawn in.

Convection may exacerbate the effect, I'd allow that.

I detected no effect at all with cold water. I'm willing to believe that you did, but as with the mpemba effect, my perception is that it is not typical, but may manifest in certain atypical circumstances that are difficult to identify. The article provides some insight:

The forces generated by this airflow are pretty weak. They are only sufficient to pull light, thin curtains inward. That explains why people with heavy plastic curtains typically don't have this problem. Also, if someone has poor water pressure or a poorly atomizing showerhead, they may not see the curtain suck in.

That's quite a bit of hedging and acknowledgment that the effect he found doesn't happen in all cases. I submit that stack effect convection should happen in more cases than the effect he describes. In particular, given his unusually high flow rate per new plumbing codes and acknowledgment that flow rate is a factor, I'd be surprised if his effect happens much at all anymore. In any case, I remember having the issue as a kid and that was with a heavy/stiff vinyl shower curtain with magnets to hold it in place: the convection was strong enough to pull the magnets loose (caveat: I take unusually hot showers).

There really is no requirement for a body in the cubicle, anyone could easily test this by turning on a cold shower and closing the curtain, staying outside.

My point was that the presence of a person should interfere with the effect, so detection of the effect when a person is not in the shower could be a false positive.

 

[klimatos]

I thought that this question was settled a long time ago. It is simply another (out of many) application of the Bernoulli Effect. That is, the flow of a fluid across a surface creates a drop in pressure on that surface. Water flowing across the wet side of the curtain causes a drop in pressure on that side. The pressure on the dry side is higher and pushes the shower curtain inward. It is the same phenomena that pulls heavy sheets of plywood out of the backs of speeding pickup trucks and causes the soft top of a convertible to bow outward rather than inward, despite (or because of) the force of moving air against the top. The same principle pulled fuel from the fuel bowl through the jets in the old-fashioned carburetor. The same effect accounts (with angle of attack) for some of the lift of an airfoil. It is a basic engineering principle.

You can test this by holding a teaspoon loosely under a running faucet so that the water flows against the back of the teaspoon bowl. The spoon will be pushed inward.

In the Earth Sciences, this same effect is used to explain part of how running water lifts heavy boulders from stream beds and winds lift sand in sandstorms. This is the same principle that makes a perfume atomizer pull the perfume out of the bowl. The flow of air in the crosspipe causes the liquid to be pushed up in the downpipe.

 

[NascentOxygen]

Water flowing across the wet side of the curtain causes a drop in pressure on that side.

That, unfortunately, is not the explanation here. The curtain billows inwards even where the shower recess is sufficiently large that it can be arranged that the curtain remains dry.

 

[russ_watters]

Indeed, we at least agree on that much. The Bernoulli effect is a consequence (of the reduced pressure) at best, not a cause. In particular, the velocity over the curtain is extremely low, and without a significant velocity, there is nothing for Bernoulli to do. Indeed, the reduced pressure due to buoyancy effect is maximized as the airflow approaches zero.

...though this "horizontal vortex" model is a Bernoulli effect explanation...[runs and hides]
[edit]
Oh, I misread -- it's even worse than that:

That is, the flow of a fluid across a surface creates a drop in pressure on that surface. Water flowing across the wet side of the curtain causes a drop in pressure on that side.

The water flow on the curtain? The water flow across the curtain is a thin film. It doesn't have "sides". Or, rather, with almost no thickness, it can't pull the curtain "inward" because it doesn't have an "inward". For that explanation to work, the water wouldn't need to just pull the curtain toward the middle of the shower, it would also need to pull itself toward the middle of the shower.

I think you are confusing what you describe with the coanda effect, which is what you see when a stream of water is pulled toward a surface and vice versa. But once they are stuck together, they don't get pulled further in one direction or another.

 

[klimatos]

Nascent,

I didn't see any post where the shower curtain remained dry. It wasn't mentioned in post #! of this thread. Was that specified in a different thread?

 

[NascentOxygen]

I didn't see any post where the shower curtain remained dry. It wasn't mentioned in post #! of this thread. Was that specified in a different thread?

I guess it wasn't seen as a relevant detail. But I'm throwing in that bit of detail, from personal experience. In a rental place I had a huge shower recess, and would throw water over the dry curtain to try and weigh it down.

In the end I gave up and stored the curtain in a cupboard, it was serving only as a privacy screen. (Though years later, it now furnishes data invaluable to this thread!) [emoji2]

 

[256bits]

though this "horizontal vortex" model is a Bernoulli effect explanation...[runs and hides]

So OK, the horizontal vortex has a low pressure in the centre the fellow says, or I am assuming he actually did say that.
But should there not also be a high pressure area pushing the curtain out, from the conversion of dynamic pressure of the swirling air as it loses velocity to a higher static pressure?

Is there an updraft ( away from the curtain ) or a downdraft ( towards the curtain ) in the centre of the horizontal vortex.

The vortex theory seems to be in need of some more explanation.

 

[russ_watters]

So OK, the horizontal vortex has a low pressure in the centre the fellow says, or I am assuming he actually did say that.
But should there not also be a high pressure area pushing the curtain out, from the conversion of dynamic pressure of the swirling air as it loses velocity to a higher static pressure?

I mentioned Bernoulli's there mostly to be funny - the article doesn't mention it and as you suggest, we really should get more details. I've looked but not found the actual paper (if one was ever actually published).

The article I read quotes him as saying the center of the vortex has a low pressure region like a sideways hurricane. Ironic choice, because the mechanism of a hurricane is convection. In the shower vortex, the low pressure would have to be due to centrifugal force pulling the air away from the center.

Would that result in high pressure or low pressure on the outskirts of the vortex? I'm actually not sure. My first thought was high pressure, but since the shower stall doesn't really constrain the vortex, it is free to expand, avoiding high pressure areas. My next thought was low pressure due to the speed and Bernoulli's, but again, if the air isn't constrained, static pressure should equalize. So my thought right now is a negative pressure in the center, with a gradient up to zero on the outskirts of the vortex.

As you say, I'd really like to see what the CFD results say.

Is there an updraft ( away from the curtain ) or a downdraft ( towards the curtain ) in the centre of the horizontal vortex.

I would think that with cold water it would be a downdraft, but with hot water it certainly is an updraft. But again, I don't know which was used in the CFD model.

The buoyancy model is more than enough to explain the observation on its own. A quick calc tells me the air in a shower has about 1.5 lb of bouyancy: not a ton, but enough that if you draped the curtain over top of the shower, you could make a hot air balloon!

 

[pyroartist]

That can't be. Besides the fact that the drain opening is tiny and a significant cross section is covered with water, the drain is trapped specifically to prevent airflow in the other direction. When you partly fill up a previously empty pipe with water, the air has to go out.

Hey Russ, you really missed the point on this reply. He did not say a word about air flowing up through the drain pipe. He said make-up air has to flow into the base of the shower, hence the inward pulling of the shower curtain. Drains were not mentioned or implied.

 

[NascentOxygen]

It should be a simple matter to sticky tape some light-weight ribbons to a shower curtain, like the streamers seen on the sails of some yachts, to highlight the air flow. Direct the water away from the curtain so the ribbons aren't immediately dampened.