Convection powered bladeless fan style ventilation?

[George Albercook]

TL;DR Summary

So as I understand it Bladeless fans use the Coandă effect of air hugging the walls of an expanding cylinder to creat a lower pressure zone that pulls additional air in through the center.

My question is about powering such a thing with convection. Also will the low pressure region be diminished or maybe even amplified by convection?

I'm still trying to understand the operation of these fans. How does the geometry downstream influence the flow. Specifically, there is an 8" ventilation tube in the ceiling of my warehouse. I'm thinking that I can extend it until it is 3/4 of the way to the floor. If I crack open an overhead door at night convection should pull air up the tube.

My questions:

1) If I place a Bladeless fan ring (no fan) at the bottom of the tube. Will the flow increase?

2) Do I even need to gap when the fan normally blows air along the inside of the ring?

3) Will the low pressure zone be disrupted by the long tube after the fan exit? Or is that what causes convection. In the first place?

 

[jrmichler]

Bladeless fans use the Coanda effect plus entrainment. The driving air for the entrainment comes from a blower mounted inside where it is not visible. The blower moves a relatively small amount of air at relatively high pressure and velocity. That air entrains a larger amount of air at lower velocity. Bladeless fans are quiet because the manufacturer spent a lot of money designing and building sound control around the blower.

Dyson is a well known manufacturer of bladeless fans. If you are interested in what is inside their bladeless fans, go to patft.uspto.gov, Advanced Search, and use search term an/dyson and an/technology and ttl/fan. They have over 100 patents on their fans.

Bladeless fans are a balanced design. They will not work well if you do anything to change the design. External restrictions, such as a long tube, are especially bad.

If you really want to move air in a warehouse, I suggest one or more ceiling fans. If it's an especially large warehouse, get a Big Ass fan: https://www.bigassfans.com/fans/#filter=.overhead-fans

 

[George Albercook]

Thanks. I don't want to just move air I want to exchange it. Particularly at night with the stack effect or natural draft.

The pressure inside of a pipe during natural draft is already lower.

Would the low pressure from the Coanda effect add to, subtract from etc. the natural effect?

 

[jrmichler]

If you want to move enough air out of a warehouse to make a noticeable difference inside, I can recommend one of these: https://aerovent.com/products/roof-ventilators/d53-roof-ventilator/. There are other manufacturers just as good. I have had good experience with Aerovent and New York Blower products.

To give an idea, I once put a 72 inch roof ventilator in a room about 300 feet long by 75 feet wide (rough guess) at the opposite end from the truck door. When we did a smoke bomb test, we saw that the ventilator not only exhausted air, but also caused a whole room air circulation. The people in that area, and their supervisor were happy with the results.

The natural draft effect of an 8" duct by itself, with or without help from a (relatively) tiny little blower will not have a noticeable effect on room comfort. Unless that duct is about ten feet diameter or larger. Roof ventilators are designed for the job. They include flaps that close when the ventilator is turned off to keep out rain and snow.

If you want a natural convection solution, look into cupolas. But think big if you want to make a difference large enough to notice. And include a way to close it off.

 

[George Albercook]

Thanks. I appreciate the advice and your experience. I don't own the building so installing those ventilators would be complicated.

I really want to understand the answers to the questions that I asked. It is a forum full of geeks after all.

Let me clarify one point. I'm not asking about including the internal electric fan that is in the body of the bladeless fan. I'm wondering is will the stack effect convection be increased by the addition of the ring with the Coanda effect.

Thanks again

 

[anorlunda]

Let us help with the engineers approach. You are starting with a favored design. In other words, the method for achieving your goal. The approach of engineers is to first clearly define the goal. This thread begins without a clear explanation of the goal.

What are you trying to accomplish? How big a volume? What shape? How much ventilation do you require? (measured in air changes per hour) Are seasonal variations important? Is fan noise a problem? Are there budget or other constraints?

One of our regulars @russ_watters , works in HVAC full time. He might be able to help if he knows what you're trying to do.

 

[anorlunda]

I really want to understand the answers to the questions that I asked. It is a forum full of geeks after all.

And the geeks are hinting that your question is the wrong question.

Bladeless fans are a balanced design. They will not work well if you do anything to change the design.

That is the real answer. In other words, "it depends on the details."

You can spend the rest of your life refining the design details, and if you succeed you may find that it's still not enough to achieve your goal.

 

[russ_watters]

Let me clarify one point. I'm not asking about including the internal electric fan that is in the body of the bladeless fan. I'm wondering is will the stack effect convection be increased by the addition of the ring with the Coanda effect.
[from the OP]
1) If I place a Bladeless fan ring (no fan) at the bottom of the tube. Will the flow increase?

No, the ring will just be an obstruction to your convection. On its own, it doesn't do anything. It requires a "real" fan to push air through nozzles in the actual tube of the ring.

So as I understand it Bladeless fans use the Coandă effect of air hugging the walls of an expanding cylinder to create a lower pressure zone that pulls additional air in through the center.

The primary principle of operation of the Dyson "bladeless" fan is entrainment or induction. The shape helps direct air into the primary airstream of the nozzles, but without nozzles there is nothing for the ring to do except get in the way.

The Dyson "bladeless" fan is aesthetically pleasing and well marketed, but it isn't a good fan if what one cares about is efficiently moving air.
[edit]
Here's a guy who did some measurements of the Dyson fan vs a normal fan:
https://sites.google.com/site/bladesvsbladeless/data-analysis

The results are clear:
1. The Dyson fan moves less air for the same energy input than the conventional fan.
2. The Dyson fan is louder than the conventional fan.
3. The Dyson fan costs at least 8x(!) what a conventional fan costs

 

[256bits]

I'm thinking that I can extend it until it is 3/4 of the way to the floor. If I crack open an overhead door at night convection should pull air up the tube.

If there is no lower opening at that height ( in the wall, a door, etc ) for air to enter the building, then lowering the 'tube' does nothing but cut off any air ventilation. So you have to have a reason to lower the tube, so that air can enter, or exit ( depending upon which way the convection ventilation is proceeding ) at that particular height, or less, and cutting off any ventilation of air at greater heights than the lower point of the tube.

The stack effect is about buoyancy of air, not about a making a stack or tube.
As an example, a fireplace has a tube to direct and remove the combustion gases ( you could run you fireplace without the tube and just have a hole in the roof, but the products of combustion would mix and disperse throughout the room before exit, rather than being directed to the outside, but the fireplace would still burn ).
Since the products of combustion are at a much higher temperature, they are much lighter than room air, and offer a greater buoyancy, hence the 'stack effect' is promoting more movement of the gases.

In your warehouse, the higher temperature, and less dense air collects nearer the roof, rather than 3/4 the way down. IMO, lowering the tube defeats the natural convection. Unless of course, you have a heat source at that lower height and you want to remove the hot air up the tube, so it does not mix with the building air.

 

[George Albercook]

Thanks 256bits. I clearly got that one wrong. The hottest air is up high.

So I looked it up and Delta P=Cah((1/T0) - 1/Ti)). So if I wanted more draw I would have to extend the stack above the root further. Correct?

 

[George Albercook]

Thanks for the reference Russ

The results are clear:
1. The Dyson fan moves less air for the same energy input than the conventional fan.

From my reading of that link, clearly the Dyson is far more costly. But he did not compare energy efficiency.

I expect that if he measured the complete Dyson fan vs just the fan in the base there would be an increase in the flow rate.

(One would have to correct for any changes in current draw as a result of removing it from the housing).

 

[George Albercook]

No, the ring will just be an obstruction to your convection. On its own, it doesn't do anything. It requires a "real" fan to push air through nozzles in the actual tube of the ring.

And in this way it is a "air multiplier." I have a strong expectation that I'm trying to pick myself up by my own boot straps, which is why I'm trying to understand where the error in my thinking is.

If I can use your term "real" fan, can I replace the air moved by the "real" fan with the air moved by the stack effect?

If, for example, we were to take a Dyson fan and remove the electric fan from the base of the unit so we have an open hole, say 4" in diameter in the bottom of the Dyson fan. Keeping the Dyson in it's intended orientation if we attached a vertical 4" stack to the bottom of the fan we would effectively replace the "real" electric fan with a real convection fan.

I think that we all agree that the Coanda effect and the entrainment effect cause additional air to move. In this case it would be air moving horizontally through the ring.

If we rotate the ring of the Dyson 90° so it is horizontal and pointing up more flow should come up through the ring than would have gone through the stack alone.

I realize that the ring will cause a restriction but that happens in the electric fan powered Dyson and flow is still multiplied.

Is there an error in my thinking?

Much thanks to everyone for your patience while I babble my way through this.

 

[jrmichler]

There are two things happening in a Dyson fan.

1) The Coanda effect is where a jet of air (or fluid) tangent to a convex curved surface will follow that surface. The simple experiment to show this is blowing on a roll of toilet paper.

2) Entrainment is where a jet of fast moving fluid comes in contact with slower moving fluid and pulls the slower moving fluid with it. Entrainment works better inside a duct than in open air, but it is always less efficient than moving the larger volume of air directly.

Stating it another way, you can move a certain volume of air by entraining a higher velocity jet, or you can move the same volume of air by moving it directly. It is always more efficient to move the larger volume of air directly than by using a jet with entrainment. The reason for this is that momentum is conserved, but kinetic energy is not.

If you have a vertical tube going up through the roof of a warm building, you can expect air flow up and out. The velocity of the air flow will be a function of the temperature difference between inside and outside, and the length of the tube. A longer tube has a larger chimney effect, but has more resistance to flow. Adding a fan blowing upward will increase the flow. It does not make a difference whether the added fan is a conventional fan or a bladeless fan. The power of the fan does make a difference. Power refers to ability to push air, not motor power.

A good experiment is to puff a little smoke into the duct, measure the duct length and time to flow through the duct, and calculate the total flow rate. Do this with and without the fan. Divide the total volume (length times width times height) of the warehouse by the flow rate in cubic feet per hour (or cubic meters per hour) to get air changes per hour. If you want to improve the temperature in the warehouse on a hot day, try for many air changes per hour. One or two air changes per hour will not make a difference.

 

[russ_watters]

From my reading of that link, clearly the Dyson is far more costly. But he did not compare energy efficiency.

Yes he did -- he said both consume 40 Watts, but the conventional fan moves more air for that same energy input.

I expect that if he measured the complete Dyson fan vs just the fan in the base there would be an increase in the flow rate.

He did measure the total flow rate.

And in this way it is a "air multiplier." I have a strong expectation that I'm trying to pick myself up by my own boot straps, which is why I'm trying to understand where the error in my thinking is.

If I can use your term "real" fan, can I replace the air moved by the "real" fan with the air moved by the stack effect?

If, for example, we were to take a Dyson fan and remove the electric fan from the base of the unit so we have an open hole, say 4" in diameter in the bottom of the Dyson fan. Keeping the Dyson in it's intended orientation if we attached a vertical 4" stack to the bottom of the fan we would effectively replace the "real" electric fan with a real convection fan.

Entrainment/induction uses a high velocity airstream to induce a lower velocity airstream. The reason it is less energy efficient than a normal fan is that a high velocity airstream by nature requires more energy input than a low velocity airstream.

In your case, you only have one airstream: a single stack effect airflow. Running some of it through the base of the Dyson fan is just getting in the way, and won't create the "air multiplier" effect at all, because stack effect is a really weak, low velocity airflow. In fact, for the Dyson setup to increase your airflow would violate conservation of energy. That's why as an engineer I'm not a fan (no pun intended) of Dyson. He's basically claiming to be selling perpetual motion machines.

I realize that the ring will cause a restriction but that happens in the electric fan powered Dyson and flow is still multiplied.

And that's exactly why the Dyson fan must be less efficient than a normal fan.

 

[George Albercook]

Thanks Russ Watters. I went back and read it more closely. I think I understand things a bit better. The Dyson Bladeless fan is an air flow amplifier but before that it is and air flow attenuator.

From the study
"The flow rates calculated are as follows:

• 0.2242 m^3/s for DAM
• 0.2414 m^3/s for fan
• 0.0365 m^3/s for DAM without duct"

It would have been interesting if they could have measured the flow out of the slot. But I assume that the slot only reduced the flow compared to the "DAM without the duct"

So the duct produces a flow multiplication of at least 0.2242/.0365 = 6.22

I wonder what the flow would have been it the fan had been ducted into the DAM duct. Would the power draw have increased?

I am kind of surprised that the DAM without duct produces so much less flow than the regular fan.

I still don't have confidence in my understanding of the relationships between, flow, velocity, power and momentum.

Is my analogy of the DAM duct acting like a step down transformer lowering the velocity but increasing the mass reasonable?

Thanks again with all the help thinking this through.

 

[russ_watters]

It would have been interesting if they could have measured the flow out of the slot. But I assume that the slot only reduced the flow compared to the "DAM without the duct"

So the duct produces a flow multiplication of at least 0.2242/.0365 = 6.22

I agree with that.

I wonder what the flow would have been it the fan had been ducted into the DAM duct. Would the power draw have increased?

Probably decreased. That's a counter-intuitive thing about fans; power draw depends more on the airflow than on the pressure or restrictions to airflow. So if you remove a restriction and the airflow goes up, the power draw goes up.

I still don't have confidence in my understanding of the relationships between, flow, velocity, power and momentum.

Is my analogy of the DAM duct acting like a step down transformer lowering the velocity but increasing the mass reasonable?

I like it, though I tend to be in the minority when it comes to analogies here...

I do think a key issue is that like a transformer, power is conserved, at best (law of conservation of energy). So the power in the airflow generated by the fan is the highest just after leaving the the fan wheel/blades. After that, anything that happens to it reduces the power due to losses. So it's just inherently impossible for an "air multiplier" fan to produce more airflow than a standard fan at the same power input. There are situations/applications where induction is useful, but improving energy efficiency is not one of its upsides.

There's some concepts that may help with your understanding of performance; the "fan laws": They are a set of equations that describe the relationship between things like velocity and power.

Momentum is not conserved in a duct system in general; the same ducted airstream (think of a plastic bag full of air to visualize it as a collection of moving objects) has a higher momentum at higher velocity, even though the energy is conserved with a change in velocity due to Bernoulli's principle.

Outside of a duct, where the pressure is zero (gauge), the same momentum can be achieved at lower velocity. An this is the idea behind a helicopter or a BigAssFan (see link in post #2 -- they are in fact an industry leader in this space). Larger, lower velocity fans move more air at the same momentum for less energy.

Thanks again with all the help thinking this through.

You're welcome!

 

[George Albercook]

Ok things were going so well now I'm probably going to throw a monkey wrench in the works.

So I estimate that I'm moving 6000CFM into the building which has roughly 6000 sq. ft. and a 20 ft ceiling. I'm blowing air in at ground level and at lease some of it is leaving out the makeup air vents in the ceiling. As one would expect by the end of the day the temperature of the ceiling is higher than the walls and floor. I have observed that after I turn on the fans in the evening the temperature of the ceiling drops to the same temperature as the walls. So I think that I am pushing the hot air up and out when I pull in air near the ground.

The disappointing thing is that even after running the fans for 12 hours (something like 36 air exchanges) the indoor temperature is still 5 to 10 °F higher than the outdoor temp. For example if it gets down to 62°F at 4:00 AM, at 7:00 AM the indoor temperature is still 72°F.

I can't find any examples of night time ventilation for cooling of building where the exchange rate is changed. To see if I'm just moving too little air because of the huge mass of the building.

Here is the part that might kick up some dust, so to speak. The air coming out of my blowers is going 30m/s (60 mph) I would like to use the Coanda effect, entrainment and pressure drop of the Dyson Bladeless fan to increase the volume of air that I'm moving into the building from outside. It seems to me that with my transformer analogy I'm not violating and conversation laws if I move more air at lower speed. (I won't venture to guess if it is a velocity or velocity squared relationship.)

It seems that I don't need to restrict the air flow inorder to get higher velocity so I should not have the attention effect just the amplification effect. I'm thinking that I use to cylinderical cardboard concrete forms. Run them horizontally across the garage door. One on the ground and one some distance above. Then I make wedges out of four sheets of plywood. The top most and bottom most are horizontal. The inner two sheets provide the increase in cross section for the pressure decrease that Dyson claims.

If I can get a four fold increase in the volume of air it seem that should provide more exchanges and be able to bring the indoor temperature down closer to the night time outdoor temperature. It would make a huge difference to start the day at 62°F rather than 72°F.

Why don't people do this already? Dyson patent?

Attached is a very poor drawing with three views to maximize confusion.

As always thanks for reading and for any input.

_______________________

 

[jrmichler]

If I can get a four fold increase in the volume of air it seem that should provide more exchanges and be able to bring the indoor temperature down closer to the night time outdoor temperature. It would make a huge difference to start the day at 62°F rather than 72°F.

This is 100% correct.

You are up against some laws of physics relating to natural convection:
1) Air flow by natural convection is proportional to the temperature difference.
2) Air flow by natural convection is proportional to the height difference.
3) Extending a roof vent upward increases the height difference. It also increases duct losses, so the increase in air flow will be less than you think.
4) To double the air flow through an existing system, you need to push four times as hard. That could be four times the temperature difference or four times the height.
5) To get double the air flow with the same height and temperature difference, you need twice the flow area.
6) Double the air flow will reduce the temperature difference, so you will need more than twice the flow area.
7) Sometimes you can get more air flow through an opening by rounding the entrance or adding a bell to the entrance. Search pipe entrance loss to learn more.

8) Check for restrictions in the roof vents. Do they have covers that open part way? Are there struts or something blocking the air flow?
9) Moving four times as much air through the same openings takes 16 times the pressure difference and 64 times the power. Better to make the openings bigger and smoother. Adding plywood that makes an opening smaller is a step in the wrong direction.

 

[Tom.G]

The air coming out of my blowers is going 30m/s (60 mph)

I know not a lot about the subject but a few things come to mind:

It would be interesting to know the volume and speed of the outgoing air at the roof vents. For instance the majority of the air may not even get there.

If there is much thermal mass in the building and contents with limited surface area, you may be fighting a losing battle.

If there are concentrations of thermal mass can you direct the airflow at them? It may not be getting there at all. The smoke bomb approach that @jrmichler noted sounds like it's worth a try.

That's my 2-cents worth anyhow.

Cheers,
Tom

 

[George Albercook]

I'm sorry, I think I was not clear.

You all schooled my on the ceiling vents and natural ventilation. I don't expect much from it.

I will open a second door to let more air out.

@jrmilcher thanks for the graphic that makes it very clear.

I have fans installed. Here is a picture of the box that holds them when it was half built. You can see the garage door track to the right of the right fan.

Here is a picture of the outside of the building.

The fans are in the door at the right. The truck well is rarely used. The larger door on the left could be opened several inches and a grid placed in the opening to prevent unwanted animals (including humans) coming in at night.

Here is the inside of the building.

Let me summarize what I think I know now.

1) Natural ventilation is not even close to enough.
2) I need more air flow. The inside of the building is still 5-10°F warmer that the outside after a full night of ventilation.
3) The Dyson duct does multiply the flow through the slot. (Not compared to a fan of the same wattage.)
4) The net result of the Dyson duct is to trade velocity for volume flow. This does not violate any conservation laws.
5) The Dyson fan restricts the flow from the fan in the base to get higher velocity. So it must attenuate before it amplified.
5) I have plenty of velocity. 30 m/s (60 mph)What I propose.
A) I make a Dyson style duct that is rectangular out of cardboard concrete forms and plywood. The existing blowers will blow into the cylinder shaped forms which will be kind of like the rounded edge of an airplane wing.
B) Air will exit the cylinder by way of a slot going lengthwise in the cylinder. The cross sectional area of the slot can be as big or bigger than the blower exit.
C) The air will have to change direction so there will be a small reduction in efficiency from that.
D) The air exiting the slot will be directed along a piece of plywood and due to the Coanda effect to will stay close to the plywood.
F) Additional air will be entrained which will increase the volume of air moving but reduce the average velocity.
E) Their will be a second "wing" like a biplane but unlike a biplane the four surfaces of the wings will i arrange differently. The top of the top wing and the bottom of the bottom wing will be parallel. The two remaining surfaces will start closer together at the front where they attach to the cardboard cylinders and will become further apart at the rear until the meet up with the top and bottom surfaces. The cross section of the space between the wings increases from the front edge of the wing to the back edge.
F) Dyson claims that the increase in cross sectional area causes a drop in pressure which pulls even more air through the duct. (I believe this was demonstrated in Computational Fluid Dynamics (CFD) models.

If I'm right this should increase the volume of air that the fans move. Dyson claims 15x, CFD models of the Dyson Bladeless fan show about 7x if I remember correctly, the data sited earlier in the thread gave 6.22x. A multiplier of only 4x would increase the air exchanges from 3/he to 12!

Have I fooled myself?

Thanks again for the civil, informative, useful and productive conversation.

 

[jrmichler]

If I'm right this should increase the volume of air that the fans move. Dyson claims 15x, CFD models of the Dyson Bladeless fan show about 7x if I remember correctly, the data sited earlier in the thread gave 6.22x. A multiplier of only 4x would increase the air exchanges from 3/he to 12!

Have I fooled myself?

Yes, because you earlier mentioned 6000 CFM and wanting to move about four times that much. Entrainment is always less efficient than directly moving the desired amount of air. Dyson uses entrainment for the purpose of hiding the blower, but as you have already seen, it is less efficient than a regular propeller fan.

The centrifugal blowers shown in your photo are designed to move air against a certain amount of back pressure, such as a duct system. They are not efficient for moving air through an open building.

The best way to move air through an open building such as yours is by one or more propeller fans such as these Aerovent panel fans: https://aerovent.com/wp-content/upl...RF-BPRV-DDP-DDPRC-DDPRF-DDPRV-Catalog-168.pdf. For example, their 48" fan, either belted or direct drive, will move about 20,000 CFM. One or two panel fans mounted similarly to your centrifugal blowers will move as much air as you want. A large fan running at slow speed will move as much air as a smaller fan running high speed. The larger fan will use less power, and will be much quieter.

Aerovent is not the only good manufacturer. I have also had good experience with New York Blower.

 

[George Albercook]

I will look into those fans.

I'm not depending on entrainment.
It just adds to the air moved by the blowers.
From russ_watters's reference he,

From the study
"The flow rates calculated are as follows:

• 0.2242 m^3/s for DAM
• 0.2414 m^3/s for fan
• 0.0365 m^3/s for DAM without duct"

It would have been interesting if they could have measured the flow out of the slot. But I assume that the slot only reduced the flow compared to the "DAM without the duct"

So the duct produces a flow multiplication of at least 0.2242/.0365 = 6.22

He agreed that there is a multiplication but only of the air moving through the slot. First there is a reduction before the slot.

 

[George Albercook]

I realize there are lots of things missing from the following calculation. I just wanted to make sure I'm not way into the perpetual motion world.

6000 cu ft of air/ min.* 0.07 lbs/ cu ft = 420 lbs. / Min. That seems huge.

I measured the air speed coming from a 30 in floor fan at ~1800 ft/min or 30 ft/sec.

420lbs/min. * 30 ft/sec. = 5874 Joules/60 sec.
or 97.9 Joules/sec.

750 J/s = 1hp

so the power in 6000 cfm going 1800 from is 98/750 = 0.13 hp.

if my math is right I'm not crazy.

at least this does not prove I'm crazy.

 

[Tom.G]

If I'm right this should increase the volume of air that the fans move. Dyson claims 15x, CFD models of the Dyson Bladeless fan show about 7x if I remember correctly, the data sited earlier in the thread gave 6.22x. A multiplier of only 4x would increase the air exchanges from 3/he to 12!

Have I fooled myself?

Perhaps.

It is unclear if those "wings" you are proposing to move more air are located in the building after the fans. Yes, entrainment will move more air, but if that new air is already in the building all you are doing is re-circulating hot air. Not very productive.

From the photos of the building interior, a major thermal mass is the concrete floor. If that is the case, attaining a high flow rate across the floor could help quite a bit. It would be informative to measure the floor temperature in early AM and at the end of day.

Cheers,
Tom

 

[George Albercook]

Thanks Tom. The intake for both the fans and the "wings" will be fresh air.

The concrete slab lags behind the temperature of the air by zero to 5°F. The temperature of the slab in front of the fans, which are at floor level is only one or two degrees colder than the air. I suspect that there is insulation under the slab because it does not seem to be cooler in the center of the building where the effect of colder ground should be bigger. It also does not seem to average out to a temperature lower than the air.