EDF Air Intake Area

[Richard Sharman]

RB:

Just to recap (and to get this thread back on track) you have:

(1) an inlet opening of 56 cm^2 (resulting from two isoscles triangles of side 10,10,6) which is feeding into

(2) a 90mm fan(?) of area about 63cm^2, with "fan swept area" of 57cm^2, which is tapering to

(3) an exit duct of 50cm^2 (resulting from an exit diameter of 80mm

-- have I got this right ? If not, please correct me.

 

If so, this is basically right - the entry duct opening is bigger than the exit opening, and not bigger than the fan swept area, so it should work. However, it is not very efficient (in terms of turning electrical watts into thrust). There are two reasons for this:

(a) the triangular entry is inefficient, and therefore the entry size is overestimated, in practice the realistic entry size is smaller than so far estimated, and

(b) the difference between entry and exit is too small, which implies that the energy advantage which the motor is trying to achieve is also small, i.e. the thrust produced will be low.

 

I think I would advise a smaller exit, but I would need to do a detailed calculation first.

Richard.

[John Timmis]

[John Timmis]

Hi all,

 

This is my first post so you will have to be patient.

 

I've been following this thread with interest because I've just built & flown ( only 3 flights so far) a Vampire. It's my first ducted fan so I'm no expert.

 

The model is a Vampire MK. 3. That's the single seat, short nose, round wing tip version.

It flies away from a bungee & seems to have plenty of thrust for a Vampire.

 

I built the model to be as scale as possible; the only deviation is a small increase in the size of the exhaust.

 

The inlet ducts are scale & are glass epoxy mouldings over foam patterns.

Main details are; Span 49", Weight 4 lbs, WeMoTec 480. 70mm fan, Battery 4s 3200, Motor 4000kv (I think), Power 670 watts from a fresh battery.

 

Inlet 3750 sq mm, FSA 3580 sq mm, Exhaust 2825 sq mm. The exhaust is enlarged from 45mm dia to 60 mm dia.

 

Hope this helps.

 

John T

 

 

[Richard Sharman]

Posted by John Timmis on 06/10/2011 21:17:40:
.....Main details are; Span 49", Weight 4 lbs, WeMoTec 480. 70mm fan, Battery 4s 3200, Motor 4000kv (I think), Power 670 watts from a fresh battery. Inlet 3750 sq mm, FSA 3580 sq mm, Exhaust 2825 sq mm. The exhaust is enlarged from 45mm dia to 60 mm dia...............

John, thanks for posting the details -- this is an interesting contribution, and it's always useful to know the basic parameters of a given model.

 

The inlet area is 105% of the fan swept area - was this deliberate, or did it just come out that way from scale considerations?

 

You enlarged the scale outlet, and I'm sure this was a good idea. How did you reach your chosen value ?

 

Do you have any idea of the static thrust you are getting? I would guess about 500g? This would help in estimating dynamic thrust, and so evaluating flight performance.

 

Do you know what speed the model flies at when cruising straight and level? Your comment that it "has plenty of thrust for a Vampire" suggests it is not over fast, but if you are consuming 670watts you ought to be shifting pretty quick. At the moment I can't see where all the power is going.

[John Timmis]

Richard

 

Thanks for your comments.

 

Your guess about the thrust was a good one. It's 530gm on a fresh battery.

 

I had seen, on the internet, a 36" span Vampire for a 50mm fan which seemed to fly well.

 

I thought that about 48" would be good for a 70mm fan. I decided to keep the inlets scale as they seemed to be within range of acceptable areas. It was really a case of "suck it & see" in both senses. In any case , the duct entry follows the profile of the wing section so it would be a fairly simple matter to increase the area by cutting back the line of the leading edge. If this was done on the lower edge it would not be too noticeable.

 

The size of the exhaust (60mm) was recommended in the WeMoTec instructions.

 

I don't know the flying speed. It's only flown 3 times so far & my first priority was to explore the low speed at a height at least 3 mistakes high. However, it climbs well & the glide is very flat. I don't think it will be underpowered. more info later.

 

John T

 

 

 

 

 

 

 

[Tony K]

Hello everybody, been lurking for many months but this is my first post.

 

Although fluid mechanics is not my field, may I add a few thoughts here.

 

As I see it, a ducted fan works by accelerating a mass of air rearwards and the reaction to this acceleration is what drives the model forwards. Therefore I believe it is the mass flow which should be considered.

 

Mass flow is density times cross section area times velocity. As we are not adding or subtracting heat, I think we can ignore density. So as mass flow is constant through a closed system, the relationship of cross section area (A) and velocity (V) is the concern.

 

Imagine the air immediately in front of the fan and the air immediately behind. A (FSA) is the same so V must also be the same. This means that air is accelerating in the upstream duct.

If we call the velocity at the intake (model flying speed) V1 and the intake area A1, similarly the velocity at the fan V2 and the FSA A2, we get V1xA1 = V2xA2. The area of the intake should theoretically be A1 = V2xA2 divided by V1. So the intake area should be larger than the FSA.

 

How much larger is in proportion to the difference between V1 and V2 (theoretically, fan pitch times RPM), ignoring intake and duct losses. Assuming the same EDF, for a relatively slow model V2/V1 is greater than for a fast model so the slower model needs proportionally larger intakes.

 

I can not imagine a system where the intake area is smaller than the FSA as the incoming air would decelerate before being re-accelerated by the fan.

 

Downstream of the fan is less important. The mass of air has already been accelerated so we just need to let it out without losing too much of the increased velocity. Constant cross section or small convergence should achieve this.

 

I hope this makes sense.

[Tim Mackey]

Hi Tony and welcome outside.... nice to see you out of the closet

Interesting first post..... I need to read it a few times before my brain can absorb it all

[Richard Sharman]

Hi, Tony, and welcome to "forum-ing" !

 

Especially nice to see that someone else is interested in the "theory" of ducted fans. In fact, so interesting that I am about to start up ANOTHER thread specifically for this. This current thread was specifically about the size of RickenBangler's air intake, and is has wandered on and off topic several times, so perhaps it's time to start over again !

 

On the subject of theory, Tony, yes you are basically right that the problem is one of understanding the mass flow difference induced by the action of the fan, but one or two of you details are not quite right, in my opinion - particularly over the exit duct size.

 

Anyone interested in "Ducted Fan Theory and Practice" please join me in the new thread.

best wishes, Richard

 

[RickenBangler]

Posted by Richard Sharman on 05/10/2011 22:07:21:

RB:

Just to recap (and to get this thread back on track) you have:

(1) an inlet opening of 56 cm^2 (resulting from two isoscles triangles of side 10,10,6) which is feeding into

(2) a 90mm fan(?) of area about 63cm^2, with "fan swept area" of 57cm^2, which is tapering to

(3) an exit duct of 50cm^2 (resulting from an exit diameter of 80mm

-- have I got this right ? If not, please correct me.

 

If so, this is basically right - the entry duct opening is bigger than the exit opening, and not bigger than the fan swept area, so it should work. However, it is not very efficient (in terms of turning electrical watts into thrust). There are two reasons for this:

(a) the triangular entry is inefficient, and therefore the entry size is overestimated, in practice the realistic entry size is smaller than so far estimated, and

(b) the difference between entry and exit is too small, which implies that the energy advantage which the motor is trying to achieve is also small, i.e. the thrust produced will be low.

 

I think I would advise a smaller exit, but I would need to do a detailed calculation first.

Richard.

 

Thanks for that Richard - your recap is (I think) spot on!

I've decided to keep the relevant dimensions very close to those specified by the original models designer.

 

When I first started this thread I wasn't expecting the subject to become quite so involved, so thanks to everyone who chipped in - I've learnt a lot from all this...

 

RB