PDA

View Full Version : low velocity nozzle designs



davidhills - n/a
15-Feb-09, 04:45 PM
Good Evening



can anyone point me in the right direction to understand how a low velocity air stream can be put thru a nozzle to maximise thrust. (or maybe it does not have much or any effect)and how you would design such a nozzle.



I have a 40mm diameter pipe with a 50kph air flow, how do I maximise the thrust?





I've looked on the net and there are plenty of explanations for rockets etc, but is any of this relervent for such a low speed situation





thanks





David Hills

Ian Brooks - n/a
15-Feb-09, 07:40 PM
High velocity air in the pipe must be slowed down in order to improve the efficiency of the nozzle. In principle, this is simply a case of constructing a divergent (increasing area) nozzle. The issue is that the maximum angle of divergence is about 8 degrees - go higher than this, and the flow separates from the wall and the nozzle stops working. Working with this angle, the nozzle will become unfeasibly long for any useful expansion ratio.



Ian

Jeremy - n/a
15-Feb-09, 08:40 PM
For good low speed thrust you only need a very modest flow velocity above your cruise speed, from whatever nozzle/duct/propulsor you're using. The key to maximising thrust efficiency is to get the required mass flow at a velocity "just" high enough to give you the thrust you need at design maximum speed.



Any excess thrust velocity is wasted, which is why large diameter props/fans are so much more efficient for low speed applications. It's also why turboprop aircraft are used for relatively low cruise speed tasks and why high bypass ratio turbofan engines are used for slightly higher speeds. The inefficiency of turbojets, as used on early airliners, was largely down to their high velocity, small jet pipe area, design, which is less well suited to sub-sonic speeds.



As Ian very rightly says, converting a high velocity flow into a low velocity one with a divergent duct is inefficient and difficult within the constraints of a small hovercraft, because of the need to keep the flow attached to the sides of the duct.



There really is no substitute for a propulsion system that has an inherently large area outlet, with a low velocity, high mass flow. This is relatively easy to do with a big prop, or even a big axial fan, but is harder with a centrifugal fan, because it takes up so much room.



Jeremy

davidhills - n/a
18-Feb-09, 05:49 PM
Thanks for your replies Jeremy and Ian



Have I got this right.



My model wants a top speed of 15kph say, so the exhaust velocity at the exit of the nozzle wants to be slightly higher say 20kph.



my centrifugal compressor has a exit area of 35cm2(and is probably designed correct for peak flow) my pipe is currently

got a exit area of 12cm3 and a velocity of 50kph.





I would guess the exit velocity of the compresssor is 25kph

(I can measure it)



So I need a pretty big nozzle.



may be two nozzles?





Ian I don't really understand this bit



The issue is that the maximum angle of divergence is about 8 degrees (how do a design this with 8 degree's?







thanks All







David































see attachment for rough diagram



Any household items come to mind ?







thanks







david







I guess the exit area of the a single nozzle needs to be about 50cm2 so diameter of 12cm.







What should the wall pofile shape be (eliptical?)





attached is a rough diagram

Paul Fitz - n/a
18-Feb-09, 11:31 PM
The simplest way to calculate your thrust outlet size is by ratio.



Fan gives 25 kph from 35cm2. you want 20 kph.

25/20 = 1.25 x 35 = 43.75cm2. this is the discharge area required to lower the velocity. (larger area for same volume gives lower velocity). This does not take into account the losses, but these will be low anyway providing that you do not restrict the fan discharge prior to enlarging it.



The divergent duct works well on an axial fan in a round duct. On a Centrifugal fan it is usually less effective because the air is forced outward in the scroll and therefore discharges on one side of the fan outlet. Increasing the discharge pipe size will still help by lowering the velocity but only if you have a long discharge duct (at least 5 fan diameters) so forget about the 8 degrees as it will be irrelevant.



It is never a good idea to split the discharge from a fan, it invariable makes the fan unstable in operation if the two discharges can be affected by pressure variation, eg in wind. In addition it is necessary to reduce the duct size to maintain velocity which increases fan loading and lowers the efficiency.



Your drawing of the profile would create high losses. Both restrictions and enlargments of a duct should be as gradual as possible. The "eliptical" discharge you show is actually an abrupt end to the pipe with a cowl over it. The air exiting any discharge is subject to a pressure loss of 1.22 X 0.5 x V^2 (where V =m/sec) so reduce V and you reduce the losses.



The general rule in a ducted system is keep the velocity as low as possible to do the work but also as constant as possible. So avoid decreasing the pipe size after the fan and then increasing it.



Read through the article "Principles of hovercraft design...." on the downloads page. The diagram shows and explains how pressure changes in a ducted system. The first 4 sections of the article "Calculation of thrust" explain how to caltulate thrust for differing discharge areas.



HTH

davidhills - n/a
19-Feb-09, 03:47 PM
Is this a more appropriate shape?





Is it a case of, smoothly constricting the flow from the compressor untill it starts to effect the mass flow rate(probably very little constriction, as the fan should have been designed for max performance with it's own exit size), then smoothly expand the pipe untill you get your required exit velocity?



thanks





David

Paul Fitz - n/a
23-Feb-09, 11:37 PM
It is not necessarily true that the fan "should have been designed for max performance with it's own exit size", as commercial fans are designed to work over a fairly wide range of duties.



Any constriction of airflow will give an energy loss. If greater velocity is required, this has to be accepted but ideally kept to a minimum value.



Re: your attached diagram. Draw a vertical line through the centre. The inlet shape (Bell-Mouth?) is ideal for an open inlet. Not so for a restriction from an existing pipe (fan discharge?). This would be better with a straight (conical) restriction, to smoothly constrict the airflow.

The discharge is essentially an eliptical or parabolic increase in duct area from the line, which would be the ideal if this is the final discharge. However, it is probably more difficult to form the compound curves than straight tapers and the overall losses would not be very different. Keep it simple the real differences will be small.



paul

Jonathan - n/a
27-Feb-09, 01:25 PM
For good low speed thrust you only need a very modest flow velocity above your cruise speed, from whatever nozzle/duct/propulsor you're using. The key to maximising thrust efficiency is to get the required mass flow at a velocity "just" high enough to give you the thrust you need at design maximum speed.



Any excess thrust velocity is wasted, which is why large diameter props/fans are so much more efficient for low speed applications. It's also why turboprop aircraft are used for relatively low cruise speed tasks and why high bypass ratio turbofan engines are used for slightly higher speeds.



There really is no substitute for a propulsion system that has an inherently large area outlet, with a low velocity, high mass flow. This is relatively easy to do with a big prop, or even a big axial fan, but is harder with a centrifugal fan, because it takes up so much room.



Jeremy


Can you define Low speed applications?

If you were wanting to build a REALLY high speed hovercraft (100mph) would you be better to use a low rpm Large fan or a faster spinning smaller fan?

Jonathan