Thread: Discussion The Truth about ducted fans
View Single Post
Old Jul 03, 2006, 07:08 PM
winmodels is offline
Find More Posts by winmodels
Registered User
Wareham, Dorset, UK
Joined Dec 2003
1,912 Posts
Originally Posted by AirX
The outlet tube is an integral part of the system if it is less than FSA, the outlet is where the final factor is realized. I stand, Klaus what say you? Eric B.
Hi Eric

where locally the thrust is created, is that really such an important question?
It is a reaction force created by the change of momentum of the air which ejects from the nozzle exit. (That's not much better either, sorry)
If one looks at the pressure development of the fan drawing of post #101 one can see that the highest total pressure is at the rear of the fan stage. This is also the location of the highest static pressure. The drawing is pretty accurately drawn in this respect and the fan produces around 11N thrust in the static case. Now if you calculate the net axial force on the fan annulus as the product of area times pressure the result is very much the same.
If you only count the blade area and the axial resultant from the lift generated by the blades (or cascade, if you use cascade calculation methods) you will find that that force is less than half the thrust.
Whichever way you look at these things you will find that only the whole provides a clear picture.

One can also get away with out any form of thrust tube. In the early years of EDFs it was general practice to use the WeMoTec MINI-Fan without thrust tube and the motor open to the air stream for cooling. Through trial and error it was then found that a slight reduction in exit diameter did provide better results.
I have include here a little drawing which is now nine years old which shows this fan fitted in a nacelle. It shows some variations of the shapes which we found advantages, but for example the end cone which I built for it (below centre line) did not provide any more thrust than the open motor variant.
The shape of the front end was very important. By changing to the longer spinner I gained around 30% thrust. I still have the moulds for the engine pod.

Which leads us nicely back to our original discussion. In the next picture I have drawn four different variations of EDF installation. By now we are reasonably familiar with the descriptions: Ac is the catchment area which defines the streamtube cross section at duct entry (including the intake lip, left out here because of the scale). Aoo is the cross section of the stream tube ahead of the duct entry at a distance where no influence of the duct can be felt; and Fe is the nozzle exit. There is still Af the fan annulus area, which is also called FSA, and Fi, an intermediate cross section area, which is neither of the others but needs a special treatment.
In A we can clearly see that the cross section areas from ahead of the plane to the very end reduce in size. We have a continuously converging duct system with the possible exception from Af to Ai, where it can be a constant area duct (but make sure that Ai is not larger than Af and please bear in mind that we talk of areas here not diameters!).
Naturally the duct causes a resistance to the air stream (from Ac to Ae) which is mainly due to friction. The stream tube from Aoo to Ac does not cause any friction because it has no wall; but we have to ensure that the stream tube slides smoothly into Ac and beyond, which is why we donít use sharp intake lips but nicely rounded ones. In this context it is sensible to know the area ratios, namely Ac/Aoo and also Ae/Af.
Several years back I once read an article in which some aerodynamicist proved mathematically that the total ratio Aoo/Ae should be 2 and the individual ones square root from 2. I think it was in connection with the Fenestron (sp) rotor which is a ducted fan employed in helicopters for the rotor torque compensation. But we want to use the EDF in a fast flying model which certainly differs vastly from a helicopter; and how do we know anyway?

Next instalment will come shortly.

Originally Posted by planemaker
Variable geometry can be useful if you plan to have a "wide" speed range. For most EDF applications we're talking about here, that is not really the case.
Theoretically variable geometry may have some attraction but in the end you see that it is not really necessary and just think of the mechanical complications

Originally Posted by planemaker
No, I did precisely what I intended and it did precisely what I wanted. The thrust generated for a given power level is not misleading in any way and is in fact one of the "best measures of merit" for some applications, especially electrics, since power is at a premium
It disappoints me that you use a variable for this purpose. I have never seen any such figures in relation to flying speed (where I could see a certain benefit). It is much better if we are provided with real information as you can find in thrust/power/exit velocity over rpm graphs, as is becoming more common now.
winmodels is offline Find More Posts by winmodels
Last edited by winmodels; Jul 03, 2006 at 07:15 PM. Reason: minor corrections
Reply With Quote