Jul 14, 2005, 09:22 AM Registered User Ontario, Canada Joined Nov 2002 1,568 Posts FSA, exit diameter, and bench-testing... After learning that efflux velocity can be derived from thrust and exit diameter (still don't have the math though...), I started puzzling about the relationship to thrust, FSA, and bench-testing. With most EDF bench-test set-ups I've seen there is no exit ducting used, it's simply the back of the bare fan assembly with the motor. In this case, with the motor usually protruding past the fan, the exit "diameter" is essentially the FSA. Now we're right back to square one, where to calculate efflux velocity from measured thrust, the actual diameter of the bench-tested fan is irrelevant, and we have to use FSA. Correct? It also seems that with "high-ratio" FSA/diameter fans like the Microfan (where the rotor hub is a substantial proportion of the diameter), would suffer more from drag losses caused by post-fan ducting, as it's nearly impossible to maintain FSA in a uniform manner past the fan. (shouldn't you have a 37mm, or so, exhaust duct on a Microfan, simply to maintain FSA?) I guess ideally, the motor should have a "bullet" on the end of it, and the exhaust duct should taper with the bullet to maintain 100% of the FSA. I've seen quite a few applications of the motor "bullet", but I don't think I've seen an exhaust tube which tapers proportionally in that area. Ideas?
 Jul 14, 2005, 09:26 AM Member Edina, MN, USA Joined Oct 1999 11,556 Posts Have sketch that might help; will post it when I find it.
Jul 14, 2005, 11:41 AM
Member
Edina, MN, USA
Joined Oct 1999
11,556 Posts
Crude sketch, but hopefully informative!

From continuity >> A*v = a*V.

This means that for a given mass flow (say what the EDF produces for some W) a big “A” (area) makes for a small velocity and a small “a” makes a big velocity.

Easy to under stand on the back end; but it also applies to the front end!!

Lets think of the velocity as the approach/flying speed of the aircraft.

When the aircraft speed and the speed the inlet is actually designed for the capture tube a head of the plane is the same diameter as the inlet – the orange flow condition in the picture.

At a low airspeed, the stream tube’s diameter is larger, and in fact at static air comes from behind the inlet – turning 180 to go in – the blue wiggly line shows this.

For airspeeds higher than the design point, the capture is smaller – see the red circle and line.

The difference is the red area & orange area presents itself as increased drag (more frontal area); thus more thrust is needed for that flight speed.

Reduce the inlet size and you have less projected area and thus drag – so more flight speed.

Of course if your go too small on the inlet, you reduce the flow of (starve) the fan.

Some good info can be found via:
http://www.lerc.nasa.gov/WWW/K-12/airplane/bgp.html
http://www.lerc.nasa.gov/WWW/K-12/airplane/thrsteq.html

# Images

Last edited by jrb; Feb 02, 2008 at 12:26 PM.
 Jul 14, 2005, 12:01 PM EDF rules... :) Joined Nov 1999 13,701 Posts Considering the microfan, you would think that would be the case but with large ratio FSA like the GWS the blade rood does little if anything to add to the efflux because of the rotational speed of the root is slow compared to the tip. So using a larger motor to carry load will not be detriment as far as lowereing the available area in comparison to the previous example except from a weight standpoint which is a minor consideration as compared to overall aircraft weight. The FSA is what generates the airflow so with a testing setup not using an outlet tube FSA is the defining area. Cheers, Eric B.
Jul 14, 2005, 12:09 PM
Registered User
Joined Nov 2002
1,568 Posts
Quote:
 Originally Posted by jrb Crude sketch, but hopefully informative!

Full link to the "book": http://www89.pair.com/techinfo/MassFlow/ductbook.htm

Their basic point is that it's a major error to "Copy Turbojet Inlets".

Apparently, the correct inlet design for an EDF is the polar opposite of any kind of jet engine! (turbojet/turbofan)

Note the shape of the inlet in the attached image...

Thoughts?

# Images

Last edited by meteor; Jul 14, 2005 at 12:24 PM. Reason: addition
 Jul 14, 2005, 12:26 PM Dieselized User Chicagoland Joined Feb 2000 7,530 Posts Even though that book has a lot of information i don't believe much of it applies to models as we use them. That image is great if you fly your jet standing still. Do some studying of inlet flows and you will find that what Jim posted is correct. Klaus has also gone into detail. http://www.rcgroups.com/forums/showthread.php?t=331157 http://www.rcgroups.com/forums/showthread.php?t=352986 http://www.rcgroups.com/forums/showthread.php?t=320001 Happy reading. Greg
Jul 14, 2005, 12:59 PM
Member
Edina, MN, USA
Joined Oct 1999
11,556 Posts
Have read the referenced document(s); though I’ve read far better references in my work and education as a propulsion engineer.

In the case of an F-86 or F-16 the inlet doesn’t care if the mass flow is established by a turbo fan, turbojet, EDF, or vacuum cleaner.

Frankly, I’ve done a lot more wind tunnel tests of inlets using a vacuum system; though quite a few with the actual engine as well.

Likewise, the exit/nozzle doesn't care either.

A rotating propulsive device that’s closer to its inlet (or exit) than a few diameters will have a 3D rather 2D flow field which will affect its performance, i.e. a B-747 nacelle.

# Images

Jul 14, 2005, 01:36 PM
Registered User
Joined Nov 2002
1,568 Posts
Quote:
 Originally Posted by gkamysz Do some studying of inlet flows and you will find that what Jim posted is correct. Klaus has also gone into detail.
Didn't say it wasn't...

Read all those articles, and Klaus has some excellent diagrams included in the threads.

Basically, I just wanted to point out that using a turbojet diagram might cause unnecessary confusion, and the article was to point out the interesting, but opposite, principles of accelerated/decelerated airflow within the fan itself.

I guess these designers are talking specifically about a very narrow "podded" DF, which has it's own set of issues, which may, or may not, exist in a "long" ducted scenario.

Also, Klaus' "full-system" diagrams show exactly what I had anticipated for the shape of the duct requirements.

The enlighting part which, from what I can tell, no-one else had suggested to date was the concept of a continually decreasing area, starting from a larger-than-FSA area to a smaller-than-FSA. (everything I'd read before suggest a constant duct area, with a quick taper at the end)

Very interesting stuff...
 Jul 14, 2005, 02:33 PM Dieselized User Chicagoland Joined Feb 2000 7,530 Posts The turbojet diagram does not cause confusion. The inlet data is the same, like Jim, said regardless if you are feeding an EDF, turbojet, or vacuum. The turbine inlets that book is referring to without specifically saying so, though I bet it's in the book if you actaully care to order a copy and read it, is any supersonic inlet, which operates on a completely different set of principles. Have you read that book or any other book on inlet aerodynamics? Greg
Jul 14, 2005, 03:14 PM
Registered User
Joined Nov 2002
1,568 Posts
Quote:
 Originally Posted by gkamysz Have you read that book or any other book on inlet aerodynamics?
Yes and yes... Perhaps you should try the one linked, it's very informative, and covers topics quite relevant to us, as their design parameters are specifically the 0-200 knot range, with a heavy emphasis on the compromise between static and dynamic thrust. (a big issue for most scale EDF'ers)

Regardless to this amusing little side-trip, I cobbled together an Excel spreadsheet to calculate efflux velocity from either FSA or exit duct size.

As another chap mentioned, it would be neat to see some sort of empirical verification of this mathematical model, and see if the "real-world" data trends up or down from this baseline.

A very good pair of related discussions, very informative.
 Jul 14, 2005, 03:31 PM Member Edina, MN, USA Joined Oct 1999 11,556 Posts Both Greg’s and my spreadsheets do this well; I’ve incorporated “constants” that let me match thrust and velocity based on empirical data.
Jul 14, 2005, 03:34 PM
Dieselized User
Chicagoland
Joined Feb 2000
7,530 Posts
Quote:
 Yes and yes... Perhaps you should try the one linked, it's very informative, and covers topics quite relevant to us, as their design parameters are specifically the 0-200 knot range, with a heavy emphasis on the compromise between static and dynamic thrust. (a big issue for most scale EDF'ers)
Then I don't understand why we have the questions. Maybe I should be sitting in your lecture hall.

The book didn't give the formula to get velocity from thrust? The books I've read did. Yes, the real world conforms to the paper world. I've done the tests.

Greg
 Jul 14, 2005, 03:55 PM Stilwell Shipyard United States, FL, Cape Coral Joined Nov 2003 2,100 Posts All I know is that I am getting a headache! Hurricane Larry
Jul 14, 2005, 06:59 PM
Registered User
Joined Nov 2002
1,568 Posts

Quote:
 Originally Posted by gkamysz The book didn't give the formula to get velocity from thrust?
Yes, but not in any kind of useful manner. The question was more for a confirmation of my assumptions, and a clarification of the concepts therein.

If you've got a large audience of experts (use that term a bit loosely), why would you not leverage it if you're not sure about a concept/"fact"?

I'm someone who likes to make sure I have the all facts right before charging ahead, and I don't mind at all to be corrected on misunderstandings or misconceptions. (done it many a time, but I always re-iterate the correction in the thread, to help others who may be equally lost)

As a general aside, I must say that some people clearly understand what it means to answer a question in direct and clear way, where some others unfortunately don't.

I found it astounding how you can get three distinctly different "answers" to a simple question. (the following examples are not an indictment of the person, just an observation to the -actual- replies garnered...)

For example, my asking for a simple formula for efflux velocity from thrust:

Quote:
 Originally Posted by jrb The following is a list of 1st principles and relations that might be interesting/useful: F = MA, but Newton actually said F = the time rate of change (derivative) of momentum dMV (mass times velocity); in our case the derivative of mass (mass flow -- airflow of prop/EDF) times the velocity of that airflow is of most interest! P = dMV^2 V = rpm * pitch dM = dia * pitch * rpm F = dia * pitch^2 * rpm^2 P = dia * pitch^3 * rpm^3 Note: appropriate units, dia/area relationship, and constants (g, pi, etc.) apply. From the above I get the calcs I need in both my PropCalc & FanCalc performance predication spreadsheets (i.e. like MotoCalc, etc.).
Kind of you to reply, but this is essentially a useless answer to a very direct question!

I can only assume you knew the exact formula that the next two posters replied with! If you're going to take the effort to respond, why on earth would you not respond with a simple succinct answer, in response to the actual question?

Very puzzling...

----------

Quote:
 Originally Posted by sguty Here's some relationships that deal just with static thrust, exhaust area, and efflux velocity. which should at least get you started: exhaust area = mass flow / (air density * efflux velocity) mass flow = thrust / efflux velocity Plugging the second into the first gives you: exhaust area = thrust / (air density * (efflux velocity)^2) which should give you: efflux velocity = sqrt(thrust/(air density*exhaust area))
Hey! Now we're getting somewhere, but the details are still a little fuzzy! (vast improvement over #1 though)

----------

Quote:
 Originally Posted by peterangus Here is a derivation of the relationship. force = rate of change of momentum Applying the above to an EDF bench test: static thrust = mass flow x efflux velocity = volume flow x air density x efflux velocity T = static thrust [newtons] A = exit area [square metres] D = air density = 1.23kg per cubic metre at sea level V = efflux velocity [metres per second] T = A x V x D x V = ADV^2 V = [T/AD]^0.5 ------------------------------------------------------------------------ I have no good data to test the above relationship. Can someone supply some data? ------------------------------------------------------------------------ Conversion factors: 1 newton = 0.225 lbf 1 metre per second = 2.24 mph
Hallelujah! Someone who understands what it means to be complete and through, yet very easy to understand!

Exactly what I was after! (actually more than above the call!)

This clear, easy to understand answer now serves the entire EDF community, not just me! (if only there way some way to gather these pearls, and make them easy to find!)

In closing, We have to keep in mind that these forums are not just idle banter between the participants alone.

There are hundreds, perhaps thousands, who will read these threads, and it behooves us all who (perhaps) know more than them to be accurate, understandable, and complete when you are trying to illustrate a concept.

There's nothing more frustrating to have these "knowledge carrots" dangled in front of you which simply don't contain enough information, or are simply too obtuse/difficult to understand the poster's point.

I have always made it a point (perhaps even my mission) to cut through the crap and try to present information in an easy to understand manner, in a way that most people will comprehend the "big picture".

If this means being apparently redundant with asking the same question, or perhaps posting apparently similar "answers/observations", fine!

If someone finally "gets it" when reading the thread, that's exactly what these forums need to accomplish. These forums are not a place to spout half-truths, or be used as a vehicle for intellectual posturing.

Soapbox has been put away...

Thanks.
 Jul 14, 2005, 10:21 PM Member Edina, MN, USA Joined Oct 1999 11,556 Posts BOL Meteor! By they way I hope you can use pen & paper should the batteries on your calculator go bad, or your PC fails to boot. As you said, plenty of folks around here that will likely respond to your postings, and a few others that won’t. Last edited by jrb; Jul 15, 2005 at 09:14 AM.