A simple experiment to try

[pezman]

I was thinking a little bit about the vortex motion in augmenters and I got to wondering whether these things behave a little like a Ranque (a.k.a. Hilsch) tube. Here's the idea:

• - Presume that the vortex is a torus
  - Presume that the fluid rotating about the concentric layers of the torus rotates at the same frequency.
  - Presume that the outside edge of the torus must therefor move faster than the inner portion of the torus.
  - Presume that the speed of the moving air reflects its temperature

If these assumptions are true, then it would suggest that mean temperatures measured near the center of the augmenter and just inside the inner edge of the augmenter (i.e. the outer shell of the torus) would be higher than mean temperatures at the "fat spot" of the torus. The temperature profile should look kind of like a sinc function (concentric "hot in center", then cooler, then hotter) or possible an inverted sinc (cool in center, hotter at inner boundary of torus, then cooler, then hotter at outer boundary of torus).

I'm not sure that this conjecture is true, and I'm not certain that long-term average temperature measurements would be able to capture these transient effects, but its a pretty simple experiment - basically, just move a thermometer across the profile of an augmenter and measure the temperature.

Said differently, the existence of a toroidal temperature profile like this would be pretty good evidence for a "Ranque" effect in augmenters, but the absence of such a profile would not rle out a Ranque effect.

If the effect exists in augmenters, then it is likely that the effect is also present in "expanded tail" designs.

If the effect is present, then it might be possible to exploit it in some way to improve engine performance.
[/list]

[Jim Berquist]

Are you thinking something like this?


Jim

[Rossco]

Im up for some cfd if you want to draw a simple picture of your envisioned layout.
Ill put together a simple gap separated augmentor in the meantime and see what sort of vortex looks to be created.

Rossco

[pezman]

I'm not sure if a CFD program would capture the effect. It would be an interesting test to see if you could model a Ranque tube using a CFD program.

BTW, the CFD posts that you've been putting up recently are interesting -- what software are you using??

At any rate, for a real-world test, putting RTDs at various points along the radius of an augmenter or across the exit of an expanded tail jet should tell if there is a cool torus in the middle of the toroidal vortex.

[Rossco]

I use Cosmos at the moment, embedded in Solidworks... It is not ideal, and i am in the process of setting up for Ansys. Still using cosmos for this one.

OK, i agree, i dont know if i can simulate, let alone capture the effect you are speaking of. I do however think that i can test your assumptions there. Its a learning experience that i am after, and if i can do something at least interesting even usefull, i will be happy.

Here is the setup as i have it.
Everything isnt dimentioned, just built as i sketch... i can change anything if you have some suggestions.

The gap is set to be 1.5 times the diameter of the outlet... again, any suggestions?

EDIT: The clear section around the gap is an environmental pressure interface... ignore it

Rossco

[Jim Berquist]

I got to get a CAD program!

Do I follow this right? Your trying to use the end cone on the jet as a Augmentation?

Jim

[Rossco]

Berq... not sure what you mean there?

Here she is... .00082 s in... forming up a votex...
Velocity is a bit up there!??? i just chucked 2atm in the cc at 1000 C.

EDIT: heres one of the temp too...
The little arrows are the induced air into the augmentor... interesting

EDIT again, and again since it wouldnt post: Forgot to add the pic.

Rossco

[Rossco]

OK, now at just over .001 she is starting to agment!

Not a lot of temp variation in the vortez... mmm
Next time i will focus more on that.

Rossco

[Rossco]

ah, 1.5ms and the vortex suddenly collapsed??? i didnt expect that, i thought that it would actualy increase in size as the press bleed off.

ill add a screen shot of what im whatching to determine where its at.

Rossco

[Rossco]

OK, pressures shifted again, now a huge big vortex has set up further back in the augmentor...crazy stuff

Rossco

Actualy, has it chainged direction too?...no i dont think so.

[Jim Berquist]

Now I'm totally confused! That don't take to much! Your program looks to be working with a closed end?

I thought you was working on a way to exstract augmentation from a conincal tail end.

I will shut up and See!

Jim

[Rossco]

OK, let me see if i can remember this right.

At 3ms the cc went atm press.

At 3.5 it hit max neg press.

At 4ms the flow has pretty much compleatly stopped out the intake. Shown below.

Now, last one, the induction. Will it pull up the flow out the augmentor???

Rossco

[Rossco]

4.5ms, and it has started pouring back into the CC...
Ive taken out some of the lines, to show the flow more clearly.
and bumped the size of the arrows up to show direction... maybe not enough?
The Augmentor still has a high V core flowing out the back!

(is this annoying anyone yet?)

Just one more... or so.

Rossco

[Mike Everman]

Very niiiice. Good work, Rossco. Is that the limited cosmos that comes with solidworks, or the full cosmos package?

Thanks for just watching for a bit, Jim. Posting your confusion was getting distracting. Sorry.

[Rossco]

A touch over 5ms and its hit max reverse(intake) velocity of about 301.5 f/s.

Rossco