First of all, thanks for the images
Dipesh1995 wrote: ↑28 Jul 2017, 19:36
I don’t think the streamlines have really showed anything with regards to the problem especially since there are no streamline seeds in the tunnel. I’ve zoomed into my vector scenes and reduced the size of the vectors in Glyph and the vectors are all facing in the right direction at the surface of the wing so there is no flow reversal. The outboard velocity contours show a separation on the pressure side of the three-element cascade wing, this has already been solved, like I said, the CFD is from a slightly older wing.
On the point of the glyph vectors etc, all of your plots you have supplied thus far for velocity are "velocity magnitude" - which is ALWAYS positive (the whole SQRT(x^2 + y^2 + z^2) etc. and if your bulk fluid velocity is a substantial speed, i.e. 240 km/hr, then of course your vectors will appear to always point backwards. You need to plot a contour (or vectors) with just the x-magnitude quantity shown to see whether there is any negative there - and that means it is separated.
On the streamlines point, you supplied surface seeded streamlines, not surface-streamlines. What the actual surface streamlines show is something akin to an oil flow visualization. See below what I mean and how it will look when flow is attached (top) and separated (bottom):
Dipesh1995 wrote: ↑28 Jul 2017, 19:36
I know the wing is not perfect, the image I’ve sent is pretty much a first generation of that style of wing. I can’t develop the wing until I’ve solved the issue with the CFD simulation. I’ve already got a few ideas regarding the development path of the wing which I’ll implement once I’ve ironed out the current issue.
Thats all fine mate - I'm not criticizing your efforts thus far, just offering advice
What you have done is great work - just needs some tweaking is all!!
Dipesh1995 wrote: ↑28 Jul 2017, 19:36
I'm not sure if I agree that the design is totally different, looking at the underside at least of the front wing of the Red Bull (shown in it's thread pg45) and ignoring that its delta shape, the vortex tunnel is very similar to what I've got however like I said, the gradient of the tunnel on my wing increases much earlier. This coupled with the approximate geometry of the tunnel compared to rest of the wing and comparing that with my wing makes me think this isn't a substantial difference certainly in the conceptual design of the tunnel if not in the finer details. I’ve supplied an image of the underside of my wing at a similar angle so you can compare.
Yeah... See what I was getting at is that yes, you have "included" a lot of F1 aero devices and techniques etc, and so (at a fundamental level) your design
is similar. What I was hinting at is that F1 aero is so sensitive, you cannot simply just "copy" it and assume that it will work because it looks similar, you know? Of course, I'm sure you know this, but yeah... important to keep in mind just how "different" a wing it is.
Dipesh1995 wrote: ↑28 Jul 2017, 19:36
This is the outboard section and the slice is straight through the vortex tunnel. This the root cause of the problem and what I am having trouble with figuring it out.
Okay, here you supply a picture of the wing out of ground effect and the vortex forms nicely - cool. Ground effect is the issue with your tunnel design. I assume you know all about the venturi effect etc. and about how a wing in close proximity to the ground generates a larger proportion of downforce yadda yadda, you know this stuff. So what is happening is that when you are in free stream, the boundary layer (under normal freestream conditions) is fully capable of resisting the adverse pressure gradient on the aerofoils - that is to say that it is easily able to follow the curvature of the aerofoil. When you place the wing in close proximity to the ground, what is happening, is that the greater suction force present due to the venturi effect, any "diffuser pumping"-like aero behaviour etc, is causing the boundary layer to not be able to resist that now BIGGER adverse pressure gradient which is generated by the larger pressure drop relative to the freestream under the aerofoil.
The fact that that slice was through the tunnel of the wing, makes it more clear to be honest... the tunnel is causing that massive re circulation behind it, because it is separating.
I edited two images (one of the velocity contour of the tunnel and the other of thje skin friction) and pointed out where it is separated. The skin friction I should have stipulated to put it in the x-direction only, but the dark blue is 0.00005 or something, which is near enough to zero to indicate it anyways. If you could redo them scaling it -1 to 1 or something small like that with x-dir only, that would show more.
Everywhere circled in red is separated; there may be more going on, but yeah - this is what is "clear" from the images you sent through.
Dipesh1995 wrote: ↑28 Jul 2017, 19:36
I am going to be testing a two-element wing shortly, similar to wings back in 1990s, to see if I’m getting the same issue with the edge vortex in ground effect. I have a feeling that the issue will be present at which point it’ll almost be certain that it’s not a design issue but a CFD issue.
Im going to be frank here... But the issue is most definitely a design issue...
OR a meshing/setup issue, but most likely the former, in my opinion. It is a great start to a wing, don't get me wrong - but there are clear markers of separation and other typical bad aero phenomena occurring which everyone designing these sorts of complex things experiences,
MYSELF INCLUDED!! But saying that the "CFD is just wrong" or words to that effect, isn't correct.