Diffuser Confusion

Here are our CFD links and discussions about aerodynamics, suspension, driver safety and tyres. Please stick to F1 on this forum.
User avatar
vonk
1
Joined: 15 Apr 2010, 04:49
Location: Virginia, USA

Re: Diffuser Confusion

Post

hollus wrote:The diffuser is using the energy from the rest of the car to create a vacuum, even if it is not powered by itself. The space that was occupied by the car in one moment will have to be occupied by something the moment after the car has passed by, and gases will only be able to fill in this void created behind and under the diffuser at a certain speed. So where the difusser is, one moment before there was no air, and all of a sudden there needs to be a lot of air. Air will have to rush in to fill this space either from under the floor, from the sides, or from behind the car.
Then which would you say is true?

Image
Techno-Babble = Meaningless use of technical terminology to feign knowledge.

User avatar
hollus
Moderator
Joined: 29 Mar 2009, 01:21
Location: Copenhagen, Denmark

Re: Diffuser Confusion

Post

If v is the speed of the car and the gray triangle is the diffuser volume, then the red dot should be at a pressure lower than ambient.
Rivals, not enemies.

User avatar
slimjim8201
12
Joined: 30 Jul 2006, 06:02

Re: Diffuser Confusion

Post

Vonk, your fundamental assumption that air pressure around the body is the same is not true. That said, pressure behind the body is most definitely lower than the atmospheric.

Please note that the first two pressure images have a different max/min legend scale than the final two pressure images.

Air Pressure (Smooth) - Range [-400 to 400 Pa]
Image

Air Pressure (Contours) - [-400 to 400 Pa]
Image

Velocity Profile
Image

Pressure Comparison - [-800 to 400 Pa]
Image

Velocity Comparison
Image

Overall Pressure Comparison - [-800 to 400 Pa]
Image

User avatar
vonk
1
Joined: 15 Apr 2010, 04:49
Location: Virginia, USA

Re: Diffuser Confusion

Post

slimjim8201 wrote:Vonk, your fundamental assumption that air pressure around the body is the same is not true. That said, pressure behind the body is most definitely lower than the atmospheric.

Please note that the first two pressure images have a different max/min legend scale than the final two pressure images.

Air Pressure (Smooth) - Range [-400 to 400 Pa]
Image

Air Pressure (Contours) - [-400 to 400 Pa]
Image

Velocity Profile
Image

Pressure Comparison - [-800 to 400 Pa]
Image

Velocity Comparison
Image

Overall Pressure Comparison - [-800 to 400 Pa]
Image
Thanks for the quick reaction. Nice images. :)

I did not mean to imply that air pressure around the body is the same. Just bad graphics (image fixed).

According to the scales, your model appears to be moving at about 58 mph. Is this a laminar flow model? What’s the Reynolds #? I can’t distinguish the turbulent wake. What am I missing?
Techno-Babble = Meaningless use of technical terminology to feign knowledge.

User avatar
vonk
1
Joined: 15 Apr 2010, 04:49
Location: Virginia, USA

Re: Diffuser Confusion

Post

hollus wrote:If v is the speed of the car and the gray triangle is the diffuser volume, then the red dot should be at a pressure lower than ambient.
Sorry about the unclear original picture.

Image

Same answer?
Techno-Babble = Meaningless use of technical terminology to feign knowledge.

User avatar
hollus
Moderator
Joined: 29 Mar 2009, 01:21
Location: Copenhagen, Denmark

Re: Diffuser Confusion

Post

Same answer. And now we have CFD to back it up. :-)
Rivals, not enemies.

User avatar
horse
6
Joined: 23 Oct 2009, 17:53
Location: Bilbao, ES
Contact:

Re: Diffuser Confusion

Post

vonk wrote:I can’t distinguish the turbulent wake. What am I missing?
Well, these are steady RANS simulations, so turbulence is averaged out in the flow field as a fundamental of the mathematics.

Also, I doubt the model is of fine enough resolution to pick up separation points and track them properly and, also, that is very unsteady phenomena anyway.

Theoretically you should be seeing some net upwash for the diffuser case, but it's a little hard to make out.
"Words are for meaning: when you've got the meaning, you can forget the words." - Chuang Tzu

User avatar
slimjim8201
12
Joined: 30 Jul 2006, 06:02

Re: Diffuser Confusion

Post

horse wrote:
vonk wrote:I can’t distinguish the turbulent wake. What am I missing?
Well, these are steady RANS simulations, so turbulence is averaged out in the flow field as a fundamental of the mathematics.

Also, I doubt the model is of fine enough resolution to pick up separation points and track them properly and, also, that is very unsteady phenomena anyway.

Theoretically you should be seeing some net upwash for the diffuser case, but it's a little hard to make out.
The downstream wake is very visible. I'll post a few images highlighting turbulence intensity shortly.

Horse, you are correct: These are steady RANS simulations but I am utilizing a fully adaptive meshing solution which does a fantastic job of locating wake separation. I shall post a few mesh shots for you as well.

User avatar
horse
6
Joined: 23 Oct 2009, 17:53
Location: Bilbao, ES
Contact:

Re: Diffuser Confusion

Post

Looking forward to them, slimjim. :)
"Words are for meaning: when you've got the meaning, you can forget the words." - Chuang Tzu

User avatar
slimjim8201
12
Joined: 30 Jul 2006, 06:02

Re: Diffuser Confusion

Post

Turbulent Kinetic Energy
Image

Turbulent Energy Dissipation
Image

Turbulence Intensity
Image

Mesh Closeup
Image

Total Domain Mesh
Image

PNSD
3
Joined: 03 Apr 2006, 18:10

Re: Diffuser Confusion

Post

Adaptive meshing? Nice. What program did you use to create the initial mesh? And what was your solver?

User avatar
horse
6
Joined: 23 Oct 2009, 17:53
Location: Bilbao, ES
Contact:

Re: Diffuser Confusion

Post

Cheers slimjim. I wonder if there is any chance you could plot the turbulence for the non-diffuser case?
"Words are for meaning: when you've got the meaning, you can forget the words." - Chuang Tzu

hardingfv32
32
Joined: 03 Apr 2011, 19:42

Re: Diffuser Confusion

Post

Would it be difficult to double the air flow (velocity?) and compare the results?

Thanks
Brian

flyboy2160
84
Joined: 25 Apr 2011, 17:05

Re: Diffuser Confusion

Post

PNSD wrote:Adaptive meshing? Nice. What program did you use to create the initial mesh? And what was your solver?
+1 What adaptive mesher did you use? Thanks.

Norway
0
Joined: 17 Jul 2011, 06:06

Re: Diffuser Confusion

Post

xpensive wrote:
vonk wrote: ...
My MS in aeronautical engineering (RPI) has been helpful in this.
...
Not really sure what that is, but if it's referring to some kind of university degree, you are definetly hiding it well.

Your posts should also benefit greatly from less of what you yourself call; "techno-babble", never ever seen that much word-dropping on F1T before; bounary-layers, separation, stagnation, vortexes, slip-conditions and velocity gradients... #-o
You know, I was mostly reading this entire thread because I found the sheer lack of knowledge certain people have about anything aerodynamic amusing. It's nothing to be ashamed of; It's a very, very complicated subject. This is rocket science.

Then Vonk just had to go and toss around the RPI christened MS in Aeronautics. As someone who went to RPI for the same degree, this is beyond embarrassing. I promise it's a good school. Scout's honor.

Ad hominem attacks aside, I'd like to add one or two points which may or may not help some thought processes. Findings in this field change on the monthly basis, so take what I say with a grain of salt.

1) There is no physical law which states that the air needs to be moving at free stream velocity upon exit of the diffuser. I know for me, it was much easier to understand downforce when I thought the air HAD to move faster under an airfoil simply because it was a longer, curved surface. In reality it's an extremely complex conservation of angular momentum flow problem at work (the Kutta–Joukowski theorem).

2) If you want to think of a diffuser as a pump, there is absolutely work being done. There's a nice big V8 powerplant driving the air through the diffuser.

3) All aerodynamic components on a formula 1 car are functioning in an open system. wings included. Simply because there are gaps connecting free stream to the manipulated flow, does not mean that the manipulated flow is by default at free stream pressure. It may be true for incompressible flows but not for air at sea level. This quote puts it very simply (and curiously unreplied to by vonk):
horse wrote: In even more simple terms your argument that an open system can not generate a pressure difference is absurd. If this were the case planes would never leave the ground.
One last point:
Arguing a point for arguing-a-point's sake is sometimes counterproductive, and almost always annoying.

Post Reply