logo
NOTICE:  This is the new PunchCAD forum. You should have received an email with your new password around August 27, 2014. If you did not, or would like it reset, simply use the Lost Password feature, and enter Answer as the security answer.
Welcome Guest! To enable all features please Login or Register.

Notification

Icon
Error

Options
Go to last post Go to first unread
Art Smith  
#1 Posted : Friday, October 25, 2019 11:30:10 AM(UTC)
Art Smith

Rank: Senior Member

Joined: 1/16/2015(UTC)
Posts: 329

Thanks: 42 times
Was thanked: 11 time(s) in 9 post(s)
does anyone have any experience or thoughts on how to "dimple" (ie: like a golf ball) a non-linear surface? I've thought about and discarded projecting circles onto the surface because of the distortion resulting from the non-linear curved surface and because at some point 3D solids (ie: spheres) will be required to subtract from the baseline solid model's surface. any help or thoughts will be greatly appreciated!

Art
[email protected]
rockyroad_us  
#2 Posted : Friday, October 25, 2019 12:44:10 PM(UTC)
rockyroad_us

Rank: Senior Member

Joined: 9/12/2009(UTC)
Posts: 546
Man
United States
Location: In my place

Was thanked: 24 time(s) in 23 post(s)
well, you can do a negative for the dimple and do a spherical pattern with the pattern command in 3D solids. You might have to get creative to take up all the empty spaces but that would be a starting point. Maybe do a pattern along projected curves on the sphere too.

maybe you can share your non-linear surface but I would make that into a solid to work with the pattern command.
FX 9-1224 Pro10-1371
OSX.11.6
2x2.8GHz Quad Xeon
32GB
Radeon HD6870 2GB
Work
Art Smith  
#3 Posted : Friday, October 25, 2019 1:32:24 PM(UTC)
Art Smith

Rank: Senior Member

Joined: 1/16/2015(UTC)
Posts: 329

Thanks: 42 times
Was thanked: 11 time(s) in 9 post(s)
see attached image of curved elliptical surface. patterns is a great place to start but I have no idea how to "wrap" the non-linear surface.............

Art
[email protected]
Art Smith attached the following image(s):
intake horn image for PunchCAD forum.jpg (45kb) downloaded 0 time(s).

You cannot view/download attachments. Try to login or register.
rockyroad_us  
#4 Posted : Friday, October 25, 2019 2:14:42 PM(UTC)
rockyroad_us

Rank: Senior Member

Joined: 9/12/2009(UTC)
Posts: 546
Man
United States
Location: In my place

Was thanked: 24 time(s) in 23 post(s)
so you want to dimple the inside of the ear cones? If so, you need to put a dimple to start.
FX 9-1224 Pro10-1371
OSX.11.6
2x2.8GHz Quad Xeon
32GB
Radeon HD6870 2GB
Work
murray  
#5 Posted : Friday, October 25, 2019 3:06:58 PM(UTC)
murray

Rank: Senior Member

Joined: 9/24/2014(UTC)
Posts: 275
Australia

Thanks: 5 times
Was thanked: 39 time(s) in 34 post(s)
I'd use the grid intersections of an isoline layout on the trumpets' inner surface as the pattern to locate the dimples, in the absence of any other definitive parameter, and a shell thickness offset to specify indentation depth. I'm interested in the idea of this: intake trumpets are supposed to tidy up airflow to maximise airflow velocity, while dimples on golf balls are said to stabilise the ball in flight by disrupting laminar flow around the rotating ball on its trajectory. These seem inconsistent aims to me.

Edited by user Friday, October 25, 2019 3:11:02 PM(UTC)  | Reason: Not specified

Art Smith  
#6 Posted : Friday, October 25, 2019 3:54:38 PM(UTC)
Art Smith

Rank: Senior Member

Joined: 1/16/2015(UTC)
Posts: 329

Thanks: 42 times
Was thanked: 11 time(s) in 9 post(s)
Murray-

thanks for your input! where to I find the tool for invoking the isoline layout on the intake horn's surface. given a "wrapped" grid, I'll likely use the Powerpack tool for distance normal to a surface at a point to locate the center of the subtraction spheres.

the top level objective is getting more air into a carburetor. that implies using all the horn's area for air flow; ie: attached flow on the short side radius. dimples are an aerodynamic "trick" to convert laminar boundary layers to turbulent boundary layers which stay attached farther around a curved surface. it's why golf balls with dimples fly farther than spherical objects the same size and weight; less profile drag as a direct result of the turbulent boundary layer staying attached further. most introductory texts on fluid mechanics or aerodynamics are likely to have illustrations and further/better explanations of what's going on.

Art
[email protected]

Edited by user Friday, October 25, 2019 6:11:41 PM(UTC)  | Reason: Not specified

murray  
#7 Posted : Friday, October 25, 2019 5:50:50 PM(UTC)
murray

Rank: Senior Member

Joined: 9/24/2014(UTC)
Posts: 275
Australia

Thanks: 5 times
Was thanked: 39 time(s) in 34 post(s)
Isoline parameters are in the display tab of object properties in the Inspector. Thanks for the explanation of the theory, I've never seen that implemented on intake horns in fifty years of involvement and observation of engine development, so there might be something new under the sun after all.
Art Smith  
#8 Posted : Friday, October 25, 2019 8:54:27 PM(UTC)
Art Smith

Rank: Senior Member

Joined: 1/16/2015(UTC)
Posts: 329

Thanks: 42 times
Was thanked: 11 time(s) in 9 post(s)
thanks for the pointer on isolines!! I'd never changed the zeros before so had never stumbled across the utility of the tool. is there any way to select a switch such that the isoline intersections become selectable?? the grid is outstanding but without a means to select the intersection of the vertical and horizontal lines I don't see how they can help me. cutting and fitting line segments to the surface is definitely doable, just a lot of work.....! line segments would be the enabling vehicle for arrays of equally spaced "points" on the surface of the horn(ie: equivalent to horizontal & vertical isoline intersections).

the reason you probably haven't seen dimpled surfaces on intake horns before is there are very few carburetors with siameased venturi. there is no need/value for dimples with straight elliptical horns. Weber 32/36 carburetors (ie: the one used in Formula Ford, Formula Continental, & Sports 2000) are one of the carburetors with the siameased "feature". to implement the geometry advocated by Prof. Blair & W. Melvin Cahoon in their paper "Best Bell", something other than two straight elliptical horns is required. the non-linear center line of the elliptical horns is the root cause of the resulting short side radius and the subsequent need for the dimples to maintain attached flow.

see custompolycast.com for an example of currently available dimpled intake horns.

Art
[email protected]

Edited by user Friday, October 25, 2019 9:22:33 PM(UTC)  | Reason: Not specified

rockyroad_us  
#9 Posted : Friday, October 25, 2019 9:53:35 PM(UTC)
rockyroad_us

Rank: Senior Member

Joined: 9/12/2009(UTC)
Posts: 546
Man
United States
Location: In my place

Was thanked: 24 time(s) in 23 post(s)
You would need to copy and paste the model, then do the isolines per the number you want and then once the iso lines are shown you can then change the object type to lines to obtain them in space.
FX 9-1224 Pro10-1371
OSX.11.6
2x2.8GHz Quad Xeon
32GB
Radeon HD6870 2GB
Work
Art Smith  
#10 Posted : Friday, October 25, 2019 11:13:26 PM(UTC)
Art Smith

Rank: Senior Member

Joined: 1/16/2015(UTC)
Posts: 329

Thanks: 42 times
Was thanked: 11 time(s) in 9 post(s)
rockyroad_us-

thanks for the input! where/how do I change the object type? assume you have a slow student......

Art
[email protected]
murray  
#11 Posted : Saturday, October 26, 2019 4:57:19 AM(UTC)
murray

Rank: Senior Member

Joined: 9/24/2014(UTC)
Posts: 275
Australia

Thanks: 5 times
Was thanked: 39 time(s) in 34 post(s)
Select and right click context menu.

Edited by user Saturday, October 26, 2019 4:59:34 AM(UTC)  | Reason: Not specified

Art Smith  
#12 Posted : Saturday, October 26, 2019 10:53:54 AM(UTC)
Art Smith

Rank: Senior Member

Joined: 1/16/2015(UTC)
Posts: 329

Thanks: 42 times
Was thanked: 11 time(s) in 9 post(s)
Murray-

thanks! unfortunately I'm still doing the two steps forward followed by one step back game...... changing the object type also changes the grid spacing?? I've confirmed after chancing the object type the grid vertices are selectable which is major progress. now the question is how to maintain control of the grid spacing; it's critical for imposing the desired dimpling pattern uniformly onto the surface. see attached image below.

Art
[email protected]
Art Smith attached the following image(s):
intake horn grid conversion challenge.jpg (82kb) downloaded 5 time(s).

You cannot view/download attachments. Try to login or register.
murray  
#13 Posted : Saturday, October 26, 2019 6:10:23 PM(UTC)
murray

Rank: Senior Member

Joined: 9/24/2014(UTC)
Posts: 275
Australia

Thanks: 5 times
Was thanked: 39 time(s) in 34 post(s)
I'd use multiple copies of the surface with different isoline density and combine those that have a distribution close to what you want, and you also have the capacity to elevate the order of curves to increase the control point density, and to convert the isocurves into polylines or line segments that might give more informative features.

Thinking on it, this is a "kissing circle/number" problem, a research topic in its own right.

I wonder what's the outcome if ridges are used instead of dimples? I'm thinking like the rippling on sand dunes.

alternatively, variation in the depth and/or radius of the dimpling dependent on surface curvature.

Edited by user Saturday, October 26, 2019 7:06:04 PM(UTC)  | Reason: Not specified

rockyroad_us  
#14 Posted : Sunday, October 27, 2019 10:00:53 AM(UTC)
rockyroad_us

Rank: Senior Member

Joined: 9/12/2009(UTC)
Posts: 546
Man
United States
Location: In my place

Was thanked: 24 time(s) in 23 post(s)
Not sure how the additional iso lines are changed. I didn't get that result. see picture.

Maybe that's version dependent. I'm on a mac.
rockyroad_us attached the following image(s):
iso1.jpg (115kb) downloaded 0 time(s).

You cannot view/download attachments. Try to login or register.
FX 9-1224 Pro10-1371
OSX.11.6
2x2.8GHz Quad Xeon
32GB
Radeon HD6870 2GB
Work
Art Smith  
#15 Posted : Sunday, October 27, 2019 2:52:10 PM(UTC)
Art Smith

Rank: Senior Member

Joined: 1/16/2015(UTC)
Posts: 329

Thanks: 42 times
Was thanked: 11 time(s) in 9 post(s)
thanks again guys!! the dimpling challenge is solved. in the end I resorted to brute force descriptive geometry to impose the dimpling geometry on the curved elliptical horn's inner surface. to avoid the science project implicit in dimpling the horn's diminishing radius surface, dimple geometry and dimple spacing in the velocity direction were fixed. dimple spacing in the circumferential direction was allowed to float as a means to maintain desired row-to-row relative geometry. circumferential spacing was defined by the smallest row radius (ie: the inner one); the initial guess of 19 vertices for 120° turned out to be too small requiring all the work to be redone with 26................

the work was done with V10 (1371) including the PowerPack on a Windows 7 Pro machine. PowerPack's "line normal to a surface" tool was instrumental to solving the challenge!!

Art
[email protected]
Art Smith attached the following image(s):
dimpled intake horn.jpg (57kb) downloaded 6 time(s).

You cannot view/download attachments. Try to login or register.
murray  
#16 Posted : Sunday, October 27, 2019 5:30:25 PM(UTC)
murray

Rank: Senior Member

Joined: 9/24/2014(UTC)
Posts: 275
Australia

Thanks: 5 times
Was thanked: 39 time(s) in 34 post(s)
Good going, Art. If it ever gets on a dyno, show off the results to us?
Users browsing this topic
Guest
Forum Jump  
You cannot post new topics in this forum.
You cannot reply to topics in this forum.
You cannot delete your posts in this forum.
You cannot edit your posts in this forum.
You cannot create polls in this forum.
You cannot vote in polls in this forum.