Wool Tuft Testing - MkIV Jetta

[trebuchet03]

So I had a 200 mile drive today (which is why I didn't do the one tank of gas challenge contest thinger :p), and before I left I added some bits of yarn.

For those wondering how to do it..... Here's some tips

* Pick a tuft color for good contrast (black and yellow in my case).
* Try to follow some sort of grid for placement with lines following what you think the air will do.
* Try to avoid tangles. If you expect an area will cause the strands to whip around, make sure there's enough clearance between each strand to prevent entanglement


Front:
http://www.gassavers.org/attachment.php?attachmentid=1051&stc=1&d=1194753261
I added a few more after this picture - and a couple long ones near the hood trailing edge to watch the hood<-->glass flow interaction/transition.

Rear:
http://www.gassavers.org/attachment.php?attachmentid=1052&stc=1&d=1194753261
Adjusted Rear:
http://www.gassavers.org/attachment.php?attachmentid=1053&stc=1&d=1194753261

45mph Hood:
http://www.gassavers.org/attachment.php?attachmentid=1054&stc=1&d=1194753261
Nice and straight.
45mph Glass:
http://www.gassavers.org/attachment.php?attachmentid=1055&stc=1&d=1194753261
Also nice and straight.
More coming in a flash, including results from drafting a semi o.0

[trebuchet03]

Rear Glass - 65mph - No Draft
http://www.gassavers.org/attachment.php?attachmentid=1056&stc=1&d=1194754319
Another Angle
http://www.gassavers.org/attachment.php?attachmentid=1059&stc=1&d=1194754319

The stills don't tell the entire picture... Some of those tufts did complete circles - wildly. While in the turbulant flow of a semi, this changed dramatically (see next post).

It looks like flow separation occurs rather quickly, something more likely to happen on a sedan with a steeper glass angle compared to a long sloping hatch lid.

Front Glass - 65mph - no Draft
http://www.gassavers.org/attachment.php?attachmentid=1057&stc=1&d=1194754319

Hood - 65mph - no Draft
http://www.gassavers.org/attachment.php?attachmentid=1058&stc=1&d=1194754319
Notice the long tufts near the trailing edge of the hood get sucked under - there was no chance of it getting pushed up onto the glass. Room for improvement methinks :)

[trebuchet03]

I have video... But it's going to take some time to get it online...

Hood - 65mph - behind a semi (104" wide)

http://www.gassavers.org/attachment.php?attachmentid=1060&stc=1&d=1194755001
A single photo doesn't say what's happening. The video shows that the string stay straight (they don't curl) but go through pulses of left and right motions. Large eddies coming off the rear of the semi truck making it easier for my car to cut through the air :)


Glass - 65mph -behind a semi (104" wide)
http://www.gassavers.org/attachment.php?attachmentid=1061&stc=1&d=1194755001
The same applies here.

Rear Glass - 65mph - behind a semi (104" wide)
http://www.gassavers.org/attachment.php?attachmentid=1062&stc=1&d=1194755001

Here's where the very interesting stuff happens. Not drafting anything, tufts wildly go in circles - occasionally completely detaching for a few moments. While behind the semi, not 1 tuft did a 360. But there was some tuft curling - indicating that the flow was either inconsistent and/or weak (likely both - especially the former). To me, this means the flow is sticking nearby.

Tuft conditions post experiment:
http://www.gassavers.org/attachment.php?attachmentid=1063&stc=1&d=1194755001

Rear tufts, post test... They are quite frayed - indicated wild changes. The fronts and some of the rear (near the sides and top) were completely fine :thumbup:

[trebuchet03]

This space for rent... Just in case I feel I need to upload more pictures for someone ;)

[ajohnmeyer]

Nice shots so far. So have you come up with any ways to use all of the subjective "data" that you collected during your trip?

[ChrstphrR]

Since I've a Mk IV like you, of course I'm interested in the results -- thanks for the pictures!

Rear Glass - 65mph - No Draft
Notice the long tufts near the trailing edge of the hood get sucked under - there was no chance of it getting pushed up onto the glass. Room for improvement methinks :)

When I was waxing my Jetta a few weeks ago, in preparation for winter (hey, it's Canada, not Florida here!), I was thinking about the big gap between the top of the hood, and the cavity where the wipers reside.

It'd be a nice spot for a fairing, that still allows for the wipers to function.

I don't want to hijack the thread here, but I did want to throw out that idea, since you were pointing to that very spot on your tuft-testing.

[trebuchet03]

Nice shots so far. So have you come up with any ways to use all of the subjective "data" that you collected during your trip?

Extra emphasis on subjective :D

Actually, ChrstphrR took the words from my mouth :p

I was thinking about the big gap between the top of the hood, and the cavity where the wipers reside.

It'd be a nice spot for a fairing, that still allows for the wipers to function.

Which is actually, if I recall correctly, a key point of an old Autospeed article.

Additionally, I want to do some research on methods to trip flow into turbulence.


I don't want to hijack the thread here, but I did want to throw out that idea, since you were pointing to that very spot on your tuft-testing.

Not at all :thumbup: It's perfectly on topic :p

[ajohnmeyer]

I have been thinking about the same thing for my Civic. The obvious solution (at least to me) is a cowl-induction style hood, as one would typically see on a camaro or other such muscle car. Unfortunately, this style doesn't find it's way into the "tuner" (read economy) car market. This is about as good as it gets....IF you have a 6th gen Civic.


http://www.bodykits.com/store/image.php?type=P&id=32098

looks like you Jetta folks have the same long odds on cowl hoods. Though there seems to be some hope on the adhesive-cowl front.

http://i239.photobucket.com/albums/ff289/ajohnmeyer/hood2.jpg

[ChrstphrR]

Which is actually, if I recall correctly, a key point of an old Autospeed article.

Additionally, I want to do some research on methods to trip flow into turbulence.

Well, digging about when you mentioned it... I think this is likely the Autospeed article you had in mind, no?

"Silencing Wiper Wind Noise" - Autospeed, Feb. 26, 2005
http://www.autospeed.com/cms/A_2445/article.html

It looks as though the fellow that did this article managed to quiet down wind noise, but didn't get any appreciable change in fuel economy.

Mind you, the modification he did is quite a bit smaller/shorter than I was envisioning while waxing my car -- the Kamm effect strikes again, I think.
It's also a stealthy change -- the prototype's noticable, but the finished deflector looks relatively stock in appearance.

Little elegant solutions like this make me regret changing away from Mech. Eng. :P

trebuchet03's post here, and Julian Edgar's article are both giving me the urge to go out and buy a ball of wool and some masking tape.

[2TonJellyBean]

treb, that's just wild seeing those pix!!!

Thanks for posting!

[ZugyNA]

The only way you'd gain from drafting that semi would be to sit 5' to 10' max off his bumper...something I don't recommend. Otherwise you are just being buffeted by the big swirls of air.

[trebuchet03]

The only way you'd gain from drafting that semi would be to sit 5' to 10' max off his bumper...something I don't recommend.

Mythbusters had some cool testing done at NASA Ames Research in a scale tunnel.... At 7 car lengths (~100 feet from the truck) - they found a drag reduction of 11%. At 10 feet, 39% drag reduction. I never got as close as 10 feet, but at the distances shown in my pictures - I was getting a fuel economy increase (37.8 versus ~41.6 - a/c off, resetting the SG on long flats using CC).

My overall FE for the entire trip (200 mi) was 39.0 - dropping in the end from 39.3 when I was forced by my passenger to turn the a/c on :thumbdown:

Otherwise you are just being buffeted by the big swirls of air.
Even very close, you're being hit by tangential air flow (but lower energy, slower). That will not be drag against forward motion - that is, instead of the energy being in the parallel direction to motion, it's tangential - and the only fluid velocity that matters (in our case) is flow parallel to the direction of motion :thumbup: I should build an anemometer to get some data :)

[Hateful]

What drafting I've done was not very close and ScanGauge does show an increase in MPG's. There is turbulence from the sides but the straight on air resistance is reduced. I've even drafted a tiny little Nissan truck with a camper shell at least a car length back and got just as good an increase.
An all or nothing attitude will normally yield the later.

I bet cats would have a ball playing on that car with the yarn.

[trebuchet03]

So, I have an interesting update :)

Unfortunately, I didn't get pictures - due to low light conditions on the ride home. BUT - the difference between cracking the window open and having it shut is huge!

With the window shut, the rear tufts flicker back and forth as I would expect. BUT, crack the window open (Even just an inch) and those tufts go crazy - fluttering back and forth much faster and occasionally separating from the glass (but only very briefly).

Just food for though :) I'm sure cracking is better than full open, but I didn't realize the effect was that great.

[ChrstphrR]

trebuchet03,

When you're taking pictures ... are you taking them yourself, or do you have someone else in the car behind the lens?

[trebuchet03]

trebuchet03,

When you're taking pictures ... are you taking them yourself, or do you have someone else in the car behind the lens?

A passenger took almost all of them... I took a couple - but really, that's dangerous :thumbdown: All the ones behind the semi were done by a passenger ;)

[ChrstphrR]

I was more worried that there was the chance it was you taking the ones of the back window :D

[omgwtfbyobbq]

When looking at the car alone versus drafting the semi, doesn't the change in behavior lend itself to the change in drag coefficient wrt speed I've read about?

[trebuchet03]

When looking at the car alone versus drafting the semi, doesn't the change in behavior lend itself to the change in drag coefficient wrt speed I've read about?

I'm cautious in saying yes... Tufts, smoke, oil etc. are used to visualize flow interaction and really shouldn't be used as a metric to compare cD values.

There is a relation between cD and Reynolds number (which is a function of velocity, length et. al.) :thumbup:

http://www.aerospaceweb.org/question/aerodynamics/drag/drag-disk.jpg

[omgwtfbyobbq]

Can't we get away w/ it for high drag/low pressure areas like behind a car? I mean, lets say I'm Ernie and I do tuft tests on my Beetle in order to see if there's a difference via the spoiler. Wouldn't less eddies as indicated by the tuft imply there's a decent chance of a drop in Cd? Or for that matter, if the tufts eddy at one speed but not another, wouldn't this imply that the flow hasn't detached at the point and some speed but may detach higher up at some higher speed, leading to greater drag?

I feel it should be like load wrt efficiency in an engine, but I suppose I'm just grossly oversimplifying it. :D

[trebuchet03]

Can't we get away w/ it for high drag/low pressure areas like behind a car? I mean, lets say I'm Ernie and I do tuft tests on my Beetle in order to see if there's a difference via the spoiler. Wouldn't less eddies as indicated by the tuft imply there's a decent chance of a drop in Cd? Or for that matter, if the tufts eddy at one speed but not another, wouldn't this imply that the flow hasn't detached at the point and some speed but may detach higher up at some higher speed, leading to greater drag?

I feel it should be like load wrt efficiency in an engine, but I suppose I'm just grossly oversimplifying it. :D

Lets say it this way.... It's not that there's a change in wake due to a change in cD. It's the other way around - there's a change in cD due to a change in wake. I know it sounds almost exactly the same - I'm just kinda anal about little details like that :p

But yes, I intuitively agree... See less wild tufts - there's a greater chance for more favorable aero.... With respect to the semi, I think it's a combination of perturbed flow AND slower relative velocity (think corridor effect) :thumbup: That's why I'd like to put an anemometer on my car :D

[Biffmeistro]

You may not want wild tufts blowing everywhere on your car, but you want all the tufts to get a nice stream of air. The lack of motion on the back window tufts indicates flow separation from the back window, which is undesirable. I'd recommend doing something along the lines of vortex generators on the rear of your roof to pull the flow back down onto the rear window.

[trebuchet03]

You may not want wild tufts blowing everywhere on your car, but you want all the tufts to get a nice stream of air. The lack of motion on the back window tufts indicates flow separation from the back window, which is undesirable. I'd recommend doing something along the lines of vortex generators on the rear of your roof to pull the flow back down onto the rear window.

Lack of motion? That's not a problem, there's too much motion as stated in my original posts - so much so that the tufts do complete 360's in certain places...

Here's where the very interesting stuff happens. Not drafting anything, tufts wildly go in circles - occasionally completely detaching for a few moments. While behind the semi, not 1 tuft did a 360. But there was some tuft curling - indicating that the flow was either inconsistent and/or weak (likely both - especially the former). To me, this means the flow is sticking nearby.

I do agree that some form of flow control could be beneficial - I have a non VG concept solution that I haven't seen applied to cars before that I'll be tinkering with within the next month ;)