VW Beetle TDI Kammback
 

I just finished an experiment with my homemade Kammback. Made of cloroplast and packing tape, my design improved my fuel economy remarkably. Typically, my car averages around 38/39 city and 43ish Highway.

With the Kammback, I achieved 55.4 mpg.

http://i29.photobucket.com/albums/c2...f/IMG_0701.jpg
http://i29.photobucket.com/albums/c2...f/IMG_0704.jpg
http://i29.photobucket.com/albums/c2...f/IMG_0703.jpg
http://i29.photobucket.com/albums/c2...f/IMG_0702.jpg



Anything relevant is welcome. Please give me suggestions or comments.

Thanks!

Nice! what was your testing procedure?

That's amazing for such a simple set up.

Well, I've been using fuelly.com for quite a while to track mpgs. (which is where I get previous #s)

For my experiment with the kammback, I drove around town with it for a few days. Then to test it on the highway, I fueled up, hit the highway and came back to the same station and refueled in EXACTLY the same manner, for a round trip of 57.6 miles @ 1.040 gallons.

Unfortunately, I don't have a scanguage at this time. I'd love to get one, though. I think that if I was able to get a "live" or "immediate" reading, it might even reveal more mpgs. I say this simply because I did have to enter/exit the freeway.

Sorry, not trying to rain on your parade, but a 33% gain from before is a bit much (considering MetroMPG only achieved a 15% gain from a full boat tail).

http://ecomodder.com/forum/showthrea...mpg-10691.html

Since you dont have a scanguage, i strongly suggest testing tankful-tankful, instead of relying on small fillups. Small fillups are very misleading because they don't average out any "miscellaneous" factors that could affect your mpg readings. Try to get as many tankful readings as possible to strengthen any mpg results you are achieving.

If you can only do short highway runs, at least try to use the cruise control to cancel out your "foot bias."

I highly recommend reading this thread
http://ecomodder.com/forum/showthrea...ery-11445.html

Beware, it begins with only a test (mine was with cardboard) and a morning you find a chloroplast car in your driveway ;)

You should consolidate the chloroplast as the gravity will make it bend.
This will be accentuated by fast speed for the top and side winds for the sides of your mod.
You should abuse strong tape and monitor it closely before the heat deteriorates it and so your mod flies away...

Does your mod always permits you to open your hatch ?
Please take a picture with the open hatch.

How much your mod reduces the rear view from the steering wheel ?
Please take a picture from inside.

Did you felt a difference under your right foot ?
I can definitively feel a difference with/without the rear wheel skirts, the car accelerates stronger and it's more difficult to keep it at a steady speed when the road isn't perfectly flat.

What about a rear wheel skirts test now :cool: ?

About the scangauge, I have 3 advices : buy it, buy it and buy it.

Have fun,

Denis.

Like jedi_sol wrote, a 33% increase is huge for a mod like this, so expect this gain to be watered down in everyday driving over an entire tank.

Then again, the Beetle's rear end is notoriously bad in the aerodynamic department. Everything and anything will likely help it ;)


What transmission is in the Beetle - I'd guess an automatic ?
50+ averages should be in there for the manual version, but the auto (even DSG) will only work really well on the motorway.


Didn't the Beetle come with an on-board computer showing the mileage ?

A ScanGauge will work on diesels, but it's not a miracle solution for them.
I've found they get calibrated to one's driving style rather than to your actual fuel consumption.
Keep the same driving style, and it'll become very accurate with calibration.
Change your driving style, and it will be off ... until it gets calibrated to your new habits.
But if the Beetle didn't come with any FE display, a SG (or the cheaper ultragauge) will help you getting better FE.

There's something in that overall shape, I think.
See the Toyota Will Vi:
http://upload.wikimedia.org/wikipedi...ta_Will_Vi.jpg
And this concept from Teijin:
http://gazoo.com/NEWS/Img/162362.jpg

These are 'closed' shapes so I wonder what the aero difference would be with a rear panel on yours?

I'm a little confused. I thought the goal was to reach a high rate of efficiency?

I agree with you, there are a lot of variables at play here. However, I wholeheartedly achieved 55.4 miles per gallon with my kammback. No ifs, ands, or buts about it.

It, of course, does make sense that a better "testing" would be conducted on longer trips or "tankfuls" at a time.

I did run into a problem concerning the theory of testing:

Without using a scanguage, theoretically, it is a must to drive the EXACT distance, terrain, speed and using the same accesories (a/c, vent, radio, etc.).

WITH a scanguage, I would assume that there would be a problem as well. While the scanguage is incredibly accurate from moment-to-moment, it doesn't account for terrain at any given moment (especially where the terrain is less obvious).

So I guess what I mean is...with a scanguage, how do you know the moment is right to take a readout?


Also, of course I used cruise control.

Yea, I agree. There is something to the overall shape. It is a semi-kammback. Dr. Wunibald Kamm, aerodynamicist created the "Kamm" design in the 1930s. You should google it.

Yes, I do think a closed shape would be better. I'm working on this!

Thanks!

You set a specific route. That route should eliminate as many variables as possible. This usually means finding a quiet country road that has a decent speed limit.

Seriously, it is a good read:
http://ecomodder.com/forum/showthrea...ery-11445.html

dieselbeetle, that Kammback looks much better than I imagined it would. I've been so hung up on developing an Ernie Rogers wing, I'd not considered the Kammback.

Nice work!

Even if there is a gradient in your route, you should do runs in both directions to average out the difference in the gradient.

Spot the differences between the picture above and the one below:

http://ecomodder.com/forum/attachmen...5&d=1253817995

Answer: The sides are angled inward at the same angle that the top is sloped down. The teardrop we are after is a three-dimensional shape, not just 2D.

When you work on the permanent version, try to get rid of the corners, like MetroMPG did:

http://ecomodder.com/forum/attachmen...1&d=1218654440

I couldn't agree more, this would improve flow greatly. However, I ran into big problems because of the convex shape of the beetle's rear window. I'm mathematically perplexed.

I WAS able to open my hatch. However, I have removed my kammback until my next road trip. I'll photograph it then for ya!

I am very happy with the amount of rear visibility. I'd estimate the kammback reduced it by only 10-15% if any at all. A picture from the inside is a great idea. I'll do this as well on my next road trip.

I'll be driving to Texas from Orlando the first week in June. I'll definitely be enclosing my results and many new pictures.

Forget the math. Just get some cardboard and a marker and start cutting. :) Once you make one you like out of cardboard, then you can transfer it to acrylic or coroplast or shower surround or whatever you like.

That's a good point. What I'd really like, though, is to have someone design a kammback and 3d print it!

CAD = Cardboard Aided Design

If you look at the pic from behind, you'll see the top corners are near to or even outside the car's outline - that's where you could improve things a bit further.

If you loosen the connection between the vertical and horizontal panels at their corners, you can angle in the side panels.

If the curve gives a problem, make a straight score a bit above the curvature of the panel, and angle it in along that straight line.
The curve would still fit the rear window, and you get to angle in the sides.

The top panel will then automagically drop down towards the rear.
Don't make the top slope down too much - say 10-12 degrees - see if it's symmetrical left and right, fix temporarily, mark where the sides touch the top, and cut the top panels to shape.

It's what you registered of course.
The only question is what the experimental error on that measurement is ;)

The smaller the values you register (fuel quantity and miles driven) the higher the error will be.
The fuel pump cut-off will introduce the biggest - unknown - variable.

The instant readings don't get used.
The recording is started (i.e. SG trip is reset) at a fixed point along the route.
The reading taken is the recorded average, again read out at a fixed point along the route (say an exit sign).


Even then, as with any road-side testing, there'll always be errors due to conditions beyond the control of the tester or his equipment.

Any more recorded mileage done with this setup?

nope, sold it.

I really only did the kammback for a school project.

feel free to read the paper on my blog:

blog

I will help you get a 3d model ready for print if you can find someone to print it.

Bingo !
I'm glad you mentioned this .
Very important to not only round off the edges, but taper the tail in the plan view and not just the side - just as you mentioned here .

i LOVE that firefly and it's respective mods.

I don't see anything related to the Beetle testing.

Why didn't you do any extended testing for a more thorough proveout?

welllllll

it was a humanities course...

"ancient engineers" was the course. the project was to make something. (emphasis on the period.)

i ran into some time constraints. really i would've liked to have done more in terms of testing, but i just didn't have the time (at the time). I was taking 5 classes and now i'm a stay-at-home-dad! living in korea and owning 1 vehicle that's an suv makes things harder now, too.

for me, the project WAS about improvement, of course. the paper in itself was more about the history of such efficiencies and engineering feats. but, ultimately, it was about obtaining a big number, which i did! that beetle got pretty good mileage to begin with and i tracked it pretty avidly on another website, which is how i knew where to begin.

i'm aware that my test wasn't as scientific as perhaps it should've been, which i guess i kind of regret, but hey, we live and we learn. and also to the scientific bit, i'm very well educated as to how scientific concepts work. I think at the time i did the experiment i was more excited than anything and very green in my understanding of aerodynamics and "ecomodding", if you will.

thanks, man. unfortunately, though...I don't own the beetle anymore. honestly, i can't say i'm all that sad about it. that car was very expensive to maintain and the interior was just completely falling apart. a great example of this was I went to visit my parents across country and my dad got in the car, pulled the door handle and it came off in his hand!

whoever the knucklehead was that owned the car before me really screwed up the interior (of course i probably didn't help matters much)

...one thing that really improves efficiency on a beetle is a good dash cover.

I don't know about efficiency, but it sure would help with glare on that monster front porch of a dash. Would love to cover it with solar panels and do something useful, but can't think what it'd be.

Well, in terms of efficiency, I was referring to the A/C's cooling ability. That dash gets absolutely hot as hell and does little for the the coolness of the cabin. I know most don't even use the A/C, but in terms of keeping the cabin cool, it's still relevant, because it's likely that the black dash will send cabin temperatures well above that of ambient temps.

my solution was to get air duct insulation, which is reflective on one side, not too thick and sticky, cover the smooth portions of the dash with this. THEN, i covered up the unsightly-ness of this with a custom new beetle dash cover from Dashmat. yes, there was still a massive glare, but you get used to it, just like the original glare issue. this really improved the feel of the cabin, plus it looked a lot better and made it a lot quieter, too!

Who the hell would buy an untested Kammback? :D

i'm not sure what you mean in regards to "buy[ing]" or purchasing. perhaps i made a fleeting comment about producing the "device" and how neat it would be (because i don't have the car anymore, i'm completely unwilling to re-read what has already been said. mostly for the reason that my personal time (with the exception of me typing this) is almost always allotted for something i want to do, rather than what someone else thinks i should be doing). or maybe you're mistaking my thread for someone else's.

either way, it's interesting that you mention consumerism. because ultimately, we buy things that have been subject to insufficient testing measures each and every day in other aspects of our lives. because this is a community of like-minded appreciators of science, we hold each other to higher standards of methodology.

chiefly, i have admitted to the erroneous ways of my testing in my kammback project. however, i refuse to apologize for actually getting off my ass and doing something i think everyone here deems worth while. i will not be the subject of scrutiny or persecution continually, as it is not warranted. i also refuse to take down this thread, because i made a few mistakes here and there. i learned something from my experience and i'm under the impression that others have as well...if not, so be it. i've been reassured that others have at very least enjoyed my project in much the same way that i have. i have made the observation that the few scrutinous readers of my project have largely been those who've not attempted projects of their own, ambitious or otherwise.

Calm down, I was just laughing at that quick answer. I could just see a guy walking down the street carrying a Kammback that he had bought at a yard sale.

Did you actually sell the Kammback, or did you sell the Beetle?

the car, not the kammback. the kammback wasn't really cohesive enough to sell.

Would love to see (unless i missed, link if so) plans or dimensions or somesuch so I can try to replicate it, if my wife will allow the monstrosity. I think it's time for the VGs to come off, since apparently keeping the air attached down to my license plate is a bad thing.....which they do according to the trails left by the combination of dust and dew/rain