Wind Drag & Turbulance

[brian_bp]

Many years ago, deflectors were popular on the cabs of big rigs - highway tractor-trailers. Now, they either have sleepers with high roof aerodynamic designs, or the cab is plain but the trailer has a "bubble" front... few deflectors left to be seen. Hmmm...

Fifth-wheel RV rigs rarely use deflectors, as well, although some older ones still have them. That may be something to do with the configuration when not towing, but I just think it makes more sense to shape the trailer appropriately.

My current trailer is not tiny, and does a good parachute imitation. I have no idea what I will do about that (other than switching to a smaller trailer), but I did notice carrying a big roof box or a canoe (only one of the two fits at a time) might not help drag, but doesn't hurt - maybe surprising, given that it's a huge canoe, longer than the van and quite wide. Perhaps it doesn't matter much whether the roof load or trailer does the air-bashing?

[gyroguy]

I design and build experimental aircraft as a hobby. "Experimental" means I fly IN them.

There are different kinds of drag.

INDUCED DRAG: With airplanes, induced drag generally decreases as speed increases. The low-speed, high-drag situation results from having the wings at a high angle of attack to get enough lift to fly at lower airspeeds. If you viewed the airplane from the front at lower airspeeds, you'd see the tail hanging down and a lot of airplane meeting the wind. As the airplane goes faster, the angle of attack of the wings needed to support the weight of the airplane becomes less. Seen from the front, the airplane would look as if you were seeing it sraight-on. This is the lowest induced drag an airplane can get. Even if the airplane goes faster, you can't lessen the induced drag any further.

FORM DRAG: The more streamlined the front of the airplane is, the less form drag it has. More things sticking out on the skin of the airplane add to the form drag -- fixed landing gear, antennas, lights, door handles, etc. Form drag increases as speed increases. This means that any airplane has a top speed in level flight at which total drag equals the total thrust the airplane can produce. To make an airplane go faster, it is easier to refine its streamlining than to add more power.

INTERFERENCE DRAG: This is a component of form drag, basically a kind of drag that results from making vortices, or horizontal whirlpools of air. On a typical wing, for example, the reduced air pressure on top of the wing causes an airflow that migrates outward to the wingtips. The higher air pressure under the wing does the same. When they join, the pressure differences cause the joined air to rotate and separate from the airplane at the wingtip. The result is a drag-causing vortice. (The wingtip vortice at the left wingtip spins in the opposite direction of the one from the right wingtip.) A way to reduce interference drag is to extend the wingtip up or down. This takes some fancy calculating.

FRICTION DRAG or SKIN DRAG: The passage of air over the surface of the aircraft also generates drag. A smooth metal surface generates less skin drag than a metal surface with rivets. The faster sheet-metal airplanes have special rivets that are flush with the surface. With a good coat of paint, they effectively disappear. The fastest airplanes have laminar-flow wings, in which the air follows the shape of the wing exactly because the skin is so smooth. Laminar flow works perfectly as long as the skin is perfectly smooth. A dead bug can mess up laminar flow.

LITTLE DRAG TRICK: Contrary to what you'd think, it is possible to reduce the drag on the rear of the airplane sinply by cutting it off -- instead of reducing its cross-sectional area gradually to a point. If you'd like an example of this, look at the tail fins of the X-15. Again, it takes some calculating.

I'll work up another post to show how this could apply to teardrop trailers. Just remember, I build airplanes as a hobby, and I don't have all the answers. I don't have wind tunnel, either.

[Arne]

I look forward to what you have to say..... as I'm in the midst of trying to build a tear with less drag (all kinds of drag)..

[Kankujoe]

Interesting post Gryoguy!!

I'm looking forward to more information. Please don't limit it to teardrops... some of us have squared off cargo trailers.

[gyroguy]

This will have to be a short post as I'm short on time. You have seen the diagrams posted earlier, showing airflow over two connected objects. Drag considerations for trailers must consider not just the trailer, but the tow vehicle's shape.

TOW VEHICLE SHAPE: Maybe it is obvious from those drawings -- the shape of the tow vehicle is a major consideration. The idea of putting a deflector on top of a car is to shunt the air upwards so it hits the trailer with less force. What is not shown is the turbulence the tow vehicle causes by its forward movement.

TURBULENCE NEGATES STREAMLINING: The turbulence caused by the tow vehicle causes -- as far as I can tell -- so much disturbance in the relative wind that the streamlining effects of a teardrop trailer are largely ineffective. The way to make the streamlining of a teardrop trailer effective is impractical. You'd have to push it in front of the tow vehicle.

ONE AERODYNAMIC UNIT: When I designed the Crocodile Tear, I considered my tow vehicle. It is a small Japanese pickup truck with a camper shell. My design idea was to make the trailer just as high as the camper shell. The second idea was to put the front of the trailer as close as possible to the rear of the camper shell. I was able to round out the top front of my Crocodile Tear to help airflow flow over it. Ideally, the truck and trailer should operate as an aerodynamic unit instead of two separate pieces.

Take a look at <www.crocodiletear.com> to get construction and shape ideas for my teardrop trailer. The information above was not included in the website discussion because I wasn't aware anyone cared much about trailer aerodynamics.

[Trackstriper]

OK. Count me in on this discussion. I have a full-sized Dodge work van that I use for business. It would be the only vehicle to tow a TTT that I'm designing. I do want to keep the drag low and am not married to any of the traditional teardrop shapes. I want to be able to have a trailer that is a near "standy" (full standy for the wife unit) without it becoming a pain to pull for many miles.

I do see the close coupling between the Crocodile Tear and the tow vehicle. The front end is tucked in close, but I don't think that the sloped back is helping any with the aero. Please look at my album to see where I have been going with my current design (10' version - it may go to 11' or 12') which has an tear drop curve on the front and a blunt rear. I want to have a rear access and with the dropped floor I can generate the required 5'-6" headroom. This does preclude dropping the roof line somewhat 'cause I want the headroom in the aft end. A 12' version can allow a slight drop in the roof line of about 4" at the end and is more pleasing to the eye, but is also more complex to build.

I've been thinking over the past few days that a close-coupled, bull-nosed trailer might also get the job done. Just a partially rounded front like in the NASA aerodynamic truck design that Andrew throws up once in a while. Totally straight sides and flush rear to keep it simple. Sort of like Andrew's "Dinette".

What do you guys think?

[Sam I am]

I've read that drag is proportional to the square of speed. Doubling your speed multiplies drag by four. Tripling speed multiplies it by nine, etc. If I did the math right, going from 50 mph to 70 mph will increase drag (and the horsepower needed to overcome it) by a factor of 1.96 - almost doubling it!
So it looks like if you don't have wind deflectors and the like, the easiest way to save fuel is to simply drive slower when towing. We're all on teardrop time anyway!

[angib]

I've been thinking over the past few days that a close-coupled, bull-nosed trailer might also get the job done.

Then I'll re-post another picture from long ago. I reckon it might be possible, at least theoretically(!), to design a trailer with either zero extra drag or possibly even negative extra drag. Here's my suggestion:



Extremely close-coupled (so close that it has to have a tongue that extends automatically for going around corners), exactly the same cross-section as the tow vehicle but a couple of inches higher and wider (to get the airflow to re-attach to the trailer), rounded on all the front edges (sides as well as top/bottom) and then gradually tapering on all four sides.

Andrew

[angib]

I've read that drag is proportional to the square of speed.

Yeah, and if you want to scare yourself, power needed is proportional to the cube of speed - it takes nearly three times as much power to do 70mph as it does to do 50mph, burning three times as much fuel (all other things being equal).

The upside is that at 70mph, you get there sooner and so stop burning fuel at such a high rate sooner - which gets you back to the square law you quoted, for both drag and fuel consumption.

Andrew

[gyroguy]

So it looks like if you don't have wind deflectors and the like, the easiest way to save fuel is to simply drive slower when towing. We're all on teardrop time anyway! :)

The easiest way to save fuel is to stay home... but that's not why we build teardrop trailers, is it?

Correct, the next easiest way to save fuel is to drive slower. It's also the easiest way to get rammed from behind by people who are on Kamakazi Time instead of teardrop time.

What I'll write next is all theoretical, since I don't know how to do the math for drag on a trailer. FYI, with airplanes, you treat the airplane as if it were surrounded by air on all sides with no interference. A trailer has built-in interference from the space between the bottom of the tow vehicle and the ground, and from between the trailer and the ground. That's all turbulent air going every which-way and bouncing off the ground and the truck and trailer bottoms. I haven't got a 3-D wind tunnel to show what that air does under ideal conditions. A road surface is far from ideal... so what I'll write is all theoretical.

We don't have a way to streamline the bottom of the tow vehicle or trailer. Both are creating turbulent air by passing quickly over a surface covered with still air. The various projections -- wheels, axles, oil pans, frame -- and the various indentations -- spaces around engines, frame, driveshaft -- all play their part. The result is turbulent, draggy airflow. Since we can't get away from it and can't guesstimate its overall effect, we'll ignore it and work on what we can change.

The next easiest way to save fuel is NOT to have the top of your trailer stick up higher than the rear top of your tow vehicle. Airflow from your tow vehicle needs to FLOW over the top of your trailer, not meet it like a boulder in the middle of a river.

The next easiest way to save fuel is to have the width of your trailer less than the width of your tow vehicle. Seen from the top, airflow from your tow vehicle will curve inward towards the centerline of your trailer. Your trailer needs to intersect this inward airlow and route it along the sides of the trailer. Think in 3-D!

[gyroguy]

Deleted 2nd copy of post.

[gyroguy]

Andrew wrote, "I reckon it might be possible, at least theoretically(!), to design a trailer with either zero extra drag or possibly even negative extra drag."

It's been done. Look at the articulated buses used for public transportation, or the connections between passenger railcars.

Okay, I think I have a way to attach a scale drawing of the Crocodile Tear and the Japanese truck tow vehicle. Here goes:



Ideal airflow: Theoretically, at one specific airspeed, with calm local winds, and on level ground -- the airflow over the tow vehicle and the Crocodile Tear MIGHT look like this. The airflow off the rear top of the camper shell doesn't disconnect. It drops due to lower pressure between the vehicles, then flows up and over the rounded front top of the trailer. I've shown it generally following the rounded rear top of the trailer and down the hatch. What probably would happen on the hatch is airflow separation into turbulent air... but I want to show what would ideally happen.

Turbulent airflow: Notice that airflow beneath the vehicles is completely turbulent air. At the rear of the trailer there is a "V" for a vortice formed when air rotating in different directions joins. The vortice would be the width of the trailer and would add interference drag.

Practical considerations: I couldn't change the angle of the rear window in the camper shell. Wish I could. The distance between the rear of the truck and front of the trailer is the minimum possible. Consider that you have to have the A-frame on the trailer almost parallel to the truck bumper, in sharp turns, without mashing the bumper into the trailer body. Under my trailer are two under-floor boxes. When the trailer hitch touches the ground, their fronts don't. The boxes cause additional drag under the trailer, and I decided to ignore it.

Airflow across top: Airflow bends around an object. At the end of a blunt oblect, the airflow tends to flow inward. With a gap, the air will flow inward, as shown in the diagram. To assure smooth airflow, I should have made the top of the trailer LOWER that the rear top of the camper shell. Axle and spring placement made that impossible. The next best thing was to make the top of the trailer as tall as the top of the truck.

Airflow along sides: If I had no gap between the trailer and truck, I could make both with the same cross-section. But there's a gap. The Crocodile Tear is only 4 feet wide, and the truck is almost 6. That leaves plenty of room for the airflow on the sides to follow the sides of the trailer. And that's considering that the edges of the trailer are square, not rounded as they are on top.

Remember when you look at this diagram, it's just theory. It shows a one-dimensional flow. It shows what I was trying to do, but there's no telling if I achieved it. Since I've said about all I can say, I'll leave the rest of the discussion to you.

[angib]

I'd agree with lots of that, but a couple of comments:

Notice that airflow beneath the vehicles is completely turbulent air.

No doubt true for a truck, but that hasn't really been designed. If you look at modern aerodynamic cars, I think you'd find this wasn't accurate - they have lots of fairings under the body to get a reasonably smooth underside and the way the flow is constrained between the underside and the ground means it can stay attached.

Airflow bends around an object. At the end of a blunt object, the airflow tends to flow inward.

Very definitely not true all of the time. Airflow does not bend around sharply-curved objects, only around gently-curved objects. Even wings stall and they are rather nicely shaped in general.

The flow in your diagram is quite unrealistic - I'll bet my savings that the airflow separates at top of the slope and the slope may as well not be there.

Andrew

[Trackstriper]

Andrew,

Regarding your post of 29 June...

Can you help me develop your thinking within what's practical? For my particular case, I have a standard full-sized American work van, about 80" high at the rear. I understand that it would be best to tuck a trailer in close to the rear of the bodywork, but I think I'd have to rule out an extensible tongue for purposes of practicality...although it could be done. The closest that I think is workable is to have the hitch ball 24" from the front of the trailer, which is about 36" from the actual bodywork. That's pretty tight but will allow for reasonable articulation. Is this too much distance, or are the benefits simply reduced somewhat?

Is a 12"-15" radius on the nose, similar to your Dinette (including bottom edge), sufficient? The Dinette nose can be built without ball-type corners. Many cargo trailers have roof bows that blend the transition of the roof into the sidewall. There is probably an aero advantage to doing this as some airflow will cascade either from the roof to the sides, or visa versa. But when tapering the profile and the planform, this ain't gonna work without a lot of compound curves....I'd rather give up efficiency. What is your instinct here? Does the taper outweigh the rolled upper edges? My guess is yes.

Does the trailer need to be several inches wider than the tow vehicle to initiate flow reattachment (which requires a compound curve on the sides to match the tow vehicle...yuk) or can we assume that all will be at least OK, but not ideal, if the trailer falls into the general shadow/wake of the tow vehicle?

I know that's a lot of questions, but I think your instincts are generally right on and I am curious.

In hopes of having a trailer that tows easily....

J.B.

[angib]

I think the lowest drag will come from a trailer that's entirely inside the wake of the tow vehicle, preferably inside it by at least 6" all round. Reattaching the flow is a very tall order and nearly impossible with a non-rectangular tow vehicle.

In drag, small is always good!

I would guess (and we're drifting into guess territory here) that if one of the surfaces (top, side, etc) was inset a long way in from the tow vehicle, then it's less important to radius that front edge - so if you had a wide, low teardrop, it would probably be better to radius the front of the sides than the top.

Remember that you don't have to radius the body itself - there's nothing stopping you adding a radius on in front of the body - that's what several trailer body fairings do over here. Some light 12" diameter plastic pipe/duct cut in half would make a good radius fairing and would be easy to mitre the top corners.

Andrew