Aerodynamic Mirrors (fluid dynamics help?)
 

Okay, so after tuft testing the side of my metro, I confirmed with hard evidence something anyone could have told me, which is that the mirrors create a big ugly wake next to my window and sends my yarn spinning at least a foot back.

I want to install video cameras or inboard mirrors, but due to various motorcycle related circumstances I can NOT hazard being pulled over! Like, jail, dig it? But let's not make that the topic of this thread...

Basically I need something reflective that sticks out of the side of my car, and preferably does double duty as a means of seeing behind me. I was just re-reading Gravity's Rainbow, and got inspiration from the V-2 rocket. This is my idea-

https://www.fuelly.com/attachments/fo...b3e8b56695.jpg

The gray semi-circle at the right end represents a 2" convex mirror. The gray triangle in the background is the mount on the car.

Is there any rule of thumb, independent of scale or viscosity, for the optimal proportion between the width, length, degree of taper, widest point, etc? Do I have a way of knowing whether my truncated 'Kamm' section will be narrow enough, or set far enough back, without testing? I tried looking through Horton's book, but briefly, and I didn't find any reference.

Also- I instinctively made the tip pointed, from whatever naively ingrained notion of what an aerodynamic shape is, then realized that I never see a solar car/HPV/modern projectile with a pointed tip. What's the lowdown? If you were to make a tip that was entirely spherical, that would have the same frontal area, no? Just at a more severe angle of attack...

So many questions!

A pointed tip catches the "wind" as it moves forward. Also, if there's any crosswind at all you end up with the airflow coming in at an angle instead of head on.

A blunt tip (I think a parabolic curve is usually used, or something close to parabolic) lets the airflow find it's own path of least resistance. Also used on torpedoes and many boats.

However at the trailing edge, a taper to a thin tip or edge is a good thing. It encourages the airflow to rejoin itself smoothly. An airplane wing is a good example of both the rounded blunt leading edge and a tapered trailing edge.

Thank you! That solidifies several unformed biases floating around in my head against the pointy tip.

My only design limitation is that the rear must be a 2" convex mirror. I've considered shrouding it with a transparent plastic cone, but I don't want to optically handicap what's already a tiny rear-view. If I can 'virtually' shroud it with Kamm's cut-off, that would be ideal...

before you get all exited about designing a super smooth shaped mirror (wich is fine of course), take into account that the drag caused by any object is the combination of drag quoefficient and frontal area!... and since you have not only the mirror but also the mount to take into account, perhaps motorcycle style mirrors are somethign to considder... most passenger cars use mirrors on big mounts to allow for mechanisms to adjust them from the inside, but if you can live without that you could mount the same mirror surface on a much smaller and thinner support that will allow the air to stay attached to the side of the car. some sportscars also use thin wide mirrors... so maybe you could reduce the height of the mirror to save frontal area as well without loosing to much usefull information from behind you.

"If I can 'virtually' shroud it with Kamm's cut-off, that would be ideal..."

I really should edit that wikipedia article that seems to be responsible for perpetuating the bs that a Kamm back can create some sort of virtual and equivalent teardrop shape. It can't. It's an engineering trade-off. You cut it sharply at the point of diminishing returns. Most of the drag of a truncated teardrop will be in proportion to the cross sectional area of the truncation.

At some point, the decreasing area becomes not worth worrying about in relation to practicality concerns, or cost, or that there are enough other big drag causes (e.g. wheels) that aren't addressed, that it is myopic just to focus on extending the teardrop.

Even the jumbo jet has a "Kamm back", i.e. a point at which the extra reduction in drag coefficient are not worth the extra weight or decreased thrust by having a longer exhaust for the auxiliary engine.

But there is nothing magical about it that enables it to cure the low pressure region at the back of the Kamm back.

Another way of looking at it if the taper is in two dimensions, with the beginning area of the taper starting at W0, length from that beginning as L, with taper angle "theta", the cross sectional area = (W0)^2 -2(W0)(L)Sin(theta) + (L^2)(Sin(theta))^2

Most of the gains occur close to the beginning, with not much to be gained by extending the taper to a point. Tapering in at the sides will have negative consequences in crosswinds more of the time than a car without such taper, since the crosswind will only need to be slight in order for the boundary layer on the surface on the lee side of the crosswind to become detached, and hence, low pressure with a component of the resulting force towards the rear of the vehicle, pulling you back.

Note that if the taper is only in one direction, the reduction in area will stay linear, meaning you might as well taper to an edge if you can.

You could make the mirror itself a teardrop shape with a pointed tail, and mount it directly to the car so that there is no need for a mount.

The V2 rocket, including many other rockets, missles, jet-powered aircraft use the pointed tip to help traverse the speed of sound. High powered rocketry used to be a hobby of mine and we followed the same guide lines. High impulse rockets intended to break the sound barrier always used pointed nose cones while low impulse/long thrust duration high altitude models used parabolic nose cones. I think the point in the nose allows for more efficient air compression in the trans-sonic phase.

I asked for this, by not being more explicit with my diagram.

https://www.fuelly.com/attachments/fo...3d8cb689da.jpg

This is the front of the mirror, it is it's own body. I'm sure the fairing between the mirror/foil/car won't be optimal to reduce clashing boundry layer drag, but if the improvements are even measurable I'll be esctatic. EDIT: I'll probably be ecstatic, too

I had a sassy retort for this one, until I realized that the rest of the post answered my questions exactly. I chose poor phrasing too, I'll admit immediately, and will watch my step from here on. I was trying to imply that I thought the taper (not the mystical void beyond, granted) on the body would aid a reduction in drag given the logistical compromise necessary to fit a flat mirror on the end, a hundred lame apologies.

I did a search, but only one thread came up critiquing my proposed mirror shape. It was this one, so I regret nothing:confused:

I'm not sure why it is instinctive that we tend to shy away from smooth, blunt shapes that approach the point of largest cross sectional area parallel to the expected direction of travel. These have the least drag associated with them in the subsonic regime.

I suspect that it's either looking at fighter aircraft and rockets designed for transonic/supersonic regimes, or that we are used to sharp points/edges with things like darts, axes, knives etc for piercing solids where the solid has fairly high forces of cohesion. Almost all of the work done in cutting something like a rope is done in severing whatever is holding a tiny area of the material together, pushing the remainder of it out of the way once cut is minimal by comparison. So for that you need a point or an edge to induce that force in a small area.

We've had edges and points for so long that our understanding of them would have to be ingrained in our genetic code by now - those who had an instinctive understanding of edges/points would have had an edge since hunter gatherer days.

Boats we have had a long time, although the difference between a blunt end and a sharp end is mainly a refinement. A dull edge will not cut, but a boat with a pointy hull will still sail.

This document on rocket nosecones is interesting.

For our application, it seems as though short, blunt, elliptical type fronts would be best for the front end of a mirror. For areas close to the ground (e.g. front end of a vehicle) you want the air to be flowing up and over the car, so at the very least it should be perpendicular to the front of the car and perhaps/probably raked back somewhat.

The alternative is to raise the body of the vehicle above the ground somewhat to reduce ground effect, but there are other issues associated with that (more torque from crosswinds, marginally more side area, need wider wheelbase for stability in cornering). The best road vehicle drag coefficient in the world is the Pac-car, with a Cd of 0.075, and utilizes this concept.

Cool, I'm confident in my revised design now... I'll post a drawing.

The mirror will be a one-off fiberglass-over-foam affair, with all the sanding and body filler that goes with that. In theory I could make a mold and pump out a few more for fellow metro-types that have inspired me. Skewbe? MetroMPG?

No problems! I don't blame you, I blame journalists and marketers either out of their depth or dishonest. e.g. What are you going to buy, a snazzy looking, practical car, or a teardrop shaped auto with better fuel economy and top end speed? Can't decide? How about if I told you that you could get all the benefits of the teardrop shape with all the practicality/style of a Kamm back?

Because your controllables are Cd * A (as someone else said), if you stick the mirror at the back of a Kamm back type shape, you are making the Cd better but making the A bigger, as well as increasing skin friction. Ultimately, this provides negative benefit.

You have to fit the mirror at the point of widest frontal area, which is the mirror. This means the only thing you can do is fit some sort of transparent taper (either full or partial) after the mirror. There is one company that sells an exorbitantly expensive mirror that does exactly that. I can't find it at the moment, they sell it for racing not the FE set.

Hey a couple of points . . .
1 Do the mirrors have to be sticking out that far since you want to see what is behind you but more to the side you could put the mirror right along the edge of the door frame.
2 If you use the wake of the windshield along the side of the door and tuck the mirror in close to it you may actually improve the air flow coming off the windshield.
3 The surface area creates drag and a sphere has the least surface per volume and the best aero dynamic shape for the leading surface.

would that be a good idea aerodynamically? you'd spoil the attached flow to the vehicle... a lot of mirrors on new cars i see are not directly attached to the vehicle but have some space between the body of the car and the mirror.

i might be wrong but it's my impression that as long as te attachment brace of a mirror is thin enough (it could be shaped like a wing) free standing mirrors add less drag

like this

https://www.fuelly.com/attachments/fo...fe715d3333.png

maybe it's just for noise reduction (most mirror developement seems to be driven by this), but a lot of mirrors i see are freestanding... some even with the support running down to the door rather than sideways to the window

I just thought I'd put up a link in case you wanted to see what I ended up doing for an aero alternative to stock mirrors :) . (Edit: you can skip all the way down to the last posts of the first page to view some images.)https://www.gassavers.org/showthread.php?t=4504
I thought about making something but was too lazy so I bought the motorcycle mirror.

https://www.fuelly.com/attachments/fo...9f54882f4a.jpg

okay, back to the faired bubble after all. If it doesn't have a hard edge, or taper to a point, it shouldn't impede my vision except in bright sunlight. It's a compromise for useability, but should definitely be an improvement on what's mounted currently...

Okay, getting speculative, but since my mirror has a rounded tail for clarity, as opposd to a clean break, do you think it may benefit from some boundry layer acceleration device, i.e. trip wire et al? I know it's worked wonders for spheres, could the same propertys (EDIT: Prauperteeies) apply to an oblong tapered ellipse?

Tangentially, has anyone heard of the golf balls with faceted dimples? Apparently they travel much farther than the smoothly dimpled ones... I just googled it and Callaway had a patent published in July of this year...

took me a while to find it, but here is co zx2's entry:
https://www.gassavers.org/showthread....1962#post61962

https://aycu33.webshots.com/image/192...1369167_rs.jpg

That's a good point, you might have something there.

I think it may be possible to do better than that. Consider the front end of some torpedos:
https://img140.imageshack.us/img140/9...pedofw8.th.jpg

There is less surface area than the sphere (less wetted area), it's still smooth, so may actually flow better.

The reason why you don't see this shape in things like cars and velomobiles is that they have to house things inline like motors, people, luggage etc, and if you have to encase such things, you may as well minimize the wetted area.

But if all we are concerned about housing is a small plane such as a mirror, I suspect that a more blunted end (but still smooth and with a plane parallel to the direction of the car's motion at the mirror's edge) would be (slightly) better.

In a word, no.

I like your mirror design, it's very similar to what I thought of when I considered how to do my mirrors (but didn't because there is a lot of lower hanging fruit than the mirror). And it's very similar to those racing mirrors I mentioned.

https://img404.imageshack.us/img404/3265/img11dd3.th.gif

As to boundary layer separation, the reason why those trip wires on a sphere help is that they enable the boundary layer to cling to a more sharply raked trailing edge. This means the area of low pressure at the back of the sphere is smaller. Unless your trailing edge is beyond about the 15-17 degree region, you don't need to trip the boundary layer.

Thanks for digging that up! Interesting that someone here tried it, I'm sure that most smart people concerned with vehicle aerodynamics have considered the obvious transparent boattail on the mirror idea, and then probably dismissed it after thinking about visibility concerns (dust, warping, etc).

I wonder how the visibility is?

like this i was thinking
Attachment 1133

ok didn't read your post thoroughly... that makes sence....

the car that has some of the nicest mirrors i know is the opel vectra B

https://www.fuelly.com/attachments/fo...970484d673.jpg
it's got a drag quoefficient of 0.29... not bad... although the previous moddel had more converntional mirrors that where jioned to the side of the car, and somewhat continued the angle of the windshield... same drag figures for that car though...

i don't know how much for looks this setup is, but given the overall attention to aero i don't think they'll hurt

still better looking that a lot of other designs imho

That's fascinating, it could simplify matters to call into service the bottom of a cylindrical container...

I tried a tiny 1.5" convex mirror inboard, and it did the job satisfactorily. I found it even better when mounted further up the A pillar toward my head, so my strategy is changing-- I think I'll mount an inboard mirror for visibility's sake, then have a doppelganger outside to deflect unwanted attention.

Idea 1- cut the profile of the stock mirror out of it's housing entirely, leaving a thin foil (no pun intended) shaped deceptively like a standard mirror but minimizing drag. ;) Rely on the optical illusion of seeing the road ahead through the mirror, or a clever paint scheme inside the housing, to make it look authentic. Chance of success is thin...

Idea 2- Make as aerodynamic a shape possible to protrude from the side of the car, and cover the tapered section with mirrored vinyl to give the impression of an unorthodox but undeniably present mirror.

I think the latter sounds more likely, even if... unlikely.

This is a good point. My time would be better spent elsewhere on the car. Right now I'm taken with the action of building the mirror, but it may fall down the list as other projects start. I have driven with no mirror for a stretch, and it's slightly quieter though, which is sweet.

I was drawing some side skirts in photoshop.. when trying to close off the backside of the front wheel wells I was dissatisifed. To account for suspension movement the gap is just too big, especially with the way I drive on occasion.

My idea, then, is to make an airdam to close off the backside of the wheel well, but attached to the hub. I think I need to draw this one... no, I'll start a new thread specific to that modification, in the name of organization

That's a really good idea. Make a COZX2 style boattail for the outside technically within minimum government specifications, but for your actual driving install permanent inside mirrors, for example something like the "wink" mirrors. At least until your government starts allowing more fuel efficient options.

Judging on screen size, I'd imagine an internal LCD reversing camera wouldn't use more than say, 10-20 Watts, and would be the most elegant solution.

A request to the mods: could the above three posts be transferred to the front skirt thread? Thx.