In cabin wind/"redneck a/c" aerodynamics
 

Any clue how to improve this to increas
aerodynamics while ventilating the cabin?
I am curious if the rear roof angle is ideal and if there is an ideal hole placement/size in the back to empty air into the vacuum behind the car from the cabin.

Edit this is **not** for an existing model car, but for a prototype I have built from scratch (see pic, which I have repost outside of Facebook to ensure it is visible)

http://ecomodder.com/forum/member-mi...ototypetow.jpg

https://1drv.ms/i/s!ApZC8QFk-svmgXIZdrsgSxXP-kV8

Not able to see the link. :(

Not touching a Facespook link. At least tell us the make and model.

Meanwhile, fabricate or simulate a wind wing.

Does not compute: http://ecomodder.com/forum/attachmen...1&d=1507143992

[crickets]

Updated/fixed image
 

Sorry about that, just updated the image link. Will upload a wind tunnel analysis sketch soon showing where internal air ducting is now (wheel well tops and upper back of car).

I just have both the windows down 2 inches on the passenger side of the car and amazed at how much the air circulates without the use of a fan or A/C. The front and rear windows must be down and equal amount though or it doesn't work.

Thanks.

If the sketch includes the window area and door cut-lines that will disallow things that couldn't work.

The wake is stagnant, low-pressure air. Look at the engine cooling on the Edison2. It is a vertical slot at the maximum camber (widest point of the body). The air is moving fastest there and you can intake or exhaust most efficiently there. For cabin ventilation the window height is prolly adequate.

http://edison2.com/storage/VLC-NCCAR...=1343934993089
Edison2 - Very Light Blog - Edison2, Local Motors & Siemens: The VLC Design Challenge

My Superbeetle's flow through ventilation exhausts in the C-pillar. Opening to the wake could allow exhaust gases (if not electric) to intrude into the cabin.

So would exhausting out the middle back of the car, literally dead center, be ideal? Or is there a reason I'd want to exhaust via the c-pillar area further forward, even if I have a tapered rear end behind that?

Still trying to wrap my head around it...my gut reaction is it would be easier
to push exhaust air into the low-pressure area behind the car than the fast moving area beside it. Or, on my car body/pic, could you circle potential areas you'd recommend to exhaust cabin air through? I'd really appreciate it.

I'm no expert, despite the tag somebody put under my user name. I think that's based on number of posts. See if you can lure aerohead into the thread.

Basically, fluids like air have a pressure and a velocity at any given point. You can no more push air than you can push a rope.

https://en.wikipedia.org/wiki/Bernoulli%27s_principle

The most you can do is pool it in a plenum and suck it out through a port.
IMHO, the stagnant low-pressure air in the wake will be no more effective than the high-velocity air moving over the sides of the body. think of a reverse Coanda nozzle:

https://en.wikipedia.org/wiki/Coand%C4%83_effect

The picture is poorly lighted. How close is the sketch? Three-views would be nice. Is it gas/diesel or electric?

Sailplane cockpit exhaust system from South Africa is at top center behind the canopy.

When there is any mention to "redneck A/C", the first thing that comes across my mind are those swamp-coolers widely used by truckers in my country. Anyway, maybe it would worth to perform some experiences with a Venturi tube and a small fan as some sort of "redneck A/C".

Freebeard and anyone still here...

This is roughly what I'm trying to do (link)

https://1drv.ms/u/s!ApZC8QFk-svmgXNKbTVyLAgQpdaF

According to Flow Illustrator it
can work...what comes out of the wheel well follows the interior roof bottom line to the back and is released into the vacuum in the rear of the car. My question is if the inlet/outlet points for air in the interior or roof angle can be improved and how.

Your CFD modeling of the interior is probably inadequate. CFD in general requires supercomputer levels of compute power. The modeling is not representative and the resolution is too coarse. Sorry.

You can click on your username, go to your profile and create an album to put you pictures in. then you can place them directly in the thread. You'll get more eyeballs on the problem.

You might CFD model the intake or exhaust port. Or you could observe best practices in the field. NACA ducts for intake, but they don't help on exhaust. Etc.

Were it I; the front wheelwells would follow this:

http://ecomodder.com/forum/member-fr...gt-one-tm1.jpg

You already have a start with the open fender backs. Think about the relative volumes of the air swirling in the underbody, the amount needed for cabin ventilation and the amount needed for intercoolers on the apparently electric drivetrain.

Have you settled on a motor or would you like to hear an opinion on that?

If the exhaust vent is on the top of the canopy where high velocity air can pull it out, that could be a good location. You could probably have cabin air exhaust into the engine compartment. It may or may not require additional inlets.

Hmm...well is there free modelling software around that can handle it?
I may not have a super computer, but do have an 8 core I7 I use for music and hobbiest scientific processing.


Here is another pic of the car/trike. Problem with my model vs. your suggestion I see is how to apply it to a car with open wheel wells in the back.

http://ecomodder.com/forum/member-mi...ototypetow.jpg


That's exactly it...I can't seem to find any best practices on cabin exhaust. It seems the top-center of the rear canopy may be a good exhaust point, but ditto for the bottom center right behind the plenum formed in the rear center where the rear wheel and (indeed electric) motor are.

My concern with exhausting it top center of the rear canopy is wouldn't that change/disrupt the flow of quick moving air at the top-rear of the car, causing the airflow from the bottom to circle back into a vacuum in the top rear area. Or, again, I'm wondering if that effect is not real, but just a problem with the Flow Illustrator wind tunnel simulator.

Would you recommend a top/center outlet over a bottom/center one? Or where on the photo, for example, would you recommend it?

Right now I'm using a Mars ME1003 motor, which delivers about 15-20 HP continuous without overheating at about 150-200 amps with 72v (using a high-end golf-cart motor controller, the Alltrax SR72550). People take telling me to get a bigger motor :snail:, but my ideal goal is to make it haul 4 people to 60 mph using just that amount of power e.g. efficiently :turtle:. It may not be an eVLC (I wish I had the skills to fabricate that sort of avian shape), but the fact an eVLC did the job with a similarly powered motor gives me some hope.

I'll probably repost this with global/album pics...soon as I get some background learning from this thread so I don't test people's patience too much with my novice-ness.

OpenVDB: An Open Source Data Structure and Toolkit for High-Resolution Volumes

When I went looking for the thread, I learned that no-one had commented on it. :(

In clean, moving air not buffeting turbulence. But seek other opinions.

http://ecomodder.com/forum/member-fr...nov-09-375.jpg

I have one of these. The Toyota engineering is indistinguishable from magic. There was some discussion of locking the differential for a motorcycle. 50hp/68NM.


Controller mods or build for E-assist altermotor

These are 20hp; which won't be sufficient uphill in a headwind.

Reynolds number is problematic
 

With a Rn of 5,000,this flow would be of a laminar boundary layer,not possible when the entire vehicle is submerged in the turbulent boundary layer of Earth.
And there's a risk that air would actually flow forwards,from the back of the car to the front,seeking fast-moving,low-pressure regions up front.

Since your other thread has zero response I'll respond here.

Putting side-by-side seating in the rear could potentially create bad vehicle dynamics. It also compromises plan taper. If you are committed to a four-seater, then — Oh, well. Else, a central rear seat, moved forward, with footroom under the elbows of those in front .

It's hard to comment without an indication of the [presumed] door cut-lines. The front fender cutouts seem low and wide, but it's hard to know without a better understanding of the cross-section through the front suspension.

You are trying to make a aerodynamic shape from flat sheets. They will fight you every step of the way. It may be better to make an eggcrate from foam strips and then carve the outer face of the eggcrate to shape.

At least make one continuous curve down the side without steps on the edges. That back-facing rectangle should be an oval.

behind the car
 

Some thoughts:
*streamlining is to reduce pressure drag.
*Pressure drag is a function of separated flow.
*To reduce the pressure drag you need to attack flow separation.
*Behind the front wheels is a major source.The fenders can be extended into side pods,like Peraves used on their Monotracer,which will add streamlining and crush zone protection in the event of a side impact.
*Behind the notched roof.
*Behind the abruptly-chopped body rear.
*The sharp corners on the upper edges are a source of vorticity,also a source of drag.
*You'll be disappointed to know that it's a dead end,attempting to fill the turbulent wake by injecting cabin air.
*The solution is always to get the air around and to the rear of whatever you have without allowing the flow to be tripped into eddies and turbulence.Once the air is turbulent,there is absolutely no way to recover any of the kinetic energy of this swirling mass.The less air which passes through the body the better.
------------------------------------------------------------------------
As to cabin ventilation,you can run a ram-air tube from the forward stagnation area of the trike and duct this high pressure air into the cabin,or harvest fast-moving air with a NACA submerged inlet,ducting this converted,high pressure air in.

Which you might well be able to get from a midrange NVidia graphics card. You might want to check their site to see if anyone has done a CFD app.

Once again, proof no-one looked at the thread I linked in Permalink #16.

Allow me to quote from there:

[relevant links in original]

There is more there about OpenVDB support in current versions of Blender, which has animation capability.