[Author Prev][Author Next][Thread Prev][Thread Next][Author Index][Thread Index]

Exhaust size and free power.

What you guys are getting perilously close to is called Ejector Theory. 
When a rapidly moving stream of air exiting a nozzle is surrounded by a
larger diameter collar which is open at both ends, the total volume of air
moved by this system across the exit plane of the collar is greater than
that of the primary stream of air.  The increase in mass flow from this
pumping action is significant enough for engineers to attempt to harness
this effect for industrial use, although the exact mechanism by which this
occurs is yet to be clearly explained.  Problem is that the effect takes
place over a rather narrow range of conditions.  It is rather like a tuned
pipe exhaust system (used on motorcycle, outboard boat, and model airplane
racing) in that it is best at a precise ratio of mass flow, tube length and
diameter.  For an Ejector the gains are greatest when a specific
relationship between primary stream mass flow and secondary flow area is

Years ago at the Arnold Engineering Development Center, worlds second
largest wind tunnel complex, we worked very hard to harness this effect for
wind-tunnel testing of turbine engines, and rocket motors, by trying to
develop a variable area Ejector, which could be useful over a range of
conditions, and as of when I left that job it had not been accomplished.  I
must mention the need to add the Ejector to these engines and motors was
not to increase their performance, but to increase the ability of the
facility to test them over a wider range of altitudes.  Warning: requests
for more information on this topic constitutes sufficient cause for instant
membership into the Nerd Club.

Problem with trying to harness this effect for automotive exhaust is that
the primary fluid, the exhaust stream... is the one to which energy is
applied to cause the effect and it is the very same system on which you are
trying to see the gain.  Bootstrapping rarely works in real-world physical

The closest application I've seen toward evacuating the exhaust system was
in the early-mid 60s some drag cars used an exhaust system that had a huge
megaphone looking thing pointed forward underneath the car.  The ram air
was funneled into a relatively narrow section at which point the engine
exhaust was introduced into this now 'low pressure' area and was suppose to
evacuate the exhaust system better.  Clearly this system had a chance to
work at the big end of the track but at low speeds there was certainly some
exhaust exiting forward, which would then reverse as vehicle speed
increased and everything started going the right way.  

The performance gain of introducing an external vacuum-inducing device to
the system, like Frederic reports, will certainly work, but the added
horsepower will likely be overcome by the extra drag/weight/energy
requirements of that added system.  I'm not suggesting that 'free power'
doesn't exist, but I am sure that 'free' and 'easy' are not the same thing.
 I'm also sure that many experts proclaimed turbocharging would never
result in any significant gains for automotive applications.  

Regards,  Gross
from Kurt Vonnegut's iconoclastic 1997 commencement address at MIT, "Sing.
Floss. Stretch. Do one thing every day that scares you."

> Date: Fri, 21 Nov 1997 20:30:03 -0500
> From: Frederic Breitwieser <frederic.breitwieser@mcione.com>
> Subject: Re: Corky Bell's "rules of thumb"...exhaust size

> Not sure where I saw it... but there is a new product being developed
> called a "negative backflow muffler".  Basically, its a tailpipe inside a
> tail pipe, with the inner one connecting to your engine.  The outside
> tailpipe, or sheath, has a (wild guess here) 1/2" gap between it's inside
> wall and the outer wall of the regular tailpipe.  This sheath then
> to a funnel or air pump (they are testing and playing with it) and
> hopefully by increasing the velocity outside of the tailpipe in a
> controlled manner, the venturi effect will take affect and suck the
> gas out.
> Frederic Breitwieser
> Homebrew Automotive Mailing List
> Bridgeport, CT 06606
> http://www.geocities.com/MotorCity/Downs/4605/index.html
> 1989 AG Hummer 4-Door
> 1993 Supercharged Lincoln Continental
> 2000 Mid-Engine Sports Car <smile>
> Seeking Info on: GM Tubular Upper and Lower A-Arms
> , comments to quattro-owner@coimbra.ans.net