Smooth bottom cowl system???

[Martin Sutter]

A little history on Van's "smooth bottom Cowl" intakes:

Van's first use of an injected engine was the blue-green RV8 prototype. It had a 200hp Lycoming IO-360 with a forward facing injector body and first flew with a ram air inlet akin to the one on the Sam James cowl. On the production version the intake system was reconfigured to the present snorkel intake. The combustion air is taken from inside the cowl in front of #2 cylinder. While on first thought a straight ram air inlet would appear vastly superior, the snorkel intake has it's own advantages and performs very well (Van likes speed and efficiency). The air entering the cowling thru the inlets pressurizes the area above the cylinders and allows the engine to breathe this pressurized air supply. This architecture allows for the placement of an appropriate air filter resulting in minimum loss of air flow. Incorporating an adequate air filter in a straight ram air inlet given the physical lack of space under the cowl in that location might have been one reason the snorkel solution was chosen. An added benefit was the smooth looks of the cowl without a ram inlet snout.

Martin Sutter
building and flying RV's since 1988

[N184JG]

Also the forward induction is cold air and adds about 5% HP gain.

[Martin Sutter]

The "cold air" advantage only applies if you are comparing it to an air intake source under the cowl that is taken from the rear baffle. Van's intake is located immediately behind the cowl inlet before the air comes in contact with the hot cylinders and is therefore cold also.

Martin Sutter
Building and flying RV's since 1988

[DanH]

Dan C's MP data is interesting. Theory says a good snorkel from the cooling air intake should get pretty close to the same MP as a straight front intake. The theory assumes good intake pressure recovery, a correctly sized filter, and no adverse twist and bends.

Dan's "snorkel" looks like it should work better than Van's. As I recall the stock fiberglass duct has a pretty sharp radius right before the throttle body. Dan used an alternate air duct from AFP which brings air to the front of the TB at less of an acute angle.

If you crank K&N's equation for filter size, the filter supplied with the Van's snorkel is undersized. Still, the pressure drop may not be significant.

I'm near enough to building a combustion air intake to be thinking about this stuff. At the moment (ha!) I'm inclined to take air from the left cooling intake, pipe it to a filter can in the lower left front cowl, and turn it into the AFP FM-200 with a 3" radius silicone hose. The filter can would accommodate a conical filter with more area than the standard flat filter, and would allow a water drain. The silicone hose would require a bump in the lower cowl; there's just not much room to work with in front of AFP's longer TB.

[Bill Wightman]

Pics tell the story best:

Machining the IO sump for frontal induction


We fitted an "IO"sump to an "O" top end. This required using IO360C intake pipes with a .250" thick adapter at the cylinder attach point. We liked the IO sump because of its smaller vertical profile, pseudo cold air induction, and frontal intake config.

"IO" sump fitted to "O" top end. Note intake pipe adapter.


The next trick was to design and fab an adapter to mate the RSA-5 throttle body to the sump.

Throttle body adapter


Now a fiberglass S-duct was designed. Although it looks sort of crooked, it actually has a linear reduction in cross sectional area from inlet to nozzle. The duct also looks clean from the top because we hid almost all the attach hardware underneath the baffle floor.

Intake duct


The filter is a cut down K&N mounted in front of the #1 cylinder. It slides out the front for cleaning.

K&N filter


The view from underneath shows how it all fits.


Just one way to do it... more pics posted at the Picasa web album.

[N184JG]

Originally Posted by Martin Sutter

The "cold air" advantage only applies if you are comparing it to an air intake source under the cowl that is taken from the rear baffle. Van's intake is located immediately behind the cowl inlet before the air comes in contact with the hot cylinders and is therefore cold also.

Martin Sutter
Building and flying RV's since 1988

The cold air I am talking about is on the forward sump the air is not running through hot oil as in a vert.sump has nothing to do with the air pick up.

[Bill Wightman]

About taking air off the rear baffle:

Interestingly, on my RV4 I used air from the rear baffle as fresh air supply to the cockpit, AND for induction air to the engine. The pick-off point for the fresh air was about 1/2 way between the cylinder fins and the top of the baffle.

Result: Same temp as outside air, as far as I could measure. There was no measurable increase in temp at the rear baffle at all. I assume all the air flowing through the upper cowl plenum went down through the cylinder fins and out the bottom. None of the heat went up, as long as the plane was flying.

The induction air was taken from just above and behind the right rear cylinder. It ran through a 3" hose to the fuel servo, which was turned around backwards under the sump and *then* through a 90 deg elbow. My manifold pressures ran consistently the same (in formation) as a friend who had a straight-through gut shot into his sump, with a snoot right below the spinner. Maybe conventional wisdom isn't so conventional.

One data point... take it for what its worth.

[DanH]

Really pretty work Bill. Good data too (the rear baffle air source).

<<Although it looks sort of crooked, it actually has a linear reduction in cross sectional area from inlet to nozzle.>>

Student question please. Is there a specific reason for a linear reduction in cross section? Some bit of aero theory that suggests an advantage?

[Bill Wightman]

Dan, I was wanting to avoid rapid changes in pressure inside the duct. As we shaped the foam around the starter, it turned out that was also close to where we had minimum lower cowl clearance. The duct cross section thus began to get too small in that area, and the next thing the flow would encounter would the the 90 degree bend into the throttle body.

So, I decided to take a closer look at the duct's cross sectional area. I found that we had a minimum cross section right before the bend, and thats not good. There could be some flow separation or instability on the inside bend surface.

I was then faced with adding cross section in the critical area and needed to know how much, so I simply fit the area -vs- duct position to a linear function.

Short of doing a CFD analysis, this is a WAG at avoiding flow problems.

Bill

[David-aviator]

Originally Posted by Bill Wightman

Dan, I was wanting to avoid rapid changes in pressure inside the duct. As we shaped the foam around the starter, it turned out that was also close to where we had minimum lower cowl clearance. The duct cross section thus began to get too small in that area, and the next thing the flow would encounter would the the 90 degree bend into the throttle body.

So, I decided to take a closer look at the duct's cross sectional area. I found that we had a minimum cross section right before the bend, and thats not good. There could be some flow separation or instability on the inside bend surface.

I was then faced with adding cross section in the critical area and needed to know how much, so I simply fit the area -vs- duct position to a linear function.

Short of doing a CFD analysis, this is a WAG at avoiding flow problems.

Bill

Which is why I decided not to mess with the FAB to make it fit the AFP intake.

Interesting to read what Dan Checkoway had to say about this issue 4 years ago. I know he flew with his own FAB at the left intake connected with scat tube to the AFP "y", but I don't know if he abandoned it later. Seems like later photos of the airplane show the intake below the prop missing but it may have not been visible due to the angle of the shot. I do know he achieved great air flow with very little MP drop using his intake device and the direct ram air "y" may have been removed.