What's new
Van's Air Force

Don't miss anything! Register now for full access to the definitive RV support community.

How I did my ram-air intake, results

scsmith

Well Known Member
Now that I have good test results, I figured I would post some pictures of how I modifed Rod Bowers' ram air intake.

Rather than draw the alternate, filtered air through reed valves from the lower cowl plenum, I wanted to supply the air from the front baffle floor in the cooling air intake. This area has nearly the same ram pressure as the full ram intake, and only minor ducting losses, plus the loss through the conical K&N filter.

I made a replacement fiberglass canister that was the same dimensions as Rod's canister, then molded on the adaptor for the scat tube connection.

Rod's reed-valve canister is designed to prevent back-flow through the filter and out to the lower pressure area, loosing valuable ram-pressure. In my case, since the cooling intake area is essentially at ram pressure, there is virtually no back flow. What little back flow there is made up for by the over-size of the ram intake, able to supply much more air than the engine needs.

I get 0.5" hg lower MAP through the alternate filtered supply than the full ram air supply.
One thing that this suggests is whether the extra 0.5" hg pressure from the ram intake is worth the installation work? As I think Dan Checkoway showed, a well-designed filter mount on the baffle, followed by the standard curved intake duct, also gets within 0.5" hg of full ram pressure. This also preserves the completely clean lines on the cowl, which is nice, although I kinda like my air inlet too. But if I build another RV, I will have to decide if all this is worth it for a half inch of MAP?

here's some pic's

100_1801a.jpg

p1010163b.jpg
 
Last edited:
is it just the picture angle...
the RAM opening seems smaller than Bower's.
Does it make a difference?
 
SNIP

I get 0.5" hg lower MAP through the alternate filtered supply than the full ram air supply.

SNIP

This also preserves the completely clean lines on the cowl, which is nice, although I kinda like my air inlet too. But if I build another RV, I will have to decide if all this is worth it for a half inch of MAP?

SNIP

Looking at the Lycoming power chart in the operators manual for my 160 HP O-320 B2B, it shows a 5 HP increase for a 1/2" MP increase.

Is it worth it?
 
Looking at the Lycoming power chart in the operators manual for my 160 HP O-320 B2B, it shows a 5 HP increase for a 1/2" MP increase.

Is it worth it?

I'm sure plenty of people would say YES! Heck, I'd bolt a mailbox to the bottom of my cowl for an additional half inch of MP!

To the OP:

What is your MP at cruise and 7500?
 
Last edited:
I bet you are getting a lot more than the equivalent of 1/2" because now you are pulling cold induction air instead of heated lower plenum air, which should be especially beneficial during takeoff.

Vic
 
I bet you are getting a lot more than the equivalent of 1/2" because now you are pulling cold induction air instead of heated lower plenum air, which should be especially beneficial during takeoff.

Couldn't agree more.

It's nice work Steve, and a pretty airplane too.

Small suggestion about the hose intake; some lip radius might be a plus. Easy mod.
 
I like the looks of the intake! Really looks good. Almost Mustang-esque.

690245238_62yEM-M.jpg


Gee - that paint scheme looks kind of familiar!
 
I have the Rod Bower set up on my rocket. I used to use the RAM air just for high altitudes and for races but after a long flight in stable air I am going to change my operations. On the way home from Pagosa Springs CO I had some time to play with systems. With the RAM air door closed my induction air comes through reed valves that surround the filter. The air is taken from the lower, warm side of the cowling, through the reed valves, the filter and into the engine. With the RAM air door open the air goes directly into the engine, unfiltered.
As we fly at 22 squared I do not really need the RAM air for MP so I elected to go with filtered air in cruise altitudes between 2500 and 8000 feet and RAM air over 8000 feet.
I had made the assumption that 22 squared would be be 22 squared regardless of the source of the air.
To test this theory in the dead calm air that we had that day I opened my RAM air valve and then reduced the MP with the throttle back to 22 ?. After letting the air speed stabilize I was surprised to note a 3 to 4 knot increase in speed. This was repeatable. As the MP seen by the engine was the same this power increase could only have been caused due to the difference in air temperature and perhaps a smoother flow of unfiltered air. After giving this some consideration it really is not that surprising as using warm induction air from the lower cowling is almost like turning on carb heat. I would not have expected this great a difference in speed for the same fuel flow so I will most certainly be using the RAM air on most cross country flights when the air is clear.
My net manifold pressure gain from using the Rod Bower system was at least 0.5? and that cost me a lot of time and money. For the kind of flying I do it was worth every penny. Now I will use it even more then before!
 
So far I've built two pressure recovery intakes. Both were filtered; this 390 cost more than my first house.

Airbox #1 was lovely, but in retrospect was dumb as a bag of hammers.

RU3120%20Intake.jpg


Cone%20Filter%20Airbox.JPG


K&N cone filter. Way too much pressure drop on the flow bench. Lesson learned; K&N's oft-quoted equation for required filter area doesn't always apply. The filter dimensions don't help much because they make the filters with many different pleat depths. What counts is media area, not filter size....if you cut the folded media out of the rubber frame and stretched it out, how much area would you have? Most of the round or conical filters don't have much pleat depth compared to the flat plate filters. Duh on me.

I used the second airbox for a how-to example in a fiberglass thread:

http://www.vansairforce.com/community/showthread.php?p=364799#post364799

The filter was selected on the basis of media area; it was the largest thick plate filter I could stuff into the available cowl space.

P5240004.JPG


P6260001.JPG


Don Rivera was kind enough to run it on the flow bench. Mounted on an FM200, pressure drop through this airbox was 1.9" H20 (0.14" Hg manifold pressure) compared to the same FM200 tested with no airbox at all. I'll venture the Bower system doesn't do much better even with the butterfly wide open.

Postscript Per the data in a VAF homepage advertising announcement (11/4/09), the above filtered system is slightly less restrictive than an unfiltered Bower setup, and 10X less restrictive than a Bower in the filtered mode.
 
Last edited:
read between the lines


As I read thru this a while back, I understood the writer to say that increasing the size of the inlet will produce additional MP. Careful sizing and shaping will minimize drag from the outflow (actually a pressure wave in front of the inlet).

I recall a well recognized aerodynamicist telling me to go ahead with larger than desired cooling inlets on my cowling for this exact reason (pressure increase), but this sort of design required throttling the cooling system with a cowl flap. I mention this only as a confirmation of what is written in the article noted above.

As "Dan the Firewall Man" Horton explained it to me also, this oversized inlet is called a 'low speed inlet', and needs less attention to detail/ducting etc behind the inlet. Have a look at a Cirrus or Cessna Corvallis for examples of such design work.

Carry on!
Mark
 
Ram air

If you have started out with the Rod Bower system, you will note poor take off performance.

This is not only due to heated air in the lower cowl.

Here are some tests I performed on my own flow bench.
flow bench results 3/23/14
set motors for 250cfm with vac hole on the flow bench open, then place several objects over hole with plenums to get relative flow drop without changing motor speed.

33-2060 std sq snorkel filter (used) with square plenum 2" deep between filter and table 243 cfm.
RU-2520 Rod Bower filter with plate on small end. 250 cfm. no restriction at all.
filter in RB can with valve open. 249 cfm. no bell on end, just square end of valve body tube.
valve closed both reeds normal 225 cfm.
one reed removed so breathing through 2 3" holes. 243cfm

one reed set removed and the 2 holes taped over 176 cfm.

The tests show a solid 24 cfm flow drop using the reed system which is very poor.

I started out with the snorkel system and was always about 1" hg below standard atmosphere at 8000 feet.

Now with the ram air open, I consistently have MP slightly over std atmosphere. That's the good news. The bad news is the unfiltered air.

This winters project will be a filter set up like Dan H. I want to use the existing Bower scoop. I will use unfiltered air for alternate air from inside the cowl.

On a recent flight here is the data:
density alt 7997
MP 22.7
the standard atmosphere charts for 8000 lists 22.22 in hg.
 
Errata explanation

I read the report (http://www.n91cz.com/InductionSizing...izing_rev0.pdf) and was initially confused.

He states earlier in the report to divide the volume derived from displacement x rpm by two because he was considering a 4 stroke engine. He inserts values into the formula but fails to show the "4 stroke" correction.

However, without explaining, he got back on track by using the value 7500 in^3/sec.
 
Last edited:
Errata explanation

I read the report (http://www.n91cz.com/InductionSizing...izing_rev0.pdf) and was initially confused.

He states earlier in the report to divide the volume derived from displacement x rpm by two because he was considering a 4 stroke engine. He inserts values into the formula but fails to show the "4 stroke" correction.

However, without explaining, he got back on track by using the value 7500 in^3/sec.

You divide (displacement x RPM) by two because half the time the pistons are stroking out that displacement, you aren't taking air in (on a four-stroke).
 
It's in there.

While we're commenting, the last two lines above the chart may confuse some readers.

For example, At 200 kts, with a 3 inch diameter inlet, the differential in the inlet comes to 148 kts or 1.0 in-Hg. This translates to 1.9 in-Hg above the zero ram condition found in the engine performance charts.

Although obvious to some, Chris didn't mention altitude in the example. A 148 knot inlet differential generating 1" Hg would require a hydroplane, as altitude for that result is sea level. All else equal, plan on something less at altitude.

Chris is not saying the example offers a 1.9" Hg rise. The difference between 1.0 and 1.9 in this example is induction system loss, the end reference points being the inlet of the servo, and the primer port, where we usually tap for a manifold pressure reading.

I use similar equations with the addition of VE, and they seem to be accurate in measured reality. From a recent report:

The airbox works fine. At 2500 ft and 201 KTAS, measured pressure recovery at the airbox inlet is 17.7" H2O. Calculated value is 16.89 for 100% VE, or 17.51 for 90% VE, assuming freestream only, no prop outflow

Lycoming's chart number for induction system loss (0.9" Hg, 12.2" H2O) may be right for an updraft engine like CZ's. I have no idea. The 390 chart also discounts MP by 0.9" Hg, which seems odd given the difference in induction system layout. I measured overall pressure loss for the entire intake tract at 14.4 or 14.5" H2O. The airbox and filter measured at 1.9" H20 loss on the flow bench, so the actual loss appears to be 14.4 - 1.9 = 12.5" H2O, which is 0.92" Hg...close enough.
 
Last edited:
Here are some tests I performed on my own flow bench.
flow bench results 3/23/14
set motors for 250cfm with vac hole on the flow bench open, then place several objects over hole with plenums to get relative flow drop without changing motor speed.


filter in RB can with valve open. 249 cfm. no bell on end, just square end of valve body tube.

That is pretty good news! Who knows, but maybe if you had put a bell mouth on the inlet, you might have gotten 250?

Anyway, this shows that the intake tube with the butterfly plate in it causes very little loss.
 
.... Who knows, but maybe if you had put a bell mouth on the inlet, you might have gotten 250?
Well, a bellmouth, a funnel, whatever, may give you more rain, more bugs, more dirt, but not a bit more pressure. Not fair is it? Not intuitive either. :D

BTW, I am a Free Diffusion advocate. It is ~100% efficient. This translates into large, well faired intakes.
 
1/8” radius ? 1/4” ? More? I am ready to act with sandpaper in hand !

I don't know of a paper dealing specifically with spillage and nacelle forebody shape in the context of a low speed intake.

Any lip radius is a compromise given the variables. Look at the lips on Steve's intake, first post here. Pretty generous, and he knows about these things.

An old paper from the era of round engine research, NACA 745, offers pressure distribution and drag for round cowls, thus a clue to a desirable forebody shape for low separation. I'll venture you want something like Nose A, Nose B, or Nose C (figs f, g, and h).

NACA%20745%20Shapes.jpg


There are ordinates, if you want to be real picky.

NACA%20745%20Ordinates.jpg


I thought the body shape was pretty good here, and it has enough interior volume for an airbox with lots of filter area.

http://www.vansairforce.com/community/showthread.php?t=173865&highlight=silicone

In retrospect, I think mine is too sharp...not enough leading edge radius, forebody a little lean, so it may suffer some external separation at higher airspeed. The cooling inlets have a better external shape and seem to work fine, but one of these days I'd like to tuft test them.

Frontal%20View.jpg
 
Last edited:
Well, a bellmouth, a funnel, whatever, may give you more rain, more bugs, more dirt, but not a bit more pressure. Not fair is it? Not intuitive either. :D

BTW, I am a Free Diffusion advocate. It is ~100% efficient. This translates into large, well faired intakes.

I meant a bell mouth for testing on the flow bench. Certainly not on the airplane.
 
KISS?

I might be seeing things wrong, but the inlets I have (THANKS DAN!) sure look like the LE of a Cub or Champ airfoil, like nose 1. Nose B and C resemble the LE of the RV wing...fairly simple to fab up a mold for any of those.

Of course our inlets will NOT have the afterbody like the round engine cowls do...a lot more expanding on our setups except on the upper sfc.

Dan: why is your air inlet for the engine so different from the cooling inlets? Seem you'd want the same recovery value for both? I must be missing something...
 
1/8” radius ? 1/4” ? More? I am ready to act with sandpaper in hand !

Hi Larry,

Mine is about 5/16 radius, and I think it is definitely overkill.

In Dan's picture, it looks like a bit under 1/8" but hard to say with the lighting.

I guess I would recommend at least 3/16" radius, with continued thickening so that if you go back 1/2" from the lip, it is at least 1/2" "thick" (distance through duct from inner wall to outer wall. As if it was a 1/4" radius, but not circular, more elliptical.

Hope that helps.
 
Last edited:
Lip shape radius

This is so timely, I?m working on the inlet lip now. So far I?ve used a short piece of 1/2? pvc (cut in half lengthways) with sandpaper cupped on the inside to create that half round shape. By more ?elliptical shape? do you mean instead of a perfect 1/2 circle it would be just a little pointed at the apex? A simple sketch of the cross section would be great.

Don Broussard
RV9 Rebuild in Progress
57 Pacer
 
Dan: why is your air inlet for the engine so different from the cooling inlets? Seem you'd want the same recovery value for both? I must be missing something...

In terms of pressure recovery, it works just fine. It is a low Vi/Vo type, like the cooling inlets....a 4" ring

I built that inlet many moons ago, with an eye on style. It may be separating externally, aft of the lip when spillage is at max, causing some external drag. Some tufts and a GoPro could be in my future ;)
 
Last edited:
1/8” radius ? 1/4” ? More? I am ready to act with sandpaper in hand !

I forgot about the profiles in CR3405, identified as Kuchemann A-series. They are much as Steve describes in post 23. Start at page 38.

The reference is Aerodynamics of Propulsion, Kuchemann and Weber. Out of print and not cheap as a used book. Anyone have a copy? Worth having?
 
Last edited:
Airbox #1 was lovely, but in retrospect was dumb as a bag of hammers.

Help me understand why your first airbox was dumb? Just not enough filter area? Not enough divergence in the box?

Use small words, I'm just a dumb pilot.
 
Back
Top