What's new
Van's Air Force

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

Help with measuring cowl pressure differential

rocketbob said:
edit: here it is: http://martelmeters.com/products.php?cat=4
Click to expand...

Cool. Looks like all you would need is the meter, two resistors, and a 9V battery clip. It's only 200mV full scale straight out of the box, but the data sheet lists the two resistor values necessary to re-scale. I guess you would have to go to a 20V scale as 2V isn't enough for the expected temperatures @ 10mV/degreeF from the LM34.

Thanks Bob.

Speaking of the LM34....any necessary filtering, etc, or is it good to go with the very basic wiring?
 
Shouldn't really need anything else with the LM34 as long as you use twisted pair.

In the -6 I used the 2V range off the Electroair since it put out 10mV/degree advance, so the 2V range was sufficient. Just make sure you use precision 1% resistors. But in this case if your measuring 300 deg. the output is going to be 3V, obviously you'll need to set the divider resistors to 20V.
 
DanH said:
So, Sonny, you started the thread. What exactly do you want to determine or accomplish?
Click to expand...

The end result is that I want to increase my cooling capability in hot weather and (hopefully...potentially) decrease my cooling drag. I want to do it without cutting holes in my lower cowl for louvers.

Hopefully, measuring the upper and lower cowl pressure shows that there isn't enough differential pressure (whatever that may be). Right now I can think of 2 potential things to improve to assist in lowering the pressure in the bottom half of the cowl. 1 - I still have a gap behind my spinner that dumps air into the lower cowl, and 2 - my FAB carb heat flap may not be closing all the way. I've done everything else I can think of to improve the airflow, and thus the cooling of my oil temps. I've closed every other gap in my plenum and baffling, and I've increased my oil cooler to a 10-row cooler and increased size of the hole that feeds air to the cooler.

Crossing my fingers it'll be that easy...
 
lostpilot28 said:
and 2 - my FAB carb heat flap may not be closing all the way...
Click to expand...

Hmmm more details here... the FAB should be sealed to the front of the cowl, lost on the flap not closing all the way, maybe I missed something somewhere. :) If you're talking about the top of the FAB to the stock Van's carb heat muff... I really don't this that's your issue. :)
 
Last edited:
DanH said:
Speaking of the LM34....any necessary filtering, etc, or is it good to go with the very basic wiring?
Click to expand...

Dan, I would at least put a .1uF bypass cap right at the sensor. Hook it between the sensors V+ input and GND. It will ensure a clean voltage powering the sensor.

-Chris
 
1 - I still have a gap behind my spinner that dumps air into the lower cowl,
Click to expand...

Actually, just the opposite is happening here. The upper cowling pressure is higher then the lower cowling pressure. The lower cowling pressure is higher then the outside air. If this was not the case then air would not flow through the cowling. Thus air from the lower cowling tries to get out where ever it can, the exit air area, around the cowling attach points, at the gear farings and around the spinner cut out.
The problem with air escaping from spinner is that aside from causing drag it upsets the airflow into the cowling inlets.
This has been discussed before and I am sure you will be able to find some different examples of how people have reduced and/or sealed this area.
 
LifeofReiley said:
Hmmm more details here... the FAB should be sealed to the front of the cowl, lost on the flap not closing all the way, maybe I missed something somewhere. :) If you're talking about the top of the FAB to the stock Van's carb heat muff... I really don't this that's your issue. :)
Click to expand...

Yep, talking about the flap that opens to the carb-heat muff. When it's in the "no-carb-heat" position, it doesn't sit completely flush. I doubt it's the problem, too...just not 100% sure.

Tom Martin said:
Actually, just the opposite is happening here. The upper cowling pressure is higher then the lower cowling pressure. The lower cowling pressure is higher then the outside air. If this was not the case then air would not flow through the cowling. Thus air from the lower cowling tries to get out where ever it can, the exit air area, around the cowling attach points, at the gear farings and around the spinner cut out.
The problem with air escaping from spinner is that aside from causing drag it upsets the airflow into the cowling inlets.
This has been discussed before and I am sure you will be able to find some different examples of how people have reduced and/or sealed this area.
Click to expand...

Hi Tom, yes I've seen the other posts on this...and you could be right. Although, I have the Sam James cowl and plenum, and I"m not sure how air flow could get disrupted flowing into those. Maybe it can...I just can't visualize it. Either way, after initial pressure testing, I plan on sealing that gap the same way Dan did...looks like an easy way to skin that cat, and I have a ton of pop-rivets left over! :D

The only other thing I can think of is that my exit air flow is restricted. I have the Vetterman 4-into-2 exhaust setup, but I also have a nosewheel gear leg and supporting mounts in the way.
 
Tom Martin said:
The problem with air escaping from spinner...
Click to expand...

Tom, consider the idea of flow both in and out. The CFD porn suggests a high pressure area below the spinner and a low pressure above it. I suspect air is entering the spinner gap at the bottom in all cases. What it does from there is dictated by lower plenum pressure and/or seal placement.

For cowls with extra large exits I suspect below-the-spinner pressure easily exceeds lower plenum pressure and a fair quantity of air enters the lower cowl.

Cowls with more conventional exits probably see some rather confused flow and a good bit of air exiting the spinner gap at the top.

Cowls with tight exits probably flow up from the bottom of the spinner and from inside the lower cowl.

And folks with sealed propshafts like you and Ken and I probably have no flow in or out of the cowl, but still have some useless flow behind the spinner....at least for now ;)

Here's the thing....at this point in time we're talking about measuring, not guessing, and that's got to be all good.

bhf2c3.jpg


lostpilot28 said:
I have the Sam James cowl and plenum...
Click to expand...

Sonny, it's going to be hugely interesting to get a set of pressure measurements on an SJ cowl.
 
Last edited:
DanH said:
Sonny, it's going to be hugely interesting to get a set of pressure measurements on an SJ cowl.
Click to expand...

Oh no! :eek: Now the pressure's on! I'm very interested, too...and despite looking like I don't know what I'm doing, I'll post it anyway. The friendly Chinese company said shipping will take around 15 business days! Doh! Stay tuned...I'll be routing hoses in the meantime...
 
lostpilot28 said:
Stay tuned...I'll be routing hoses in the meantime...
Click to expand...

You can run them through the cabin heat box or rig cheap bulkhead fittings.

I grabbed some 3/16" (10-32) all-thread rod from the scrap bin, cut short lengths, chucked them in the lathe, cut off the threads on each end to form a hose barb, and center-drilled them. Add two nuts and two washers and you have a miniature steel bulkhead fitting.

The yellow tygon tubing is available at the auto parts store as "small engine fuel line". It has good heat resistance. I don't know if clear vinyl tube will hold up but somebody might want to try it...it's cheap.

2wp106g.jpg
 
DanH said:
The yellow tygon tubing is available at the auto parts store as "small engine fuel line". It has good heat resistance. I don't know if clear vinyl tube will hold up but somebody might want to try it...it's cheap.
Click to expand...

Thanks, Dan...I was wondering what the yellow tubing was. Since it's cold up here this time of year my heater will be open anyway so I'll probably route them through there.
 
DanH said:
Tom, consider the idea of flow both in and out. The CFD porn suggests a high pressure area below the spinner and a low pressure above it. I suspect air is entering the spinner gap at the bottom in all cases. What it does from there is dictated by lower plenum pressure and/or seal placement.

For cowls with extra large exits I suspect below-the-spinner pressure easily exceeds lower plenum pressure and a fair quantity of air enters the lower cowl.

Cowls with more conventional exits probably see some rather confused flow and a good bit of air exiting the spinner gap at the top.

Cowls with tight exits probably flow up from the bottom of the spinner and from inside the lower cowl.

And folks with sealed propshafts like you and Ken and I probably have no flow in or out of the cowl, but still have some useless flow behind the spinner....at least for now ;)

Here's the thing....at this point in time we're talking about measuring, not guessing, and that's got to be all good.

bhf2c3.jpg




Sonny, it's going to be hugely interesting to get a set of pressure measurements on an SJ cowl.
Click to expand...

Im inclined to agree with this depiction especially considering there will almost always be a positive AOA which will most likely ensure lower pressure above the spinner in much the same profile as the leading edge of our wing.

I think I need to adjust my upper cowl spinner seals as they do not appear to be wearing while the lower set shows the feathering you've had on yours. I'm sure they are very very close but probably leakier than they can be.

It would be great to be able to video tufting inside the cowling but I'm not so sure any of the common cameras can handle the environment. maybe rig up an enclosure with some dedicated cooling air for the camera??? Uh oh...here comes another project :)

or how about a rubber bulb filled with a viscous colored oil with a few well placed nozzles throughout the cowl. Go fly, squeeze the bulb injecting some colored oil inside the cowl (hopefully it doesn't explode :) ). Land, remove cowl and check where the oil trails lead? I'll work on the camera enclosure, Dan you can squirt the oil in your cowl :)
 
RV8R999 said:
I think I need to adjust my upper cowl spinner seals as they do not appear to be wearing while the lower set shows the feathering you've had on yours. I'm sure they are very very close but probably leakier than they can be.
Click to expand...

That seal design is imperfect anyway. Obviously I'm thinking about a seal for the spinner backplate, but I'll probably measure there before doing anything.....back burner stuff.

I'll work on the camera enclosure, Dan you can squirt the oil in your cowl :)
Click to expand...

Naaa, let's get fundamental pressure and temperature information. Here's a proposal......I'll make up three identical sets of piccolo tubes, one set for me, one for you, and one for Sonny. We install them in identical locations. One goes across the top of the engine. A pair goes under the cylinders, connected together with a tee. We'll need two lines to the cockpit manometer and a third to tap the aircraft static system. Then we all go fly the same test card and record pressures at specific speeds and altitudes.
 
DanH said:
Naaa, let's get fundamental pressure and temperature information. Here's a proposal......I'll make up three identical sets of piccolo tubes, one set for me, one for you, and one for Sonny. We install them in identical locations. One goes across the top of the engine. A pair goes under the cylinders, connected together with a tee. We'll need two lines to the cockpit manometer and a third to tap the aircraft static system. Then we all go fly the same test card and record pressures at specific speeds and altitudes.
Click to expand...

Speaking of this - when I run my tests, I'd like to make the environment consistent with everyone else. I think I need to be careful of where I put the ends of the tubes, how they're dampened on the end, the test airspeed and density altitude. Am I thinking of this right? Do all of those criteria matter much? Any that I'm missing?

The Chinese instrument allows the user to use it for differential, or gauge pressure (I'm assuming just capping off one of the lines). I lke the simplicity of gauge pressure and just listing the numbers for each location.
 
lostpilot28 said:
Speaking of this - when I run my tests, I'd like to make the environment consistent with everyone else.
Click to expand...

That's the idea.....a standardized installation and a specific test card to fly.

The Chinese instrument allows the user to use it for differential, or gauge pressure (I'm assuming just capping off one of the lines). I like the simplicity of gauge pressure and just listing the numbers for each location.
Click to expand...

You'll use it as a differential gauge. Run two lines from the engine compartment, then tap the aircraft static system for a third line. Bring all three lines to some easily accessible spot in the cockpit. Might want to incorporate a trustworthy shutoff valve at the static system tap point. No leaks please!

Ken, you in? We would have comparison data for a James cowl, a stock cowl, and my heavily modified cowl.
 
DanH said:
That seal design is imperfect anyway. Obviously I'm thinking about a seal for the spinner backplate, but I'll probably measure there before doing anything.....back burner stuff.



Naaa, let's get fundamental pressure and temperature information. Here's a proposal......I'll make up three identical sets of piccolo tubes, one set for me, one for you, and one for Sonny. We install them in identical locations. One goes across the top of the engine. A pair goes under the cylinders, connected together with a tee. We'll need two lines to the cockpit manometer and a third to tap the aircraft static system. Then we all go fly the same test card and record pressures at specific speeds and altitudes.
Click to expand...


sounds like a plan. I'm in.
 
temp and pressure sensors

I am working on a board to mount Freescale MPXV7007DP differential pressure sensors. It would output a .5 to 4.5vdc signal that would equate to a max reading of ~27 inches of water (1psi).

my question to you all, would any of you be interested in getting a circuit board? The way I have it configured now is a board 1.4" x 2.25". a 9 pin db connector, and set up for up to 4 of the sensors.

I am looking at using it with the skyview ems and using 5vdc from the ems.

I have not committed to making the circuit boards, so there may be the option of 2 or 3 pressure sensors and a couple or more temp sensors.

My thought is to sell the circuit boards and give out the schematic and parts list. The components are fairly inexpensive and available at digikey.

When we finally get datalogging (are you listening Dynon??) this should be fairly useful.

I have not verified that these sensors will work with the 10k pull up yet, but intend to do that shortly.
 
DanH said:
You'll use it as a differential gauge. Run two lines from the engine compartment, then tap the aircraft static system for a third line. Bring all three lines to some easily accessible spot in the cockpit. Might want to incorporate a trustworthy shutoff valve at the static system tap point. No leaks please!
Click to expand...

Why tap into the static system? I don't have a problem doing this, but just curious why we would do this...wouldn't straight differential pressure be preferable?
 
lostpilot28 said:
Why tap into the static system? I don't have a problem doing this, but just curious why we would do this...wouldn't straight differential pressure be preferable?
Click to expand...

Connecting the manometer to the upper plenum piccolo line and the lower plenum piccolo line would indeed tell you the pressure drop across the engine baffles. That would be the simple approach if it was the only reading we wanted.

However, we would also like to know how much dynamic pressure is being converted to increased static pressure in the upper plenum, primarily a function of inlet shape and size. That will be a differential measurement between the upper plenum piccolo and freestream static, ie, your static port.

Since we already have one leg of the manometer connected to aircraft static we'll use it to also get the pressure drop.

You'll make measurements at 10 knot intervals and fly with a helper. He or she connects the upper piccolo, notes the reading, then removes that line and replaces it with the lower piccolo line, again noting the reading. You fly to the next faster speed and your helper repeats.

The upper piccolo to static connection is plenum pressure rise. Upper piccolo minus lower piccolo is baffle pressure drop.

Anyone have thoughts on eliminating aircraft static error?
 
Last edited:
Here is my Ps data:
Vi dVpc Vc dPs
(mph) (mph) (mph) psi
80.56162965 1.102969718 81.66459936 0.003198794
107.0318794 0.215233839 107.2471132 0.000827989
133.5021291 -1.092441066 132.4096881 -0.005243053
156.5197376 -0.986773021 155.5329646 -0.005589557
179.5373461 -4.121590828 175.4157552 -0.026719488

I think when measuring Pt the aircraft dPs can be added algebraically as errors are generally very small for pitot style measurements. However when measuring static using the piccolo I'm not so sure adding aircraft dPs is valid since we have no idea what the dPs of the picollos are which can be sensitive to flow orientation. I dunno..I need to think about this more and it hurts right now.
 
RV8R999 said:
Here is my Ps data:
Click to expand...

If I read correctly Ken's worst error is 0.0267 psi at 179.5 mph indicated, or 0.739 inches H20, which is significant when compared to the pressures we wish to measure.

BTW, Ken is a professional test pilot with the necessary advanced education. To be useful here, us schmucks prefer a "Static Error For Dummies" written in plain English....limited abbreviations, no sigma, no epsilon, just step 1, step 2, step 3.
 
Frank: How did you get the website to take your excel table? I've not been successful in getting it to do that. Do you have to scan it first and insert it as a picture, or what?

Thanks,


Lee...
 
logansc said:
Frank: How did you get the website to take your excel table? I've not been successful in getting it to do that. Do you have to scan it first and insert it as a picture, or what?

Thanks,


Lee...
Click to expand...
Hi Lee, I used a screen capture (windows snipping tool) then posted as a picture. Some cool free stuff on line I use--snippy, Zoner13 free photo studio.:)
 
Last edited:
Frank, please tells us about your physical setup. Did you use a piccolo tube on the end of each hose, or an aquarium bubble rock, or careful placement 90 degrees to airflow? Where exactly were the probes placed in the engine compartment? Simple water manometer for as a gauge? What else can you tell us?
 
DanH said:
If I read correctly Ken's worst error is 0.0267 psi at 179.5 mph indicated, or 0.739 inches H20, which is significant when compared to the pressures we wish to measure.

BTW, Ken is a professional test pilot with the necessary advanced education. To be useful here, us schmucks prefer a "Static Error For Dummies" written in plain English....limited abbreviations, no sigma, no epsilon, just step 1, step 2, step 3.
Click to expand...

Ha! It just means I sound good when I'm saying something dumb and my education was courtesy of the US Navy so I'm not sure how advanced it was.

I'm sitting down with some of our bright test engineers and instrumentation folks today and will ask some of these questions.

edit: interesting read related exactly to this thread: http://www.n91cz.com/Pressure/PlenumPressure.pdf

I'm going to directly measure exit velocity as a function of cowl flap position as well as all the other pressures discussed.
 
Last edited:
DanH said:
Frank, please tells us about your physical setup. Did you use a piccolo tube on the end of each hose, or an aquarium bubble rock, or careful placement 90 degrees to airflow? Where exactly were the probes placed in the engine compartment? Simple water manometer for as a gauge? What else can you tell us?
Click to expand...
Dan,
I Used A rollyourown H2o manometer teed to the static line. On the line ends I used some homebrew foam mic muffs. the high pressure side was hooked to the engine lifting lug, the low was under and aft of the mags. Once I found this location satisfactory, I burned a whole bunch of gas.
I tested a bunch of contraptions on the botton of the cowl, one of the few that worked (a little) was a length of split vinyl hose clipped to the bottom edge of the cowl. The method I used - A printed a fill in the blanks chart. Yup one flight = what are all these numbers anyway:confused:. use cruse control and swap the labeled tubes & repeat--- For my temp readings I used a Omega HH-73RTD ($5 at yard sale:))
My plane RV-9A, o-320 d2j, standard baffling, fp prop.
Why - Im mental:p:p
1io7s0.jpg
[/IMG]
 
FrankK90989 said:
I Used A rollyourown H2o manometer teed to the static line. On the line ends I used some homebrew foam mic muffs. the high pressure side was hooked to the engine lifting lug, the low was under and aft of the mags. Once I found this location satisfactory, I burned a whole bunch of gas.
Click to expand...

My compliments sir.

To date my "probes" are much like yours, just the hose end wrapped in something to null dynamic pressure. Miley and colleagues (NASA 3405) tried a variety of measurement probes, including point probes and piccolos. Chris Zavatson used piccolos when he made measurements following cowl work on his Lancair. The idea is pressure averaging across a larger area in the measured cowl volume rather than pressure at a single point, so we'll go that way next. Seems like the right approach for comparison of airplane to airplane or for different cowl types.

I tested a bunch of contraptions on the botton of the cowl
Click to expand...

I assume you're not happy with cooling capacity. Your pressure drop numbers look a little low, but it's too early to be sure. Hopefully we'll have some comparisons for you in the near future.

For my temp readings I used a Omega HH-73RTD ($5 at yard sale:)
Click to expand...

Lucky dog!

Did you by chance record OAT, CHT, and cowl air exit temperature at any or all of your test points?
 
DanH said:
My compliments sir.

To date my "probes" are much like yours, just the hose end wrapped in something to null dynamic pressure. Miley and colleagues (NASA 3405) tried a variety of measurement probes, including point probes and piccolos. Chris Zavatson used piccolos when he made measurements following cowl work on his Lancair. The idea is pressure averaging across a larger area in the measured cowl volume rather than pressure at a single point, so we'll go that way next. Seems like the right approach for comparison of airplane to airplane or for different cowl types.



I assume you're not happy with cooling capacity. Your pressure drop numbers look a little low, but it's too early to be sure. Hopefully we'll have some comparisons for you in the near future.



Lucky dog!

Did you by chance record OAT, CHT, and cowl air exit temperature at any or all of your test points?
Click to expand...

This all started when I replaced my front cylinder dams (aluminum tape) with the same size as the calibrated beatup tape, 2024 aluminum.
Went for a ride and my Dynon gives notice that my cht is over 400° ?
havent seen this since break-in. whoa touchy front end!
Should have just replaced the tape---no I have to go off the deep end and-----
Yes I have saved lots of Dynon data, it does a good job and easy to graph. my oil always runs cool. I mostly close off my oil cooler.
If I was on the ball I would add a couple more thermocouples to my dynon, it does have the capability.:eek:
 
Last edited:
FrankK90989 said:
If I was on the ball I would add a couple more thermocouples to my dynon, it does have the capability.:eek:
Click to expand...

If you find your ball, get an accurate OAT, CHT and exit air temperature and we'll be able to see how efficient your baffling might be, ie how effectively it transfers heat to the air.

Related but not the same.....back to your pressure drop numbers. Do you have a Lycoming cooling air chart for your O-320? If not, email and ask for one.

This is the chart for an IO-360 angle valve:

o79j46.jpg


I've penciled in an example assuming 4" H2O across the baffles. At a baffle face pressure equal to a 5000 pressure altitude (ie the actual aircraft altitude would be a little higher) the horizontal line tells you the mass flow rate is about 1.5 lbs per second. Extend the horizontal line until it intersects the vertical line for the outside air temperature (60F here) and note where the intersection lies in relation to the three power curves.

In this case it is below the 140HP@400CHT line; at 140HP your CHT will be higher. It is above the 140HP@435CHT line, so at 140HP your CHT won't be that high. Looks like about 420F if you squint hard and judge the position of the intersection between the two lines.

Run a line up from 6" H2O and over to the 400CHT line. If you had 6" you could stay below 400 even with OAT at nearly 80F.

In our example the lines happen to intersect right on the 200HP@475CHT curve. IF you could make 200HP at 5000 ft with an IO-360 (a miracle, or a turbo?), having only 4" across the baffles (1.5 lbs per second cooling flow) would leave you right on the edge of a meltdown. You would need more cooling air mass flow, meaning you need more pressure difference across those baffles.

Anyway, get the 320 chart and see what it says using your pressure numbers.
 
Last edited:
The chart doesn't say anything about baffle configuration, ie. is this the pressure drop from top to bottom in a test cell? Any inlet turbulence is going to affect energy losses due to decreasing velocity and increasing pressure and vise-versa. So I don't see what the point is of the graph since its only practical to measure the system as a whole. This is just the measurements of one configuration of inlets/baffling/outlets.

I'm of the opinion that constructing a large plenum chamber is wrong, like we typically have.
 
rocketbob said:
The chart doesn't say anything about baffle configuration, ie. is this the pressure drop from top to bottom in a test cell? Any inlet turbulence is going to affect energy losses due to decreasing velocity and increasing pressure and vise-versa. So I don't see what the point is of the graph since its only practical to measure the system as a whole. This is just the measurements of one configuration of inlets/baffling/outlets.

I'm of the opinion that constructing a large plenum chamber is wrong, like we typically have.
Click to expand...

Bob - I don't think it matters much how the pressure drop was accomplished, which as you point out is affected by many things, but given a pressure drop across the cyl the resulting mass flow will result in cooling conditions as indicated in the chart.

Dan - do you have a chart for the standard O-360?

Ken
 
rocketbob said:
So I don't see what the point is of the graph since its only practical to measure the system as a whole.
Click to expand...

Yep. A favorite design reference calls it "semiempirical in nature". However, it is a handy yardstick.

RV8R999 said:
Dan - do you have a chart for the standard O-360?
Ken
Click to expand...

No, but I need to ask again about a fresh 390 chart since certification is done now. Lemme see what I can do.

Frank, I'll ask for the 320 chart too.
 
Last edited:
Ahhhh, morning....sunshine, fresh coffee ;)

I have an email in to Jeff Schans at Lycoming asking for charts.

Let's consider an example of the yardstick. As we saw above for the IO-360 example, cooling 140 hp at 5000 feet is marginal with only 4" drop across the baffles.

Assume a cowl with a very large exit area. Lower cowl pressure is probably close to freestream static. Thus we can reason that upper cowl static pressure must be at least 4" higher than freestream static, minimum. Obviously 5" would be much more desirable.

You might not guess those numbers without the yardstick chart. Let's apply them.

Freestream dynamic pressure (q) at 5000 ft (std day) and 100 knots (a self-imposed yardstick) is 29.2 lbs per sq ft, or 5.61" H2O. The handy equation is 0.00339 (a constant allowing V in knots) x density ratio (from a standard atmosphere chart) x V^2 = dynamic pressure (q) in lbs per sq ft.

Our cowls are not closed-end pitot systems; we don't convert 100% of q to static pressure. The percentage we do convert is called a pressure coefficient; (upper cowl static - freestream static) / freestream dynamic = Cp.

In this case we must have a Cp of (4 / 5.61) = 0.7 for the marginal system. If we can design a cowl with a Cp of (5 / 5.61) = 0.89 the CHT should stay under 400F on a standard day. Can we? Is it realistic?

Time to pull out another yardstick. A prime feature of good 'ole NASA Report 3405 is the actual measurements of Cp for various inlet configurations. They ranged from below 0.6 to above 0.8 (see below). Yep, it's possible, but difficult; 0.89 Cp is above the recorded range for the best tested inlet configuration on the modified Aztec. Clearly we do need a cowl with high performance (a high Cp) if you expect to use all the performance of the RV airframe. It will be fun to see where standard RV cowl installations and popular alternatives score against the yardsticks.

t62d1z.jpg
 
Last edited:
DanH said:
Ahhhh, morning....sunshine, fresh coffee ;)

I have an email in to Jeff Schans at Lycoming asking for charts.

Let's consider an example of the yardstick. As we saw above for the IO-360 example, cooling 140 hp at 5000 feet is marginal with only 4" drop across the baffles.

Assume a cowl with a very large exit area. Lower cowl pressure is probably close to freestream static. Thus we can reason that upper cowl static pressure must be at least 4" higher than freestream static. Obviously 5" would be much more desirable.

You might not guess those numbers without the yardstick chart. Let's apply them.

Freestream dynamic pressure (q) at 5000 ft (std day) and 100 knots (a self-imposed yardstick) is 29.2 lbs per sq ft, or 5.61" H2O. The handy equation is 0.00339 (a constant allowing V in knots) x density ratio (from a standard atmosphere chart) x V^2 = dynamic pressure (q) in lbs per sq ft.

Our cowls are not closed-end pitot systems; we don't convert 100% of q to static pressure. The percentage we do convert is called a pressure coefficient; (upper cowl static - freestream static) / freestream dynamic = Cp.

In this case we must have a Cp of (4 / 5.61) = 0.7 for the marginal system. If we can design a cowl with a Cp of (5 / 5.61) = 0.89 the CHT should stay under 400F on a standard day. Can we? Is it realistic?

Time to pull out another yardstick. A prime feature of good 'ole NASA Report 3405 is the actual measurements of Cp for various inlet configurations. They ranged from the high 0.5 area to above 0.8. Yep, it's possible. Clearly we do need high performance (a high Cp) if you expect to use all the performance of the RV airframe.

It will be fun to see where standard RV cowl installations and popular alternatives score against the yardsticks.
Click to expand...



from the report referencing typical rubber baffle seals:

"The implication drawn was that approximately
55 percent of the cooling air entering the intakes was bypassing
the engine and leaking through the external engine
baffle system. "

and with a plenum cover:

With the "dog house", the same
engine baffle pressure drop is obtained with 38 percent less
cooling flow enterin_ through the inlets.

what I find interesting is the number of folks who switched from rubber baffles to a plenum cover and reported little affect.

obviously a lot more to the problem and I suspect the devil is in the detail of the tranistion from inlet to plenum no matter what system is used.

still reading through the report..very interesting. Thanks!

and thanks for asking about the O-360 charts...

Ken
 
It has been a few years since I went through this exercise but at the time I used the information taken from an excellent article in Sport Aviation August 2003 called "Cool It" by Neal Willford.
He included a spreadsheet that all information for your aircraft cooling system could be added to. It allowed you to play with inlet and outlet sizes, made allowances for different engines etc. He drew on the work in the 30s and 40s to optimize inlet size vs. aircraft speed and related everything to the horsepower that it took to cool the airplane; interesting stuff.
Ultimately I was able to make modifications to get my cooling system to work at the temperatures I wanted to run at. It thought it was one of the best articles written on the subject. I suspect you can find this on the eaa archives
It worked for me.
 
Tom Martin said:
.... I used the information taken from an excellent article in Sport Aviation August 2003 called "Cool It" by Neal Willford.
Click to expand...

I like everything Willford writes. Two quick notes on the article, which I just read.

Did you see the expression "Heat removed = Mass Flow x Cp x Air Temperature Change"? Here we see a detail missing from the cooling chart approach. The chart assumes a particular fixed air temperature rise. However, anything you can do to increase air temperature rise is an increase in cooling capacity....and it's more or less drag free.

In my previous post I used the term "Cp" for Coefficient of Pressure, as defined in NASA 3405. In Neal's article he also uses "Cp", but it has a different meaning, Constant Pressure Specific Heat of Air, BTU/lb per degree F in English units. Now you're not confused, right?

Ultimately I was able to make modifications to get my cooling system to work at the temperatures I wanted to run at.
Click to expand...

Tom, didn't you also do some pressure measuring? Care to share methods and results?
 
Dan
I used a homemade water manometer, much like the ones seen here, and had a number of 1/4" plastic air brake hoses coming through heat box. I just disconnected the hose, rerouted it a bit and used aluminium tape to seal around the hoses.
I do not remember specific numbers, basically there was a 5" difference between the back of the top plenum and the bottom of the engine in the lower cowling area. I believe it was 12" on top and 7" below.
What I really wanted to test was the pressure in the crankcase and I had removed the dip stick and had a hose taped into a plug at that location. I had another line down by the outlet of the crankcase vent. It dumped on the exhaust like many of the RVs. This works ok as the drips get burned off on the exhaust pipe. What I was finding is that as I closed off my outlet I was blowing a bit more oil out the vent and my theory was that I was increasing the airflow at that location and creating a vacuum in the crankcase.
After engine start, at idle on the ramp, there was one inch of positive pressure in the crankcase.
during the flight I measured 8" of pressure in the crankcase and I thought, there we go 8" of vacuum that explains the oil loss. But it was 8" of POSITIVE pressure and I just could not believe what I was seeing, most certainly not what I expected. However there is no coincidence that the 7" lower cowl pressure plus the 1" positive noted at idle added up to 8". My crankcase was being pressurized by the lower cowling pressure.
I rerouted the breather a couple of inches aft of the cowling and went for another flight and I now had a Negative 2", or a vacuum in the crankcase.
Apparently the auto race guys like to create a bit of vacuum in the crankcase as the pistons do not have to fight pressure on the down stroke.
Immediately I noticed a complete stoppage all the tiny oil leaks in my engine.
The oil consumption did not change but it turns out that was related to a faulty set of rings that were discovered when the cylinders were returned in the ECI cylinder recall.
Any ways, my engine is a lot happier with the breather out of the cowling. Perhaps that is why all the certified aircraft have them hanging out there.....

ps my last pressure test was with a hose that was located at the cowl inlet. I did not get a reading there but I did blow all the water out of my manometer all over the cockpit!
 
Although, I must say My RV-9A, Out of the box following the plans meets or exceeds Van's published preformance numbers. Engine cooling is easy to manage (I have been to so. TX. in the summer) To add any significant improvements to Van's design is going to take lots of magic.
Let the experimenting continue---:)
-11°F this AM
 
Last edited:
Tom Martin said:
Dan
I used a homemade water manometer, much like the ones seen here, and had a number of 1/4" plastic air brake hoses coming through heat box. I just disconnected the hose, rerouted it a bit and used aluminium tape to seal around the hoses.
I do not remember specific numbers, basically there was a 5" difference between the back of the top plenum and the bottom of the engine in the lower cowling area. I believe it was 12" on top and 7" below.
What I really wanted to test was the pressure in the crankcase and I had removed the dip stick and had a hose taped into a plug at that location. I had another line down by the outlet of the crankcase vent. It dumped on the exhaust like many of the RVs. This works ok as the drips get burned off on the exhaust pipe. What I was finding is that as I closed off my outlet I was blowing a bit more oil out the vent and my theory was that I was increasing the airflow at that location and creating a vacuum in the crankcase.
After engine start, at idle on the ramp, there was one inch of positive pressure in the crankcase.
during the flight I measured 8" of pressure in the crankcase and I thought, there we go 8" of vacuum that explains the oil loss. But it was 8" of POSITIVE pressure and I just could not believe what I was seeing, most certainly not what I expected. However there is no coincidence that the 7" lower cowl pressure plus the 1" positive noted at idle added up to 8". My crankcase was being pressurized by the lower cowling pressure.
I rerouted the breather a couple of inches aft of the cowling and went for another flight and I now had a Negative 2", or a vacuum in the crankcase.
Apparently the auto race guys like to create a bit of vacuum in the crankcase as the pistons do not have to fight pressure on the down stroke.
Immediately I noticed a complete stoppage all the tiny oil leaks in my engine.
The oil consumption did not change but it turns out that was related to a faulty set of rings that were discovered when the cylinders were returned in the ECI cylinder recall.
Any ways, my engine is a lot happier with the breather out of the cowling. Perhaps that is why all the certified aircraft have them hanging out there.....

ps my last pressure test was with a hose that was located at the cowl inlet. I did not get a reading there but I did blow all the water out of my manometer all over the cockpit!
Click to expand...

Very interesting... right after I installed my cowl flap I noticed more oil on the bottom of the fuse than ever before and oil consumption was up. My breather lines terminate right at the stock exit but with the cowl flap installed they are now upstream. Hmmm..

Thanks for posting!
 
DanH said:
I like everything Willford writes. Two quick notes on the article, which I just read.

Did you see the expression "Heat removed = Mass Flow x Cp x Air Temperature Change"? Here we see a detail missing from the cooling chart approach. The chart assumes a particular fixed air temperature rise. However, anything you can do to increase air temperature rise is an increase in cooling capacity....and it's more or less drag free.

In my previous post I used the term "Cp" for Coefficient of Pressure, as defined in NASA 3405. In Neal's article he also uses "Cp", but it has a different meaning, Constant Pressure Specific Heat of Air, BTU/lb per degree F in English units. Now you're not confused, right?

Tom, didn't you also do some pressure measuring? Care to share methods and results?
Click to expand...

Pretty sure Cp is BTU/lbm-R if you are going to look up the value in a table. Just convert to F if needed.
 
longranger said:
Same thing (1deg F = 1deg R) if you're looking at a temperature difference, which you're doing here.
Click to expand...

huh? Cp is not used for a delta-T but rather to calculate the fluid properties at a given location within the control volume in question. The measured temp will not be a delta-T but a property value and must be converted to an absolute scale (either Rankine or Kelvin) to use seemlessly with the other thermo equations. You can convert Cp to deg F but then the units of all the other constants will need to be changed too. Much easier to stick with the convention but this is just my opinion.
 
At risk of further exposing my ignorance, how much does Cp vary over the few hundred degrees temperature difference from the top to bottom of the engine? I was assuming it was constant enough that you could just multiply Cp by the mass flow and temperature difference to derive a heat transfer rate, rather than have to calculate the total heat at each condition. (I'm only an instrumentation EE, and it's been 35+ years since my last thermo class...:eek:)
 
Easy gents.

Specific Heat of Air is simply the amount of heat it takes to raise the temperature of a given mass by one increment of temperature. In the given English units the amount is BTU, the mass is one pound, and the temperature can be stated as F or R because they are the same increment.

The rest of the world likes kJ/kg per K. Let's not go there.

Anyway, it doesn't matter for purposes of this thread. I only brought up Cp in the context of specific heat in order to prevent confusion. Boy am I dumb.....;)

The jury is instructed to ignore all comments related to specific heat. Henceforth "Cp" shall mean Coefficient of Pressure.
 
RV-7A with an IO-360-L2A with 56 hours.

I used a simple water-based manometer. Positioned an end of a tube above the cylinders and moved the other tube end to various places below the cylinders. Could do this with just the top cowling off. Ran both tubes through the firewall to the cockpit (and the manometer ports) using my cabin heat inlet.

End result was an 8" differential. I believe that is about .3psi. My cylinders stay cool.
 
R or F I was in error...

Its been raining all day here in West Palm Beach so no flying which gave me a chance to dust off my trusty old Zucker - Fundametals Of Fluid Dynamics.

Don't try to impress the girls at the bar talking like this though it won't work :)

1st Law - Conservation of Energy: Q=W+change in Energy

a whole bunch of calculus to get the control volume equations:

station 1 is upper plenum, 2 is lower

h1 + (V1^2/2gc) + (g/gc)Z1 + q = h2 + (V2^2/2gc) + (g/gc)Z2 + Ws

Where: h = enthalpy = Cp(T) (well really Dh= Cp(dt) for a perfect gas) but since we will consider the flow in the cowl 1 dimensional and steady we can drop the diferential form.) Steady only means the flow at a point isn't changing with respect to time - it can change from point to point and still be considered steady.

V^2/2gc= Kinetic Energy (KE)
(g/gc)Z = Potential Energy
q = heat transfer to or from the control volume
Ws = Work done on or by the system

For our case: Ws=0

and across the cyl change in height will be very small ~ 0 (or we can calculate it if we want from inlet to exit which is more significant)

so:

h1 + KE1 + q = h2 + KE2

I believe we can look up heat transfered from engine tables or charts (Dan??)

h1=Cp(T1)
h2= Cp (T2) (Cp is for perfect gas at constant pressure which we know isn't exactly correct here but helps understand the process)

h2-h1 = Cp(T2-T1) so YES, you are correct either R or F is insignificant for this calculation assuming we can look up heat transfered. Sorry my bad :)

KE2-KE1 = (V2^2 - V1^2)/2gc

Therefore:

q = Cp(T2-T1) + (V2^2 - V1^2)/2gc

which may provide some uselful insight but we'll need to introduce the pressure-energy equation which includes concepts of entropy and the stagnation process to really understand what is going on under the cowl.


Area changes, friction and heat transfer are the 3 most important factors which affect flow properties. Its been a long time for me so I'm digging back in the books to understand Fanno Flow propeties, which include losses, and understand how this might or might not apply to under cowl flow characterization.

Ultimately we are limited by properties we can either measure during flight or look up in tables - I'm always trying to ensure my flight tests are done efficiently with the minimum of configuration/instrumentation changes. Understanding the theory helps ensure we set-up each test to this end.
 
No problem Ken!

I did shoot from the hip a bit by assuming that Cp would be constant enough for our purposes over the temperature range involved. Thanks for doing the research to verify! My thermo books haven't been unpacked from at least two moves (15 years) ago.:rolleyes:
 
RV8R999 said:
I believe we can look up heat transferred from engine tables or charts (Dan??)
Click to expand...

In very general terms, something like 1/3 of fuel energy or 1/2 rated HP. 0.4 HP seemed to work in past models. I saw a chart yesterday (in the Willford article Tom mentioned) which suggests 0.4~0.5 hp for engines in our size class.

However, let's not drift. Our interest here is simple, repeatable pressure measurement.
 
You must log in or register to reply here.
Back
Top