Kahuna

Kahuna

Moderatoring
It was with great dissappointment that I read Bobs results of his cowl mod . How depressing.

Bob you did a ton of work, great docs... Very sorry it did not heed the results you anticipated. Im sure we have all looked at our engine installs and seen opportunities for improvement. I once contemplated what Bob has done. I never would have done as good a job. I never did follow through with my ideas figuring that if it had significant improvement, we would see this on other piston installations. Folks like Mooney and Cirrus fight tooth and nail for every knot. Their engine installs look just like mine.

Thanks to Bob for ignoring what others are doing and trying this out. While the results are disappointing, the efforts are extrodinary. Thanks for the great pics, details, follow through, and results Bob.

Best,
 
Kudos as well

Bob
Your efforts are what experimental aviation is all about. Your design and results are valuable data. Thanks for sharing it with us.
 
I Agree!

Good work Bob! Every result of an experiment is, in a way, a positive - even when you don't get the results you expected.....it adds to the overall knowledge base, and helps everyone. I don't know how many times I've tried somethign in asimulation and the (useful) result was "well, let's make sure not to try THAT again!!" :rolleyes:

I followed your saga with great interest - I admired the creativity. It showed great promise in the idea...

So...what are ya' gonna try next?! :D

Paul
 
Cooling drag is created because the air that comes into the cowling must slow down to match the aircraft's speed (using the aircraft as the frame of reference). Or, if we switch our frame of reference to the outside world, that air must go from a standing stop to the same speed as the aircraft. It takes a force to accelerate that air, and this force comes from the aircraft pushing forward on the air.

If you want to decrease cooling drag, you need to reduce the amount of air that is being used for cooling. The only way to do that is to reduce the size of the cowl inlets and/or outlets. If you reduce the amount of cooling air, the engine temperatures will increase, so there may need to be improvements in the baffling to make better use of the cooling air. Adding devices like this mod to control the air inside the cowling could help make better use of the cooling air, and thus allow the amount of cooling air flow to be reduced by other cowling mods, thus leading to a drag reduction.

Don't give up. There may be some speed to be had by playing around with cowling inlet and exit areas. But first, get a good bunch of baseline speed data using a more accurate flight test technique. The amount of inaccuracy in the U.S. Air Race, Inc technique could be greater than the speed changes you are trying to measure. I strongly recommend the four leg option on the NTPS spreadsheet. And do several runs on different days, to be sure you have a repeatable result.

Ideally, all testing would be done at the same altitude and temperature. But that is not possible in the real world, so the next best bet is to always be at the same density altitude. The U.S. Air Race, Inc handicapping test procedure tries to compensate for temperature, but they seem to assume that there is a standard lapse rate between 6,000 ft and the test altitude. The test altitude they suggest is a good first guess, but once you get there, check the temperature, and adjust the test altitude as required to get the desired density altitude.

If you want 6,000 ft density altitude, the following combinations of pressure altitude and temperature will do it (temperatures in deg C):

9,000 -26.7
8,500 -21.9
8,000 -17.1
7,500 -12.1
7,000 -7.1
6,500 -2.1
6,000 3.1
5,500 8.4
5,000 13.7
4,500 19.1
4,000 24.6
3,500 30.2

Plot those points on a graph, and draw a line through them. Take it with you when you do the testing, and find an altitude where the temperature falls on the line.

Note: engine power is not exactly proportional with density, so using density altitude to try to correct non-standard day data to predict standard day speed is not an exact science.
 
Conclusion

The factors you are working against are that the air flow inside the cowling is significantly slower than free-streem velocity. Items causing parasitic drag inside the cowling do not significantly effect overall cooling drag because of this low velocity. It was certainly a valient effort though, and very interesting.

Can we all conclude that changes to the upper deck and baffling more directly effect cooling and cooling drag than what happens after the air passes through the fins?

-Bruce
 
Excellent results

Kahuna said:
It was with great disappointment that I read Bobs results of his cowl mod . How depressing.
Click to expand...
I am not depressed. Now we know that does not work. OK

Good for Bob, he is the man. He had an idea and excellently executed and documented it, from start to finish. When Edison was asked if he was depressed about the 1000's of failures he had finding a working light bulb, he said no. I recall his quote was to the extent of now I know 1000 things that don't work. Bob is following a great tradition like Edison and the Wright Bros.

I don't think anyone could say for certain if this would have improved speed without trying it. Now we know thanks to Bob. Hats off. :D

PS. I don't think he is going slower and not sure how he got his average but I would agree it did not make a large difference.
 
You guy's said it

Sometimes you just have to try and idea to clear your mind for something better. Thanks to Doug for posting my e-mails. I got a lot of ideas from you folks before during and after this little experiment and I appreciate them. I have thought about the mod and the results and I am not quite ready to give up yet. I'll let you know what comes up. Paul good luck with Discovery tomorrow. There is a young man out there that turned 50 today - Happy Birthday Donald.

Bob Axsom
 
All -

I am fortunate to know and work with Chris Zavatson.
He's a great person, very unassuming... and probably the most brilliant experimental airplane guy you'll ever meet.
My only regret is he's a "fast glass" person instead of a "Hershey Bar wing" advocate.
Please - Don't hold that against him.

Chris built an award wining Lancair 360 - and has spent the last several years working to improve it.
One of his most recent endeavors was aimed at cooling efficiency / performance increase.

Check out his website.... it's worth it!!
www.n91cz.com

Lorin D
9A Wings
N194LD reserved
 
..Oops (my unintentional bad)

... PS Bob -

I hope you didn't take my last post the wrong way.
We appreciate the blood, sweat & tears you put into your bird.
I can't imagine how tough it must be when things don't quite turn out as we hoped.
I know I learned someting.... and that's a good thing..

So, on behalf of all of us - Thank you for all your extraordinary work and willingness to share!!

Lorin D
9A Wings
 
Bob

Well done! Lots of work...and good results, though as Kevin said, there are better test procedures, and you may find you have gained a little after all, after you fly it a while.

Kevin H -- I never see reference in the test proceedures to barometric pressure. But, using the NTPS spreadsheets, and carefully controlling density altitudes and power settings, I feel sure I see a variation due to barometric pressure. I often hear glider pilots complain that the local high is causing a general sinking of air, and there is no lift to be found. But, when under a low pressure area, there is lift to be had with ease. Similarly, I think I see that I get higher speeds under low pressure, and lower under high pressure. This occurs repeatedly when I am testing for repeatability with no changes to the airplane. Do you see such variations in your testing?

I went through the same program Bob did, at the same time, but used a different physical implementation. Unfortunatly, I am unable to repeat my test program at this time of year -- trying to fly at 9,000' d.a. puts me too close to the ground with summer temps. So, I don't know the result of my work, but later when things cool I will let everyone know.

See you at OSH! And, Bob, see YOU at Dayton!

John
Race 18
 
Nuisance said:
Kevin H -- I never see reference in the test proceedures to barometric pressure. But, using the NTPS spreadsheets, and carefully controlling density altitudes and power settings, I feel sure I see a variation due to barometric pressure.
Click to expand...
The NTPS spreadsheet is designed to answer the question "how accurate is my pitot-static system?". In other words, it is designed to determine the relationship between IAS and CAS. One of the steps in the method is to determine the TAS at each test point, using GPS data. We are hijacking that part of the spreadsheet to determine TAS at full throttle.

Whether the air is rising or descending slightly isn't a problem if your intent is to access pitot-static system accuracy. You adjust the power as required to get the same IAS on each leg. But, if you want to determine max speed, then you need air that is neither rising nor descending.
I often hear glider pilots complain that the local high is causing a general sinking of air, and there is no lift to be found. But, when under a low pressure area, there is lift to be had with ease. Similarly, I think I see that I get higher speeds under low pressure, and lower under high pressure. This occurs repeatedly when I am testing for repeatability with no changes to the airplane.
Click to expand...
Interesting observation. How much of a change in barometric pressure does it take to see a difference in your data? How much speed change do you think you see for a given amount of pressure change? How does the OAT compare at these different conditions?

It is hard to know the cause of the effect you are seeing. It could be due to rising or descending air. Or, if you are using a technique of doing all the tests at a particular density altitude, it could be due to the fact that engine power does not vary exactly with density altitude. I.e. if you have two tests at different temperatures, but you vary the altitude to get the same density altitude, the engine power at the two conditions is not quite the same.

For example, let's assume that you have established 9,000 ft as your reference density altitude. Let's assume that your air box performance is such that at full throttle the MP is 0.5" lower than the ambient pressure. Let's look at three different conditions that give a density altitude of 9,000 ft (power data for an IO-360A):

Code: 
pressure     temp   rpm    MP      power
  alt
 (ft)      (deg C)       (in HG)   (hp)
 8,000       7.7    2700   21.7    153.5
 9,000      -2.8    2700   20.9    150.1
10,000     -13.1    2700   20.1    146.9
If you calculate the power at each test point, then you can make an analytical correction for the difference in power at the test condition vs the power at 9,000 ft on a standard day. The speed should vary approximately as the cube root of the power, for small changes in power or CAS. For example, if the speed at the 10,000 ft test was 180 KTAS, you would correct it by the cube root of (150.1/146.9) to get 181.3 kt.
Do you see such variations in your testing?
Click to expand...
To be honest, I haven't done enough level flight performance testing in any one aircraft to have had the chance to see this effect. I'm looking forward to when my aircraft is flying and I can be generating my own data.
 
Kevin Horton said:
<clip>if you want to determine max speed, then you need air that is neither rising nor descending.Interesting observation. How much of a change in barometric pressure does it take to see a difference in your data? How much speed change do you think you see for a given amount of pressure change? How does the OAT compare at these different conditions?

It is hard to know the cause of the effect you are seeing. It could be due to rising or descending air. Or, if you are using a technique of doing all the tests at a particular density altitude, it could be due to the fact that engine power does not vary exactly with density altitude. I.e. if you have two tests at different temperatures, but you vary the altitude to get the same density altitude, the engine power at the two conditions is not quite the same.

For example, let's assume that you have established 9,000 ft as your reference density altitude. Let's assume that your air box performance is such that at full throttle the MP is 0.5" lower than the ambient pressure. Let's look at three different conditions that give a density altitude of 9,000 ft (power data for an IO-360A):

Code: 
pressure     temp   rpm    MP      power
  alt
 (ft)      (deg C)       (in HG)   (hp)
 8,000       7.7    2700   21.7    153.5
 9,000      -2.8    2700   20.9    150.1
10,000     -13.1    2700   20.1    146.9
If you calculate the power at each test point, then you can make an analytical correction for the difference in power at the test condition vs the power at 9,000 ft on a standard day. The speed should vary approximately as the cube root of the power, for small changes in power or CAS. For example, if the speed at the 10,000 ft test was 180 KTAS, you would correct it by the cube root of (150.1/146.9) to get 181.3 kt.To be honest, I haven't done enough level flight performance testing in any one aircraft to have had the chance to see this effect.
Click to expand...

I generally seem to run variations of +/- 2 kts or so from the mean, on different days and even times of year. It looks like I have several causes of these errors. I think the high/low pressure is a factor. So, apparently is power output variation with temperature (Lycoming says approximately 1% power with a 10 dF change), something I hadn't thought of. You are saying above that generally I will get faster speeds in the summer than the winter, flying at the same density altitude.

I spent a while looking through my old data, and though it could be, and though barametric pressure may contribute, in general I don't have enough data in any one configuration to say for sure. Unfortunately, it is not practical to test each configuration on the same day and same temps when there is a month's work involved. Maybe I should just declare victory :D

So, Bob, you shouldn't write off all your hard work yet -- wait for a cold front and try again!

Kevin Horton said:
I'm looking forward to when my aircraft is flying and I can be generating my own data.
Click to expand...

So are we all!

Thanks, John
 
DAYTON IT IS!!!!!

Man I am getting up for this! I also got a wonderful e-mail from Chris Zavatson that was very enlightening. Tell you about it after we recover at Fond Du Lac (ha ha). Actually he said that my mod probably increased the mass flow for more effective cooling but it also slows the plane down. He said if the CHTs are properly controlled I should be able to reduce the exit area and get the speed increase I'm looking for. At this rate I may bring tools to Dayton to finish the work on Saturday before the race on Sunday (joking again - I think!). Good luck with yours.

Bob Axsom
 
I had the same thought Chris did after I wrote yesterday. When I did my inside-the-cowl mods, my temps went down a little. I have some rings I can put in my intakes to choke them down a little. Maybe you can make more of the air go around, and less through your engine compartment.

Dave Anders says that your output area should be 78% of your intake area. Did you notice that Dave will be racing with us? He has done a lot of work in this area, and is always generous with his ideas and theorys. That is extra good since he beat me by over 20 kts when we raced the Sun 100 together a few years ago.

Otherwise...YOU NEED A PLENUM!! (you said in for a penny....right?)

John
Race 18
 
Yes I Noticed

I saw that Dave Anders had entered. I remember the article in Sport Aviation about his RV-4 before he started racing at Reno - the only builder I ever heard of that temperature controlled his rivets until installation. The field is very impressive I never expected to see 5 SX-300s in any one place at the same time. I just ordered a C4 CHT system from Aircraft Spruce with overnight delivery. $543 plus $26.57 for overnight. I'm thinking of a tuning "plug" on the center support of the cowl outlet (maybe that is an "A" model feature) to adjust the outlet size based on CHT. I'm thinking that I will start with one that results in a 78% of the air inlet area size at the outlet. That will go in after the C4 system is installed. I may put a cot in the hangar next week.

Bob Axsom
 
You must log in or register to reply here.