What's new
Van's Air Force

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

MT 3 blade vs Hartz. 2 blade performance

Kahuna

Moderatoring
I finally got around to publishing my data on my head to head flight test of the MT 3 balde vs the Hartzell paddle blade 2 blade.
A lot of work went into the flight testing data captures and the resulting data and graphs. The nut of it is this. The 3 blade is smoother, but much slower. Enough so that I pulled it off. AND I was able to ge the 2 blade smoothness in line with the 3 balde. Thought you rockets guys might like this info.
Performance comparisons here
How I solved the 2 blade smoothness by clocking the prop
newprop1_small.jpg

Best,
 
Last edited:
3 blade clocking??

So, I guess someone has to ask this, did you "clock" the MT to get the smoothness, or is it just a "bolt it on and go" fact of life that the MT will be naturaly smoother??

Thanks,
 
As a rocket guy it is obvious from Mikes data that if both planes had the same prop then the F1 would be 10 knots faster :)
 
Prop vibes and performance

Kahuna: This is great data and obviously took a lot of effort. Thanks for that and for making it available to the rest of us. Bill
 
Kahuna said:
I finally got around to publishing my data on my head to head flight test of the MT 3 balde vs the Hartzell paddle blade 2 blade.
A lot of work went into the flight testing data captures and the resulting data and graphs. The nut of it is this. The 3 blade is smoother, but much slower. Enough so that I pulled it off. AND I was able to ge the 2 blade smoothness in line with the 3 balde. Thought you rockets guys might like this info.
Performance comparisons here
Mike - that is very interesting data, and quite depressing for anyone who wanted to run that particular MT prop on a Rocket or Super RV. Thanks for sharing it with us. I appreciate the effort you have made to measure the performance of your aircraft.

Prop efficiencies vary with power and TAS (among other variables), so it isn't clear how applicable these results would be to MTs and Hartzell props that would be run on RVs, at power outputs produced by engines typically seen on RVs. But, it certainly suggests that Hartzells are probably a better bet than MTs, if level flight performance is considered important.

A few comments/questions:

"Ind TAS" - Your graphs are labelled "Ind TAS". What is that? I assume that it is a TAS indication on an EFIS, which while labelled TAS, would not be corrected for instrument error, static source position error and OAT error. This would certainly be a useful indication of performance changes, so I understand why you would record it and report it. It is much more practical to do a single run at each condition, and record the TAS indication on an EFIS system, than it is to do four legs at each condition. Only having one test point at each condition leads to a bit of error, as you can see from the waviness in the graphs. But the overall trend is quite clear, and that is what you are interested in here, rather than the exact performance at any one condition.

TAS vs average ground speed - Looking at the spreadsheets, it seems like you did a cross check by taking GPS ground speeds on the four cardinal headings, averaged the results and compared against the "Ind TAS". A lot of people like to average four GPS ground speed that way, and it certainly is simple, but there are a couple of errors that can bite you. First, the average of those four ground speeds will only equal the TAS is the wind is zero. The error is small if the wind is light, but the error can be several knots if the wind is strong.

I played around with a spreadsheet to look at the effect of wind on average ground speed on the four cardinal headings. The error introduced varies a bit with the wind direction, but you can make a rough correction for the effect of wind by taking the standard deviation of those four ground speeds, dividing it by the average ground speed. Then, subtract the following values from the average ground speed for a rough wind error correction:

std dev/.....error
ave GS
5%..........0.1%
10%..........0.4%
15%..........0.9%
20%..........1.6%
25%..........2.6%

For example, if the std deviation of the ground speeds is 15% of the average ground speed, the actual TAS would be about 0.9% less than the average of the four ground speeds.

Heading error - Also, small errors in the actual heading can have a significant effect. If your compass is out by a few degrees, or you don't accurately fly the cardinal headings, the average ground speed could vary by several knots. For example, for the condition where the std deviation of the ground speeds is 15% of the average ground speed, a 5 degree error on two of the legs would change the average ground speed by over 0.5%

NTPS Spreadsheet - If you are going to go to the trouble to fly four legs and record data, I strongly recommend you also record GPS track. Then you can use the NTPS spreadsheet to calculate TAS, and not be affected by winds or heading errors. Use the "Four legs" tab of that spreadsheet.
 
I think the Hartzell is so superior in this test that any small potential errors in TAS are irrelevant. We are not talking a few knots here, we are talking lots of knots. In any case by taking so many readings, errors are likely to mainly cancel. Occam's razor Effect here.

I'll say again in flight testing, I use IAS at the same altitude and close OAT for comparisons. Way quicker and it avoids all the posts on how you came up with TAS figures from GPS or from glass, whiz wheel etc.

I personally prefer the side by side flight comparison method for true performance testing of modifications and welcome Kahuna's inclusion of that data here as well. This negates other variable such as humidity and hp output- something which can have significant effect and which is rarely measured.

Thanks for posting your real world results and impressions. Also interesting was your clocking trials. I've always wondered if I could get my package even smoother or if I'd be wasting my time.
 
Last edited:
rv6ejguy said:
I think the Hartzell is so superior in this test that any small potential errors in TAS are irrelevant. We are not talking a few knots here, we are talking lots of knots. In any case by taking so many readings, errors are likely to mainly cancel. Occam's razor Effect here.

I agree 100%. I was not suggesting that the potential small errors in the data invalidated the conclusion. Those potential errors are probably on the order of one or two knots, which is much, much smaller than the difference that was observed between the two props.

rv6ejguy said:
I personally prefer the side by side flight comparison method for true performance testing of modifications and welcome Kahuna's inclusion of that data here as well. This negates other variable such as humidity and hp output- something which can have significant effect and which is rarely measured.
This is a good approach, if you have another closely comparable aircraft to test against, and no changes are made to the other aircraft. But, if one aircraft had a Lycoming, and the other had an alternative engine, it is possible that the two engines may have different variations of power with ambient conditions, which could possibly pollute the results.
 
I'd bet if you were comparing the MT against the Hartzell with 8475D-4 paddle blades, or the new 8068 blended airfoil blades which are more efficient at Rocket speeds, there would be an even greater speed difference.

 
This whole prop-clocking thing is amazing. I wonder how much of it is airframe-specific. I also wonder how Lycoming came up with the original clocking in the first place.

Mooney did the same thing with the M20J with the IO-360-A3B6D where they clocked the bushings 60 degrees to increase smoothness. There's even a Lycoming service bulletin on how to do it which changed the model designation of the engine from -A1B6D to -A3B6D. Has anyone here tried it with their (4-cylinder) RV and noticed a positive difference?

Heinrich Gerhardt
 
Kevin Horton said:
This is a good approach, if you have another closely comparable aircraft to test against, and no changes are made to the other aircraft. But, if one aircraft had a Lycoming, and the other had an alternative engine, it is possible that the two engines may have different variations of power with ambient conditions, which could possibly pollute the results.

Yes and in the case of propeller testing, if we can get relative differences side by side in the same piece of air then trade props and re-test. We can use simple GPS readouts to confirm differences accurately since both aircraft are subject to the same variables in the atmosphere.
 
rocketbob said:
I'd bet if you were comparing the MT against the Hartzell with 8475D-4 paddle blades, or the new 8068 blended airfoil blades which are more efficient at Rocket speeds, there would be an even greater speed difference.


I don't see much difference in "Rocket speeds" here. I think there are more than a few RVxs capable of 175-180 knots TAS at 8000 with 4 bangers. Kahuna's Super 8 is quite impressive with the bubble canopy in my view.
 
Kevin Horton said:
A few comments/questions:

"Ind TAS" - Your graphs are labelled "Ind TAS". What is that? I assume that it is a TAS indication on an EFIS, which while labelled TAS, would not be corrected for instrument error, static source position error and OAT error. This would certainly be a useful indication of performance changes, so I understand why you would record it and report it. It is much more practical to do a single run at each condition, and record the TAS indication on an EFIS system, than it is to do four legs at each condition. Only having one test point at each condition leads to a bit of error, as you can see from the waviness in the graphs. But the overall trend is quite clear, and that is what you are interested in here, rather than the exact performance at any one condition.

Kevin,
THe 4 headings were gps ground tracks. I did average them but could easily plug into the formula link you had to get it more exact. Or it would be even better to know what the formula is and stick that into the sheet.
What is the formula? I fooled with it a while ago and my memory says the difference between using the formula and doing average gps track was insignificant. But I could be wrong.
Best,
 
Mike S said:
So, I guess someone has to ask this, did you "clock" the MT to get the smoothness, or is it just a "bolt it on and go" fact of life that the MT will be naturaly smoother??

Thanks,
It was smooth out of the box and bolted on.
Best,
 
Thanks

That is what I was hoping to hear-----------MT prop came with my project.

It sits, patiently waiting in its triangular box, stored under the fuse.
 
hgerhardt said:
This whole prop-clocking thing is amazing. I wonder how much of it is airframe-specific. I also wonder how Lycoming came up with the original clocking in the first place.

Mooney did the same thing with the M20J with the IO-360-A3B6D where they clocked the bushings 60 degrees to increase smoothness. There's even a Lycoming service bulletin on how to do it which changed the model designation of the engine from -A1B6D to -A3B6D. Has anyone here tried it with their (4-cylinder) RV and noticed a positive difference?

Heinrich Gerhardt

Heinrich. You are exactly correct. The original idea came from testing they did on the Mooneys. When I showed Hartz the vib data Les said. "Hey, Ive seen this before... On the Mooneys..Lets try clocking.." And on we went. He was also VERY interested in solving this problem as they have had complaints over the years about objectionable vibs on the rockets and Super 8's. I was the only guy with a test fixture (ie spetrum analyzer) and a willingness to run all those tests. It takes a lot of time and gass$$ to accomplish this. Not to mention pulling props on and off. I had fun doing it and I love to capture realistic data on stuff like this. To have an 85% improvement in vibration from clocking the prop was unbelievable. As a result of this flight testing data, Hartz is supposed to be putting a hub together for a 9-3 clock but I dont know where that sits today.

Doing this on top of the MT vs Hartz performance tests meant I was doing a lot of flight tests. Like others have mentioned, these tests are only useful when done on the same airframe in order to do apples to apples comparisons.

I tried very hard to capture a lot of data to be analyzed. Tried to be methodical and accurate. It was a lot of fun to see this come together.

Best,
 
I'm not sure that many people realize how much time and fuel it takes to gather data like this. I appreciate it and would encourage others to do a bit of this type of work and share it. Thanks again! :)
 
I was thinking of doing this with my 7a with it's WW rv200 2 blade and then going down to dave's hangar and borrowing his new in the box Hartzell blended prop.

maybe run one at 8:00am one day and then the other at 8am the next.
 
Mike's Hartzell propeller is a "special" 76" dia. propeller. I believe the minimum Hartzell diameter is 77" for his propeller.

Mike spent a great deal of time collecting his data and developing his charts. He has every right to be proud of his efforts.

The validity of any data is only as good as the accuracy in the data gathering method.

The last time I talked with Mike, density altitude and power settings were not used in obtaining his data.

As a point of reference in using fuel flow instead of power setting, I am providing the following data taken from a recent flight in my 180 hp RV-6A at 8,000' density altitdue, 2300 RPM and 23.9" MAP.
Best Economy (peak EGT) Fuel Flow = 10 gph.
Best Longevity (50 degrees F ROP) Fuel Flow = 11 gph with 2 knots increase.
Best Power (100 degrees F ROP) Fuel Flow = 12 gph with 1 knot decrease.
Both the EGT and Fuel Flow gages are my least accurate instruments. My fuel flow gage has a 1 gph fluctuation, and the EGT has a serious lag in the readings. (Finding peak EGT on a carburated engine is probably within 20 degrees F, and other two EGT settings kept floating by 10 degrees.)

Jim Ayers
 
I've hesitated to jump in here since anything said will be construed as either chest thumping or an attack. I will just make a couple of brief points:

1. Great job Mike and thank you for taking the time and money to conduct this comparison.

2. It's a known fact that a two-bladed Hartzell is faster than a three-bladed MT.

3. If Mikes airspeeds are accurate in his chart, my Rocket is at least 10 knots faster. (I fly with a three-bladed MT).

I think the data shows that for Mike's airframe, the Hartzell offers him a significant improvement. Be careful about extrapolating that data to other airframes. Data collected by the F1 crowd suggests that the speed difference betweem the two props still favors the Hartzell, but the margin is much less that what Mike found.
 
f1rocket said:
I've hesitated to jump in here since anything said will be construed as either chest thumping or an attack. I will just make a couple of brief points:

1. Great job Mike and thank you for taking the time and money to conduct this comparison.

2. It's a known fact that a two-bladed Hartzell is faster than a three-bladed MT.

3. If Mikes airspeeds are accurate in his chart, my Rocket is at least 10 knots faster. (I fly with a three-bladed MT).

I think the data shows that for Mike's airframe, the Hartzell offers him a significant improvement. Be careful about extrapolating that data to other airframes. Data collected by the F1 crowd suggests that the speed difference betweem the two props still favors the Hartzell, but the margin is much less that what Mike found.

Interesting. Are you using the same prop as Mike?

Certainly prop performance can vary on different airframes. Mikes data, even if off a bit, shows that this MT on his airframe is quite inferior. I'd agree with Jim's post that other MT props may not perform the same way and previous reference has been made to several pilots running 2 blade MT props showing their heels to Hartzell equipped Rockets.

Keep testing those combos guys.
 
I believe RVJim has good data on the Hartzell 2 blade vs the MT 3 blade (tested on a single airframe), tho I'm not sure the 3 blade used in the test was the std 79" dia unit -- I think it might have been a special racing 76" version -- Jim, can you confirm?

In any case, we have not found significant, repeatable cruise speed differences on the Rocket airframe in testing 3 different props: Hartzell blended foil 3 blade; Hartzell Ol' standby paddle blade (2 blade version) and the Std F1 MT. Mike's results (faster with less disc area) may be due to the differing flat plate area of his airframe compared to the Rocket's.

Carry on!
Mark
 
Kahuna said:
Kevin,
THe 4 headings were gps ground tracks. I did average them but could easily plug into the formula link you had to get it more exact. Or it would be even better to know what the formula is and stick that into the sheet.
What is the formula? I fooled with it a while ago and my memory says the difference between using the formula and doing average gps track was insignificant. But I could be wrong.
Mike,
The formula is as long as my arm, so I won't try to post it here. You can find it in the NTPS spreadsheet that I linked to earlier. The basic formula needs GPS ground speed and track from three legs. It was derived by Australian RV-6 builder Doug Gray. I told the NTPS about it years ago when I was doing a refresher course with them - they were using a less accurate means to get TAS from GPS data at the time. They came up with the very clever enhancement of gathering data on four legs, even though you only need three. Their spreadsheet does four TAS calculations, using a different mix of three legs each time. If all four results are similar, that indicates that you have nice clean, consistent data. If the data from one or more of the legs is questionable, then the calculated TASs will differ from each other - you don't know which leg is bad, so you throw out all the data from those four legs.

I took a quick look at using four cardinal tracks, vs four cardinal headings. If you average the ground speeds from four legs on cardinal tracks, the result is a bit less than the TAS, unlike the result if you use cardinal headings, where the average is greater than the TAS. Given the same TAS and wind, the error is greater if you use track than if you use heading. For example, if the wind is 20% of the TAS, the average ground speed is about 0.4% higher than the TAS if you use four cardinal headings. The average ground speed is about 1% less than the TAS if you use four cardinal tracks.
 
F1Boss said:
I believe RVJim has good data on the Hartzell 2 blade vs the MT 3 blade (tested on a single airframe), tho I'm not sure the 3 blade used in the test was the std 79" dia unit -- I think it might have been a special racing 76" version -- Jim, can you confirm?

In any case, we have not found significant, repeatable cruise speed differences on the Rocket airframe in testing 3 different props: Hartzell blended foil 3 blade; Hartzell Ol' standby paddle blade (2 blade version) and the Std F1 MT. Mike's results (faster with less disc area) may be due to the differing flat plate area of his airframe compared to the Rocket's.

Carry on!
Mark

Mike Stewart's flight test data was with a 76" dia. Hartzell. I don't know if the blade thickness met the Hartzell specified minimum thickness.

I made several flights in Ted Rutherford's HR2 with the Hartzell "D" 2 blade propeller (80" dia.), the standard Rocket 3 blade MT Propeller (198cm or 78" diameter) and the Hartzell "J" 2 blade propeller (80" dia.). This was with a stock 260 hp engine.

This was in preparation for racing at Reno, so the flights were at what we guessed would be the Reno density altitude. Ted Rutherford used the "J" blade Hartzell at Reno.

As a summary:
The Hartzell "D" blade and the Rocket 3 blade MT Propeller both had a peak airspeed at 2600 RPM and full throttle. (At 2700 RPM, the airspeed was slightly lower than at 2600 RPM.)
(Also, the 3 blade MT Propeller was about 2 mph faster than the "D" blade Hartzell.)

The Hartzell "J" blade propeller matched the performance of the 3 blade MT Propeller up to 2600 RPM. However, with the Hartzell "J" blade propeller, the airspeed kept increasing with the increase in RPM above 2600 RPM.

Another Reno Rocket pilot told me he had a peak airspeed at 2600 RPM with the "J" blade Hartzell on his 330 hp engine.

As a side note for the 7666 blades on a Lycoming 360 180 hp engine.
The RPM for peak airspeed at full throttle is given below for the different density altitudes.
2,500' - 2700 RPM
7,500' - 2550 RPM
12,500' - 2400 RPM

Perhaps something else to think about. The peak torque on the Lycoming engine is at 2200 RPM.

Jim Ayers
 
RVjim said:
As a side note for the 7666 blades on a Lycoming 360 180 hp engine.
The RPM for peak airspeed at full throttle is given below for the different density altitudes.
2,500' - 2700 RPM
7,500' - 2550 RPM
12,500' - 2400 RPM
Electronic ignition or mags? EI would make sense...the EI is likely advancing more with lower RPM and lower MP.
 
I'll count the MT and Hartzell users at Reno next week on Rockets and talk to some of the guys. Me thinks most will be using the fastest prop or at least one that is the same in speed.
 
MT's better for racing because.....

rv6ejguy said:
I'll count the MT and Hartzell users at Reno next week on Rockets and talk to some of the guys. Me thinks most will be using the fastest prop or at least one that is the same in speed.
f1rocket said:
This is probably a two-way street.
That is true and I can see Reno guys using the MT for a very important reason, metal props are very picky about the engine they bolt to.

Any mods to an engine or even mods to the airframe with out a vibration survey is a big unknown. A prop on a racing plane/engine, using an untested Hartzell (with that combo) may be risky, from a harmonics stand point. Clearly the Glass/Wood blades are much more forgiving in regards to vibration and fatigue. Metal props have to be tested.

Thankfully Hartzell has done a lot of testing to support the homebuilt market with some mild mods, eg electronic ignition. However get into HC pistons, elect ignition advance at high power and radical turbos, forget it, all bets off. Each new mod needs to be tested to cover your bases with metal props.

So MT is best in my opinion for radical engines and even ultimate aerobatics. It may lose a some MPH's but it also is more forgiving vibration wise.
 
Last edited:
Sorry I'm jumping in late on this forum thread. I made my origianl reply on the Matronics list, since the posting was on both, and didn't realize this is where the discussion ended up.

Although it seems you folks are making a lot of conclusions based on comparison "data," please look carefully at it. I responded to Mike Stewart off-line, but his reply leads me to believe that the data does not exist to do an accurate comparison.

I have read the charts, as Mike suggested, and the data connot be compared - sorry. It's not done at repeatable or comparable conditions. You need to do the tests at constant density altitude and at an engine power setting that can be repeated, such as WOT and leaned EXACTLY the same each time. The data presented is therefore unusable. At the test pilot school where I teach we would throw it out. What has been done is very subjective and cannot be relied upon for the results that are trying to be presented. Therefore the conclusions are unsubstantiated.

I'm not advocating one prop over another, but am constantly amazed at how gullible people are on these lists. If you want an accurate comparison then you have to use the proper flight test techniques to compare apples to apples. Right now you're all looking at apples and oranges.
Mark S.
 
gmcjetpilot said:
That is true and I can see Reno guys using the MT for a very important reason, metal props are very picky about the engine they bolt to.

Any mods to an engine or even mods to the airframe with out a vibration survey is a big unknown. A prop on a racing plane/engine, using an untested Hartzell (with that combo) may be risky, from a harmonics stand point. Clearly the Glass/Wood blades are much more forgiving in regards to vibration and fatigue. Metal props have to be tested.

Thankfully Hartzell has done a lot of testing to support the homebuilt market with some mild mods, eg electronic ignition. However get into HC pistons, elect ignition advance at high power and radical turbos, forget it, all bets off. Each new mod needs to be tested to cover your bases with metal props.

So MT is best in my opinion for radical engines and even ultimate aerobatics. It may lose a some MPH's but it also is more forgiving vibration wise.

Actually you will see far more metal props at Reno than composites. Speed is king, vibration is secondary as there is already plenty of that usually. There are some major prop checks mandated by the sanctioning body however now after some serious failures. With tweaked governors and sometimes 3200 rpm in the Sport Class, metal props can be hand grenades.

Hartzell has been at the front on most Sport Class entries at Reno for years now. Rockets are not competitive in this class being about 100mph off a winning pace. I just wanted to count the two different props there on these airframes.
 
The Lycoming peak torque number at 2200 RPM was provided to me by the manager of Lycoming's Thunderbolt engines. I choose to believe him.

I have been obtaining flight performance data in my RV's for the last 19 years.

I used a single point flight test for my RV-3. 7,500' density altitude, full throttle and 2700 RPM. I was trying to document changes to the airframe debris. I also tried 5 different fixed pitch propellers, two Lycoming O-290 engines, a LOM 243 Cu. In. engine, an electric variable pitch propeller, and an electric CS 3 blade MT Propeller.

The flight test data on my www.lessdrag.com website was taken in a friends RV-6A with a Lycoming O-360-A1A engine with one Mag and one LSE electronic ignition. We made three different flights with the Hartzell 2 blade propeller. Two flights at the beginning of the flight test effort and one flight at the end of the flight test effort. These was done to verify the repeatability of the test method. There are a couple notes on the website identifying were I felt there were descrepancies in the flight test data, and why they were there.

George.
There is absolutely no way to make a valid comparison of my flight test data with Van's Aircraft flight test data.
I understand that Van's Aircraft uses GPS ground speeds and averages the number over a number of flights with different pilots.
We used the uncalibrated airspeed indicator in my friends RV-6A for my data.

Also, only the Hartzell 2 blade propellers were of the same type tested.
Van's Aircraft tested the 3 blade MT Propeller MTV-12-B/183-59.
I tested the 3 blade MT Propeller MTV-12-B/183-59d.
I am testing the current production 3 blade MT Propeller MTV-12-B/183-59b.
(The blade design number is the last dash number in the MT Propeller part number.)
The -59, -59d and -59b identify the blades designs as being unique and different. The -59b blade design does not have mid range RPM restrictions, and has a better propeller efficiency than the -59 blade design.

Flight testing is a system test. And it is a very complex system.

I bought a flying RV-6A with a Lycoming O-360-A1A engine with dual Mag's and a Hartzell 2 blade CS propeller with 7666-4 blades.

I have recently been refining a repeatable flight test procedure for cruise performance testing. Eventually I will add test proceedure information into my website.

I have been doing the flight testing at a specific DENSITY altitude, using full throttle and leaning very carefully to the same EGT point of 50 degrees ROP (Best Longevity). I vary the RPM from 2000 RPM to 2700 RPM in 100 RPM increments. So I get eight data points at each altitude. (Power settings get more complicated below 8,000')
It takes me about an hour of flying time at each altitude for a single set of data points. And I have to do it a second time to establish the validity and repeatability of the test method and test data.

Since I am running the engine at full throttle, there is no need to have a calibrated manifold pressure gage. I calibrated the Tach using an optical tachometer. My Tach reads 10 RPM high up to 2600 RPM, and 20 RPM high at 2700 RPM.
The testing has to be done in smooth air. The test area has to be free of standing waves. Every change in power setting requires a change in the mixture setting. And there is a ten foot altitude window that has to be held for about 3 to 5 minutes to stabilize the airspeed.

The actual proceedure to make this happen repeatedly is even more involved.

So, if someone thinks I am creating Parker Pen flight test data to sell propellers, then they don't know me and they don't know what I am about.

Jim Ayers
Less Drag Products, Inc.
 
Hummmm

Thunder engines? What is that. No one is calling you out Jim. I called Lycoming and they don't have torque data (but did not talk to the manager of thunder engines). At what altitude? Test stand? As you know torque is work and HP is the rate at which you do that work (torque). It's important to have torque to accelerate (in cars) but HP is needed to go fast (top speed). Unless you have a prop MADE to work at 2,200 rpm (big paddles you see on turboprops) than its a moot point. HP = speed and the props are made to use that HP not low RPM torque, even MT's. I can tell MT's are not low RPM props by their shape and design. Even if they had more efficiency at low RPM it would trade high RPM (and speed) performance. That old no free lunch deal.

"There is absolutely no way to make a valid comparison of my flight test data with Van's Aircraft flight test data." Not true Jim relative difference is relative difference on a given same plane. It also matches other test by other people. If 3 out of 4 people constantly find the MT is slower, noticeably and your data says its not than who to believe. It's not about Hartzell v. MT, just good data. That is all we want. People decide to buy MT props on other factors than top speed. However if a builder buys a MT to go faster or thinks they will not lose speed, they will be disappointed. Most are willing to trade off the speed for smoothness and less weight. Your data shows the MT can be as much as 6 mph faster which I have never seen in any other test. Altitude is a factor in prop efficiency, but 0.5% not 16% or a factor of 32.

This thread by the way started with a test that showed the MT was 13 Kts slower than a Hartzell! :eek: Granted different props, engine and plane but its consistently the same over and over, just not that much. I hear 5-9 mph all the time.

The basic characteristics of a PROP at sea level are the same at 12,500 feet and the altitude and RPM factors figure in but they are small factors not major factors. Sure you can fall off the "J" curve or efficiency curve of a prop, but we operate in a fairly small band that is no where near the far back side of the efficiency curve. The prop designer should be shot if they made a prop that lost 15.5 mph by just changing the altitude 5,000 feet, or may be given a medal, since its impossible to even do that intentionally.


Below is an except from your data. One big example of a problem is the Hartzell 2700 rpm curve, it's not right. We can agree to disagree, the 3 blade MT Propeller MTV-12-B/183-59d has been tested by others and the data does not match your data. I choose to believe even the "d" model is at least 5 mph slower or more at almost ANY altitude or RPM.

Jim I believe you believe your data, but as I pointed out there is no way the curves are valid. I have actual Hartzell data, actual Coefficients of thrust, "J" factor for different altitudes and rpms and your data is not correct.

Lycoming makes less and less HP as you climb (thus you go slower). Its linear or straight line. Even your own data contradicts it self for different altitudes at a given RPM. Bottom line no trends and large deviations (statistically), which don't match known physics of props and engines means the data is not valid. I am sorry it makes no sense.

A Hartzell that is more efficient at 2,500 feet (any RPM) is not going to become super inefficient at 12,500 feet. As well a MT (d model or not) is not going to become hyper efficient at altitude. Its physics. There is an altitude factor in the prop equations, but your data is many orders of magnitude beyond impossible. Since the engine and plane is the same, that is eliminated. The only answer is your test method, data collection and reduction has errors. I worked in flight test, I know about it, errors happen.

Yes there may be greater or less performance gain or loss for different props at different altitudes and RPM's, but Jim, we are talking a fraction of one percent, NOT 16%. It's impossible unless the blades actually transformed, changed and morphed in a new shape (cross section / profile) and total area.

Wood props are thicker = less efficient
Three blades = less efficient than two, for lower HP/slower top speeds

(eg, less than 250-300 hp and 250-300 mph)

A 180 hp, 190 mph RV does not NEED three blades. However there are advantages to three blades but speed is not one of them. Its just physics, math and engineering they taught me at school.

Attention all Hartzell drivers request, go up and fly at 7,500 and 12,500 feet. Measure true airspeed at 2,700 rpm. Let me know if you lose 15.5 mph!!! by climbing, 7,500 verses 12,500 feet @ 2,700 rpm. Never happen. Jim I think your data is wrong, sorry, it does not match other's test and experience, including the 'd' model which is not significantly different. And if 2,400-2,500 rpm is so good (better than 2,700 rpm) why is 2,300 rpm worse? It's closer to peak torque right? It just does not make sense or match known data I have from Hartzell.

Even you MT data is all over, look at the 2,300 and 2,400 rpm data. It's all over the place with all due respect Sir.

Cheers, Jim I am NOT against you and appreciate your efforts and wish you great luck and safe flight with your future flight test. People that want MT's on the West Coast, I give'em your name, because your customer service is good. You claim the the MT is not much slower and actually is faster at altitude, that's what your data says.

All the best George


PS: OK easy fly-off, two RV's one with Hartzell, one with the MTV-12-B/183-59d. Fly side by side at the start. Both use the same distance way point and GPS to track. After so many minutes, note lat/long or distance left to the way point. Use the same type of handheld GPS even. Repeat fly off at different RPM's and altitudes. It will not give an absolute number but it will show if there is any dramatic gain or loss of speed for either prop, based on changes in the RPM/altitude factor. Per your data Jim, after 5 min at 12,500 ft, 2,700 rpm the MT would be 1/2 mile ahead or gained by that much over lower altitudes. Of course its relative to the speeds measure at lower altitudes. Side by side, less chance for error or debate, justification and dispute of results.

There's no debate if one pulls away from the other. Very simple and fool-proof, WOT of course at 7,500 feet & above. Not about a put-down of any brand prop or making a sale, just good data. ANY TAKERS?


jimspropdata2qeadn7qs8.jpg
 
Last edited:
gmcjetpilot said:
I called Lycoming and they don't have torque data
They do publish power charts, and torque is simply power divided by rpm (with an appropriate conversion factor). My interpretation of the O-360A power chart has the torque essentially constant at 391 ft-lb from 2000 to 2200 rpm, then decreasing to 362 ft-lb at 2700 rpm.

gmcjetpilot said:
Jim I believe you believe your data, but as I pointed out there is no way the curves are valid. I have actual Hartzell data, actual Coefficients of thrust, "J" factor for different altitudes and rpms and your data is not correct.
We need to be a bit careful how we interpret prop efficiency data. As I understand it from discussions with Les Doud at Hartzell, typical prop efficiency charts are analytical predictions. They are not generated from actual flight test data. And, the flow field through the prop will be affected by the shape and size of the engine cowling. So, I wouldn't discount flight test data just because it disagrees with a prop efficiency map.
 
Great points and agree 100%

Kevin Horton said:
I wouldn't discount flight test data just because it disagrees with a prop efficiency map.
Excellent points and agree 100%. Thanks for the torque data. I called Thunderbolt engines (Lycs custom engine product). They are at Reno and than Red Bull San Diego. The answer was there are no documents with torque, but they where going to get me something when I get back. I suspect they will do what you did, derive it.

Reading between the lines, Jim seems to be saying running at 2,200 RPM is more efficient. He very well be right, but more efficient and fast are two different things. Also more efficient compared to what? I don't think the MT is inherently more efficient at 2,200 rpm than say a metal Hartzell at 2,200 rpm. In general the slower you turn the more prop, the greater the efficiency, but than we get into less HP. I have some data below.



Also, I agree 100% about Hartzell's prop data and actual flight test being different. I was told the same thing, engine and airframe have an affect on the theoretical predictions, certainitly. That is why we do flight test. However I'm talking about large differences between predicted and reported prop efficiency. Since Jims data some what takes out the engine/airframe, it was the same for each prop test.

Going to higher altitudes you expect to go slower (with out a turbo). Climbing = less HP; Less HP = less speed; Forget that the airframe has less drag at altitude or the prop has different efficiency, disregard it. I estimate you lose 10 mph due to less HP alone. Including the reduced airframe drag with altitude, you would not lose that much TAS.

To lose 15.5 mph, assuming the extra 5.5 mph loss was from the prop, would mean a huge increase in prop inefficiency. Looking at the Hartzell data, at 2,700 rpm changing only (estimated) HP, airspeed & air density, reflecting the change in altitude, prop efficiency loss is less than 0.25%; overall prop efficiency is about 83.3%. That is 1/4th % change in efficiency.

To lose 5.5 mph (or more) due to the prop alone, means an 8% efficiency loss, or less than 75% overall. That's not going to happen, at least from the prop predictions from Hartzell. Considering airframe drag was disregarded that means the prop would have to go into the tank. A 2 by 4 would do better. I know for a fact Hartzell, in the real world, the C2YK/F7666 is a better prop than that; thus my contention the data is not correct.

That's why I say, an efficient prop at sea-level is still going to be efficient at altitude, in the range of RPM, airspeed and tip Mach# we run. If a prop is on fire at sea level, its not going to suck suddenly at 12,500 ft



Kevin I see no magic or gremlins in a Hartzell prop or any prop that makes it significantly better or worse at altitude. It's just an efficient prop and can't wrap my mind around the prop dying at 12,500 feet at any RPM.

RPM & prop diameter does have an affect on efficiency. Diameter being fixed we only control RPM (and pitch). From Jim's data you see the general correlation, as you climb speed drops and higher RPM = more speed. However there inlays my complaint, all the changes in slope (a line with lots of curves, hills and valleys) and cross-overs, when the data is plotted It should be a fairly straight line with may be a slight overall curve but not a sine wave.



RPM affect?
Lowering RPM from 2,700 RPM to 2,500 RPM, efficiency prediction says you gain 1/2%. That's about 3/4 HP more thrust say at 8,000ft. Efficiency is simply prop thrust (HP) divided by HP at crankshaft into prop. 20% loss is the generic number people, but it varies based on flight condition (air density, airspeed, rpm, HP at crank). The down side of lower RPM, is most engines make less HP. In the case in point about 6 HP less power to gain 1/2% prop efficiency. You expect speed to decrease with RPM. 2 mph less speed in exchange for better miles-per-gal is fine, but again efficiency and speed are different. If you could keep the HP up at lower speed that would nice.

If a prop is optimized for lower RPM, like at 2,200 rpm which was mentioned to take advantage of peak torque, you may gain more efficiency. However efficiency and speed are not synonyms. Going faster is often not efficient, but who wants to go slow.

As Ross has pointed out many times, gear reduction and turning the prop slower has efficiency advantage. Of course a gear reduction has some disadvantages. Regardless if you want to "gear down" and run slow prop RPM's, the prop has to be optimized for lower RPM for a given HP & airspeed. Hartzell's are not made to run real slow. Turbo-prop props do turn as slow as 1,900 rpm. Of course their fwd speed is higher, so slower RPM keeps tip speed's down. Over Mach 0.90 tip efficiency goes to heck.

Hartzell prop guys know how to balance efficiency and speed for a direct drive Lycoming for a plane with RV performance. The above was in reference to the older but still efficient HC-C2YK/F7666-4, not the newer Hartzell blended airfoil prop, which is optimized even more for the RV. It's 3 mph faster than the C2YK/F7666. That's pretty amazing. (Are you listening Hartzell where is my free prop, give me give me. :D .)
 
Last edited:
Why flight test data doesn't relate to propeller effeciency.

Propeller effeciency is propeller thrust (force) divided by propeller drag (force).

Being an airfoil, propeller drag has two major components, profile drag (basically, air density times velocity squared) and induced drag.

As with any airfoil, increasing the propeller RPM increases profile drag and decreases induced drag. Plotting out the combined drags will give us the typical drag curve with a minimum drag at one airspeed.
From the minimum drag point, if the propeller RPM is decreased, or increased, the drag will increase.

What I am suggesting is that the propeller effeciency AND the propeller DRAG need to be considered in a theoretical analysis of propeller performance. Of course, they will vary with air density.

George:
Jack Norris has published his book on propellers. There is a great deal of information and the 183 year history of propeller design in his book.

Jim Ayers
 
Yesterday I took my good friend John Marshall's Super-8 up for a spin. I believe this one was the first or one of the first of its kind flying. Workmanship is top-notch. You guys may have seen it or its twin at a flyin, it has the dragon painted on it. This one has a stock carb'ed 260 and the 8477 blades. Went up to 7K to do some GPS runs. WOT 2300, 23.5", 11.7GPH (could have leaned some more) I got these groundspeeds: S 192kts, E 193kts, N 165kts, W 182kts. The TAS/GPS spreadsheet with the first three numbers says I was cruising at 183kts. Pretty cool. 540's rule! Still ain't a Rocket (I'm biased, I'll admit) but I will say that the 540 really transforms the airplane for the better.
 
Thats a couple knots faster than mine at that FF based on my charts. Must be the fire breathing dragon pushing her along a little:)
However, if you were WOT and leaned 50ROP, then I would expect you to be able to push ~14gph, not 11.7. Can you acccount for the difference?
Here is my data below for 8k' 2350rpm, WOT, 50ROP. which is pretty close to what you ran. My ITAS is w/i 1 knot of measured card heading numbers. I would have to pull back to ~18" to get your FF.

FF 14.1 13.6 13 12.4 11.4 10.4 8.7 7.4
ITAS 196 192 188 182 177 164 148 128
IAS 170 167 164 159 154 163 129 112
MP 21.6 20.6 19.6 18.7 17.3 15.3 13 10.9
RPM 2350 2350 2350 2350 2350 2350 2350 2350
KTS/Gal 13.90 14.12 14.46 14.68 15.53 15.77 17.01 17.30

Good report Bob.
 
Last edited:
I leaned it until it got rough then backed it off, 11.7 was the number on the EIS. Another point of data I got at that altitude, WOT 2700 RPM full rich it was burning 21.6gph, which is a bit less than other Rockets I've flown which consume 24-25gph at full power. I did not note speeds at that power setting. I wouldn't think the FF k-factor is off but I suppose its possible.
 
I leaned it until it got rough then backed it off, 11.7 was the number on the EIS. Another point of data I got at that altitude, WOT 2700 RPM full rich it was burning 21.6gph, which is a bit less than other Rockets I've flown which consume 24-25gph at full power. I did not note speeds at that power setting. I wouldn't think the FF k-factor is off but I suppose its possible.
hmm....:confused:
Scratching my head. I do not know what my FF at full rich full RPM at altitude ~8k' is. Ill try that on Wednesday night and report back. 24gph seems like a high number at 8k'. Ill see what I get. I know my FF is accurate.
The only data point I have captured is this.
RPM to 2700, 50ROP, 15gph, MP 20.5, ITAS 200, IAS 171, 10k' WOT.
 
Actually I don't know if the full-power FF should be that high at altitude, but I do know that on takeoff its typical to see 24-25gph. Whatever the case is, I can't say the data I collected is 100% accurate. I just went up to 7K and noted the numbers. It is what it is.

As a side note Whirlwind has introduced a new RV-10 prop that they aren't advertising yet. Basically a 540 version of the 200RV. It would be interesting to get one of those and do some comparisons.
 
Actually I don't know if the full-power FF should be that high at altitude, but I do know that on takeoff its typical to see 24-25gph. Whatever the case is, I can't say the data I collected is 100% accurate. I just went up to 7K and noted the numbers. It is what it is.

Yes this makes more sense. I would expect to see ~25gph with 29" at 2700RPM full rich.
 
Just got back from Reno, hoping to count MT vs. Hartzell props on Rockets there. Unfortunately, not a single Rocket to be seen! I guess the Sport class speeds now have just made them stay home.

I did note that DGs Lancair now sports a custom MT 4 blade in place of the Hartzell 3 blade of previous years and one other Lancair also now has an MT prop. Looks like MT is getting into racing more seriously now which can only help improve their products. They are a major sponsor of DG now.
 
Well, the fastest of the bunch--sports-- at Reno was Nemesis NXT, and it had a 3 bladed aluminum prop------I couldn't tell the brand, entire prop looked like it had been sanded with a DA.

Second place also a NXT with a 3 blade prop.

These two were faster than about half of the unlimiteds, or so the announcer said.
 
Well, the fastest of the bunch--sports-- at Reno was Nemesis NXT, and it had a 3 bladed aluminum prop------I couldn't tell the brand, entire prop looked like it had been sanded with a DA.

Second place also a NXT with a 3 blade prop.

These two were faster than about half of the unlimiteds, or so the announcer said.

Yep, Sharp did a 390+ mph lap on Saturday. This is the most awesome sounding Lycoming I've ever heard. What do you expect for $93K. Lovely to see and hear anyway.
 
One other data point as I just got back from Reno and will post more tomorrow. I crewed for Lynn Farnsworth who as some know, runs a super Legacy similar to DG's. We flew our best time this year at 350MPH which only finished us 4th in the SS Gold, but that's an entirely different story.

Anyway, The 2 NXT's fly a Hartzell racing prop. As well, we switched to the same this year. Let's you run at 3k rpm, but the total surprise was how good it was over the same 3 blade standard prop at standard operating modes. On the trip out at 16.5K this year, this airplane was almost 20kts faster in TAS than it was last year with the standard prop at the same altitude and power settings.

Makes ya want to switch to one of those, but alas, they are close to $20K! ... Drats...

More later.....Last fun note. Lynn said on the way back today, he had to climb to FL230 to get over some weather and while there, even with a headwind was 330kts TAS on 30" of MAP at 2500 rpm in a twin turbo Legacy.... yeeeee haaaa, that's cruising... said, he even had his Mach indicator showing .5!... Cool stuff, I'm going to have to get the engine data and take a look at the flight data.
 
Yesterday I took my good friend John Marshall's Super-8 up for a spin. I believe this one was the first or one of the first of its kind flying. Workmanship is top-notch. You guys may have seen it or its twin at a flyin, it has the dragon painted on it. This one has a stock carb'ed 260 and the 8477 blades. Went up to 7K to do some GPS runs. WOT 2300, 23.5", 11.7GPH (could have leaned some more) I got these groundspeeds: S 192kts, E 193kts, N 165kts, W 182kts. The TAS/GPS spreadsheet with the first three numbers says I was cruising at 183kts. Pretty cool. 540's rule! Still ain't a Rocket (I'm biased, I'll admit) but I will say that the 540 really transforms the airplane for the better.

I was flying for business yesterday so I duplicated this in my Rocket. WOT at 7,200' 23.6" 2300 RPM leaned to peak, 72% power, FF was 11.4 GPH and 192 KTAS.
 
I was flying for business yesterday so I duplicated this in my Rocket. WOT at 7,200' 23.6" 2300 RPM leaned to peak, 72% power, FF was 11.4 GPH and 192 KTAS.

Just curious. Is 7,200' Station pressure altitude, Pressure altitude or Density altitude?

It's good to see you used WOT. Using WOT and density altitude is key to obtaining repeatable flight test cruise data.

Just a side note, Lycoming identifies three "standard" EGT leaning temperatures.

Peak EGT - Best economy

50 degrees F ROP - Maximum longevity

100 degrees F ROP - Maximum power

I have seen about 2 knots difference in airspeed at the same power setting, depending on the leaning method. Leaning to Peak EGT was the slowest.

Jim Ayers
 
One other data point as I just got back from Reno and will post more tomorrow. I crewed for Lynn Farnsworth who as some know, runs a super Legacy similar to DG's. We flew our best time this year at 350MPH which only finished us 4th in the SS Gold, but that's an entirely different story.

Anyway, The 2 NXT's fly a Hartzell racing prop. As well, we switched to the same this year. Let's you run at 3k rpm, but the total surprise was how good it was over the same 3 blade standard prop at standard operating modes. On the trip out at 16.5K this year, this airplane was almost 20kts faster in TAS than it was last year with the standard prop at the same altitude and power settings.

Makes ya want to switch to one of those, but alas, they are close to $20K! ... Drats...

Makes me wonder why Hartzell doesn't start producing more of these props and getting the price down. How much better is this prop at say 200-250 knots where a Rocket might operate? 20 knots is huge. Was this prop any faster at race speeds?

That burst of speed by Mike in the last lap was my fault. Sorry. When he saw the white flag, he turned up the wick a bit.
 
Last edited:
hmm....:confused:
Scratching my head. I do not know what my FF at full rich full RPM at altitude ~8k' is. Ill try that on Wednesday night and report back. 24gph seems like a high number at 8k'. Ill see what I get. I know my FF is accurate.
The only data point I have captured is this.
RPM to 2700, 50ROP, 15gph, MP 20.5, ITAS 200, IAS 171, 10k' WOT.

Bob, Today I got the following
DA 8500', WOT, 2700RPM, 50ROP, 206ITAS, 16.0GPH
 
Back
Top