What's new
Van's Air Force

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

'Best' Prop RPM

nigelspeedy

Well Known Member
So after spending the last 50 hours breaking in the engine and fixing all the little things that turned out not to be as perfect as I thought they were during construction I have finally set my mind to experimenting with the performance of the aircraft.
My first thought was to try and reduce some of the variables involved in cruise performance. So what is the best RPM to use in level flight with a constant speed prop, in my case a Whirlwind 74RV. How does one define best? My priorities were speed, efficiency, vibration & noise, roughly in that order.
So the aircraft is an RV-8, the exterior is pretty much built to plans with the exception of the Sam James Cowl. The engine is a 380 CID stroker parallel valve engine with 10:1 and dual P-Mags and fuel injection.
I did my prop RPM experiment by flying straight and level at 8,000' pressure altitude, on my test day it was ISA +10C. I had the throttle wide open for each point and did a mixture sweep from full rich (approx 300F ROP) to as lean as it would run without shaking apart (approx 80F LOP). After each mixture adjustment I let the airspeed stabilize. I corrected all the recorded IAS values to TAS by adding the instrument error and position error (I assumed compressibility error was zero). After each mixture sweep at a constant RPM I changed the RPM and repeated the exercise. I flew 2600, 2500, 2400, 2300, 2000 RPM. For this prop 2700 RPM is the max for takeoff and can't be used continuously so I did not test this as it can't be used for cruise. The prop also has an avoid range from 2050 to 2300 RPM, which is the reason I did not test 2200 or 2100 RPM.
What I was expecting was that as prop RPM was decreased the efficiency would increase by virtue of less friction in the engine. Turns out this was not the case.
As shown in the graph below for any given fuel flow between best economy and best power mixture setting the highest airspeed is with the highest prop RPM. So what I suspect is happening is that the change in ignition timing with RPM has a bigger effect on the engine efficiency/power than the change in RPM has on the efficiency of the prop.
So what RPM to choose then? For max speed choose 2600 RPM, but this is a bit noisy. At 2000 and 2300 RPM there is some perceptible roughness, not bad just not as smooth as 2500. The engine/prop were dynamically balanced at 2500 and it just a little smoother here than at 2400 RPM.
So based on efficiency, speed, vibration and noise 2500 RPM is my cruise prop setting.
Prop%20RPM_zpsdoqcqqeg.png
[/URL][/IMG]
 
Last edited:
Engine efficiency as a function of pressure altitude

After determining the best prop RPM my next quest was to determine the best altitude. One metric of interest was the specific fuel consumption of the engine (fuel flow in pounds per hour per horsepower). I did this test in conjunction with level flight performance testing by flying a range of altitudes from 2000' PA to 18000' PA. For each test I used 2500 RPM. I started out with throttle wide open and the mixture set to either 100F ROP or 25F LOP. I then varied power using the throttle to control MAP and thus airspeed. My aircraft has the Dynon Skyview system which calculates % power based on MAP, RPM, OAT and Mixture. I have calibrated the fuel flow sensor and it is accurate with 1%. I have to assume that the max power provided by Lycon when they dyno'd the engine is accurate and that the Dynon calculated % power is also accurate. For each data point I calculated the SFC and plotted it versus the horsepower as shown in the two graphs below.
Based on calculated figures for the best SFC both LOP (0.36) and ROP (0.4) I think that the provided max horsepower figure and the calculation of % power by the Dynon are pretty good, perhaps one or both of my assumptions are a little on the optimistic side.
Interestingly the SFC seems insensitive to pressure altitude. The data from all tested altitudes lies pretty much on top of each other. So while different altitudes require different throttle settings to achieve a given MAP it does not seem to have a measurable effect on SFC.
Not surprisingly LOP is more efficient than ROP.
SFC%20ROP_zpsbtcg7s3i.png
[/URL][/IMG]
SFC%20LOP_zpsiepkpgze.png
[/URL][/IMG]
 
Best Altitude for Speed, Endurance, Range

To determine the best altitude to fly when trying to maximize speed, endurance & range I flew the aircraft at 2000', 6000', 10000', 14000' & 18000' pressure altitudes under ISA +10C conditions. At each altitude I used 2500 RPM. I initially set the throttle wide open and adjusted the mixture to either 100F ROP or 25F LOP from that point I reduced throttle to control MAP and speed. So each altitude had a ROP and LOP run. I varied the throttle in approximately 1" MAP increments and allowed the airspeed to stabilize for about 1 minute before recording data. Instrument and position errors were added.

In my aircraft the results can be summarized as follows:

Best power mixture or ROP naturally gives the highest speeds. The best efficiency mixture or LOP gives lower speeds but better endurance and range.
As altitude increases the highest possible speed in level flight reduces. So to go as fast as possible go as low as possible and use best power mixture.
SR%20ROP_zpsygtgywxk.png
[/URL][/IMG]

One surprise to me was that as altitude increased the specific range (nm/gal) or fuel economy increased all the way up to 18000'. I thought that with a normally aspirated engine it would increase up to about 8000' - 10000' and then reduce. Now I'm intrigued to know what altitude will give the best range, another opportunity for education and recreation I guess. In automotive terms being able to travel at 195 mph while getting 34 mpg is fantastic. The RV-8 makes up for the fact that I don't own a Prius.

Another surprise was how flat the specific range curves were, so over a large range of speed you get pretty much the same efficiency, so you may as well fly fast.

SR%20LOP_zpss92s4jhs.png
[/URL][/IMG]

For longest time aloft or best endurance set mixture for 25F LOP and throttle as required to achieve 110 KTAS. The endurance seems to be quite insensitive to altitude and the min fuel flow varies by only a couple of tenths of a gallon per hour regardless of altitude. So if you have to hold for an extended period there is no great advantage to changing altitude.

To maximize range and go as far as possible go as high as possible and use 25F LOP and throttle as required to achieve 150 KTAS

Fuel%20Flow%20LOP_zpsbookkwho.png
[/URL][/IMG]

If in the circumstances speed and efficiency are of equal importance (i.e. I want to fly at the Carson speed) then I should fly as high as possible, use 25F LOP and set throttle to achieve 170 KTAS. But this is pretty close to the max speed at higher altitudes so if you were lazy just flying with wide open throttle does not incur much of an efficiency penalty at the gain of a little speed.

Carson%20Speed%20LOP_zpsslmz1lyp.png
[/URL][/IMG]

How lucky we are to be able to build such fantastic aircraft.
 
Last edited:
Carson speed is best lift/drag times 1.316.
A Carson speed of 170 knots means a best L/D of 129 knots? Seems high.
 
198 knots at 10.5 GPH. Doesn't jibe with experience.

That's what I thought!:eek: I must have built a slow poke! Best I got with my 7 at 8,000 with OAT of 75f was 174k or 200mph! Are the 8's that much faster or is my plane just that slow? Doesn't say what engine but mine is IO360 180hp.
 
Note to self: "self don't post graphs with numbers on them". Rather than the absolute values what was of more interest to me was the trends with altitude and the sensitivity of the performance with speed changes.

Perhaps I miss used the term "Carson Speed", I'll have to go and re read his paper. If it is defined in terms of L/D then I am in error, as I used it as the speed that achieves the maximum value of the product of SR and TAS. The speed that gives the maximum SR in level flight is probably not the same as the one that gives the best L/D when gliding. In gliding flight the L/D of the airplane is not dependent on the efficiency of the prop or engine. In level flight the speed for maximum SR is the speed where the airplane as a whole (airframe, engine and prop) give the best efficiency. The SFC of the engine improves with increased power so speed for max SR in level flight is likely quite a bit faster than the speed for best L/D gliding. So I doubt if the best L/D speed of my airplane is 129knots, probably closer to 100 knots.
Cheers
Nige
 
Carson Speed correction

So just re reading Carson's paper he does in fact define the cruising efficiency as the product of L/D and TAS, so my use was incorrect. Sorry for any confusion. That said, I buy fuel by the gallon (unfortunately at a unit cost of $5.29) so I am more interested in how far and fast any one unit of this fuel gets me than L/D.
Cheers
Nige
 
How is the 4 going?

G'day Axel,

How is the four going? You got your airworthiness certificate a week before me and I think the last post I saw you were working through some CHT issues, how is that going?
Cheers
Nige
 
So just re reading Carson's paper he does in fact define the cruising efficiency as the product of L/D and TAS, so my use was incorrect. Sorry for any confusion. That said, I buy fuel by the gallon (unfortunately at a unit cost of $5.29) so I am more interested in how far and fast any one unit of this fuel gets me than L/D.
Cheers
Nige

Interesting testing and data. As you say, the numbers could be scrutanized a bit, but the trends are nteresting. If ecomomy is what you are looking for, I think that your numbers support 2300 RPM at about 9GPH is the best for economy. It would be interesting to see how that worked out at altitude.
 
Carson's

Carson speed is best lift/drag times 1.316.
A Carson speed of 170 knots means a best L/D of 129 knots? Seems high.

Carson speed is usually expressed in CAS.

Carson defines it but in fact, it is 1.316 x V L/D max. This is well established.

My beef is with the difference between a Carson speed derived from zero thrust glide data and a realistic number for in-flight. I don't think they are equal but I don't want to hijack this EXCELLENT thread.
 
Last edited:
Engine assumption made by Carson

On page 3 of Carson's paper he makes the following statement:

"By neglecting minor variations in propeller efficiency and specific fuel consumption that may exist at different airspeeds and power settings
for any given aircraft, it is not difficult to calculate the theoretical amount of additional fuel that would be consumed on a given stage length if an aircraft were flown at an airspeed different from the optimum."

I think that this is a pretty rash assumption. In my plane the variation in airspeed at a given fuel flow due to changes in prop RPM are in the order of 7%. The variations in specific consumption at different airspeeds and power settings are over 100%. This is why the speed for max L/D in gliding flight and the speed for max SR in level powered flight are quite different, with max SR occurring higher.

So if you base your concept of "least wasteful way of wasting fuel" on fuel efficiency (nm/gal) rather than aerodynamic efficiency (L/D) you end up with a higher speed than predicted by Carson. Interesting that the CAFE foundation formula is based on mpg not L/D.

I have the Lycoming Operators Manual which has as Figure 3-1 Representative Effect of Fuel/Air ratio on CHT, Power, SFC at constant RPM & MAP in cruise range operation. While the EGT, CHT and % Power scales are numbered there is no scale on the SFC graph. Does anyone have data for a fuel injected 4 cylinder Lycoming that has actual SFC values as a function of mixture? I'd like a sanity check for the ones I calculated.

Cheers
Nige
 
CAFE and Carson

from: http://cafefoundation.org/v2/aboutcafe_orderfromchaos.php

(emphasis in red added by me)

Carson Speed: ?The Least Wasteful Way of Wasting?

Professor Bud H. Carson of the U.S. Naval Academy, independent of the CAFE Foundation?s derivation of the CAFE Formula, developed the mathematical description for a new characteristic airspeed that he called ?the least wasteful way of wasting? (time or fuel). In honor of Dr. Carson?s work, CAFE dubbed this new airspeed as Carson Speed, and its importance was that it defined the speed at which the CAFE Formula optimized for any aircraft. Quite interestingly, Steve Williams proved that Carson Speed could be found using Jack Norris? zero thrust glide test and its resulting drag polar graph (a plot of airspeed versus drag). By drawing a line from the origin on such a graph (0,0) to a point tangent to the underside of the ?J? shaped drag polar curve, it was always the case that the tangent point represented Carson Speed.
 
SFC suggestion

Finding SFC

1. Figure out how to read the Lycoming power charts including corrections for altitude and temperature. For me, this is the hardest part.

2. Pick a flight condition around 65% and make steady-state observations. Use ?best power? mixture. Record RPM, MAP, OAT, Pressure Altitude and so on.

3. For the same flight condition, match the airspeed while LOP and observe fuel flow. You will need more MAP but keep the RPM equal.

4. Find the BHP for #2 from the charts.

5. Compute SFC for #3. For the same atmosphere and the same speed, using the same prop speed, the BHP must be the same for both flight intervals.

Just a suggestion.
 
Going back to Nigel’s original post, I was very surprised that efficiency, defined as miles per gallon, was highest at high RPM. That is contrary to what I thought was true.....the gospel of over-square operations, Lindbergh teaching P-38 pilots how reduced RPM increases range, because frictional losses are positively associated with RPM, right? I PM’ed Nigel on this, and he explained that his electronic ignition and a couple of other factors unique to his airplane may play a role. He also reminded me that the behavior of complex systems can be difficult to predict.....you really need data. I haven't tested the effects of RPM on efficiency, because I mistakenly thought I knew the answer. However, I enjoy collecting and analyzing data, and the G3X logs some 84 parameters once a second. Today I took advantage of a cross country flight with the family to do a pilot study – a few mixture sweeps at 2400 and 2600 RPM. This evening I plotted it up, and this is what I have:



Shown is true airspeed as a function of fuel flow for 2400 and 2600 RPM. The isoclines are miles per gallon of fuel. This is a plain denim RV-10 with the standard IO-540, slick mags, Hartzell blended airfoil. Conditions were throttle wide open with a density altitude of 10,700 ft and an OAT of about -1C. My mixture sweep was narrower than Nigel’s, but included the peak EGT, and I think I got the peak efficiency (MPG) as well. Leaning improves efficiency up to a point, but with more leaning, efficiency apparently begins to decline. For my system, 2400 RPM yields the higher efficiency. The difference is small at the rich end of the mixture range, but large at the lean end. This is just a start, I need more RPM values and a wider mixture sweep to really nail this down. The take home lesson......there are so many variables in terms of how we build and equip our RV’s that we can’t take data from someone else’s airplane and assume it will be true for ours. We really need to conduct these sorts of performance tests on each individual airplane. The recent generation of EFIS systems makes this analysis a pleasure, with their ability to accurately record flight performance data. And the result is data that allows us to operate our aircraft more efficiently. Education and recreation, right?
 
Last edited:
Nice job Andy,
I'd be interested to see how much of an affect cruise altitude has on your specific range (nm/gal). On my 8 higher was better. I would be curious to see if standard mags with no variable advance react the same way.
Cheers
Nigel
 
I haven't tested the effects of altitude in a standardized manner, but I have done a post hoc analysis of efficiency on our cross country flights.



Fuel efficiency (MPG on y-axis) is true airspeed divided by fuel flow. Each data point is a different flight, and represents a 0.5-1.0 hour segment of the flight flown at a constant altitude and power setting. The G3X records parameters once per second; airspeed and fuel flow data shown here are averaged over the entire segment, so there is no cherry picking. Typical cruise for me is lean of peak and low RPM, but there is variation here in power settings, aircraft weight, and fuel/air stoiciometry, hence the scatter. Fuel flow for these flights ranged from 8.5 to 10.9 GPH, and TAS ranged from 140 to 171 kts.

So, does an airplane powered by an engine with Slick mags lose its efficiency at higher altitudes? Despite the scatter, you can see that there is no suggestion that efficiency is dropping off with altitude. The slope is 0.29 - in other words, you gain three-tenths of a MPG for every thousand feet. I am guessing that the beautiful semi-laminar flow wing is having a big effect on these results, providing an example of how the complete system affects performance.
 
Interesting to see how big the efficiency gains are with a little experimenting with RPM, mixture and altitude. You can get a handy 6% or so just by reducing prop RPM by 200 RPM, and then a whopping 25% or so by choosing your altitude. Admittedly cruising in the high teens comes at the expense and complexity of requiring oxygen. I have had heaps of folks tell me 8,000' is the optimum altitude for NA engines but perhaps this is just an OWT. Or perhaps it is more true of under powered aircraft where they have insufficient power at even moderate altitudes to stay on the front side of the power curve.
Cheers
Nice
 
Last edited:
Back
Top