![]() |
The Back-Side of the Power Curve (or lack thereof)
This article popped up in a Google search and was ironically linked back to this site. There are some fundamental errors I wanted to point out. These errors are not unique to this article, but have been around for a long time and unfortunately keep getting repeated.
![]() The article describes the make-up of the drag curves and shows a plot of these. The error comes in when these curves are equated with power curves. These are not at all the same. In the text it states: ?We know that in steady, non-accelerated flight, Thrust equals Drag, so the Thrust (power) curves must be the same.? This is not true. While drag does equal thrust, it does not equal power. Power is obtained by multiplying the drag curve by velocity. This has a dramatic effect on the curves. It pulls down the left side of the curve. The left side can be pulled down so far that the upward hook disappears altogether. The consequence is that there is no region of reversed command in the power curve for the aircraft. Now if we were flying jets and we had thrust levers, then the thrust/drag curves are the ones to use. A piston/prop aircraft, on the other hand, has a power lever and so we need to look at power curves. Below is an example of an actual thrust and power curve (my apologies, happens to be a Lancair) both in the clean and landing configurations. You?ll see the hook on the left completely disappears. What this means is that there is no back side of the power curve for this aircraft. What drives this phenomenon? For a constant power input, propellers produce higher thrust as velocity is reduced. This automatically compensates to the increase in drag shown in the drag curve. If the increase in thrust keeps up with the increase in drag, the power required curves does not climb. It is absolutely critical to separate jets (thrust lever) and props (power lever) because you need to look at different curves to determine if there is a region of reversed command. Here is a link to a video showing the lack of a back-side in the Lancair. If power required was on the rise as the aircraft is slowed, sink rate would increase. Instead it either decreases (landing config) or remains the same (clean config) as stall speed is approached. https://www.youtube.com/watch?v=RfEv4EiIA4c&t=338s I have talked to many pilots who thought they had experienced the back side of the power curve in aircraft that did not have one. There are dynamic events (not steady state) that can easily lead one to falsely believe one was on the back-side of the power curve. It is a simple thing to check in any aircraft. Replicate what was done in the video and see if sink rate starts climbing. |
Slowing my 7 down on final has a noticable effect on sink rate. Nose up to go down faster works every time.
|
Quote:
|
Quote:
Perhaps the video could have nibbled a little closer to stall (it was within 4kts). It was certainly far below normal approach speed. I am looking for aircraft to document that can reach reversed command in the normal approach region. |
Curious
I would be curious to know what five different models have no backside, and your testing method...
The speed at which the power available and power required cross in the low speed regime is the minimum level flight speed, where they cross in the high speed regime is the maximum speed in level flight. I would suggest running your experiment at a zero rate of descent and monitor your power setting for each airspeed, up to and including the power on stall. You will likely find that the generic curve is accurate... It is also a false assumption that an "increase in thrust as velocity decreases" compensates for the drag increase equally... It is very easy to demonstrate in most training aircraft, however, in order to get a good demonstration, you must be well below the normal approach speeds |
Quote:
Cessna 152s can demonstrate this region of reversed comand: Carefully adjust power and trim for level flight, at a speed just above where the stall horn comes on. Remove hands from yoke, use rudder only to keep wings level. Do not touch power. Apply a small amount of nose up trim. After an initial pitch up, the plane will settle into a steady state descent. (The nose up trim trims the airplane for a lower airspeed. But the power required for level flight increases, not decreases(?reverse command?). Since you left the power unchanged, the plane will descend.) |
Quote:
Was this clean of flaps down? Higher drag configs tend to push the min power point to the left. |
other models
Quote:
Aircraft recorded were: Lancair 235, 360, Legacy, IV, and PA28-180 When others are flying I simply have them record steady state descent rate at idle and full flaps across the entire flap extended speed envelope - down to as close to stall that individuals feel comfortable. Increments vary by aircraft, but enough points to generate a good trend line. What you end up seeing is that the slope at the slowest point is still rather steep indicating the aircraft will stall before reaching minimum power speed. For the 360, I have full drag polars, a propeller map, and engine map. Those are from a prior airframe characterization project. |
more
I have no experience with the Lancair series of aircraft, however, I have many, many hours in the PA28 series...and demonstration of reverse command in this series of aircraft is easily accomplished.
If your data indicates that the PA28 does not indicate an area of reverse command, I submit that your method or assumptions are flawed... |
Quote:
I have personally flown many aircraft that do exhibit an increase in decent rate with a decrease in IAS, and I imagine Van (author of the article you posted) has as well. |
Good post, Chris. I was one of the ones who originally thought that sink rate increased at lower speeds in my Lancair, but you (and 200 hrs. of flying) have convinced me otherwise. I'm glad you're here.
|
Quote:
|
yep
You are correct, it should be the minimum speed for level flight...looking at too many charts tonight...
|
Quote:
Not all data that comes back from folks is pristine, for sure. They are not all flight test engineers. For the PA28 it was the owner and his CFI/aerospace engineer son that gathered the data, so I am resaonably confident they followed instructions. Here are some of the curves. They still show a significant downward slope at the last point taken. |
Quote:
The popular GA media/authors would have you believe that anything under approach speed gets you to reversed command. I just posted a plot of curves from different aircraft. The margin to stall varies, but are generally just a few knots away. Does the trend look like any of the curves are about to reverse and increase? In my plane if you actually compute the CL required for min power in the landing config, you see that it is unachievable. Others may vary. I am a little surprised that Van himself wrote the article. He should no better than to mix drag and power, jets and props. |
Quote:
Here is an interesting segue on the topic of power vs stall AoA. While mapping out Pitot-Static errors and the extreme low end of the speed envelope, I spent a lot of time in the buffet region at different flap and power settings. What emerged was a trend that showed how stall speed and stall AoA were affected by the prop wash as it transitioned from driving to back-driving. If the prop is driving it is adding energy to the air stream and it is relatively well behaved. If back-driving the air coming off the prop is low energy very turbulent. This affected the inboard section of wing. One could actually feel the difference sitting in the plane. Note the units are KIAS and not KCAS ![]() ![]() |
Quote:
The original drag and power curves were actually obtained in this way. The aircraft was instrumented with data acquisition and calibrated test equipment. With help from Lycoming and Hartzell we were able to get shaft power and then thrust values at each data point. Engine and prop efficiencies really start to fall off at these extreme low power settings. |
Chris, I think what may be confusing people is your use of the term "reversed command". I'm not sure what you mean by that. In my plane in the landing configuration, pitch controls airspeed and power controls sink rate. I demonstrate that on every landing. What am I missing?
|
Quote:
Not sure what the debate is here...it's pretty easy to demonstrate this "back side of the power curve". Just slow down and hold altitude, reducing power, until you can't hold altitude anymore. Continue pulling back *and adding power back in* to hold altitude. Voila. I've had 172s, e.g., practically hanging on the prop, showing something like 15 or 20 kias IIRC (but in any case WAY below stall speed) and full power, holding altitude. If that isn't the back side of something, I don't know what is. Come to think of it, I've only done it a time or two with an instructor on board in my RV, but guess what? Same thing happens. |
Quote:
It really wasn?t so much a debate as just clarifying that there were errors in the article shown where drag curves where equated to power curves. Many articles in the likes of AOPA and Flying magazine warn against getting slow on final approach. The claim is that a high sink rate can develop as one is supposedly flying on the back-side of the power curve. The question posed is this: Is a piston/propeller aircraft susceptible to the back side of the power curve on final approach ? and this of course assumes configured for landing. From the discussion it appears some have managed to fly on the back-side of the power curve in the clean configuration. While this is certainly possible it is not the configuration that is relevant to the warnings in those articles. The physics of the situation would tend to make it less likely to be achievable considering the relationship of CL and CD at min power. For many aircraft it places the required CL too high to be flyable. A number of aircraft were qualitatively checked to see if this assertion is generally valid. During engine idle descents, gravity is the propulsion and descent rate is an indicator of magnitude. If power required was increasing at progressively lower airspeeds, one would expect sink rate to start increasing. Instead, in all cases the trend is that sink rate decreased throughout. The curves all begin to flatten but one would be hard pressed to argue that with another 3 kts they would reverse and look anything like the hypothetical curves being published. With respect to the published articles, when you look at the author bios, you will see lots of jet time, typically retired airline pilots. I suspect they end up transferring the characteristics of jets onto GA aircraft without understanding the differences. They all describe the increase in drag at low speed correctly and then make the mistake of equating that with power. I would love to see a video of a 172 in level flight indicating 15-20 KIAS. Unfortunately, once you hit buffet or stall you are not operating on the power curve being discussed. |
Quote:
The Clift Notes version. Generally when one wants to fly slower (assume no altitude change) a power reduction is needed. In the case of a jet, a thrust reduction. Flying in the region of reversed command means that flying slower requires more power, or in the case of the jet, more thrust to maintain altitude. This is a common event for jets. This is because jets have thrust levers and their Vref approach speed can be very close to minimum thrust required. i.e. going faster or slower requires more thrust. In propeller aircraft we have power levers. Our reversed command region starts once we get below minimum power required speed. This occurs at a much lower airspeed than typical approach speed. In fact, it is close to or can even below stall speed. |
I will check my 8 in 20 flaps next time I have a chance.
For the poster asking about backside, another myth commonly believed is you add power to slow down (or a region of reverse command). Typically most aircraft will have more thrust or power available to get out of this region. If you are straight and level at some point you’ll find the least power or thrust required setting. To go slower pull some power, add drag, or g and slow a little more. If now you want to maintain straight and level the power/thrust setting will be higher than the minimum power setting you found before that was at a faster airspeed. And again you have almost always you have more power available than required to fly at this condition, firewall it and you speed up; you don’t slow down with added power (not reverse command). I have found backside in turbo props, I know it exists; haven’t tried a piston GA aircraft yet. I am aware it is called a region of reverse command, but this is a misleading term. Command is something you do, like add power or pull aft on the stick (command more thrust or command more g/aoa). A region of reverse command should be used when it actually means what it says. For example the mig15 exhibits regions of reverse command at elevated g, you push on the stick at some point to get more g (and vice versa); or many none fly by wire aircraft when transonic exhibit reverse command in roll (stick left and aircraft rolls right; true also at very slow speeds in some due to adverse yaw). |
Quote:
|
Quote:
The only time my 235 develops a high sink rate is when I am in the flare and pull the power. Unless I'm doing a short field landing, I usually leave a little power in and ease it out when stabilized 6" over the runway in order to gently set down. Like so (Note: There's a 1 second encoding lag in the displayed ground speeds): https://www.dropbox.com/s/5kfrgmuds4...Flare.mp4?dl=0 |
Glider guider
In my sailplane days, I would thermal at minimum sink speed, which was just before the onset of the stall buffet. And as it sounds, this is the speed with the minimum sink. As Paul Mcready proved with the Gosemer Condor, this is also the speed of least power to maintain level flight.
I think there is a nomenclature issue here in that many power planes can fly straight and level below this minimum sink speed if they are hanging on the prop. Helicopters do this all the time. And even in the 172, this is easily demonstrated by maintaining level flight with full power in the stall buffet region. JMHO |
Video
"I would love to see a video of a 172 in level flight indicating 15-20 KIAS. Unfortunately, once you hit buffet or stall you are not operating on the power curve being discussed."
You are never going to get a video of a 172 flying at 15-20 knots. If I get out to the airport this week, I will try and get a video of the 172 requiring more power to fly at a slower airspeed AND maintain level flight... |
Quote:
|
Quote:
|
Yes
Yes, you can...and you can do it in a PA28, too...
|
never seen that before - engine fell off
|
Other interesting bits from the Van's article
Interesting and thought provoking discussion. The OP's primary point is that Power is Thrust times velocity (N m/s) is worth noting. It will flatten the plot but does not eliminate the existence of the backside behaviors we all noticed early in our private training days, that to maintain altitude during slow flight with flaps extended, a significant amount of power had to be added and that increasing pitch caused a descent and decreasing pitch caused a climb.
Van goes on to say in the OP's first post: ?Back Side? operation is primarily for lower power aircraft of aircraft with high lift systems which reduce stall speeds but produce very high drag in the process. If the total thrust available is equal to or less than the drag of the airplane, it will be unable to climb or accelerate out of this condition. ? Most RVs have enough power that they, at normal operating altitudes, can ?power out? from behind the power curve.Perhaps the root of the discussion in this thread is that the Lancair 360, I assume optimized for cruise, does not have enough lift to exhibit a "backside" behavior because it stalls at or near the minimum sink speed, as represented by the descent rate vs airspeed plots for that aircraft. My 9A will drop out of the sky if I lower the speed below 60 knots, to 55 or even 50 knots on final, but I must confess, I don't know if if the sink rate increases. The glide slope certainly gets steeper as the speed decreases. This technique is as effective as a slip when high on the approach. It's a good excuse to go flying and collect power vs airspeed curves for constant altitude in my 9A. It does take full power to stall the airplane during power on stalls. That, by definition, is a backside behavior. |
Quote:
https://www.dropbox.com/s/p14jwd07ns...ckpit.mp4?dl=0 |
You can fly front side and still use pitch to maintain airspeed and throttle/power for glide slope.
I flew some test points today. 30.17, OAT 16, 2300MSL. Wind was rather smooth. Fastback RV8, 180hp io360, constant speed blended airfoil prop. Fuel flow is broke. Mixture prop full forward for all points. Technique for front side was set power let aircraft settle into speed maintaining zero on the VSI (looked outside for this mainly to see trends quicker), use feet to center the ball. Backside, use slight G to increase drag to get slow and then set zero vsi and gradual power to maintain airspeed and zero sink/climb. Again feet for the ball. KIAS/%power/mainfold/rpm 20 flaps 67/31/15.2/2030 65/30/14.9/1990 58/35/16.4/2060 Full flaps 67/37/16.8/2190 54/39/17.3/2110 Appears in both configurations I got backside. This isn?t professional quality data, just a quick eval. I probably got on conditions for about 20-30 seconds. I suspect I am close. |
Good data!
Quote:
I went up today and did a brief flight test to confirm some of the Lancair 235 numbers I gave Chris a couple years ago (Gray line on Power Off Descent Rate - Landing Configuration chart) . Conditions were kind of gusty today so I was reluctant to go below 65 KIAS lest I fall out of the sky. At 90 KIAS my descent rate was still 1,200 fpm, but at 70 KIAS I was getting 950 fpm. Maybe Chris can adjust his gray Lancair 235 line accordingly to show it a little flatter on the bottom end. https://www.dropbox.com/s/c3kginhloy...scent.mp4?dl=0 Everyone should do this stuff to learn the numbers for their own plane. |
Nice landing Snopercod!
As I mentioned, I am guessing that the Lancair does not have high enough lift in the wing to fly on the backside of the power curve, which is what I think is at the root of the discussion. It appears to be stalling near the minimum sink rate, of the bottom of the U shaped drag vs airspeed for constant altitude plot in the first post. Van's referenced article specifically mentioned that this was "primarily for lower power aircraft of aircraft with high lift systems", which I suspect does not describe the Lancair.
Time to go fly and test the -9A |
Well - I was not expecting this!
Just got back from flight testing on a perfectly still air day in the PNW. I got the same flat power curve that has been shown for the other aircraft.
1560 lbs, 40°F OAT, IO-320, FP Sensenich GA, Density altitude between 1350 and 2580 ft, full flaps For constant altitude I got: IAS HP RPM MP 80 67 1830 19.5 75 59 1750 18.8 70 52 1690 18.2 65 44 1620 17.5 60 40 1570 17.2 55 40 1560 17.3 50 41 1555 17.3 45 42 1560 17.4 For constant airspeed with full flaps and 9 inHg MAP I got: IAS Descent Rate 80 1400 75 1090 70 900 65 730 60 600 55 550 50 500 45 450 42 300* 1 knot above stall - Verified from Skyview data log For the constant altitude test I was sure I would have to add power to fly at slower speeds. The data shows a slight increase, but that is only 2 HP. The power curve at high lift was nearly flat just as Chris originally posted. I do only get 30 degrees of flaps out on the -9A. But still it was pretty amazing to slow down the plane and essentially not have to add power between 60 and 45 knots. |
Quote:
|
RV-8 Data
I ran my own set of tests this afternoon in my RV-8.
Tests were run at 3500' +/- 30', 47F OAT, 30.15 in Hg Baro. The tests were flown in extremely smooth air over the ocean, with very low winds aloft (< 8 kts). The airplane configuration was: Weight: 1610 lbs Flaps: Full Prop: Full Forward (Hartzell BA, 74") Mixture: Full Forward (IO-360-A1B6 Angle Valve) IAS MAP Eng. Spd. FF KTS in Hg RPM GPH _____________________ 85 17.5 2490 10.4 80 17.1 2400 10 75 16.5 2300 9 70 16.3 2220 8.5 65 16.3 2150 8.3 60 16.5 2140 8.3 55 16.8 2140 8.5 51 18.1 2230 9.8 I ensured that the plane was completely stabilized in speed and altitude at each test point before recording the data. In my plane, at least, there is quite a noticeable increase in power required just above stall speed. I will try to run these tests again someday closer to gross weight. I won't be able to fly the 51 knot test point at heavier weight but I'm pretty certain that in my plane at gross the speed at which increased power is required moves up to somewhere around 60-65 kias. Skylor |
A couple of these tests seem to stop just when it's getting interesting...just a tiny bit below Vs0 or Vs1, when they had to add a touch of power to slow further. Why not keep going? Increase pitch further while adding power, and continue until either you're uncomfortable with the pitch attitude (it'll be very nose up) or run out of power to give.
It's an interesting flight regime, just keep your turns coordinated and don't stall a wing! |
| All times are GMT -6. The time now is 08:05 AM. |