JP256 wrote:So, after reading a LOT of information on the topic, I went out a couple of days ago and experimented with MY plane (a 1965 Champion 7ECA with 100 HP C-100 on "regular" tires). It should be noted that there is no POH for this exact aircraft - the closest is the one published by Bellanca when they took over manufacture in 1967, but that POH is for the O-235 powered 7ECA (115 vs my 100 hp). So I don't really have any "book specs" to compare my current performance against. But since the airframes are identical (only the engine and oleo landing gear differ on my plane), I figure that's the "book" to use...

My instructor and I both observed that my 7ECA accelerates pretty slowly while the wheels are on the ground. Before starting this session, I made sure all three tires were filled to the middle of the POH recommended range.

Previous testing had convinced me that raising the tail as soon as possible (POH procedure) is really necessary, as trying to accelerate in a 3-point attitude takes FAR more distance to get to flying speed. (And here I'm talking maybe DOUBLE the ground roll!) So for the testing described below, I brought the tail up as soon as I could. Then I went out and did 8 takeoffs: 4 using the "Vx technique" and 4 doing the "ground effect" technique.

Weather for the test: There was a 10-15 knot wind about 20* right of the runway heading. It remained consistent for this experiment. The temperature was about 95* F, resulting in a DA of about 3,000 feet. Book figures for the Bellanca 7ECA (115 HP) show Vx as 58 mph, Vy as 69 mph, and Vs as 51 mph. On a previous flight, power-on stall speed was observed to be 43 mph IAS, while power-off stall occurred at a lower airspeed (ASI was bouncing between 0 and 45, so totally useless).

Vx Technique:
For short field takeoff, the book calls for liftoff between 50-55 mph, and climbout at 58 mph (Vx). Using this technique, I consistently saw ground rolls of about 500 feet, crossing abeam the control tower even with the top of the tower (~50' AGL). Once abeam the tower, I'd lower the nose slightly, accelerate to Vy (69 mph) and continue the climb. I noted my altitude at the end of the 7000' runway, which was a pretty consistent 800 feet MSL.

Results -- Vx technique:
* Takeoff ground roll ~ 500 feet
* Altitude at runway mid-point (3500 feet) ~ 50 feet AGL (630 feet MSL)
* Altitude at runway end (7000 feet) ~ 220 feet AGL (800 feet MSL)

Using the "low ground effect" technique, I accelerated to 40-45 mph IAS, and eased the stick back to break the ground, immediately leveling off at 1-2 feet AGL. The ground roll was significantly shorter (300-350 ft). Once in low ground effect, I allowed the plane to accelerate to 65-70 mph, and initiated a "zoom climb." Interestingly, I found the aircraft initially continued to accelerate (probably meaning that the ASI lagged slightly behind actual airspeed). I continued the zoom climb until the airspeed started to drop, and tried to maintain exactly at 69 mph (Vy). Interestingly enough (and I believe confirming Jim's point about low ground-effect takeoffs), I was consistently well above the tower's elevation before I crossed abeam the tower. By the time I actually was abeam the tower, I was roughly 50 feet above the tower (100' AGL), and by the time I got to the end of the runway, I was at 900' MSL.

Results -- Low Ground Effect technique:
* Takeoff ground roll ~ 300-350 feet
* Altitude at runway mid-point (3500 feet) ~ 100 feet AGL (680 feet MSL)
* Altitude at runway end (7000 feet) ~ 320 feet AGL (900 feet MSL)

I need lots more practice to see if I can improve on these performance numbers, but am happy with the progress I'm making after only 22 hours of tailwheel time!

Good for you for trying to figure out what works for your airplane. That said, there are a couple of serious flaws in your approach.

First, your assumption that the Vx speed for your airplane is the same as for an airplane that has somewhat more (like 15 %) horsepower, not to mention that the Lycoming powered model 7 is a lot more nose heavy. What you should test FIRST is to go out and find out what the actual Vx speed is for YOUR airplane. This is the case for any of the older airplanes that don't provide that information in a flight manual, and also for any airplanes that have been significantly modified (larger engine, STOL kit, etc). Once you have that number, then you can start a valid evaluation of climb methodology.

Second, this entire discussion has been about CLIMB. Why did you use a different TAKEOFF procedure for the two different tests? Now you're comparing apples to oranges again, and in the process, you invalidate any data gathered. I can't imagine why you'd use a "known" longer takeoff run for one or the other procedure. Again, this is a discussion about climb performance. Obviously, if you're in a tight spot, you're going to use the TAKEOFF procedure that gets you airborne as short as possible.

Third, you changed the climb from Vx to something else when you passed through ~ 50 feet. So, anything after that really doesn't represent a Vx climb.

Fourth, there's something amiss in your stall speed figures. You noted that "On a previous flight, power-on stall speed was observed to be 43 mph IAS, while power-off stall occurred at a lower airspeed (ASI was bouncing between 0 and 45, so totally useless), yet the flight manual lists stall speed for your airplane (Vs) as 51 mph. How do you explain the large difference in Vs between book and what you're seeing? Also, you state that stall speed power off (Vs) is lower than the power on stall speed. There's something wrong there....I've never met an airplane that stalled at a higher speed power on than with power off.

Finally, what weight and balance were you using for the tests? Exact same fuel load/CG? The factory uses maximum gross weight, and typically tries to get the CG as far aft as practical within the limits for the aircraft. Weight and and CG is important primarily in the context of trying to put your tests into the real world of flying.

I appreciate the fact that you are trying to actually test the performance of your plane, but if you're going to do that, you need to do it right, and there are far too many flaws in this "test" to suggest anything about performance.

MTV

Trying to respond to MTV, because I really want to understand what I could do to improve my testing.

Point 1: Vx

My airplane is "bone stock" with no modifications whatsoever (except that I've removed the wheel pants - presumably more impact on cruise speeds than ROC, but ???).

The POH for the 1967++ Citabrias shows Vx and Vy to be the same number for all of the "no flaps" airplanes, regardless of the HP. May not be "truth", but that's what it says. What it doesn't say is whether that's IAS or CAS (which is a fairly large difference: 50 IAS = 58 CAS, 60 IAS = 66 CAS, 70 IAS = 75 CAS).

It is extremely difficult to actually measure Vx. Too many variables for this pilot to compensate for without extremely accurate instrumentation, measured ground courses, etc. So I chose to confirm Vy instead, with the assumption that if Vy was accurate, Vx would likely be accurate as well. Of course, since there is no VSI in the airplane, the best I've been able to do is to use ForeFlight & Stratus to obtain GPS-based climb info, and confirm that data with "time-to-climb" measurements. The test flights I made to confirm the "book" Vy (all done at max gross weight) seem to argue for the book numbers being CAS, because my best ROC appears to be right around 62-63 mph (IAS, which would be 68-70 mph CAS - pretty close to the "book" Vy of 69 mph).

Base on those test flights (repeated three times on three different days), I'm reasonably comfortable that the book Vy (and by proxy, the book Vx) numbers are in fact shown as CAS, and are at least reasonably accurate.

Point 2: Takeoff vs Climb

I used two different takeoff techniques for the simple reason that I wanted to compare the performance obtained by following the book "obstacle clearance takeoff" technique and a "low-ground-effect" technique. My goal was to determine which was better when trying to take off and clear tall trees at the end of the runway in MY airplane.

I'm not sure where you got the idea that I used a "known longer takeoff run" method for one or the other procedure. I DID mention that I had ruled out takeoffs from a 3-point attitude, just letting the airplane fly itself off the runway when ready, because it took a LOT more runway to get airborne that way. This technique (3-point takeoff) is NOT shown in the POH, for what I now believe is the obvious reason that it simply does not work well for this airplane. 'Nuff said.

In all the testing I did, using both Vx and LGE techniques, I raised the tail as soon as it was aerodynamically feasible to do so. The difference between the "ground roll" portions of the two techniques is simply at what point you "rotate" the airplane to lift off the ground...

The book "obstacle clearance" procedure calls for rotation at 50-55 mph (presumably CAS, so 42-47 IAS), and immediately pitch for Vx (58 mph - again presumably CAS, so 50 IAS), and when clear of the obstacles to increase airspeed back to Vy (69 mph CAS, or about 63 IAS). In my plane, it takes about a "4-potato" count to get the tail up, then another 4-potato count to obtain the requisite 42-47 mph liftoff speed. At that point, you rotate, pitch for 50 IAS, and climb at that airspeed to 50 feet AGL...

The "low ground effect" procedure says you should get the airplane off the ground at the lowest airspeed at which it will fly in ground effect -- usually 5-10 mph below power-on stall speed, which was 42 mph IAS for my plane -- then accelerate in ground effect before performing a 'zoom climb' to gain the altitude.

Side note: Since my airplane has lower power than "typical bush planes", I would guess that the propwash effect is lower than experienced in a Super Cub, so my plane probably can't fly as slowly in low ground effect as some others might. That's something I hope to explore in a future test flight.

And I probably mis-spoke in my original post when I said that I accelerated to 40-45 mph IAS before lifting off. What I tried to do was to force the plane into the air as soon as it got "light on it's feet" (which by the time I would cross-check, was showing 40-45 mph on the ASI). It basically took the same 4-potato count to get the tail up, but probably only a 2-potato count before it was light enough on it's feet to be able to force it off the ground, in low ground effect. That 2-potato reduction in the ground roll is where I believe the 150-200 foot ground roll distance reduction was gained. At this point, I could quickly glance inside, and that's when I'd see about 40-45 mph on the ASI.

Point 3 - Changing to Vy at 50 feet:

You are 100% correct that the "altitude at runway end" is not reflective of continued flight at Vx (versus Vy). But the "book" obstacle clearance procedure I was comparing against does NOT call for maintaining Vx beyond the point of obstacle clearance. It calls for resuming "normal takeoff" procedure once the obstacles are cleared.

I made note of the "Altitude at runway end" data point just because it was an easy data point to capture, and to some extent showed the longer-term effect of the "zoom climb" versus Vx (or even Vy) climb. On ForeFlight, I observed Vx climb rates of around 275-300 fpm, and Vy climb rates of around 300-325 fpm. During the zoom climb portions, I saw peak climb rates of 800+ fpm. The net effect was to put me over the "obstacle" at about 50 feet higher with the LGE+Zoom climb than with the Vx+Vy climb, and to wind up about 100 feet higher at the end of the runway...

Point 4 - Explaining Vs "actual" vs "book" stall speeds

I apologize for the confusion. In my hasty editing of my previous post, I inadvertently dropped an entire paragraph where I explained the large IAS versus CAS correction factors for this model. In fact, the "book" Vs of 51 mph (CAS) is pretty much identical to my observed Vs of 43 mph (IAS), given the 8-mph correction factor involved.

As for the "power off" stall speed being slower than the "power on" stall speed, I have no good explanation other than potential ASI error at very slow speeds. At 43 mph indicated, power off, the airplane was mushing around quite a bit, but not dropping a wing until the ASI dropped fairly dramatically (well beyond that which would have been commensurate with the change in angle of attack). Power on, and at 43 mph IAS, a fairly "crisp" stall occurred any time the AOA was increased by the slightest amount.

Point 5 - Weight and CG

On the test flight, I was 150 lbs below max gross weight, with a forward COG (13.7", range is about 12.5 to 19.2 at that weight). Citabrias are flown solo from the front seat, which means that without a bunch of ballast, you'll be somewhat "forward" when solo. And since I'm above average weight, I tend to be more "forward" than most people would be.

Please continue the feedback. I'm trying to sort all this out, and learn as much as possible about the airplane. Especially since I'm basically having to write my own "POH" for the airplane, given the fact that no factory data for the O-200 powered 7ECA seems to be available...

Thanks!

motoadve wrote:actually there are some urban airports that are a lot worse than any backcountry strip.

On the most typical "downwind departure" route assigned by ATC from my home field, once you're out of gliding distance from the runway you have about four miles of nearly solid houses/industrial/shopping centers before you reach anything that you'd even want to consider. A mechanical failure most likely means that you'll be in someone's back yard or on top of their car in the street.

Within this "densely populated area", you're looking at railroad tracks, freeways and city parks as the only option to not go through someone's roof. The freeways are never empty, the Wal Mart parking lot has enough trees and barriers and crap in it to carve up your airplane even if it was empty, and the city parks have people and children in them a lot of the time. The tracks have been used more than once in this area, with the expected total loss of the aircraft from the bumpy ride over the RR ties. We do have flat concrete river wash/drainage channels, and the LA river system, that would provide a good chance of not damaging the airplane, but the headache of getting the airplane back out is probably a big deal.

Considering the danger to people in the airplane and on the ground, sudden stop involving a really solid object like a truck or building, the insurance liability, and the visit from your local FSDO inspector,,, it seems like trees might be more desirable to settle into after an engine problem than a city street.

Don't know for 100% sure, never landed in a tree or a city street, but I have been in the man-made river wash with an RV-3

JP256,

Thanks for the further explanation of your approach. A couple of more thoughts:

You noted that: "It is extremely difficult to actually measure Vx. Too many variables for this pilot to compensate for without extremely accurate instrumentation, measured ground courses, etc." Not true. It's just as easy to verify Vx as it is to verify Vy. Go out in the plane in a "known condition--CG/weight/etc" and start at a known altitude (verified from your altimeter, NOT Foreflight or GPS), and initiate a climb at whatever the book suggests Vx should be. Measure the amount of time it takes to climb 1000 feet--ONCE THE CLIMB IS ESTABLISHED. Now try it at a speed a bit (like a couple mph) faster, and then a few mph slower than that first speed. See what you come up with. You say your airplane is bone stock....but I'll bet that engine isn't brand new, etc. Lots of things can change performance of the airplane.

Thanks for explaining the logic for using two different takeoff methods. That said, once you know that one takeoff procedure works that much better than the other, I would hope that in a true short field takeoff, you'd use the better of the two. So, again to compare apples to apples, I would use the same takeoff technique on both.....again, the discussion of this thread is CLIMB performance, and your use of a different takeoff technique, even though it was based on the manufacturer's recommendation, still biased the result. I'd try the same takeoff technique on both climbs and see what you come up with there.

I understood your logic regarding measuring height at the end of the runway. My point was that again, you're comparing apples to oranges.... If there's no obstacle beyond the initial obstacle, who cares? If there IS another, higher obstacle, why change the climb procedure on one process?

Again, I'm not sure I understand why power on stall for your airplane is substantially higher airspeed than power off stall. That makes no sense whatever to me. Understand that stall speed in testing is measured in a steady, gradual deceleration, not from a fixed speed, then an increase in pitch.

MTV

JP256 wrote:Trying to respond to MTV, because I really want to understand what I could do to improve my testing.

{SNIP}

Please continue the feedback. I'm trying to sort all this out, and learn as much as possible about the airplane. Especially since I'm basically having to write my own "POH" for the airplane, given the fact that no factory data for the O-200 powered 7ECA seems to be available...

Thanks!

Quick question, then a long tip and a quick tip:

Question: Where'd you get your Indicated to Calibrated correction factor? Was there a table in your POH somewhere? That typically requires a significant amount of testing to obtain, using a "truth" airspeed sensor or a GPS method and some math (there's a pretty good resource on the National Test Pilot School website under "downloads"). Local flow fields (which affect your pitot-static system) can change at different angles of attack, so it'd be prudent to check an even spacing of airspeeds from just above Vs to Vh, concentrating with smaller speed spacing at areas of particular interest to you and/or areas that show significant change from one point to the next. I recommend you use the NTPS GPS method, they've already built excel charts with the math in there for you. After you run the numbers, you'll get a calculated true airspeed from your groundspeed vectors. The difference between this and the calculated true airspeed from your indicated airspeed during the testing is your combined position error correction factor and instrument error correction factor. It will likely change with airspeed/AoA. Graphs are not accurate for different configurations, so if you had flaps, you'd want to do a graph in a clean configuration and then another one with 1/2 flaps and another one with full flaps or however you'd care to characterize the error.

Here's what that might look like for you:



Long tip: It's actually really easy to figure out your Vx and Vy with some simple testing.

Pick an altitude band (~1,000' is a good number, so say from 1,000 to 2,000 MSL, or something like that) and conduct a series of full power climbs through that altitude band and different airspeeds. Make sure your climb is stabilized before you enter the band, start a stopwatch when you cross into the band and time your climb through the band. Even airspeed spacing through your areas of interest is the best way to start. If you see that you have an area where the curve changes, take a tighter airspeed spacing in that area to define the shape of the curve. Recommend you start at a slow airspeed and work your way out to Vh to get a detailed curve. Areas of interest might be plus or minus 10 knots in 5 knot increments from where you currently think Vx and Vy are, and maybe 10 knot increments outside of that. Keep track of your aircraft gross weight while your testing, as fuel burn can be a large weight change in some airframes.

You're going to get "test day data," which can be used to pretty well infer answers, but not to build tables/truly extrapolate over different conditions, so keep that in mind. If you want to know what you might see at a known max gross weight loading/CG range, test at that range. Temp and pressure altitude will also play a factor, so record your test conditions.

If you've already figured out your position error, this testing can be done using calibrated airspeed but it probably means more to you in the cockpit in indicated airspeed.

Anyways, draw a graph with rate of climb on the Y axis and airspeed on the X axis. Place your data on the graph and fit a curve to your data. You'll get a concave downward curve, and your peak value is your Vy at those conditions. From the origin (0 RoC, 0 Airspeed), draw a line that touches your curve on a tangent. That's your Vx.

Here's a pictorial example:



Sources cited: I stole both of those pictures from kilohotel.com after a quick google search, and did not produce them myself. However, this is something I'm trained to do, and can help you with the methods and math if you'd like.


Quick tip: to get good, quality data, you need to be as stable as possible on the parameters that matter for the particular test at hand. For both of these tests, that's airspeed. The ACTUAL number doesn't matter as much as your holding on to the speed you stabilize on for the entire duration of data collection. If that's 60 KIAS, it's 60 KIAS. If it's 62 KIAS, then it's 62 KIAS. The actual number will become insignificant after you fit your curve to the data. However, any change in that airspeed during your data collection is going in present scatter into your plots that you may or may not be able to recognize depending on how many data points you get at each speed. Remember that a recognized poor data point can be weighted low or deleted in data processing, but an unrecognized poor data point can skew your answer without your realizing it.

Like most old farts, I hope this data supports what I've done all my life. I very much look forward to your results.

mtv wrote:It's just as easy to verify Vx as it is to verify Vy. Go out in the plane in a "known condition--CG/weight/etc" and start at a known altitude (verified from your altimeter, NOT Foreflight or GPS), and initiate a climb at whatever the book suggests Vx should be. Measure the amount of time it takes to climb 1000 feet--ONCE THE CLIMB IS ESTABLISHED. Now try it at a speed a bit (like a couple mph) faster, and then a few mph slower than that first speed. See what you come up with.

CamTom12]It's actually really easy to figure out your Vx and Vy with some simple testing. Pick an altitude band (~1,000' is a good number, so say from 1,000 to 2,000 MSL, or something like that) and conduct a series of full power climbs through that altitude band and different airspeeds. Make sure your climb is stabilized before you enter the band, start a stopwatch when you cross into the band and time your climb through the band. Even airspeed spacing through your areas of interest is the best way to start. If you see that you have an area where the curve changes, take a tighter airspeed spacing in that area to define the shape of the curve. Recommend you start at a slow airspeed and work your way out to Vh to get a detailed curve. Areas of interest might be plus or minus 10 knots in 5 knot increments from where you currently think Vx and Vy are, and maybe 10 knot increments outside of that. Keep track of your aircraft gross weight while your testing, as fuel burn can be a large weight change in some airframes.[/quote] (Chart omitted for brevity

Thank you both for that approach and explanation. That is MUCH simpler than what I was thinking it would require, and so I'll give it a try next time out... However, I will say that it's going to be a LOT tougher than you make it sound, due to two factors:

1) The simple fact that this low-powered airplane has a very slow rate of climb (at best) in our summer temperatures. First 14 days of August ranged between 101 and 107 degrees F. That impacts both the density altitude and...

2) The tremendous thermal impact of differential heating of the ground surfaces. It's not unusual to have 300' fluctuations in altitude within a few seconds as you fly from rising air to falling air or vise-versa. I tried to determine my best rate of climb using 60 IAS as my target a couple of weeks ago using a similar technique to that described above. During the 1000-foot "stabilized climb" portion, I saw VSI (ForeFlight) fluctuate between zero and 600 fpm as we flew in and out of thermals, and we "averaged" about 350 fpm at gross weight, 100* F, climbing from 1500' to 2500' altitude. The open fields around here would probably make for good sailplane flying, but we can basically take the word "stabilized" out of the vocabulary for a few more months!

[quote="mtv wrote:You say your airplane is bone stock....but I'll bet that engine isn't brand new, etc. Lots of things can change performance of the airplane.

Yep, the engine is not new (about 900 SMOH) and fabric has some "rough edges" where patching has been done, so that will have an impact on the results, but that's why I keep emphasizing the results are "for my airplane" under the conditions tested. I'm not suggesting anyone accept the data from my testing to use with their airplane. The 'trends' will hopefully have meaning to other pilots, but I would never suggest, for instance, that your airplane will take off with 150' less ground roll using the LGE technique ContactFlying suggests. That's just what I saw on my plane that day...

PS - I'm a novice tailwheel pilot (22.2 hours and counting), so my technique can certainly use improvement!

mtv wrote:Thanks for explaining the logic for using two different takeoff methods. That said, once you know that one takeoff procedure works that much better than the other, I would hope that in a true short field takeoff, you'd use the better of the two. So, again to compare apples to apples, I would use the same takeoff technique on both.....again, the discussion of this thread is CLIMB performance, and your use of a different takeoff technique, even though it was based on the manufacturer's recommendation, still biased the result. I'd try the same takeoff technique on both climbs and see what you come up with there.

Excellent suggestion, and I'll put it to the test next time around. It should give a more "direct" comparison between the two climb methods. When I first started doing this testing, I was more interested in reducing the ground roll, and it was only after doing a few "book" takeoffs (and recording my results) that I decided to test the "accelerate past Vx in LGE and zoom climb" approach and began documenting those results as well. I'll try doing the LGE takeoff, but vary the airspeed that I accelerate to before climbing on my next round of testing. Should be interesting.

mtv wrote:I understood your logic regarding measuring height at the end of the runway. My point was that again, you're comparing apples to oranges.... If there's no obstacle beyond the initial obstacle, who cares? If there IS another, higher obstacle, why change the climb procedure on one process?

My answer to this one is a very selfish one... The traffic pattern altitude at my home airport is 1600' MSL (field elevation is ~585' MSL). In flying a reasonably tight traffic pattern, I have been unable to get the airplane above about 1200' MSL on downwind before arriving abeam the landing point where I would typically reduce power to start the next approach. That puts me 400 feet below the other traffic in the pattern, making it much harder for them to see me (even with the orange and white sunburst paint scheme). The easiest solution might be to extend the upwind portion of the pattern, but that takes me away from open fields, etc. and over a fairly densely populated area where the residents are noise sensitive. So for both safety and good-neighbor reasons, I'd prefer not to extend upwind much beyond a 1/4 mile past the end of the runway. Getting there 100 feet higher puts me one step closer to TPA on downwind, where the other traffic in the pattern expects me to be...

mtv wrote:Again, I'm not sure I understand why power on stall for your airplane is substantially higher airspeed than power off stall. That makes no sense whatever to me. Understand that stall speed in testing is measured in a steady, gradual deceleration, not from a fixed speed, then an increase in pitch.

I'm not sure where you're getting that idea - other than my somewhat confusing mix of IAS and CAS in my first post. Power On stall occurred at 43 mph IAS. Power Off stall occurred at basically the same airspeed, although I was frankly out of aft stick travel at that point. The ASI, on the other hand, was fluctuating a lot more on the power off stall than power on. For that, I have no explanation at all...

[quote=" CamTom12]Question: Where'd you get your Indicated to Calibrated correction factor?[/quote]

It's in the POH for the '67 Citabria 7ECA, and a bunch of other "no-flap" Citabrias. The tables lists values for every 10 mph between 50 IAS and 170 IAS, along with what they call "True Indicated Air Speed" in my earlier POH (but shown as CAS in later ones that show the same values).

Thanks, and keep the comments coming. Next flight test will be early next week, because our grandkids will arrive in a couple of hours to spend a few days with us. (Priceless!) So I've still got some time to refine what I plan to do...

MTV and Cam have demonstrated the difficulty of any mathematical comparison of the normal Vx takeoff to the basic low ground effect takeoff. JP256 made a fine effort. Many on here have posted positive experience with getting tail or nose wheel off as soon as possible, mains off as soon as possible, and level acceleration in low ground effect as long as possible before engine and zoom thrust climb to just over the obstacles.

Zane or Mountain Mat , could you set up a survey to see how many of the thousands on this site take advantage of the basic low ground effect takeoff or any soft field type technique for short fields with obstacles?

The important elements to survey for would be :

Nose or tail wheel off as soon as possible.

Main wheels off as soon as the aircraft will fly in low ground effect.

Leveling the aircraft soon after lift off.

Acceleration in low ground effect as long as practicable.

Engine and zoom climb to just over obstacles rather than well over obstacles.

Something I would point out if you intend to use Contact's departure procedure:

This type of transition requires a significant bit of proficiency and practice to retain proficiency. For several reasons:

1. First, it's a judgement call as to when the airplane will be able to "hang" in ground effect, and airspeed instruments are notoriously unreliable at these speeds. So the point at which you initiate the "pull" is really a "feel" thing, and based on experience.

2. Density altitude can and will change the airplanes ability to hang, and again, feel will be key, even more so. So experience at different DA is important.

3. Hanging in ground effect takes a fine touch on the controls to prevent drift, and especially to avoid inadvertent touchdown after liftoff. On floats, inadvertent touch often means you shut down and start over. On wheels, the drag penalty isn't as great, but it's still significant in a tight area. So again, practicing ground effect is essential.

4. Determining the point at which you opt to depart ground effect and transition to climb can be tricky, and again reqiures some practice. Trust me when I tell you that looking at the bases of trees as you accelerate in a genuine tight takeoff LZ will get your heart rate up. Too soon on the pitch up and you shoot your procedure in the foot, too late and you may have to demonstrate your tree climbing skills, getting out of the wreck.

5. Pitching to the climb is also a critical maneuver. What is the pitch attitude that your plane needs to clear the obstacles, but avoid a stall or mush? Too little pith attitude, and the plane won't climb well. Again, practice to proficiency in a known environment is essential.

So, not suggesting you shouldn't go there, just that if you plan to do so, practice the procedure thoroughly, to proficiency, BEFORE you try it in a tight spot. The point is, there are a lot of factors and decisions here that require big doses of judgement and skill.

Which is why the manufacturers recommend a Vx climb....only judgement required is whether you have enough room for it to work, and the factory has given you the data to make that determination, at least in newer planes.

FWIW.

MTV

contactflying wrote:....Engine and zoom climb to just over obstacles rather than well over obstacles.

If you're even just a tad late with your "zoom", you just might be fucked.

If someone has the manufacturer's performance or self collected reference data for a particular aircraft and the variable data of the day e.g. temperature, pressure, wind, surface condition .... everything performance related can be calculated.

In the old days it was a lot of work to calculate those numbers as it was all done by hand. Nowadays in the computer age is pretty easy to set up an Excel spreadsheet or similar and calculate those numbers for many different scenarios or even to calculate a safe operating envelope for someones aircraft.

As for climbing out over an obstacle using Vx, this gives an aircraft most distance to the closest obstacle. The reason for this is that in the event of the worst case scenario, an engine failure, it gives someone most clearance, hence time available to maneuver the aircraft around such obstacles and make a decision how to deal with this situation.

Engine and zoom climb to just over obstacles rather than well over obstacles.

If an engine failure would happen at the same critical point, excessive airspeed can be used to climb further. However it most likely will result in less height gain then climbing at Vx. It also requires more difficult maneuvering (for the average pilot). If an engine failure happens at Vx, initial pilot input is to change the aircraft attitude to best glide, trim and make a decision where to glide to land or crash.

Engine failure at

zoom climb

requires the pilot to stay in a climb attitude until most remaining safe airspeed energy is traded for altitude. Then setup the aircraft for best glide. This is more difficult for the average pilot and could task saturate the average person, delaying the decision making process.

Depending on the speed achieved in ground effect and the climb out point it can also result in a higher G-load attitude change as well as a high pitch attitude for the initial climb out. A high pitch attitude also obstructs forward view, not ideal when climbing close to an obstacle.

Excessive airspeed can be useful in situations when negative performance is expected like downdrafts, windshear and mechanical turbulence, however I personally can not see an practical reason why to use this to clear an obstacle.

What MTV said is true and wise as always. Pusher's evaluation of the poor forced landing prospects with energy maneuverability are wrong; at least in my experience. My many low altitude but high speed forced landings didn't make the data base but worked out well. I see several in the data base that were much higher and slower on takeoff who didn't fare so well. Energy maneuverability is just that : maneuverability.

It's not for everyone but with training and practice it is very safe and much more efficient than the Vx model. Patrick, Learntolandshort, teaches it for mountain and backcountry work, Blackwater teaches it, several here in the Midwest, Todd Underwood in Prescott, etc. I think it would be useful to many of you flying into short fields.

The Airworthiness Directive system works well because mechanics use it. Through communication they find out what is problematic and by deduction what is working well. Why is this system so taboo in pilot training?

Energy maneuverability assuming (always assuming engine failure right? ) engine failure on takeoff : use either potential gravity thrust of altitude or kinetic energy of pressure airspeed to rapidly transition from the takeoff and departure objective to the maneuver to a survivable landing zone and land objective. The entire operation will have to be accomplished in six seconds.

Where and with how much energy maneuverability will you be?

After a year flying in the PNW I found I prefer low ground effect to accelerate and zoom past the tall trees at speed.

Reason is many times I have found turbulence above those trees , and when doing a VX climb the plane was slow when reaching tree altitude and is more affected by turbulence or drafts.

Pusher wrote:As for climbing out over an obstacle using Vx, this gives an aircraft most distance to the closest obstacle. The reason for this is that in the event of the worst case scenario, an engine failure, it gives someone most clearance, hence time available to maneuver the aircraft around such obstacles and make a decision how to deal with this situation.

It depends on the situation, but I think it may be a mistake to assume that there is additional time and distance to maneuver. Really the only maneuver that will be available is to get the nose down and try to avoid a stall. If that works out, then you are still going to have to build additional airspeed to allow whatever maneuvering is necessary. This may not leave you with better options.

Again, the primary reason that manufacturers use the "climb at Vx till clear of the obstacle" recommendations is simply because it requires minimal judgement and only basic skills. Accelerate to Vr, rotate, lift off accelerate to Vx, and climb at Vx until obstacles are cleared. But also, in my experience, this procedure also offers the best obstacle clearance performance in most situations.

In my opinion, if you've got yourself into a situation that truly DEMANDS such a procedure (and I have on more than one occassion) you've already screwed the pooch. If it's that tight, your best bet is to wait till cooler evening or morning temps, or other options. If it's windy at treetops, think carefully before you depart, regardless of your chosen technique.

Now, if you're not REALLY in that truly tight spot, use the takeoff technique that you think will work the best, but understand that the factory TESTED the takeoff and climb performance of your airplane using a Vx climb. So their data is valid for that technique.

And also understand that, if you're in the midst of a Vx climb and it becomes obvious that you'll clear all obstacles nicely and with lots of room, lower the nose a bit.....all you need to do is clear the trees, you don't have to clear them by 100 feet, and there's no need to sustain Vx if it's obvious the jobs been handled.

So, if you want to use the ground effect followed by zoom climb procedure being discussed here, you need to practice it, AND you need to verify its performance in your plane, because that procedure isn't evaluated in any POH I've seen. Maybe yours is different.

And, at the end of your testing, I suspect that you'll find that the factory recommended Vx climb will in fact get you over obstacles in a shorter distance than the ground effect/zoom climb will. And, THAT is the point of an obstacle climb. You should be aware that there may be risks involved with operating at Vx, which is why I don't operate at Vx unless I really NEED to. Which likely means I screwed up. But bear in mind that there are also several potential traps in the zoom climb program.....and YOU are now the test pilot.

On the other hand, if you have plenty of room for takeoff, the ground effect/zoom will work fine and remove some of the risks involved in a true Vx climb......assuming you've done the procedure properly.

But in a true tight spot, I'll fly Vx every day, unless weather suggests otherwise. If I screw up, I want to be able to point to something other than an Internet forum as to why I used a particular procedure......and I believe a Vx climb will indeed get me out shorter in most situations.

Your mileage may vary.

MTV

For normal short field take offs, I fully agree that a "normal Vx climb" per the POH works as well as anything. My use of a low ground effect take off has been at very high density altitude strips (+/-10,000'), where Vx and Vy are pretty close to one another, and I want to get the airplane going as fast as possible in as short a distance as I can, because of the lack of power and lift--the obstacles aren't the major bugaboo; just getting a decent safe climb rate is. It's too easy to forget that at high DAs, Vx is at a higher airspeed, so that if "normal" Vx is used, the airplane will climb out of ground effect and then mush, because it's behind the power curve.

So I've used the low ground effect take off, such as in this video, which gets the airplane up to flying speed more quickly by lifting it off at minimum airspeed and allows it to accelerate to the climb out speed, which is near to Vy. I typically use 10 degrees of flaps, raise the nose to just barely hold the nose gear off as soon as the elevators take hold, and then stay low while it builds speed. As I've mentioned, I use a little nose down trim, because as the airplane accelerates, it tends to climb, and I don't want it to climb out of ground effect before it has attained that climb out speed--I find it easier to pull to keep the airplane from descending than to push to keep it from climbing, and if I've added just the right touch of nose down trim, it's pretty neutral by the time I'm ready to pull to climb out. Then once I've cleared any obstacles, I raise the flaps and continue at that fabulous 200 fpm climb rate!



Cary

mtv wrote:In my opinion, if you've got yourself into a situation that truly DEMANDS such a procedure (and I have on more than one occassion) you've already screwed the pooch. If it's that tight, your best bet is to wait till cooler evening or morning temps, or other options. If it's windy at treetops, think carefully before you depart, regardless of your chosen technique.

MTV

x2! I'm fortunate enough to be able to keep my GA flying for pleasure and being in a situation that requires max performing the airplane is beyond my pleasure threshold. I've had to max perform my aircraft for work enough to know I have no business going there for fun, and should have found a better option for work, too.

MTV, does the POH in new airplanes reflect performance based on the pre 2013 roll on the ground and then pitch up to Vx or Vy as appropriate or the far better get off when it will fly in ground effect and accelerate to Vx or Vy as appropriate?

When I was spraying from short fields the old Vx or Vy short field over obstacles technique just wouldn't get the job done. My objective was never to get higher faster. My objective was to get over with the outcome not being greatly in doubt.

What you, the POH, and the test pilot are saying about getting highest over the shortest distance is true. The problem for those using short and or high strips is the hung out to dry part. If the nose or tail comes up reasonable for our DA and the mains come off reasonably, we know how much room we have left to accelerate in low ground effect before pitch up. To not use all available distance just doesn't make sense. At least it doesn't give us the extra assurance of the outcome of the maneuver.

Yes doing what we are taught in a sanctioned course and just avoiding getting in a position where better is necessary is safest. But it doesn't give the same assurance of the outcome of the maneuver. POH gives official assurance. Investigating and learning better techniques gives self assurance. Worry about compliance is a double edged sword. It generally increases safety through avoidance. It also generally decreases self confidence.

Contact,

The real problem with low ground effect is that, unless you're pretty in tune with the airplane, there are situations and airplanes where you can get the airplane into low ground effect, and it'll hang there....right till you hit something, or opt to land. Working seaplanes live in this world a lot, and I've been there more than once. I've spent what seemed like a week in ground effect in a Beaver before it felt like it'd climb out of ground effect. A bit heavy? A bit warm? Etc.

Temperatures and altitude can create a situation where the airplane simply doesn't have the ooomph to climb out of ground effect. Take off in that kind of situation, get the plane in low ground effect, and you may not be able to climb out. A classic example was the video of the Stinson crashing in Idaho a few years ago.

So, again, I'm not saying the technique doesn't work, because it can. What I'm saying is that that technique requires significant judgement and skill in a truly confined area (though the Idaho example video was most certainly NOT a confined area). And, if you take just a bit too long to figure out that the plane's not coming out of ground effect, you'll likely become a statistic.

On the other hand, following the POH procedure, you can figure out pretty quickly whether the plane is going to depart ground effect and actually climb. No guarantee of success, of course, but unless your head is up and locked, you'll know pretty early whether it's going to work.....and give you a bit of time to shut it down.

MTV