PDA

View Full Version : C-GMLQ Testing Summary


SMO
08-11-2014, 10:50 PM
Disclaimer
These tests were performed on a homebuilt aircraft by an amateur pilot with very little time in the model being tested. These results are not indicative of what another pilot would achieve in the same airplane. This information is being posted solely for its entertainment value.

C-GMLQ is an unpainted F1 EVO Rocket weighing in at 1351 lbs. She is powered by a stock 260 HP IO-540-D4B5, newly rebuilt by Aero Sport Power in Kamloops, BC. First flight was May 31, 2014, and by the end of July she had accumulated 25 hours. During this period a variety of test flights were flown and the results are summarized below.

Climb Tests
Testing indicates the best rate of climb speed is between 90 and 100 knots. Extended climbs at 90 knots results in cylinder head temperatures of over 400*, therefore the full gross climb test was conducted at 100 knots.

A climb test was conducted at full gross weight of 2100#. Takeoff was from a runway at 1750’ MSL. The climb was measured from 2000 ft. MSL for a period of 3 minutes; airspeed was maintained at 100 knots. Altitude gain was 5100 ft. for an average of 1700 ft./min. The density altitude at 2000 ft. was ~3275. The maximum RPM was 2670 (red line 2700).

Glide Test
Glide testing was done with the power at idle, tests were performed for 7 airspeeds starting at 122 knots and declining to 75 knots. The airplane was lightly loaded with a gross weight of ~1700# and CG well forward at ~88” forward of datum.

The rate of descent ranged from 2500 ft/min at 122 knots to 1000 ft/min at 75 knots.

Glide ratio was observed on the Garmin 696. The best glide ratio with the prop full fine was 8.0:1 at 75 knots. Pulling the prop to full coarse increased the glide ratio to 8.5:1

Airspeed
Airspeed was tested at 8000’ (7800’ Pressure Altitude), MAP was 22.3”, and engine was running ~125* ROP. A tri-directional test was conducted for each RPM setting, IAS was noted, GPS readings taken, and True Airspeed was calculated from the results. Results were as follows:
2000 RPM – 187 knots
2200 RPM – 193 knots
2400 RPM – 200 knots
2600 RPM – 206 knots

Stalls
Stalls were performed with no flaps and full flaps. The airplane was stalled with light loading and at various bank angles (g loading) and power settings. Stalls were also performed at various CG loadings as well as full gross. In all cases the stalls were preceded by an unmistakeable buffet, and if held the nose would finally drop off, often dropping a wing at the same time. With the exception of aft CG loading, the stall was easily broken by releasing pressure on the stick and adding power. If done immediately upon the nose dropping altitude loss could be held to ~200 feet. At aft CG a significant push on the stick was required to get the nose down in order to avoid a secondary stall.

Adding flaps reduced the stall speed by ~5 knots, going to full gross weight increased the stall speed by ~5 knots.

Stall speeds were as follows:
Power off – 59 knots
2200 RPM – 53 knots
Full power – 50 knots
Power off, 30* bank – 65 knots
Power off, 45* bank – 75 knots
Power off, 60* bank (2g) – 90 knots

Note that the bank angle above is an estimate. Also note that at higher bank angles the time between buffet and stall is significantly reduced, and if the low wing stalls first the airplane ends up on its back.

Stability
With a forward CG the airplane exhibits positive longitudinal stability. As loading moves aft the longitudinal stability decreases. It goes negative somewhere between 92.5” (still positive) and 93.7” (negative). At ~96” the airplane became more difficult to control, with gusts creating large divergences from straight and level. This instability increased with speed. Testing further aft of 96” was abandoned and 96” will be the aft limit for this aircraft. However the autopilot did not have any problem controlling the airplane in this flight regime.

During one test at ~94” aft, while flying in a slight descent at 130 knots, a power reduction initially slowly lowered the nose (as expected) and speed decayed. At 120 knots the nose came back up. As speed continues to decay the nose continues to rise. This test was repeated with the same result. Further testing would be required to determine the CG this phenomena begins to occur.

Lateral stability was tested (125 knots) by entering a level sideslip and releasing the aileron while holding the rudder. The low wing should rise to level, however this was not achieved. The airplane is not stable laterally.

Directional stability was tested (125 knots) by entering a level sideslip and releasing the rudder while hold the aileron. The aircraft should return to a no yaw condition, and this was achieved. The airplane is stable directionally.

Spiral stability was tested by banking the aircraft 15 – 20 degrees and releasing the controls. The bank angle should decrease. Lightly loaded and forward CG resulted in neutral spiral stability (bank angle stayed the same) however adding 80# and moving the CG slightly aft to 91.5” resulted in negative spiral stability (bank angle increased).

Autopilot
The airplane is equipped with a TruTrak EFIS IV which includes an autopilot. Roll and pitch servos are installed. The roll servo is a capstan model and the pitch servo is a “C” model. The autopilot works very well, both in smooth and choppy conditions. It will hold descent and climb rates as commanded, and rolls into and out of turns very smoothly, anticipating the roll out to end up within a couple of degrees of the commanded direction.

Shimmy
Many F1 Rockets have reported shimmy at certain ground speeds. After ~40 flights no shimmy has been detected. Most of the landings have been attempts at a 3-pointer, with a few wheel landings as well. Wheel landings have all been squeakers; 3 point landings have tested :eek: (but not exceeded :cool:) the strength of the gear.

Toe-in on this airplane was set as close to zero as possible. The gear is full length (some have cut the gear legs down) and wooden leg dampers have been installed, covered with 2 full wraps of fiberglass and a 3rd wrap part way down from the top.

flyboy1963
08-12-2014, 10:06 AM
Good to hear all's going well with the rocket.....enjoy!

For further testing, I volunteer myself as rear-seat ballast! :)

( smoke - on!)

Norman CYYJ
08-12-2014, 02:23 PM
Nicely done Mark. Nice to see that you finally got it completed.

F1Boss
08-13-2014, 02:51 PM
Hey Mark:

Those stall numbers are very different form what are normally reported. My early testing showed that Ole 84 stalls slower than an RV4 flown next to it; my current Evo will show ~40KIAS before it breaks.

So, please do a 'dynamic pitot/static system check'. Here is how:

Fly down the runway at your home drome at 50'AGL or so and check your altimeter. It should read very close to the field elevation. My guess is yours will read about 200-300 feet high.

Make adjustments to your static port to correct any deviations, and re-run your tests. If the static port needed significant work, you will find your speed runs may require re-testing too.

Please run your stall tests to current NTPS standards, or as close as you can get to those procedures.

In any case, this is a very interesting report!

I have found that longitudinal stability is negative at any loading; in other words, if you point the nose down, it stays there, and the reverse is also true. The ship seems to have the perfect setup between the wing center-of-pressure shift and empennage incidence and sizing to produce this effect.

This has demonstrated no problems to Evo pilots - a few of them prefer that tendency.

I think you were doing a pitch stability test, which will show the same issues with CG shifting in any (most?) aircraft.

See: http://en.wikipedia.org/wiki/Longitudinal_static_stability

Carry on, and good luck!
Mark

SMO
08-13-2014, 06:52 PM
Those stall numbers are very different form what are normally reported.

I initially thought the problem would be the static port (SafeAir). I ran a number of speed tests and used Kevin Horton's spreadsheet to determine if there were any static errors. The first tests revealed ASI errors of ~5 - 8 knots depending on the speed. I then put a dam behind the static port (layers of electrical tape - .030"). This got the ASI into an acceptable range - less than 2 knots error. Finally I changed out the electrical tape for a piece of .032 aluminum and the last tests showed an error of less than 1 knot. However when the ship comes back from the paint shop I will redo some of the tests to confirm, and will do the overshoot test as well.

I think you were doing a pitch stability test

The test I was running was to trim out straight and level flight, then pull to decrease the airspeed 10%, stabilized, without adjusting trim. Then I let go of the stick. If the airplane oscillated with decreasing amplitude, I called that positive. I then did the same test decreasing the airspeed 20%. Finally I did the same tests with a push instead of a pull, increasing airspeed 10 and 20%.

As mentioned, this test was positive at initial CG loadings (forward) - initial oscillations were large but each successive oscillation was less than the last one - but as the aircraft was progressively loaded aft at some point this test went negative.

SMO
02-24-2015, 03:44 PM
I have been adjusting the static port in order to get the ASI to read accurately within a knot or so across the speed range, and doing an altimeter check ~100' above the runway. Happy with the results now.

Stall speeds now read 55 knots clean, 50 knots with flaps.

Cruise and top speeds increased 3 - 4 knots after paint.

Best glide distance speed is 105 knots and lowest rate of sink speed is 80 knots.