Manifold pressure dropping during manouvers
Join Date: Apr 2003
Location: USA
Age: 75
Posts: 3,012
Likes: 0
Received 0 Likes
on
0 Posts
Delta3,
While I love simulation, and am very impressed with your work, Delta3, I am quite skeptical of this particular result! It is just too wonderful.
While in a 40 degree bank, you build 20% more thrust (11200/9307 newtons) you spend 5% less power (111/106 Hp).
This result is about 20% better than IGE vs OGE!
Some probable ways to get that result:
Decel at 2 knots per second, or about 0.1 G of Nx. This is no where near trim, and probably accounts for the power drop and the inflow effects.
Climb 100 feet in 4 seconds (1500 fpm). This will also throw off the calculations, and make some nice inflow of its own.
Can I suggest developing some maneuver controllers that feed back the set parameters to the controls to maintain airspeed and altitude. this might help to produce more trimmed results.
While I love simulation, and am very impressed with your work, Delta3, I am quite skeptical of this particular result! It is just too wonderful.
While in a 40 degree bank, you build 20% more thrust (11200/9307 newtons) you spend 5% less power (111/106 Hp).
This result is about 20% better than IGE vs OGE!
Some probable ways to get that result:
Decel at 2 knots per second, or about 0.1 G of Nx. This is no where near trim, and probably accounts for the power drop and the inflow effects.
Climb 100 feet in 4 seconds (1500 fpm). This will also throw off the calculations, and make some nice inflow of its own.
Can I suggest developing some maneuver controllers that feed back the set parameters to the controls to maintain airspeed and altitude. this might help to produce more trimmed results.
Passion Flying Hobby Science Sponsor Work
Join Date: Apr 2004
Location: Belgium
Age: 68
Posts: 461
Likes: 0
Received 0 Likes
on
0 Posts
Nick
I fully agree that simulation requires carefull preparation and checking, which I am doing at this moment as stated in previous post.
I will check your remarks. One deduction you make is however seems not adequate perhaps due to a misleading picture in the publication : it is not climbing 1500 f/min : heli is quite level starts at 400 and stays at 400 (+- 10 feet) the initial pucture shows 500 but I mixed up pictures from several attempts, see the time plot : there you see altitue (solid red) stays at 40 decafeet = 400 feet (again sorry for that trick that makes life easier for drawing this multichannel plot)
What is true is that there is first a small climb of less then 100 f/min, and this descent is stopped at the end of the simulation(see time plot, climb rate stay within 10 decafeet/min). So at the end, at the observation there remains climb rate is near zero (16 feet/min, which would not be noticible in real life). I cannot fly more precisely manually.
I'll check the power at every point and plot it timewise. I have two ways of calculating the power : one fast less precise, used online, one more precise, off line. This plot is also a way to average out, and provide a tendency. Because the maneuver may not always be precisely on trim, variations can occur, of which arbitrarily one -perhaps non representaive- point is taken in my above post, possibly leading to wrong quantitative conclusion.
I fully agree with your comment about maneuver controllers. Flying on a computer screen with just visual input is not easy and not very reproducible (by the pilot that is, the program records precisely all parameters and can replay precisely).
This is why I built already a tail rotor governor in. I also developped a full 2D flight path controller, which I used to check power in all level flight conditions, including hover, ground effect, autorotations etc, and these figures all same to be OK.
I am thinking now as you suggest of more complex flight path defintions executed by a controller.
Delta3
I will check your remarks. One deduction you make is however seems not adequate perhaps due to a misleading picture in the publication : it is not climbing 1500 f/min : heli is quite level starts at 400 and stays at 400 (+- 10 feet) the initial pucture shows 500 but I mixed up pictures from several attempts, see the time plot : there you see altitue (solid red) stays at 40 decafeet = 400 feet (again sorry for that trick that makes life easier for drawing this multichannel plot)
What is true is that there is first a small climb of less then 100 f/min, and this descent is stopped at the end of the simulation(see time plot, climb rate stay within 10 decafeet/min). So at the end, at the observation there remains climb rate is near zero (16 feet/min, which would not be noticible in real life). I cannot fly more precisely manually.
I'll check the power at every point and plot it timewise. I have two ways of calculating the power : one fast less precise, used online, one more precise, off line. This plot is also a way to average out, and provide a tendency. Because the maneuver may not always be precisely on trim, variations can occur, of which arbitrarily one -perhaps non representaive- point is taken in my above post, possibly leading to wrong quantitative conclusion.
I fully agree with your comment about maneuver controllers. Flying on a computer screen with just visual input is not easy and not very reproducible (by the pilot that is, the program records precisely all parameters and can replay precisely).
This is why I built already a tail rotor governor in. I also developped a full 2D flight path controller, which I used to check power in all level flight conditions, including hover, ground effect, autorotations etc, and these figures all same to be OK.
I am thinking now as you suggest of more complex flight path defintions executed by a controller.
Delta3
Last edited by delta3; 19th Apr 2005 at 08:33.
