[Flightgear-devel] YASim solution solution?
OK, I *think* I have nailed the platform-dependant YASim solution failures. What I found was that the solution heuristics hid a bunch of pseudo-chaotic instabilities that were introduced when the approach elevator trim feature went in a few weeks back. This was deterministic, but weird -- increasing the value of a parameter by a tiny amount could throw the solution into an oscillation. Since tiny differences in calculation results ncould (I suppose...) be due to compiler differences, I'm willing to believe this was the culprit; the older gcc had a rounding mode bug or somesuch that caused the solver to pick up on a different oscillation mode. I validated with a command line YASim compiler that the results generated by 2.95.2 were identical to those from 2.96. Unfortunately, my FlightGear 2.95.2 build has begun crashing while loading tiles, so I can't test that until tomorrow. Anyway, try the new code and see if that works for you. You'll also want new planes, as some of the old ones went pretty wacky once the fix went in: http://www.plausible.org/andy/yasim-aircraft-052902.tar Hopefully we can bury this one. Andy -- Andrew J. RossNextBus Information Systems Senior Software Engineer Emeryville, CA [EMAIL PROTECTED] http://www.nextbus.com "Men go crazy in conflagrations. They only get better one by one." - Sting (misquoted) ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] YASim solution solution?
Andy Ross <[EMAIL PROTECTED]> said: > Anyway, try the new code and see if that works for you. You'll also > want new planes, as some of the old ones went pretty wacky once the > fix went in: > >http://www.plausible.org/andy/yasim-aircraft-052902.tar Andy that works. Haven't been able to download your tar ball (is the link correct?). If you can get it to me I'll test them and sync fgfsbase with your latest patch. As for the sinking 747 problem, I'm a little uncertain about the problem being in the "lift" now. It appears that the thrust/altitude curve is a bit too steep. For example even at 1500ft ASL, mach 0.94 IAS, and full throttle I'm only getting 51400lbs. Now I realize those are unusual conditions for a flight but given the rating (and configuration in 747.xml) the thrust should be a lot closer to 6lbs at that altitude (56000lbs or more). A little higher up, I'm seeing thrust figures in the range of 34000 at 12000ft which seems low as well (would guess it'd be above 4lbs anyway). Also there seems to be a greatly exagerated "ram effect" (not sure of correct term). It seems that airspeed changes might be affecting the thrust value too greatly, but I don't have a feel for this at all. Best, Jim ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] YASim solution solution?
Jim Wilson wrote: > Andy that works. Haven't been able to download your tar ball (is the > link correct?). Sigh. Long story: plausible.org is a box in my closet. We lost power last night. It has a dumb (Asus P5A) ATX motherboard that doesn't know how to power on following a power loss (it comes up in soft-off mode instead). The file system had errors. I didn't have time to wait for the fsck this morning, so I went to work with it still down. I'll go home at lunch and get this fixed. My fiancée is complaining that the wedding site is down too. :) > It appears that the thrust/altitude curve is a bit too steep. [...] > Also there seems to be a greatly exagerated "ram effect" (not sure of > correct term). It seems that airspeed changes might be affecting the > thrust value too greatly, but I don't have a feel for this at all. OK, this doesn't surprise me too much. I haven't examined the Jet stuff very closely. The way the code works is that it matches some performance curves I got out of McFarland for a 707 engine. The turbofans on the 747 actually won't be too terribly far off in their thrust performance, I'd think. (With a little scaling of the minor output numbers like N1 and TIT to ones appropriate for the engine.) Most likely, I've got an interpolation bug in there somewhere. I'll take a look. Andy -- Andrew J. RossNextBus Information Systems Senior Software Engineer Emeryville, CA [EMAIL PROTECTED] http://www.nextbus.com "Men go crazy in conflagrations. They only get better one by one." - Sting (misquoted) ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] YASim solution solution?
On Wed, 29 May 2002 10:05:07 -0700 Andy Ross <[EMAIL PROTECTED]> wrote: >last night. It has a dumb (Asus P5A) ATX motherboard that doesn't >know how to power on following a power loss (it comes up in soft-off You sure you don't have a BIOS setting for that? I've got one on my m/b. Maybe you need to upgrade your BIOS? Jon ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] YASim solution solution?
Andy Ross wrote: > Jim Wilson wrote: >>It appears that the thrust/altitude curve is a bit too steep. [...] >>Also there seems to be a greatly exagerated "ram effect" (not sure of >>correct term). It seems that airspeed changes might be affecting the >>thrust value too greatly, but I don't have a feel for this at all. > > > OK, this doesn't surprise me too much. I haven't examined the Jet > stuff very closely. The way the code works is that it matches some > performance curves I got out of McFarland for a 707 engine. The > turbofans on the 747 actually won't be too terribly far off in their > thrust performance, I'd think. (With a little scaling of the minor > output numbers like N1 and TIT to ones appropriate for the engine.) > Most likely, I've got an interpolation bug in there somewhere. I'll > take a look. > According to this site you're pretty close: http://www.bh.com/companions/aerodata/appendices/data-b/default.htm Erik ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] YASim solution solution?
Jon S. Berndt wrote: > Andy Ross wrote: > > last night. It has a dumb (Asus P5A) ATX motherboard that doesn't > > know how to power on following a power loss (it comes up in soft-off > > You sure you don't have a BIOS setting for that? I've got one on my > m/b. Maybe you need to upgrade your BIOS? Sadly, it has current BIOS. It's an older motherboard, and my guess is that it's a hardware deficiency. In order to make the "decision" to power the motherboard on following application of power, some circuit on the motherboard (CPU, whatever) needs to receive power. If no such circuit is powered, we're toast. No amount of software can make a dead lump of silicon compute. I have the same motherboard in my machine at work, and it has the same problem. I've thought about wiring up a switch circuit that will close the soft-power switch circuit for 0.1 second or so following the application of power... Andy -- Andrew J. RossNextBus Information Systems Senior Software Engineer Emeryville, CA [EMAIL PROTECTED] http://www.nextbus.com "Men go crazy in conflagrations. They only get better one by one." - Sting (misquoted) ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] YASim solution solution?
Andy Ross writes: > OK, this doesn't surprise me too much. I haven't examined the Jet > stuff very closely. The way the code works is that it matches some > performance curves I got out of McFarland for a 707 engine. The > turbofans on the 747 actually won't be too terribly far off in their > thrust performance, I'd think. (With a little scaling of the minor > output numbers like N1 and TIT to ones appropriate for the engine.) This is probably a dumb question, but are you sure that your 707 numbers were for a turbofan engine and not a turbojet engine? I think that many of the 707 models shipped with turbojets. All the best, David -- David Megginson, [EMAIL PROTECTED], http://www.megginson.com/ ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] YASim solution solution?
David Megginson wrote: > Andy Ross writes: > > The way the code works is that it matches some performance curves I > > got out of McFarland for a 707 engine. The turbofans on the 747 > > actually won't be too terribly far off in their thrust performance, > > I'd think. > > This is probably a dumb question, but are you sure that your 707 > numbers were for a turbofan engine and not a turbojet engine? I think > that many of the 707 models shipped with turbojets. The 707 did indeed have turbojets. But in the flight regime of a jetliner, there's not a whole lot of difference in thrust performance between a jet and a fan. The numbers (fuel consumption, N1/N2 speeds, etc...) will be different, of course, but their qualitative behavior is basically the same. There is provision in the configuration files for scaling the output numbers to a range appropriate for a given engine. Jets have higher TSFC and EPR values than fans, so they get bigger numbers in their configuration files. Where jets and fans differ most is in the transsonic regime. The higher exhaust pressure of a jet leads to higher exhaust velocities, and thus less thrust dropoff at speed. YASim tries to model this (I don't know how successfully) with an exhaust-speed tunable. [At supersonic speeds, lots of stuff is happening that won't be modelled well by the current code at all. The F-15C that I did for Gene, for example, was reading 130% N1 RPM at mach 2 or so. :)] In keeping with the YASim philosophy, the point is not to mimic any single engine perfectly (by the use of giant lookup tables, for instance), but to come up with a model that works well qualitatively for any engine. Andy -- Andrew J. RossNextBus Information Systems Senior Software Engineer Emeryville, CA [EMAIL PROTECTED] http://www.nextbus.com "Men go crazy in conflagrations. They only get better one by one." - Sting (misquoted) ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] YASim solution solution?
> [At supersonic speeds, lots of stuff is happening that won't be >modelled well by the current code at all. The F-15C that I did for >Gene, for example, was reading 130% N1 RPM at mach 2 or so. :)] > Say what?! When did you do this? Did you take the DECC/ECC into account? :) g. ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
re: [Flightgear-devel] YASim solution solution?
Andy Ross writes: > Anyway, try the new code and see if that works for you. You'll also > want new planes, as some of the old ones went pretty wacky once the > fix went in: > >http://www.plausible.org/andy/yasim-aircraft-052902.tar I'm checking these in now. All the best, David -- David Megginson, [EMAIL PROTECTED], http://www.megginson.com/ ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] YASim solution solution?
I wrote: > Jim Wilson wrote: > > It appears that the thrust/altitude curve is a bit too steep. [...] > > Also there seems to be a greatly exagerated "ram effect" (not sure > > of correct term). It seems that airspeed changes might be affecting > > the thrust value too greatly, but I don't have a feel for this at > > all. > > OK, this doesn't surprise me too much. I haven't examined the Jet > stuff very closely. OK, I've examined the jet code a bit more closely, and it actually looks pretty good to me. Attached is a graph of available thrust vs. speed and altitude for the engines as modelled on the 747-400. I threw together a little program that looped over the Jet object, and played with gnuplot's really nifty contour feature a bit. (Apologies for the binary attachment -- it's only 9k, which puts it right at the border of acceptability, IMHO). All the features look about right to me: Thrust falls off linearly as speed increases; this is due to the reduction in exhaust velocity relative to the aircraft. But at speed, it starts increasing again due to the V^2 scaling of the static air compression as it enters the engine. The numbers might be different for the real thing, but the appearance of the graph is spot on. So let's fix the numbers. One thing I can see is that the CF6 engines on the 747 are listed as "60k lbs" of thrust. But they're also flat rated at 33 degrees C (306K) (that is, below this temperature, you can't use any of the extra thrust). As it happens, YASim calibrates to a standard atmosphere of 288K. Air at the flat rated temperature is 6% less dense than the air YASim is calibrating to. So if the engine can get 60k at the hotter temerature, it should really be listed as a 63706 lb engine in the configuration file. This will give you too much power at takeoff, as YASim doesn't yet implement an engine governor. But the performance at altitude might match what you want to see more closely. Try it out and see how it feels. Andy -- Andrew J. RossNextBus Information Systems Senior Software Engineer Emeryville, CA [EMAIL PROTECTED] http://www.nextbus.com "Men go crazy in conflagrations. They only get better one by one." - Sting (misquoted)
Re: [Flightgear-devel] YASim solution solution?
Andy Ross <[EMAIL PROTECTED]> said: > > This will give you too much power at takeoff, as YASim doesn't yet > implement an engine governor. But the performance at altitude might > match what you want to see more closely. Try it out and see how it > feels. > Actually I tried all the way up to 80,000lbs and still ran into problems in the 25000ft range. There is a little uncertaintly in just what I'm observing. Basically there is a steady decrease in attainable airspeed. IIRC the "required thrust" should fall off as well, resulting fairly good performance all the way up to 35,000 to 40,000ftASL with performance gradually decreasing up to max altitude. It could be the lift or it could be the drag, not sure, but something definately isn't right. With my most recent tests I'm kind of thinking the drag is too much at higher altitudes, because even with the nose pitched down I'm not able to recover much airspeed at full throttle. On the other hand it could be lift. A possible clue: when I'm having trouble the "mach" reading seems to be way too high as compared to the KAIS reading just above. Examples: @ 19000ft 419KIAS MACH=0.91 @ 23000ft 344KIAS MACH=0.83 Both of these examples are at full throttle and cruise attitude...just as an indicator of how bad the problem is. With time the max speed at 23000 falls to 327KIAS and then stays there. BTW folks can test this quickly by starting up flightgear with 747-yasim, KSFO. Hit the throttle full. Set Autopilot altitude to 19000. After reaching 50ft altitude hit CTRL+H to lock the heading. Hit "g" to retract the gear. Hit "a" four times to speed up time. You should be at 19000ft in about 2.5 minutes. Best, Jim ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] YASim solution solution?
> OK, I've examined the jet code a bit more closely, and it actually > looks pretty good to me. Attached is a graph of available thrust > vs. speed and altitude for the engines as modelled on the 747-400. I > threw together a little program that looped over the Jet object, and > played with gnuplot's really nifty contour feature a bit. (Apologies > for the binary attachment -- it's only 9k, which puts it right at the > border of acceptability, IMHO). This may or may not have anything to do with the jet code, but with the 747-yasim, I cannot slow the plane below about 280kt in level flight at 3000ft ASL with throttles at minimum and full flaps, which makes the plane rather hard to land... This is just an observation, I'm not a pilot, please forgive me if I'm talking total nonsense... Andras === Major Andras e-mail: [EMAIL PROTECTED] www:http://andras.webhop.org/ === ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] YASim solution solution?
Major A <[EMAIL PROTECTED]> said: > > > OK, I've examined the jet code a bit more closely, and it actually > > looks pretty good to me. Attached is a graph of available thrust > > vs. speed and altitude for the engines as modelled on the 747-400. I > > threw together a little program that looped over the Jet object, and > > played with gnuplot's really nifty contour feature a bit. (Apologies > > for the binary attachment -- it's only 9k, which puts it right at the > > border of acceptability, IMHO). > > This may or may not have anything to do with the jet code, but with > the 747-yasim, I cannot slow the plane below about 280kt in level > flight at 3000ft ASL with throttles at minimum and full flaps, which > makes the plane rather hard to land... > > This is just an observation, I'm not a pilot, please forgive me if I'm > talking total nonsense... Legally you shouldn't be up to 280kt at 3000ftASL :-). If I'm not mistaken you really need to get under 220 or so some 15 miles out. Below 10,000 ft slow down to 245 or so (250 is the speed limit) to prepare for an approach. Half a million pounds packs a lot of momentum. Oh yeah, if your fuel tanks are half full you might not be able to land without crashing anyway. Not sure what the recomendation is for that. There should be speed brakes which would have helped a lot, but they might not be implemented yet. There are also a bunch of flaps on a real 747 and I'm not sure which ones are actually modeled. Best, Jim ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] YASim solution solution?
> > This may or may not have anything to do with the jet code, but with > > the 747-yasim, I cannot slow the plane below about 280kt in level > > flight at 3000ft ASL with throttles at minimum and full flaps, which > > makes the plane rather hard to land... > Legally you shouldn't be up to 280kt at 3000ftASL :-). If I'm not mistaken > you really need to get under 220 or so some 15 miles out. Below 10,000 ft > slow down to 245 or so (250 is the speed limit) to prepare for an approach. Actually, he's off the hook, if operating in the USA and subject to FAR 91.117: (a) Unless otherwise authorized by the Administrator, no person may operate an aircraft below 10,000 MSL at an indicated airpseed of more than 250 knots. -> This could be a problem, if he didn't ask for permission first, but ... (b) Unless otherwise authorized or required by ATC, no person may operate an aircraft at or below 2,500 feet above the surface within 4 nautical miles of the primary airport of a Class C or Class D airspace area at an indicated airspeed of more than 200 knots (230 mph.). This paragraph (b) does not apply to any operations within a Class B airspace area. Such operations shall comply with paragraph (a) of this section. -> Irrelevant unless terrain above 500 feet MSL. In any case, if flying a standard approach to a commercial airport, he was probably in Class B so the whole paragraph wouldn't apply either. (c) No person may operate an aircraft in the airspace underlying a Class B airspace area designated for an airport or in a VFR corridor designated through such a Class B airspace area, at an indicated airspeed of more than 200 knots (230 mph). -> Probably not relevant. (d) If the minimum safe airspeed for any particular operation is greater than the maximum speed prescribed in this section, the aircraft may be operated at that minimum speed. -> He already said he couldn't make the aircraft fly more slowly. So he's fine. In any case, assuming the pilot considered being unable to slow the aircraft to be an emergency situation (I know I would), our favorite FAR 91.3 steps in: (a) The pilot in command of an aircraft is directly responsible for, and is the final authority as to, the operation of that aircraft. -> Unless the simulator is running on Windows, of course 8-) (b) In an in-flight emergency requiring immediate action, the pilot in command may deviate from any rule of this part to the extent required to meet that emergency. -> He decided to go fast and try to land. (c) Each pilot in command who deviates from a rule under paragraph (b) of this section shall, upon the request of the Administrator, send a written report of that deviation to the Administrator. -> He did that, and Andy said he'd look into it. Always fly the simulation as though it's the real thing. 8-) ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] YASim solution solution?
Jim Wilson wrote: > Actually I tried all the way up to 80,000lbs and still ran into > problems in the 25000ft range. There is a little uncertaintly in just > what I'm observing. Basically there is a steady decrease in > attainable airspeed. I found one bug. There was a property name typo* in David's changes that hooked the global environment into YASim. This resulted in all temeratures at runtime (but not at solution time) being exactly 0 degrees C. This is far too warm for the flight levels, which resulted in an air density that was significantly lower than it should have been. Effectively, the airplane was performing as it it were at a higher altitude than it was. This improves things a bit, as does the increase of thrust to 63737 pounds (we should probably check that number in, btw). I did discover, however, that the airplane has a very sensitive "back of the power curve" behavior. I had it trimmed for climb at about 230 kias (about 7 degrees of AoA), and topped out at a service ceiling of FL220. Gently easing down on the trim, I had it climbing at 500fpm again at 260 kias and 4 degrees. Then I had to go to bed, so I didn't get a complete set of numbers. But certainly "climbing too slow" is part of the problem here. You need to keep the AoA down to avoid burning all your thrust working against induced drag. Does anyone have good, hard climb numbers for this plane? I mean stuff like: "At NNN pounds gross weight, XXX feet MSL and YYY knots TAS, the 747-400 can climb at ZZZ feet per minute." My suspicion is that we're being bitten by a combination of bad performance numbers being fed to YASim, and bad pilot climb technique. Andy * "degC" instead of "degc". Norman, you can fire away at the lack of symbol safety in the property system. This is one circumstance where it's deserved. :) -- Andrew J. RossNextBus Information Systems Senior Software Engineer Emeryville, CA [EMAIL PROTECTED] http://www.nextbus.com "Men go crazy in conflagrations. They only get better one by one." - Sting (misquoted) ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] YASim solution solution?
Jim Wilson wrote: > On the other hand it could be lift. A possible clue: when I'm having > trouble the "mach" reading seems to be way too high as compared to the > KAIS reading just above. Examples: > > @ 19000ft 419KIAS MACH=0.91 > @ 23000ft 344KIAS MACH=0.83 Those numbers look correct to me. As altitude increases, the speed of sound is decreasing due to the drop in temperature and the absolute speed corresponding to "indicated" airspeed is increasing due to the drop in density. At sea level, mach one is about 700 KIAS. At the tropopause, it's only something like 370. Andy -- Andrew J. RossNextBus Information Systems Senior Software Engineer Emeryville, CA [EMAIL PROTECTED] http://www.nextbus.com "Men go crazy in conflagrations. They only get better one by one." - Sting (misquoted) ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] YASim solution solution?
Major A wrote: > This may or may not have anything to do with the jet code, but with > the 747-yasim, I cannot slow the plane below about 280kt in level > flight at 3000ft ASL with throttles at minimum and full flaps, which > makes the plane rather hard to land... By way of disclosure: there is a known bug in the YASim computation of drag from flaps, which results in far too little drag being applied at full flaps. But that's not your problem. :) You seem to be expecting the aircraft to slow instantly when you pull back on the throttles. It won't. It's a *big* jet, and it takes a *long* time to resond to speed changes. I can verify that it trims nicely for approach at 136 knots when fuel is at 20% (you have to do this manually; the default is 50%, which is much heavier). This behavior is guaranteed by the YASim solver, in fact. Andy -- Andrew J. RossNextBus Information Systems Senior Software Engineer Emeryville, CA [EMAIL PROTECTED] http://www.nextbus.com "Men go crazy in conflagrations. They only get better one by one." - Sting (misquoted) ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] YASim solution solution?
Jim Wilson wrote: > There should be speed brakes which would have helped a lot, but they > might not be implemented yet. Sure are: /controls/spoilers > There are also a bunch of flaps on a real 747 and I'm not sure which > ones are actually modeled. All of them; YASim models flaps symbolically as a property of the wing object; there's no need to put each surface in the configuration file They're not individually controllable anyway, so there'd be little reason to split them out. Although be aware that the flap drag bug prevents them from being very useful for speedbrakes. Andy -- Andrew J. RossNextBus Information Systems Senior Software Engineer Emeryville, CA [EMAIL PROTECTED] http://www.nextbus.com "Men go crazy in conflagrations. They only get better one by one." - Sting (misquoted) ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] YASim solution solution?
Andy Ross <[EMAIL PROTECTED]> said: > Does anyone have good, hard climb numbers for this plane? I mean > stuff like: "At NNN pounds gross weight, XXX feet MSL and YYY knots > TAS, the 747-400 can climb at ZZZ feet per minute." My suspicion is > that we're being bitten by a combination of bad performance numbers > being fed to YASim, and bad pilot climb technique. Think I saw something that was maybe at a fixed weight. Not the full Flight manual table. When I get home I'll look for it. But I was suprised at the data. At lower altitudes it was over 4000fpm and was at least 2000fpm up to and over 3ft. Finally dropped off to about 400fpm at 4ft, reaching 0 somewhere around 43000ft. Best, Jim ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] YASim solution solution?
Andy Ross <[EMAIL PROTECTED]> said: > Jim Wilson wrote: > > On the other hand it could be lift. A possible clue: when I'm having > > trouble the "mach" reading seems to be way too high as compared to the > > KAIS reading just above. Examples: > > > > @ 19000ft 419KIAS MACH=0.91 > > @ 23000ft 344KIAS MACH=0.83 > > Those numbers look correct to me. As altitude increases, the speed of > sound is decreasing due to the drop in temperature and the absolute > speed corresponding to "indicated" airspeed is increasing due to the > drop in density. At sea level, mach one is about 700 KIAS. At the > tropopause, it's only something like 370. Yeah but look at the values again...we're getting close to tropopause value at 23000ft. Mach should be well over 600knots at 23000ft, unless it's _really_ warm. Best, Jim ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] YASim solution solution?
Jim Wilson wrote: > Think I saw something that was maybe at a fixed weight. Not the full > Flight manual table. When I get home I'll look for it. But I was > suprised at the data. At lower altitudes it was over 4000fpm and was > at least 2000fpm up to and over 3ft. Finally dropped off to about > 400fpm at 4ft, reaching 0 somewhere around 43000ft. I found this X-Plane site: http://webpages.charter.net/rtpete/html/747.html Which agrees with you for the most part: > ROC Rate Of Climb > [...] > Above 10,000 ft to Cruise Flight Level FL > * 2200fpm from 10,000 - 20,000ft @ 280 - 340kts > * 2000 - 1500fpm from 20,000 - 26,000ft > * 1500 - 400fpm from 26,000 - 35,000 ft depending on weight But note the speed: 280-340 knots (it doesn't say indicated or true, sadly). That's much higher than the 230 knots that I was flying last night. I think what's happening is that for the initial climb out, the aircraft wants to be in a high-AoA attitude; otherwise you'd have a liftoff speed of 300+ knots and the wheels would incinerate. Once off the ground, the 250 knot speed limit is still on the back side of the power curve. If the autopilot is engaged there, the aircraft will get stuck on the back side, and never find the high-efficiency climb regime at lower AoA. Try this (since I'm at work and can't): trim for 250 knots only up to 1 feet, and then push the nose down and accelerate to something like 300 before engaging the autopilot again (or better yet, trim for 300 knots and don't engage the autopilot at all). See if the climb performance in the flight levels improves. I'll see if I can throw together a "climb rate finder" program, along the same lines as the jet thrust analyzer I did yesterday. This would be generically useful -- being able to hand it a YASim description and get back a chart of best climb speed/AoA at each altitude. Andy -- Andrew J. RossNextBus Information Systems Senior Software Engineer Emeryville, CA [EMAIL PROTECTED] http://www.nextbus.com "Men go crazy in conflagrations. They only get better one by one." - Sting (misquoted) ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] YASim solution solution?
Jim Wilson wrote: > Yeah but look at the values again...we're getting close to tropopause > value at 23000ft. Mach should be well over 600knots at 23000ft, > unless it's _really_ warm. Mach 1 at the tropopause and above is just about exactly 295 m/s, which is 573 knots *true* airspeed. The numbers you quote are "indicated" airspeed (that's the "I" in KIAS). That means that they are corrected for density (basically by the square root of the density ratio at subsonic speeds) and are much lower than true speeds at low densities. At 23000 feet MSL, the correction is about 1.5, which agrees very closely with the numbers you cite. Andy -- Andrew J. RossNextBus Information Systems Senior Software Engineer Emeryville, CA [EMAIL PROTECTED] http://www.nextbus.com "Men go crazy in conflagrations. They only get better one by one." - Sting (misquoted) ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] YASim solution solution?
--- Jim Wilson <[EMAIL PROTECTED]> wrote: > Andy Ross <[EMAIL PROTECTED]> said: > > > Jim Wilson wrote: > > > On the other hand it could be lift. A possible > clue: when I'm having > > > trouble the "mach" reading seems to be way too > high as compared to the > > > KAIS reading just above. Examples: > > > > > > @ 19000ft 419KIAS MACH=0.91 > > > @ 23000ft 344KIAS MACH=0.83 > > > > Those numbers look correct to me. As altitude > increases, the speed of > > sound is decreasing due to the drop in temperature > and the absolute > > speed corresponding to "indicated" airspeed is > increasing due to the > > drop in density. At sea level, mach one is about > 700 KIAS. At the > > tropopause, it's only something like 370. > > Yeah but look at the values again...we're getting > close to tropopause value at > 23000ft. Mach should be well over 600knots at > 23000ft, unless it's _really_ warm. Be careful here. Andy is, I believe, putting out calibrated airspeed as IAS. Soundspeed is usually calculated in terms of true airspeed, which will be considerably higher CAS at high altitudes and/or mach numbers. > > Best, > > Jim > > ___ > Flightgear-devel mailing list > [EMAIL PROTECTED] > http://mail.flightgear.org/mailman/listinfo/flightgear-devel > > __ Do You Yahoo!? Yahoo! - Official partner of 2002 FIFA World Cup http://fifaworldcup.yahoo.com ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] YASim solution solution?
Andy Ross <[EMAIL PROTECTED]> said: > Jim Wilson wrote: > > Think I saw something that was maybe at a fixed weight. Not the full > > Flight manual table. When I get home I'll look for it. But I was > > suprised at the data. At lower altitudes it was over 4000fpm and was > > at least 2000fpm up to and over 3ft. Finally dropped off to about > > 400fpm at 4ft, reaching 0 somewhere around 43000ft. > > I found this X-Plane site: > > http://webpages.charter.net/rtpete/html/747.html > > Which agrees with you for the most part: > > > ROC Rate Of Climb > > [...] > > Above 10,000 ft to Cruise Flight Level FL > > * 2200fpm from 10,000 - 20,000ft @ 280 - 340kts > > * 2000 - 1500fpm from 20,000 - 26,000ft > > * 1500 - 400fpm from 26,000 - 35,000 ft depending on weight Yes I've seen that, which is why the other table suprised me, it's numbers were generally higher. Showing the 747-400 capable of climbing up to 43000ft and cruising at 40,000ft. > But note the speed: 280-340 knots (it doesn't say indicated or true, > sadly). That's much higher than the 230 knots that I was flying last > night. I think what's happening is that for the initial climb out, > the aircraft wants to be in a high-AoA attitude; otherwise you'd have > a liftoff speed of 300+ knots and the wheels would incinerate. Once > off the ground, the 250 knot speed limit is still on the back side of > the power curve. If the autopilot is engaged there, the aircraft will > get stuck on the back side, and never find the high-efficiency climb > regime at lower AoA. > > Try this (since I'm at work and can't): trim for 250 knots only up to > 1 feet, and then push the nose down and accelerate to something > like 300 before engaging the autopilot again (or better yet, trim for > 300 knots and don't engage the autopilot at all). See if the climb > performance in the flight levels improves. I've run many tests on that theory, and trying to find the right way. Even stepping up a couple thousand feet at a time keeping the pitch very gradual and the airspeed up it still dies out in the mid 20kft range. Can you commit that air temperature fix? That sounds like it might be important. If the air is too thin for the altitude, that AoA margin could be very small indeed. Best, Jim ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] YASim solution solution?
Andy Ross <[EMAIL PROTECTED]> said: > Jim Wilson wrote: > > Yeah but look at the values again...we're getting close to tropopause > > value at 23000ft. Mach should be well over 600knots at 23000ft, > > unless it's _really_ warm. > > Mach 1 at the tropopause and above is just about exactly 295 m/s, > which is 573 knots *true* airspeed. The numbers you quote are > "indicated" airspeed (that's the "I" in KIAS). That means that they > are corrected for density (basically by the square root of the density > ratio at subsonic speeds) and are much lower than true speeds at low > densities. At 23000 feet MSL, the correction is about 1.5, which > agrees very closely with the numbers you cite. > Ah...oh. Sorry about that. :-) Best, Jim ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] YASim solution solution?
Andy Ross writes: > Does anyone have good, hard climb numbers for this plane? I mean > stuff like: "At NNN pounds gross weight, XXX feet MSL and YYY knots > TAS, the 747-400 can climb at ZZZ feet per minute." My suspicion is > that we're being bitten by a combination of bad performance numbers > being fed to YASim, and bad pilot climb technique. >From the BADA site I posted a couple of days ago: ftp://bada.eurocontrol.fr/bada/3.3/B744__.PTF All the best, David -- David Megginson, [EMAIL PROTECTED], http://www.megginson.com/ ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] YASim solution solution?
David Megginson <[EMAIL PROTECTED]> said: > Andy Ross writes: > > > Does anyone have good, hard climb numbers for this plane? I mean > > stuff like: "At NNN pounds gross weight, XXX feet MSL and YYY knots > > TAS, the 747-400 can climb at ZZZ feet per minute." My suspicion is > > that we're being bitten by a combination of bad performance numbers > > being fed to YASim, and bad pilot climb technique. > > >From the BADA site I posted a couple of days ago: > > ftp://bada.eurocontrol.fr/bada/3.3/B744__.PTF > That's it...the one I saw the other day. And you can see why the numbers suprised me... e.g. with nominal weight ROCD is 1110fpm at FL 350, 1850fpm at 280. A lot higher than other sources suggest like the ones that Andy found. Best, Jim ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel