[1.4.x-1.8.x] Airplane Plus - R26.4 (Fixed issues/Github is up to date) (Dec 21, 2019)

[blackheart612]

Welp, that's another way of designing it  

[raxo2222]

Is this compatible with FAR and AJE mods?

[blackheart612]

Not necessarily built for FAR but I believe it works just fine. AJE isn't a dependency so it won't function as intended by AJE. I haven't studied how it works too but if somebody wants to patch, it's most probably possible. When I made this, I checked AJE but it was out of date so that was the reason I couldn't take a look at it.

[theonegalen]

I've looked into the AJE config-ing of these engines. Most of them already have configs for the right kind of engine, but several would need brand new configs and apparently the only people who understand how to do new AJE propeller configs are @camlost (who has moved on) and @NathanKell (who is very busy with official SQUAD business).

[NathanKell]

Making a config is fairly simple if you have a BHP-vs-altitude curve for static and for at-speed (i.e. with ram air).

However, all the prop data needs to be updated now that FAR is updated, since many of the props have custom thrust based off old-FAR drag.

[theonegalen]

A moment about propeller efficiency.

Currently, the Bumblebee prop has this velocity curve:

		key = 0 1
		key = 0.15 1.8
		key = 0.35 2
		key = 0.5 2.25
		key = 0.533 2.5
		key = 0.571 2
		key = 0.6 1
		key = 1.2 0

while most of the other props have curves like this (this one is the Whirlwind):

	{
		key = .7 0
		key = .6 .1
		key = .5 .4
		key = .4 .5
		key = .3 .6
		key = .2 .7
		key = .1 .8
		key = .05 .9
		key = .01 1
	}

Real propellers have efficiency curves like this, much closer to the Whirlwind and other non-Bumblebee engines:



This gives widely varying performance across the different engines, when all of them should get their greatest thrust around Mach 0.075m, tailing off to various degrees after that.

Finally, I would like to point out that all the engines except the Marlin and the Bumblebee have "useVelCurveIsp" and "useAtmCurveIsp" set to false, while those two engines have them set to true, which again gives inconsistent performance. If this is intentional, I apologize, but I feel like they should all be consistent.

[Pds314]

Whoa. Attaching a 9J Baron to anything that reaches about ~430 mph results in a short and violent burst of thrust blowing the craft apart.

Edited August 18, 2016 by Pds314

[Pds314]

Uh. Hey. Um... This should not be possible.



A 9J Baron engine with infinite propellant on, at supersonic speeds, at 9 kilometers, at 0.5% thrust, is generating about the same thrust as a Rocketdyne F1 It's also generating 270 times as much electrical power as normal full thrust would.

useVelCurveIsp is a thing? Why don't the stock jets have it? If this works without a bunch  of mods, then I should mod the stock Goliath to have realistic performance, since, IRL, it should be less efficient at high speeds, since it is a high bypass turbofan.

Edited August 18, 2016 by Pds314

[blackheart612]

@Pds314 I've fixed that in a hotfix, this is what happens if I tolerate a joke on the configs  To be fair it isn't supposed to happen because the engine is so weak but I guess people always find a way.

[Pds314]

7 minutes ago, blackheart612 said:

@Pds314 I've fixed that in a hotfix, this is what happens if I tolerate a joke on the configs  To be fair it isn't supposed to happen because the engine is so weak but I guess people always find a way.

LOL. Sorry. Didn't realize it was a joke. I just noticed that it gave a huge number for the maximum thrust, but at Mach 0.7, which would be highly impractical.

Although, come to think of it, maybe I could build a "legit" version that would climb to like 10k using the engine, then use extremely good aerodynamics to reach Mach 0.7 in a dive.

Hmm. I'm doing the math on some of the engines and something is very wrong. The Bumblebee appears to have ludicrous thrust. Like, the sort of thing that would require many megawatts of mechanical power.
 

Edited August 18, 2016 by Pds314

[blackheart612]

The joke was right in the config because the thrust goes "over 9000!" at Mach 0.7. It was already done manually I think, check out a few posts back. 

[theonegalen]

15 minutes ago, Pds314 said:

LOL. Sorry. Didn't realize it was a joke. I just noticed that it gave a huge number for the maximum thrust, but at Mach 0.7, which would be highly impractical.

Although, come to think of it, maybe I could build a "legit" version that would climb to like 10k using the engine, then use extremely good aerodynamics to reach Mach 0.7 in a dive.

Hmm. I'm doing the math on some of the engines and something is very wrong. The Bumblebee appears to have ludicrous thrust. Like, the sort of thing that would require many megawatts of mechanical power.
 

One part of that problem is the thrust curve (velCurve) I posted above. Here's a good replacement:

		key = .7 0
		key = .6 .1
		key = .5 .4
		key = .4 .5
		key = .3 .6
		key = .2 .7
		key = .1 .8
		key = .05 1
		key = .01 .5

 

[Pds314]

Holy bleep! This bleep of an equation is a real bleep.

Where:
x was supposed to be the exhaust speed, y is supposed to be intake speed, k is supposed to be power input, l is supposed to be air density, and m is supposed to be disk area.

I think... I think I may have done goofed something. The equation is supposed to compute the exhaust velocity of an actuator disk given the input power, speed, air density, and disk area. There's no way the result should be such an indecipherable mess.

If anyone can confirm that this is or isn't correct, that would be nice.

EDIT: hmm. Well, it appears to give correct results. It's gonna be a pain to cram that whole thing into an excel spreadsheet.

Hmm... maybe if I stipulate that l (air density) and m (disk area) MUST be positive, the equation will simplify?

Well, it helpfully gives me a particular solution that if there is no power, but the speed, density, and disk area aren't zero, then the exhuast velocity must equal plus or minus the intake velocity... that's rather stating the obvious.

Hmm. It yields a static exhaust velocity equation. That's somewhat helpful.

Edited August 18, 2016 by Pds314

[Pds314]

Here's a 2 Megawatt ideal actuator Disk's thrust curve based on the hideous formula. You'll note that it doesn't peak at 2.5 times static thrust at 180 m/s, but instead dips down to about 11 kN, or about 25% of its static thrust.

What this leads me to believe is that the bumblebee is a monstrously overpowered prop, maybe 38000 HP or something, easily 15 or 20 times what a real double wasp had, with a static exhaust velocity of 180 m/s, but consequently, extremely poor efficiency until 180 m/s, with prop blades forced far into a stall until half the speed of sound. Basically, it sounds like a mini version of the engines on the plane in my profile pic, but only 1 part instead of nearly 200.

(Even then, that doesn't explain why it has a tiny amount of thrust ABOVE that speed, or why it doesn't generate F-35 LiftFan-like static thrust, or why it isn't designed as a blisk instead of a prop, since they can evidently achieve high exhaust velocities. so many questions...)

Edited August 18, 2016 by Pds314

[theonegalen]

 

3 hours ago, Pds314 said:

Holy bleep! This bleep of an equation is a real bleep.

Where:
x was supposed to be the exhaust speed, y is supposed to be intake speed, k is supposed to be power input, l is supposed to be air density, and m is supposed to be disk area.

I think... I think I may have done goofed something. The equation is supposed to compute the exhaust velocity of an actuator disk given the input power, speed, air density, and disk area. There's no way the result should be such an indecipherable mess.

If anyone can confirm that this is or isn't correct, that would be nice.

EDIT: hmm. Well, it appears to give correct results. It's gonna be a pain to cram that whole thing into an excel spreadsheet.

Hmm... maybe if I stipulate that l (air density) and m (disk area) MUST be positive, the equation will simplify?

Well, it helpfully gives me a particular solution that if there is no power, but the speed, density, and disk area aren't zero, then the exhuast velocity must equal plus or minus the intake velocity... that's rather stating the obvious.

Hmm. It yields a static exhaust velocity equation. That's somewhat helpful.

CRIPES

Let me know if this yields results faster than my guess-and-check method of creating thrust curves.

[Pds314]

2 hours ago, theonegalen said:

 

CRIPES

Let me know if this yields results faster than my guess-and-check method of creating thrust curves.

It does. The only problem is that they're for ideal actuator disks, not real propellers, so you need to overlay a prop-efficiency curve with a finite RPM limit, engine RPM/Torque curve and prop blade L/D and take into account mach effects. Note: the output is propwash speed, to get thrust, you must find the mass flow, or you may use the bottom formula.

Exhaust speed (relative to plane)=1/3*((2^(1/3)*(3^1.5 * (27*A2^2*D2^4*E2^4 + 8*A2*D2^5*E2^5*C2^3)^0.5 + 27*A2*D2^2*E2^2 + 4*D2^3*E2^3*C2^3)^(1/3))/(D2*E2) + (2^(5/3)*D2*E2*C2^2)/(3^1.5 * (27*A2^2*D2^4*E2^4 + 8*A2*D2^5*E2^5*C2^3)^0.5 + 27*A2*D2^2*E2^2 + 4*D2^3*E2^3*C2^3)^(1/3))-C2/3

Thrust=((1/3*((2^(1/3)*(3^1.5 * (27*A2^2*D2^4*E2^4 + 8*A2*D2^5*E2^5*C2^3)^0.5 + 27*A2*D2^2*E2^2 + 4*D2^3*E2^3*C2^3)^(1/3))/(D2*E2) + (2^(5/3)*D2*E2*C2^2)/(3^1.5 * (27*A2^2*D2^4*E2^4 + 8*A2*D2^5*E2^5*C2^3)^0.5 + 27*A2*D2^2*E2^2 + 4*D2^3*E2^3*C2^3)^(1/3))-C2/3)^2-C2^2)*D2*E2/2

Where A2 is the power input, C2 is the speed, D2 is the air density, and E2 is the disk area. Copy-and-paste into excel or similar and organize your table appropriately.

A quick and dirty way if you don't want to worry about RPM or torque or swept area or prop blade geometry is as follows:

First, cut off 10-20% of your power. You'll be loosing this due to propellers, bearing assemblies, etc, never being perfectly efficient.

Then, take the thrust at your prop wash speed. Make this the static thrust. At half your propwash speed, increase this by a small amount and interpolate between them.

Next, we need to consider mach effects. Above Mach 0.5 or so, bad things^TM will happen to your prop's efficiency. Remember that the prop has to rotate, and so the tip speed will reach well into the transonic at high speeds. Mess with the efficiency curve so that it reduces gradually from about 80-90% efficient at Mach 0.4 to very poor efficiency at Mach 0.9 or so. In fact, if your prop blades are not very thin or swept, the efficiency will probably be sufficiently low by Mach 0.9 or so as for the blades to produce more drag than thrust due to mach effects, so if you're making a normal prop, and not, say, something with prop blades that have area-ruling taken into account, cut off all thrust before mach 1. If you're making something with exotic blade geometry (really sharp, thin blades with significant sweepback, for example), decrease the efficiency slowly starting at Mach 0.8 to Mach 3 or so.

There is also the effect of advance ratio, but since this only really applies significant problems with props that already reach well into the mach regime, and indeed, is the principle reason for the inefficiency there, it's not really worth modeling in most cases. TL;DR version is that props cannot work efficiently when the forward speed is more than a few times the horizontal speed. If your prop is pitching forward 80 degrees, it'll generate almost as much drag as thrust. This is one of the reasons props avoid the speed of sound like the plague. Combine an L/D of something terrible like 2-4 with the entire prop being at an advance ratio between 2 at blade tip and infinite near the hub and you run into a situation where the prop generates essentially zero thrust. One solution is to just spin the prop faster and faster (like a turbofan blisk!) but there are other issues that basically result in a "prop" that looks suspiciously like a jet intake. Again, see the F-35's LiftFan, which is actually technically a turboprop/turboshaft, and quite possibly the most powerful one ever, at 29000 HP and 89 kN of thrust from a device 1.3 meters across that shoots exhaust out at >330 m/s.

Edited August 18, 2016 by Pds314

[ABalazs]

Strange, I never met this kind of stuff in KSP before, but "axis side" engines are way weaker then their "alliance side" counterparts. Is this some kind of accident, or really, the modmaker tries to implement some kind of "losing side had inferior tech" lie, like ex-soviet gamemakers used to lately? I mean DB-601 clone does 29 kN versus 40 kN M-105P clone? Really? IRL the Klimov engine was 1050-1100 PS vs. 1175-1350 of the DB-601 of the same time. And in this mod the 601 engine is barely stronger than a 150 PS Hispano 8V (25 kN)? Really? How? Why?

[halowraith1]

28 minutes ago, ABalazs said:

Strange, I never met this kind of stuff in KSP before, but "axis side" engines are way weaker then their "alliance side" counterparts. Is this some kind of accident, or really, the modmaker tries to implement some kind of "losing side had inferior tech" lie, like ex-soviet gamemakers used to lately? I mean DB-601 clone does 29 kN versus 40 kN M-105P clone? Really? IRL the Klimov engine was 1050-1100 PS vs. 1175-1350 of the DB-601 of the same time. And in this mod the 601 engine is barely stronger than a 150 PS Hispano 8V (25 kN)? Really? How? Why?

probably just a mistake but then again a lot people like to puff up the reich with superior tech, which actually isnt true.

[ABalazs]

11 minutes ago, halowraith1 said:

probably just a mistake but then again a lot people like to puff up the reich with superior tech, which actually isnt true.

I am not one of those ppl. There is data available about all of these engines, btw., mostly provided by allied tests made during and after WWII. German and Jap. technology was roughly equal to allied, the war won by supplies and strategy.

Edited August 20, 2016 by ABalazs

[blackheart612]

1 hour ago, ABalazs said:

I am not one of those ppl. There is data available about all of these engines, btw., mostly provided by allied test flights made during and after WWII. German and Jap. technology was roughly equal to allied, the war won by supplies and strategy.

There is data, and the data is not simply typed in directly from the source, well, math happens from data to game configuration. So it's not very easy to make an accurate replica. I recognize that data you present from the engines but I dislike comparing by country; if they are different engines and comparable to each other then that's enough for me. There are no sides in making this mod and it would be pointless to take sides in a war that was long over. Notice this was for aircraft enthusiasts (in the OP). 

That said, I apologize for the most recent release's problems, there are issues in performance for the newest engines which are caused by some problems in the configs-this especially applies to the Hispano-Suiza 8. There's also the fact that the older engines that were released were already rebalanced too so the configs change as releases go by. Complaints regarding performance like that are valuable because of this, since we aim to improve them as much as we can. Though on a sidenote, there are different versions of same engines too, which get improved along with the aircraft.

[blackheart612]

Also, Jap is quite racist.

@Artfact

It's quite hard to find references from the real Hawker Fury. Anyways, I was able to make it (although sound effects might be a problem), this is to represent the design of Fury, Hart and Nimrod but maybe we'll base the performance on the Fury.

[kiwinanday]

16 hours ago, blackheart612 said:

Does this mean we're getting a Rolls-Royce Peregrine?

[blackheart612]

@kiwinanday It will have to wait if you want it. Since we're already past its time in the roadmap

[kiwinanday]

18 hours ago, blackheart612 said:

@kiwinanday It will have to wait if you want it. Since we're already past its time in the roadmap

I'm not even sure what the RR Peregrine's role would be—it was a smaller displacement V12 than the Marlin (21l vs 27l), but it wasn't small enough to be a 0.625m engine, so it'd just be a low-powered 1.25m.

I can't think of what gaps need filling—NK-12? 

[blackheart612]

@kiwinanday That's modern technology It is yet to come.