FS2004 Modeling icing (and its sim lethality)

[Heretic]

I'm pondering about implementing a rudimentary icing simulation into my 737.

I know that FS9,FSX and P3D all model airframe icing, but according to the documentation of the "Ice Gauge", it's a tad flawed.

Summary of the document:
The "icing" flag for the waether basically just adds weight to the airframe and comes in several strengths ("light" to "severe"). Which is good in itself, but there's a problem, namely that the default "real world" weather hardly sets it even if the conditions (humidity and temperatures next to freezing) would be suitable. Third party weather tools mostly have a "random icing" function, but that doesn't always guarantee even a slight impact on aircraft performance.

The ice gauge itself tries to alleviate this by adding "ice weight" to the airframe via a custom L: var and then deflecting the elevators and/or deploying the flaps and/or adding elevator trim at a certain percentage of the current weight or MTOW (IIRC) to provoke an unrecoverable stall and thus a crash.

While being great for "aftermarket" modification of panels and hence aircraft, I, as a developer and master of aircraft source files, can tackle the spoiler system as a starting point for simulated icing drag.
(The user would not notice that the airframe spoilers are serving as a stand-in for icing.)

Its visual implementation isn't much of a problem, but figuring out the right conditions and performance impact, however, is.

So far, I think these conditions are well suited to trigger icing:

(A:AMBIENT PRECIP STATE, mask) 4 ==
(A:AMBIENT IN CLOUD, bool) 0 !=
or
(A:AMBIENT TEMPERATURE, celsius) 1 <
and
(A:STRUCTURAL DEICE SWITCH, bool) 1 !=
and

But:
At what rate would the ice contamination increase?
Should aircraft speed be taken into account when adding icing?
How much drag would be generated by the ice?
Follow-up for the latter:
What would the maximum drag be, i.e. should I aim for a 100% spoiler deflection at maximum icing or is there a usable "cap" (i.e. a negligible increase of ice accumulation) somewhere at maybe 50 or 60% deflection?
Also:
How long would it take for the structural deicing system to get rid of the ice on the wings?


I hope you guys can help me, as it would shorten a potentially long trial and error process.


(With a working system like above, implementing an animated, working ground deicing system into the aircraft model is a possibility.)

 

[Roy Holmes]

For something like a 737, the ice problem is more a case of lost lift than increased drag. Also, the worse effects usually concern the engines. Ice accumulations in the intakes can break away and damage the engines and sensors can give incorrect readings.

Good example here

http://en.wikipedia.org/wiki/Air_Florida_Flight_90

Another factor is that jet airliners do not usually cruise at altitudes where icing is prevalent unlike the older prop types. They usually pass through ice levels quickly in a climb or descent so prolonged build up is unlikely.

However as in the Air Florida story you can get a lot of snow or ice while on the ground and it is highly detremental to airfoil lift coefficients.

As for the ambient conditions the ones you stated are spot on.

Roy

 

[n4gix]

For window glass icing, you can have a series of "ice masks" that will display sequentially based on a timer (say 10-13 bitmaps of increasing "ice."

Toggling on the defrost would take the last value of the timer and count back down from there, displaying the "ice masks" in reverse order to simulate the defrosting.

 

[jx_]

At what rate would the ice contamination increase?

Strictly dependent on conditions. Severe icing can pack a few inches on in a few minutes. Depending on where the ice accumulates, an aircraft can experience symptoms from none at all, to severe instability issues. Sometimes, there's no perceivable change in performance until the pilot applies an input.


Should aircraft speed be taken into account when adding icing?

Only to find the volume of air passing over the frame. The volume of super cooled water is what is most important. Super cooled water is basically water that is cold enough to freeze, but does not contain any pollutants to crystallize on the surface of. Think of the smoke from dry ice. The amount of this type of water in a given unit of air * the amount of air being passed through would give a rough accumulation rate.

A side note: When calculating air temperature for icing, be sure to take into consideration surface cooling. Metal in cold airflow will be a lot colder than ambient air.


How much drag would be generated by the ice?

Drag is usually unnoticeable so long as the aircraft can maintain flight. Lift is usually lost at a much faster rate when accumulation is on the wings. Asymmetrical increases in stall speeds is what leads to most control losses, where one wing is stalled and the other is still flying. When you hear about incidents where the pilot changed something and lost control, this is usually why. There is no way to control an airplane in that condition. As in the case Roy pointed out, airliners are also susceptible to ice which affects thrust output during critical flight phases, which can lead to a thrust deficit.

Lift loss, whether from thrust or aero, is usually more prominent and dangerous than drag increase, although the increase in angle of attack due to lost lift will lead to a higher drag feedback loop.



What would the maximum drag be, i.e. should I aim for a 100% spoiler deflection at maximum icing or is there a usable "cap" (i.e. a negligible increase of ice accumulation) somewhere at maybe 50 or 60% deflection?

What the ice usually does is reduce all safety margins. For example, under CFR 14 part 121, your approach speed must have a margin of 23% over stall speed. That means if your normal dirty stall speed is 122kias, your approach speed is no less than 150kias, but with the ice on the wings, you could be losing enough lift to be in stall conditions at 150kias, or worse yet, you could be nearing stall conditions on one wing. So, while the airspeed looks normal at 150kias and the pilot believes he has 28 knots of buffer, he could be a few knots from stalling. The pitch angle or AoA gauge would be his only real hint.

On the flip side, if the pilot is aware he has lost lift and increases airspeed, the plane will "usually", with some rare exceptions, continue to fly normally at a higher speed (AS IF weight were added, but not because weight was added.)

So if I were you I would look up some icing accidents and figure out the likely safety margin loss for the type of plane you are working on in severe conditions and calibrate your panel to that, just remember different types all react differently.



How long would it take for the structural deicing system to get rid of the ice on the wings?

That depends on the airplane's certification. Different certificates have different requirements and strengths.
The rules are outlined in part 25 and a few others. I will look for a reference. In airliners, it usually takes less than a minute for the ice to melt or break free. I don't know what the actual certification criteria is, but it would be more than what is demonstrated on your aircraft. In other words, if you can't find the specifics on your airplane, but find the regulations, your airplane will have to melt the ice faster than the regulations, 100% of the time, to be certifiable. Please note, there is currently no airplane flying that has ever been certified to fly in severe known icing.


Found a good reference: http://rgl.faa.gov/Regulatory_and_Guidance_Library/rgAdvisoryCircular.nsf/list/AC%2091-51A/$FILE/AC91-51A.pdf

The entire document will probably interest you, but page 4-paragraph (f) references the regulations you may need.

 

[kevintampa5]

At what rate would the ice contamination increase?

Please note, there is currently no airplane flying that has ever been certified to fly in severe known icing.

Well, thats partly correct. The FAA defines severe icing as "a descriptor used operationally by flight crews reporting encountered icing intensity to traffic control. The rate of ice buildup results in the inability of the ice protection systems to remove the buildup of ice satisfactorily. Also, ice builds up in locations not normally prone to icing, such as areas aft of protected surfaces and any other areas identified by the manufacturer. Immediate exit from the condition is necessary."

If you have 2 aircraft, plane 1 and 2, both maxed out with all available anti-ice equipment, they are certified into icing conditions = air bladders, windscreen warmers,
now lets add 2 more aircraft, plane 3 and 4, with maxed out equipment = heated leading edges, heated windscreen.
The first 2 are certified into a set degree of icing, capable of repelling a select limitation, as determined in certification. The same for the other 2 aircraft.
However, lets say the pilot in plane 1 flies into icing, and he determines the icing level is fine, moderate at best. Plane 2's pilot comes in a mile behind, but reports the same conditions, but determines it looks severe to him, now that he acknowledged it is severe, he must exit.
Same scenario could be played in planes 3 and 4.
The level of severe icing really depends on a number of factors, not to exclude the pilots interpretation.
Therefore, if the pilot in command determines the ice is not severe, he does not have to exit, and should be fine.
Once the pilot determines it is severe, even if the level does not yet meet the FAA limitations of severe icing, he must exit, as regulations state that an immediate exit is necessary.

So technically your correct, but its really at pilots discretion as to what severe icing is.

There are only 2 certifications for aircraft, Certified for flight into known icing conditions, and Not certified for flight into known icing conditions. Beyond these certifications, if its certified for flight into ice, its not severe until the PIC determines it is.

Its like VFR flight, in class C, D or E, you must remain 500ft below, 1000ft above, 2000ft horizontally from clouds. But its up to the pilot to determine the distance, as you cant get out and measure it with a ruler. Once the pilot determines he is 450ft below the clouds, he is required to descend to make a 500ft distance, even if he was really 600ft below them when he determined he was only 450ft..

 

[jx_]

Yes Kevin you are absolutely correct. The PIC decides what is severe and thus can override the whole system. If the plane is falling out of the sky and the PIC calls it light icing, he can continue through it if he so chooses.

But I'm sure the FAA would say a PIC is responsible to consider any icing condition where the icing system does not maintain tolerable levels as severe. So part is his discretion and part is knowing his equipment well enough to know what the limit of tolerable ice build up in that specific type is designed to be.

That being said... you are still correct. If he crashes, there's no proof... and if he lands safely, there's no proof.

 

[kevintampa5]

Yes Kevin you are absolutely correct. The PIC decides what is severe and thus can override the whole system. If the plane is falling out of the sky and the PIC calls it light icing, he can continue through it if he so chooses.

But I'm sure the FAA would say a PIC is responsible to consider any icing condition where the icing system does not maintain tolerable levels as severe. So part is his discretion and part is knowing his equipment well enough to know what the limit of tolerable ice build up in that specific type is designed to be.

That being said... you are still correct. If he crashes, there's no proof... and if he lands safely, there's no proof.

Exactly my point, with the exception of the pilot being foolish in waiting til loss of control of aircraft part.

I was merely stating that level of icing is determined by the PIC, but defined by the FAA.
This is for fixed wing aircraft. Rotorcraft and others are covered under different chapters of regulation.
Even though an aircraft is cleared for flight into known icing, it is prohibited from flight if wing frost is present, applying directly to 14-CFR part 91, ops with part 121 , and/or Sub-chapter G parts 125 and 135 operations, unless the frost has been 'buffed' or polished smooth(an approved method of frost polishing as indicated in AC 91-74A).
Taken from AC 91-74A

"Severe Icing. No pilot may operate into known or forecast severe icing conditions unless one or more of the following apply:
(1)
The aircraft has ice protection provisions that meet part 125, appendix C."

continued

"c.
Severe Icing. No pilot may operate into known or forecast severe icing conditions unless one or more of the following apply:
(1)
The aircraft has ice protection provisions that meet part 135, appendix A, paragraph 34.
(2)
The aircraft has ice protection provisions that meet the requirements for transport category airplane type certification.

NOTE: Even airplanes approved for flight into known icing conditions should not fly into severe icing. Many Airplane Flight Manual Limitations Sections require an immediate exit when these types of conditions are encountered. Airplane certification for flight into known icing conditions does not include freezing drizzle and freezing rain. In fact, some airplanes are prohibited from flying into freezing drizzle or freezing rain, regardless of its intensity. These conditions are very dangerous and can cause ice to form behind the protected areas."

So, after reading that, we see that its not prohibited, and that flight into severe icing conditions are legal, as long as specifics are met (as determined by operational status under specified part) and required equipment is used. This is because the AC declared SHOULD NOT and not prohibited. It is also to be determined by PIC the level of icing, based on the definition by the FAA. However, if the icing is Severe, it is covering areas of the aircraft NOT visible to the pilot in most cases (along the fuselage).

I am not here to make argument, just to point out what the FAA has available that kind of contradicts itself or leaves an area for interpretation. And we all know that i see the glass half empty, but you see it half full. Its about interpretation.
If I have a mug of the greatest beer, I will see it half empty for the purpose of a refill. But if its a nasty, gutter flavored beer, the glass is half full; its not empty yet so I can get something different?

 

[jx_]

No argument here, you are correct.

And as we saw in the Colgan incident in 2009, sometimes the PIC doesn't know how bad it is until it is too late.

 

[Heretic]

Okay...FAA regulatory talk from the last three replies aside, thanks for the input guys.

I think I'll just shoot for 100% "spoiler" deflection after 40 to 50 minutes of continuous, unhindered, severe icing*.
That has to be enough to provoke conditions similar to the ones experienced by AF90.

(Note: Since I don't have access to the source code, I can't convincingly fake engine icing.
I could impose an upper "cap" on the throttle levers, but the EPR and N1 gauges would still give away that something isn't right.)

Airspeed will be taken into account for the icing part to simulate cooled air flowing over the wing and the deicing part of the code to simulate ice breaking off.

All in all, this is all I can do without FSUIPC or SimConnect.


*AF90 had ~30 minutes between deicing and take-off. Extrapolating from a deicing fluid efficiency of about 15 minutes, they had a corresponding "safe take-off" window of said quarter hour. The powered pushback, lack of anti-ice system use and the DC-9 spreading the ice over the entire wing effectively nullified this window and made the situation way worse to boot.

 

[kevintampa5]

Okay...FAA regulatory talk from the last three replies aside, thanks for the input guys.

*AF90 had ~30 minutes between deicing and take-off. Extrapolating from a deicing fluid efficiency of about 15 minutes, they had a corresponding "safe take-off" window of said quarter hour. The powered pushback, lack of anti-ice system use and the DC-9 spreading the ice over the entire wing effectively nullified this window and made the situation way worse to boot.

Thats why the FAA changes the rules for ground deicing process after AF90.

 

[jx_]

Yea I thought it was completely off topic too...

 

[Heretic]

Thats why the FAA changes the rules for ground deicing process after AF90.

Well, it would have been unfortunate if they hadn't.

 

[Roy Holmes]

Have a look at this report and the referenced accidents.

http://en.wikipedia.org/wiki/Category:Airliner_accidents_and_incidents_caused_by_ice

Most of the reported accidents were due to ice accumulations when on the ground before take-off.

Not one accident due to airframe icing on a jet airliner, because they do not fly in icing conditions for any length of time. One accident to a jet airliner because the crew left the engine anti icing on at high altitude and the resulting thrust loss meant it could not fly as high as the crew wanted. It ran out of speed on autopilot and stalled.

My flight experience with icing was no engine ice ever, mainly because we flew fast. No airframe icing ever seen, but since the wings were swept you could not see them easily and for sure you would not look out at wings in cloud, because there is no easier way to get disoriented.

Lots of ice inside the canopy when descending from high altitudes. No effect on airplane performance, but it made landing somewhat exciting. Lots of flying ice cubes in the F-4 cockpit on the ground in tropical conditions, when you closed the canopy and tried to demist. Provided you had your visor lowered the experience was quite pleasantly cooling.

The only nasty experience I had with icing was in a Robin DR400 on a cold, clear day when the carb iced up, I was slow to recognise what happened and when I did use the carb heat, it killed the engine, so I cancelled it and landed in a field.

Attached is the article from the local newspaper. Great headline!
As usual, you can not believe everything you read in a newspaper. The statement "Mr Holmes was able to take-off again from the field" while hypothetically correct did not, in fact, occur. The flying club chief pilot did it after explaining to me in great detail why I should have used carb heat earlier.

My point is that, while you may do what you plan to do with drag, I have to say it is unrealistic for a 737. What would be realistic is misuse of K:ANTI_ICE_TOGGLE/SET commands for the engines.

Just a thought.

Roy
UK CPL/IR not current

 

[kevintampa5]

lol.

Well, AF90 was due to airframe and engine icing, at pilot error being the critical chain that lead to it.

If the deicing of the wings had held, the 737 was producing enough thrust for a normal single engine takeoff, but only just, as it is able to continue departure even after rotation and V2, which is usually seconds after liftoff, if one engine had failed.
Although regulations did state that in the event of an anomaly, before or at V1, the aircraft must abort the/rto-rejected takeoff in order to use the accelerate-stop distance available to safely stop the aircraft, which did not happen.
The crash was due to icing that inhibited lift and the engines not being fully spooled providing the necessary thrust, having an epr of 1.70 instead of the 2.04 normally used for the 737-200.
Even at 1.70 (the epr difference to normal was only .34)was not enough to carry an airliner up with some 20-60% of his lift diminished by icing, but is in normal atmospheric conditions (even though it would take about 7000ft to get airborne at that %epr).

"As the plane became briefly airborne, the flight recorder picked up the following from the cockpit, with the sound of the stick-shaker (an instrument that warns that the plane is in danger of stalling) in the background"

"The National Transportation Safety Board determined that the probable cause of the crash was pilot error, citing the flight crew’s failure to use engine anti-ice during ground operation and takeoff, their decision to take off with snow/ice on the airfoil surfaces of the aircraft, and the captain’s failure to reject the takeoff during the early stage when his attention was called to anomalous engine instrument readings"
From the wiki article.

Im no expert, but what I see is two pilots who made a bunch of wrong decisions, all they had to make is one right decision to prevent this accident.

 

[Heretic]

My point is that, while you may do what you plan to do with drag, I have to say it is unrealistic for a 737. What would be realistic is misuse of K:ANTI_ICE_TOGGLE/SET commands for the engines.

Well, I'd put it into the conditional stack, but still: Performance loss by spoilers, not engines.


Unless, of course, you want to help out by coding an entirely new 737 panel.

 

[jx_]

Some interesting facts...

From 1970-2010 there were 140,000 icing incidents in the USA, of which 300 turned into accidents. These accidents included aerodynamic events (degraded performance, loss of control) and terminating events (hard landings, collisions, precautionary landing).


Of those events, jetliners contributed to:

(5) In flight collision with terrain
(6) hard landings
(1) Loss of control

One was lost from stall at cruise, and only AF90 was lost from thrust. The rest were lost due to airframe icing on the ground before takeoff and loss of lift or ice related issues on landing (braking/traction).

The biggest cause of all icing incidents for jet airliners has been air data errors; the second biggest, airframe icing. AF90 falls into the air data category.

The majority of all non-accident jet incidents, occured at cruise, from which the pilots recovered safely and continued. Those are almost all air data or airframe icing, however, severe airframe icing or airframe icing leading to a loss of control is extremely rare in jetliners.


Also I found this, which may help your simulation:

"½ inch of ice can reduce the lifting power of an aircraft by 50% and increase
frictional drag by the same amount."

To calibrate that, figure out your CL at 10,000 altitude, 250 kias, and a set weight, under ISA conditions:


CL = gross weight / q*S

where S = wing area in aircraft cfg

and where q = dynamic pressure

q = mach of 250kias at 10,000ft * mach * 1481.4 * delta ISA of 10,000ft
208.412 = 0.4523 * 0.4523 * 1481.4 * 0.6877


Once you know what your CL will be, multiply that CL * 1.5, then look at the CL vs AoA chart in the air file and find the AoA radian for CL * 1.5.


To convert the radian to degrees, radian * 57.2957.


This is the angle of attack you would get with ½ inch of ice, which is severe. I would call this maximum icing. Write this number down.

Lastly, calculate the difference of CL * 1.5 - CL. Let's say this number is 0.38 * 1.5 - 0.38 = 0.19


Use that result in this formula:

0.19 * ( max spoiler angle degrees / 57.2957 )
0.19 * ( 60 / 57.2957 ) = 0.199

This is the value of the CL_ds spoiler lift entry in the config that would produce the above AoA. If you plan to use the spoilers beyond the amount calculated here, you would need to mark the deflection percent that your value (my 0.19) represents.


If higher accuracy is important to you, you will need to map the highest KIAS and the lowest KIAS in the same manner and map the spoiler extension to that. Otherwise, this should get you what you're looking for.

Hope it helps.

 

[Heretic]

Okay, let's see if the number crunching adds up...

(Note: I'm no Imperialist *Chuckle*, hence I didn't verify if the units of measure work out)

L (gross weight): 100000 lbs (117000 is the maximum)
S (wing_area): 1097.919 ft²
q: 208.412 lbs/ft² (taking your value)

Cl = 100000 lbs / (208.412 lbs/ft² * 1097.919 ft²) = 0.437

Cl_new = Cl * 1.5 = 0.437 * 1.5 = 0.65554

Diving into the .air file and going after table 404 and doing a bit of calculus, I get:
Cl_new_AoA = 0.9229 rad = 52.878 ° (maximum icing)

deltaCl = Cl_new - Cl = 0.65554 - 0.437 = 0.21854

sl (spoiler_limit): 60°

Cl_ds = deltaCl * ( sl / 57.2957 ) = 0.21854 * ( 60 / 57.2957 ) = 0.28855 rad

So that would be the spoiler deflection angle at maximum icing?
(0.28855 * 57.2957 = 13.11°?)

(The .air file states:
Cl_ds = -0.025
Cd_ds = 28)

TBH, I'm not quite sure what to make of this. Shoot for ~22% spoiler deployment (13.11*100 / 60) at maximum icing (and double the value when engine anti-ice is off)?
Mod the .air file accordingly (might still need the "real" spoilers...)?

 

[kevintampa5]

Some interesting facts...

From 1970-2010 there were 140,000 icing incidents in the USA, of which 300 turned into accidents. These accidents included aerodynamic events (degraded performance, loss of control) and terminating events (hard landings, collisions, precautionary landing).


Of those events, jetliners contributed to:

(5) In flight collision with terrain
(6) hard landings
(1) Loss of control

One was lost from stall at cruise, and only AF90 was lost from thrust. The rest were lost due to airframe icing on the ground before takeoff and loss of lift or ice related issues on landing (braking/traction).

The biggest cause of all icing incidents for jet airliners has been air data errors; the second biggest, airframe icing. AF90 falls into the air data category.

The majority of all non-accident jet incidents, occured at cruise, from which the pilots recovered safely and continued. Those are almost all air data or airframe icing, however, severe airframe icing or airframe icing leading to a loss of control is extremely rare in jetliners.

I do not refute this, but the NTSB report citing icing as being a major factor. Even if the crash itself was an air data category of accident, if icing was not present, the aircraft would have been able to safely depart a little further down the runway then normal.
Let me load up 2 aircraft with the same weight, fuel grade, and using that proportion of thrust of 1.70eprv/2.04norm, 1 clean and 1 contaminated (iced up) and I can guarantee you it will crash with contamination and soar without. Lift is the factor that gets you off the ground. Thrust adds the ability to hit that factor. Ice cuts the lift factor. On takeoff, the arcraft is usually 20-30 knots above stall speed at liftoff. If you cut an aircrafts lift by even 30% at that moment, you wont get off the ground without using almost double the distance of a normal takeoff, in standard conditions (59 degrees F/15C) and pressure set to 29.92 at sea level is SAC).

Also I found this, which may help your simulation:

"½ inch of ice can reduce the lifting power of an aircraft by 50% and increase
frictional drag by the same amount."

This really depends on the type of ice and where its located. If its located on the airframe inside the bubble, it will only cause a few % in drag.

Wind tunnel and flight tests have shown that frost,
snow, and ice accumulations (on the leading edge or
upper surface of the wing) no thicker or rougher than
a piece of coarse sandpaper (1/32 or 1/8 of inch) can reduce lift by 30
percent and increase drag up to 40 percent. Larger
accretions can reduce lift even more and can
increase drag by 80 percent or more."
"A NASA study (NASA TM83564)
showed that close to 30 percent of the total drag
associated with an ice encounter remained after all
the protected surfaces were cleared. Nonprotected
surfaces may include antennas, flap hinges, control
horns, fuselage frontal area, windshield wipers, wing
struts, fixed landing gear, etc."
"The wing will ordinarily stall at a lower angle of attack, and thus a higher airspeed, when contaminated with ice. Even small amounts of ice will have an effect, and if the ice is rough, it can be a large effect. Thus an increase in approach speed is advisable if ice remains on the wings. How much of an increase depends on both the aircraft type and amount of ice. Stall characteristics of an aircraft with ice contaminated wings will be degraded, and serious roll control problems are not unusual. The ice accretion may be asymmetric between the two wings. Also, the outer part of a wing, which is ordinarily thinner and thus a better collector of ice, may stall first rather than last"

So we see here that icing is bad, unless its in a nice glass for drinking
Again this isnt to refute, but to elaborate that icing is a factor.

 

[jx_]

Hey,

The CL you came up is correct, however, the AoA radian looks off. Most airfoils stall around 17 degrees and 1.4 CL. That's about 0.30 radians. Confirm your AoA radian at 0.65554 CL is correct.

Also, I see something I may have said in a confusing way. What's listed as CL_ds in the airfile is a number that FS will multiply by the max spoiler deflection radian to output CL_spoiler, the actual coefficient.

If max ice effect = max spoiler deflection, the actual spoiler lift coefficient we would need in flight is CL - CL*1.5.

0.437 - 0.65554 = -0.21854 CL_spoiler

To convert this number to CL_ds, you would divide by max radians (I said multiply in error). Max radians for 60 degrees = 1.0471.


But the two parts where I think I confused you are

(1) CL_ds represents max spoiler deflection (60 degrees), not 13 degrees as quoted below.

Cl_ds = deltaCl * ( sl / 57.2957 ) = 0.21854 * ( 60 / 57.2957 ) = 0.28855 rad

So that would be the spoiler deflection angle at maximum icing?
(0.28855 * 57.2957 = 13.11°?)

You started right by saying CL_ds = deltaCL * (rad) but finished by saying = rad.

Should've been

CL_ds = deltaCL * (sl / 57.2957)
CL_ds = 0.28855


That would be the CL_ds value at 60 degrees (1.0471 radians), had I not told you to multiply in error...

-0.21854 CL_s / 1.0471 max radians = -0.20870977 CL_ds

Because this is negative lift, the result should be negative. Enter this into 1101 CL_ds.


(2) We only need to figure AoA for confirmation tests. To find the value of CL_ds it is simply [ CL - (CL * 1.5) ] / 1.0471.

If you care to confirm the math, the simplest way to find AoA if it falls between lines is to grab the two adjacent lines. Let's say our table 404 indicates:

0.525 CL falls at 0.069813 radians
0.865 CL falls at 0.139627 radians

convert to degrees...

0.525 CL @ 4 degrees
0.865 CL @ 8 degrees


bigger CL - smaller CL = range CL
0.865 - 0.525 = 0.34

CL_1.5 - smaller CL = difference CL
0.65554 - 0.525 = 0.13054

difference CL / range CL = % of difference
0.13054 / 0.34 = 0.38394118 = 38.4% of difference


bigger deg - smaller deg = range deg
8 - 4 = 4

% of difference * range deg = deg to add
0.38394118 * 4.00 = 1.53576471 deg to add

range deg + deg to add = target deg
4 + 1.53576471 = 5.53576471 deg at CL_1.5

So in this example, after applying the correct CL_ds, your AoA at 0.437 CL (approx 2.5 degrees in my example table 404) would change to 5.53 degrees pitch at same airspeed.


(You can check your math by simply taking the CL per degree * degrees to add plus smaller deg. (0.865 - 0.525) / (8 - 4) = 0.085 * 1.5358 = 0.130543 + 0.525 = 0.65554)


Now you can load up FS, maintain airspeed and slowly extend (maintaining altitude with pitch). At full deflection the aircraft should be pitched up by the correctly calculated AoA at same airspeed.


So in plain English, we are forcing the sim to produce 0.65554 CL + (-0.20871 * 1.0471) = 0.437 CL Lift at AoA = 0.65554 CL



Side note: CL_ds, the entry in 1101, is the pre-deflection coefficient value. CL_s is the actual modeled coefficient that affects performance in sim.



FS will multiply CL_ds * max radians to get CL_s at max deflection; CL_s is then converted to lift at 90 degrees * sine of deflection... As you can see, it is important that max spoiler deflection numbers listed in the cfg and airfile:320 are correct. CL_ds * table 320:spoiler = CL_s. Cfg file value is same value as 320 (overrides).



Now... this is all just step one. Step two is drag.... same process, minus AoA and radian conversions.

Cd0 * 1.5 = Cd0_1.5

(remember Cd0 in the airfile is aircraft's CD0 * 2048. As long as you keep that in mind there should be no conversions needed, and all numbers will be whole or whole plus half. Round up if necessary.)

Cd0 1.5 - Cd0 = Cd0 difference


(CL_1.5 - CL) * q * S = Lift difference

Lift difference / ( pi * Aspect Ratio * Oswald efficiency ) = Cd induced

where aspect ratio = span * span / area


Cd0 difference - Cd induced = Cd0 adjusted


Cd0 adjusted / max deflection (1.0471) = 1101 Cd_ds

Fairly simple.


To confirm, set CD0 to CD0 * 1.5 and check thrust settings in sim, then set CD0 back to normal, and check thrust settings with high AoA and spoiler deflected.





Now for step 3...

how much spoiler does the pilot need?

Up until now, we are at a point where max spoiler = max ice, without any consideration for pilot's use of spoilers. Since this can get complicated I would suggest having two inputs that sum. One input from ice system plus one input from pilot. Then you can calibrate the inputs based on your actual CD and CL _ds values, and tie your visual model strictly to the pilot input side. If this was the system used, you would definitely need higher "_ds" values for cases where, for example, the pilot extends spoilers while iced (double effect).


The good thing about using spoilers for this is they general dump more lift than drag created, and icing drag is more negligible, so you may be able to find a happy medium. Other options include using flaps instead of, or in conjunction with, the spoiler trick. You could set your panel to always load with flaps set to slot 3, where slot 3 = clean, then set up a Slat set where the lift scalar is negative, and deflection is zero at slot four, and increases up to slot 0. As ice builds up, your invisible slats deflect from positions 3, 2, 1, to 0. Of course, the visual model would need to work with this as well.


I would need more input as to how your system is set up before I could be of any real help with step 3.

 

[Heretic]

Wow, this is slightly getting out of hand...and I'm really confused.


All I want is:

Max ice = 50% loss of lift (possibly more; accounting for engine icing) = spoiler deflection in ? ° (or rad)


I'm not going to touch the flap system.