By the Numbers: Carrier Landing Geometry & the IFLOLS

I can't believe I did an hour on this. This might be a cure for insomnia. I thought some of you would find it interesting though.

After the first attempt to visualize the landing geometry envelop using 3D models resulted in the discovery that the carrier model is too small and ill proportioned, I took another stab at it using the real numbers. It's not as pretty to look at but I think this is correctly depicting what's really going on during a landing.

EDIT: Fixed youtube tags.

Video is not working...

Video is not working...

I actually went through the whole pot of coffee watching this... man! you engineers and your analytical brains:book: Tedious work but somebody has to do it. Thanks.

Wow, impressive work...thank you for the research. If your data is correct, which it seems to be...this is a serious problem.

Can one of the forum moderators merge this with the other posts regarding this topic and move this to the appropriate bugs forum?

I can only hope that we get this corrected with the new aircraft carrier module and that they aren't using the same data from the old one.

On of the comments I’ve heard repeated from the real pilots is that the DCS Hornet feels draggy or that it’s difficult to energize the ball, IE get the ball moving in response to throttle inputs, during the approach. I’m wondering if this could be related to the implementation of the meatball.

If we basically have 12 lights that light up linearly depending on the glideslope, a lot of definition of the system is lost. From the way it sounds, when you’re calling the ball at 3/4’s of a mile, to move the ball up or down you need to move the aircraft vertically 12 feet.

Thus when you’re behind the boat adding power, it looks like nothing is happening because the meatball is a kludge. Thus it makes it seem like there is something intrinsically wrong with aircraft, While it’s simply the feedback system causing the perceived issue.

This also sheds some light on why rolling out on a high ball tends to lead to more ramp strikes. Since the top ball is lands you so far forward. The amount of sink rate you need drop on the center ball from the top could exceed the craft’s ability to recover.

Ie you come in high ball, you have a lot of vertical space to get down, but the craft is configured for max lift, so it really doesn’t want to drop. Thus you have to take a lot of power out to get the ball moving, by the time you’re seeing center ball, it’s all ready to late for recovery, and you ramp strike. You could probably make an interesting case for making the top 2 lights on the IFLOLS instant wave offs like the bottom two.

From the way it sounds, when you’re calling the ball at 3/4’s of a mile, to move the ball up or down you need to move the aircraft vertically 12 feet.

This is accurate. To put the hook on the target point, the dimension of the space the pilot needs to put his head through is 11.4 feet at 3/4 miles, and 1.1 feet at touchdown.

Thus when you’re behind the boat adding power, it looks like nothing is happening because the meatball is a kludge. Thus it makes it seem like there is something intrinsically wrong with aircraft, While it’s simply the feedback system causing the perceived issue.

On the way to the deck, power movements have to decrease in magnitude and increase in quickness, as their effect on the ball increases.

This also sheds some light on why rolling out on a high ball tends to lead to more ramp strikes. Since the top ball is lands you so far forward. The amount of sink rate you need drop on the center ball from the top could exceed the craft’s ability to recover.

Only if the pilot attempts to re-center the ball too aggressively (too soon.) In general, higher on the ball is considered safer than lower on the ball.

Ie you come in high ball, you have a lot of vertical space to get down, but the craft is configured for max lift, so it really doesn’t want to drop. Thus you have to take a lot of power out to get the ball moving, by the time you’re seeing center ball, it’s all ready to late for recovery, and you ramp strike. You could probably make an interesting case for making the top 2 lights on the IFLOLS instant wave offs like the bottom two.

Not really, they're just going to earn a bolter. The key is to gradually bring the ball down rather than attempt to center it aggressively. There's a point on the glideslope at which 4 balls high (top of the lens) is just too high to recenter, and a stable ball at 2 balls high is the safe option. The fear is a "fly through down" or settle in close.

This is accurate. To put the hook on the target point, the dimension of the space the pilot needs to put his head through is 11.4 feet at 3/4 miles, and 1.1 feet at touchdown.

 

 

 

On the way to the deck, power movements have to decrease in magnitude and increase in quickness, as their effect on the ball increases.

 

 

 

Only if the pilot attempts to re-center the ball too aggressively (too soon.) In general, higher on the ball is considered safer than lower on the ball.

 

 

 

Not really, they're just going to earn a bolter. The key is to gradually bring the ball down rather than attempt to center it aggressively. There's a point on the glideslope at which 4 balls high (top of the lens) is just too high to recenter, and a stable ball at 2 balls high is the safe option. The fear is a "fly through down" or settle in close.

I think he was referring the lack of fidelity within each light bulb.

A very well done video thanks.

This is accurate. To put the hook on the target point, the dimension of the space the pilot needs to put his head through is 11.4 feet at 3/4 miles, and 1.1 feet at touchdown.

The issue is that in real life the ball will move up and down that within 11 foot space. In the sim it's a binary. So it's; center light on, then when you climb 12 feet, the light above the datam goes on. There is no half ball. While the real device has infinite resolution. The easiest solution would be to subdivide each virtual lenses into 3 sections and allow the tops and bottoms to overlap. The only issue would be that center balls would appear dim compared to half balls.

On the way to the deck, power movements have to decrease in magnitude and increase in quickness, as their effect on the ball increases.

 

Only if the pilot attempts to re-center the ball too aggressively (too soon.) In general, higher on the ball is considered safer than lower on the ball.

 

Not really, they're just going to earn a bolter. The key is to gradually bring the ball down rather than attempt to center it aggressively. There's a point on the glideslope at which 4 balls high (top of the lens) is just too high to recenter, and a stable ball at 2 balls high is the safe option. The fear is a "fly through down" or settle in close.

“More ramp strikes occur when the pilot is correcting for a high deviation in-close than for a low deviation. “

“High at the ramp with less than optimum rate of descent can lead to a dangerous long bolter”

These statements come from an comprehensive analysis of carrier landing accidents. Part of the reason why this aforementioned conditions tended to result in accidents was that the LSO’s tended to wait till in close to issue glideslope corrections.

https://ia800109.us.archive.org/8/items/DTIC_ADA135823/DTIC_ADA135823.pdf

Also here is a nice pic of IFLOLS up close, you can see how only a small portion of lens is actually lit.

The angle of the lens seems to be a little off as well. Very rarely in DCS will you trap with a high ball, when irl if the ball is even full tall on the lens you will trap on a 3 wire boat.

The angle of the lens seems to be a little off as well. Very rarely in DCS will you trap with a high ball, when irl if the ball is even full tall on the lens you will trap on a 3 wire boat.

Do you have anything that backs up that statement with respect to real life? I'd love to correct my model to it if so.

Matthew, thank you for providing that very thorough explanation of the IFOLS and how it works. Now that I have a much better understanding, not to mention appreciation, of the concept of the "hook to eye" dimention, it ocurrs to me that when HB delivers their Tomcat, it is going to be quite an adjustment for us to consistantly get a 3 wire while flying it. Folks have already mentioned the problems they are having with the Scooter mod. I wonder how, or even if ED will deal with this???

One correction that I would recommend is that the entire model of the IFLOLS be raised by 1.792 feet along the models vertical axis to correctly position the hook to strike between the 2 and 3 wires.

Right now, the IFLOS position relative to the wires will put the hook down 10 feet before the 2 wire.

The wires are incorrectly positioned relative to each other, the IFLOLS, and the ramp due to the incorrect scaling of the model.

Come find us on Discord if you ever want to discuss any of this - https://discord.gg/eVtu42R

I always felt the Stennis was too small, excellent research!

Thanks for your hard work! Hopefully a DEV will see this.

Not to belabor the point but even if ED corrects the dimensional errors, it still seems as though there will be a significant difference between the hook to eye distance between the Hornet and the Tomcat. I am just curious if the ED folks have given this any thought.

thank you for posting - great video.

I once had a French F-8 pilot explain to me that to trap on the old French CVs in the 80's & 90's the F-8 had to dump fuel to a suitable landing weight that ONLY gave enough fuel for 2 looks at the deck. In a Sea State 4 and above, even the most perfect flown approach (this is where your geometry makes good sense) the "vertical" height of the 4 wires was only about 5 feet (3 degree glidepath) and the ship deck would be pitching thru' more than 10 ft!

That earned my respect :)

AG Razor (& others) - good spot, I forgot that the lens is corrected for different aircraft types - and with the promise (F-14, A-4 even an A-6 rumoured to be in the pipeline) poor old DCS developers might need to consider this in developing their excellent simulator.

Not to belabor the point but even if ED corrects the dimensional errors, it still seems as though there will be a significant difference between the hook to eye distance between the Hornet and the Tomcat. I am just curious if the ED folks have given this any thought.

I realized that too. I wonder how they will work around that. Currently the IFLOLS lights are based on player eye point. Just F2 when your on the deck and you can move your head through the cone and see what it does. This means every player sees the same thing. They'd have to have the ability to dynamically adjust the tilt angle for planes with different H/E values. Maybe just somehow make the IFLOLS lights a client side thing so that it "self adjusts" to the model the player is occupying. I really don't know how they'll do it.

Maybe the a/c type could be put in the comms menue when checking in with the boat?

Is it sad that i thoroughly enjoyed every minute of that video??!!

Well done on all that work. Lets hope ED do correct some of the modelling errors for, as you say, an authentic experience.

Maybe the a/c type could be put in the comms menue when checking in with the boat?

IRL they must land the F-14's altogether then the F/A-18's then E-2 Hawkeye etc. To have the correct height setup (When the F-14 was still in service).

That might be hard to do in DCS multiplayer, when there's F-14's and F/A-18's all doing Case 1 around the ship together.;)

As you can see there is quite the difference of 3 Feet between the F-14 and F/A-18. Thanks Curly

That is outdated. The Legacy Hornet H/E is 16.35.

Enjoyed the video, thanks.

On the A-4 I have the impression that a perfect meatball and on speed results in just missing #4 wire.

Enjoyed the video, thanks.

 

On the A-4 I have the impression that a perfect meatball and on speed results in just missing #4 wire.

Do you know the on speed H/E value; I can tell you where it will currently land on the Stennis.