F-16.net

Senior member Joined: Jul 01, 2007 Posts: 368

No silly questions here. Anything you don't know is worth asking. Nice loaded reference by the way. Wink

F-16.net Sponsor

Enthusiast Joined: Oct 01, 2007 Posts: 29

OK well then I guess I should just keep going Smile

Today's topic of confusion is autopilot behavior and in particular how ALT HLD works.

Firstly, given reasonably calm air, how well does the autopilot actually hold altitude when you have that mode engaged?? The T.O. says something about plus/minus a couple of hundred feet but it's not clear if it means that it will wander in that range or if it wil maintain a single value that is somewhere close to the altitude when you actually engaged ALT HLD mode. Intuition says that the jet would fly more or less level (presuming enough power dialed in on the throttle lever) without hunting around in altitude much because that's what you'd want. So how good is it at maintaining a set altitude in practice??

Secondly, I imagine that in the normal run of things, you'd only want to engage the autopilot when you have the jet more or less set how you want it held...so if we're going level flight effect with autopilot, you'd reach for the ALT HLD switch (with the left switch centered for roll ATT HOLD) with the jet more or less wings level and FPM on the horizon line etc. in level flight. The interesting part is how far away from the parameters you are aiming for can the jet be??

By which I mean to ask...let's assume you want level flight at 10k feet...jet is presently at 9k climbing with 20 degree nose up attitude, if I flip the ALT HLD switch as the jet passes through 10k and let go the stick do I get level flight or did I just confuse the autopilot terminally?? If it will find level flight from there I presume it must oscillate a bit as it corrects the initial overshoot above 10k feet and then probably has a similar overcorrection to fix on the way back down to 10k again...well, that's a guess...does it settle and if it does, how quickly (say in terms of number of oscillations about the target altitude)?? The question I guess is whether the autopilot can cope with being engaged and asked for a set of parameters via the two control switch positions that are relatively far from present flight condition and still make sense of those to deliver the result.

Senior member Joined: Jul 01, 2007 Posts: 368

The jet maintains a set altitude very well. On recoveries, I set up the autopilot and if I level off at 6K then it'll hold 6K. However, if you set the autopilot with more than a 500 foot per minute climb or descent, then it'll hunt for a while. The best technique I found is to set up the autopilot switches the way I want them while holding the paddle switch. Then I get the jet set up at the parameters I want and let go of the paddle.

I actually use the autopilot all the time. In a CAS wheel it works great because you can set the roll attitude and altitude hold and it'll maintain that as long as you have gas. It works great for holding a 5-7 mile wheel because the jet will just fly a continuous circle. That allows me to work with the TGP and radar sanitization.

So, the short answer to your question is to set the jet where you want it and then engage the autopilot. I don't have the books with me right now but there are limits to which the autopilot will even engage. Also, if you have the autopilot engaged and you do any aggresive maneuvering (in the block 40s at least) then the autopilot will fail and you can't get it back. The autopilot in the Viper is a "hold what I got" kind of system.

Newbie Joined: Nov 07, 2007 Posts: 3 Status: Offline

boxer wrote:

That makes me inclined to think that some of the other "surprises" from the math are probably right too.

Senior member Joined: Jul 01, 2007 Posts: 368

Enthusiast Joined: Oct 01, 2007 Posts: 29

Yes, JP, is a master of the dark side to be sure Razz

Here's a different sort of Q. The dash one seems to be clear about the fact that during start the only round gauges on the right side console that are active are the RPM and FTIT (presuming you have flipped the ELEC main power switch out off the OFF position). There's at least one clip on the net (youtube and other places) that show an engine start sequence where in the NOZ POS gauge is alive and well. In the video I'm thinking of, the jet is in a HAS so its possible he has shore power I guess.

Well anyway, there seems to be some debate about which gauges one can rely on as the JFS runs and the main engine spools up to idle so I guess it's time to ask an expert what they see... I'm guessing you usually power up without a cart or other shore power. What needles are alive and reliable before the main engine powered generators come online??

Enthusiast Joined: Dec 05, 2006 Posts: 42 Status: Offline

Ive got a question which i hope you wont mind answering as well.
When you roll the jet does it roll around the fpm or the gun cross? or somewhere in between?

anyway, boxer, is the dash one public release and if so where did you get one?

Senior member Joined: Jul 01, 2007 Posts: 368

The only gauges I look at are RPM and FTIT. I honestly couldn't tell you what the NOZ is doing. It really shouldn't be moving anyway. The turkey feathers are full open until you are above idle. Also, a common misperception is that the gauge reads what the feathers are actually doing but instead the gauge reads what is commanded.

For Phreotho, I believe that if there is AOA on the jet then it'll roll around the FPM but if there is no AOA (0G) then it'll roll around the gun cross. I honestly never thought about it, that's a good one. Whatever it does in the sim should be correct. The -1 is not public but it's not classified either. I'm sure you can get one from governments outside the US and even some older versions of ours as well. If you call the Govt Printing Office the FOIA should cover the older stuff. Never tried it though. Crazy Pilot

Enthusiast Joined: Oct 01, 2007 Posts: 29

Ok, well if those are the only ones you look at, that would maybe support the idea that those are the only ones awake...maybe Smile

The thing about noz position commanded is interesting -- didn't know that. I think that would be possible to model since that's what the engine code does Very Happy

Next silly question. If you set up with AA TACAN between you and a wingie (say) my understanding is that you only get DME info, not bearing. That leaves an interesting question about what the HSI would show if you place the mode knob for it in one of the TCN positions. I found one reference to F-15A avionics on the web that suggests the bearing pointer will swing at roughly 30 degrees per sec in this case. Is that what happens with the Viper too?? The bonus question in this case is what about tankers -- I heard that KC-10 may have capability to give bearing and DME but the KC-135 can only do DME -- true??

@Phreotho: www.eflightmanuals.com has Viper dash one info that you can purchase...old and/or export model jet info but pretty interesting nonetheless.

Newbie Joined: Nov 07, 2007 Posts: 3 Status: Offline

Phreotho wrote:

Ive got a question which i hope you wont mind answering as well.
When you roll the jet does it roll around the fpm or the gun cross? or somewhere in between?

anyway, boxer, is the dash one public release and if so where did you get one?

One of the main particularity of the F16 airframe is that it has a huge coupling between roll and pitch.

Which means that when you roll, the AC has a tendency to pitch up by itself this is added to the natural coupling inertia coming from the EOM.

On top of that, the more Side slip angle you have , the more roll => pitch coupling you have.

What does that mean ? when you fly at 15 AOA for instance, if you roll around your gun cross axis, that would mean that at 90 bank, the AOA will be converted in side slip angle (beta 15), therefore Increasing dramatically the coupling.

This could then end up in a pitch departure that the FLCS would be unable to catch / prevent.

So the answer of the engineers has been to developp the ARI (aileron rudder interconnect) so that during a roll the AOA is NOT converted to side slip angle but stays AOA.

That means that the AC is rolling around it's velocity vector and not it's body axis.

This is one of the FLCS purpose to do that job through ARI to prevent what we call roll departure (roll finishing in pitch up and AOA > 30 deg)

So the answser is : the FLCS will do everything possible to turn around the FPM.

This is why it is strongly not recommanded to touch the rudder of the F16 during roll because you disturb the ARI and makes the AC turning around a bad vector, increasing the probability of roll departure. At a certain point of roll rate, the actual input rudder pilot is zeroed to make the AC roll safely.

Enthusiast Joined: Dec 05, 2006 Posts: 42 Status: Offline

Wow... Thanks for the info guys...
Btw, what is EOM?

Roll/pitch coupling is more severe when high roll rate is combined with high angle of attack. The F-16 can roll faster at higher AOA than most other airplanes, so the roll/pitch coupling is simply due to that high performance.

ip is correct that the F-16 rolls around its flight path axis, but his reasons are wrong. To begin with, the difference between the flight path axis and the fuselage axis is the angle of attack. So long as the pilot maintains the same g command during the roll, the AOA with be essentially constant, as will the g. So the airplane wants to roll around the flight path axis naturally. If there is significant pitchup, the g command system will keep the g near-constant. Since the AOA is near constant, it is not converted to sideslip angle,so the ARI is not used to prevent the AOA from converting to sideslip.

The job of the ARI is to maintain the sideslip angle near zero. During a roll, up and down deflection of the ailerons and horizontal tails causes an unwanted yaw moment on the airplane, so the ARI deflects the rudder to balance that moment and keep sideslip near zero.

Both of the previous posts have some good and accurate points - but both also have some misleading and/or not-quite-so-accurate statements.

Yes, the job of the ARI is to (attempt to) maintain low sideslip during rolls. When you try to explain exactly how the ARI works, you must be careful with your axis-systems. For simplicity, lets say we have an aircraft with only ailerons and a rudder. Deflecting the ailerons (differentially) gives a moment which acts (primarily) create roll-rate in the aircraft body-axis (the fuselage axis). Similarly, deflecting the rudder tends produce primarily a moment that acts to create yaw-rate in the aircraft body axis (a line perpendicular to the aircraft). (Yes, the ailerons due create some body-axis yaw, and the rudder does create some roll, but we can safely disregard those parts for this basic explanation.)

So why does this matter? Well, sideslip rate-of-change is directly proportional to yaw rate in the stability axis (i.e. velocity vector-based axis). The conversion from our aforementioned body axis to the stability axis is simply angle-of-attack (AOA). Specifically, Yaw_stab=Yaw_body*cos(aoa)-Roll_body*sin(aoa). In general, if we want coordinated (low sideslip) rolls we can assume we want Yaw_stab to be zero. At zero AOA, sin(aoa) is zero, so no matter how much roll rate we create, we need zero yaw rate to coordinate. But, skip up to higher AOA. Let's assume 20 degrees AOA. Doing the trig, we discover that for every 1.0 deg/sec of roll rate (body axis) we have, we need 0.36 deg/sec of yaw rate (also body axis) to balance it out. So, if we are rolling 200 deg/sec, we need about 73 deg/sec of yaw rate for coordination. That's where the ARI comes in. While our ailerons are busy churning out roll rate, the ARI sends a command to the rudder - attempting to generate just the right amount of body-axis yaw rate.

All that brings me to the parts of jp's post that I found a little misleading or incomplete. Specifically, inertial coupling. Since airplanes are typically not built in the shape of a sphere, simultaneous rotational rates in any two axis tend to couple, creating a moment in the third axis. The nature of this coupling comes down to the aircraft mass distribution, but for a typical modern-day fighter, the more signficant terms are roll*yaw into pitch, and pitch*roll into yaw (pitch*yaw into roll is typically less important). Let's go back to our 200 deg/sec roll. At 5 degrees AOA (doing the trig) we find we need only about 17 deg/sec of yaw rate to coordinate, as opposed to the 73 deg/sec we computed at 20 degrees AOA. So, our "roll times yaw" inertial coupling factor (creating moment in the pitch axis) is over four times greater at 20 AOA than at 5 AOA, for the same roll rate. The roll*yaw coupling typically is in the direction to increase pitch rate (and therefore AOA), so the stabs have to work that much harder to keep the aircraft from pitching out - hence, the departure concern. Too much roll rate at too high an AOA can lead to the stabs hitting their stops - which is followed by really bad stuff.

Back, briefly, to the basic ARI discussion. Specifically, why is it important that we coordinate (have low sideslip) in our rolls? There are many reasons - I'll just hit three. First, pilots don't like them. A significant amount of sideslip in rolls tends to make them feel sloppy, loose, or downright uncomfortable. Second: aerodynamics. Left to its own devices (no ARI) an aircraft will usually tend to create sideslip in what is called the adverse (as opposed to proverse) direction. What this means is that the sideslip is in the direction such that the dihedral effect creates a rolling moment that opposes the commanded roll. Too much adverse sideslip and the aircraft can stop rolling completely, and even reverse. (pilots hate that). And thirdly, there is the "trading AOA for sideslip" effect that both previous posts alluded to. There is a term in the buildup of AOA-rate which is "Roll rate * tangent of sideslip". Simply, if you are rolling at non-zero sideslip, you are creating a change in AOA that is not tied to pitch rate. For our 'non-ARI', adverse sideslip case, the signs work out so that the AOA rate is negative, so this is typically more a pilot comfort issue (uncomfortable unload during rolls) rather than a significant departure issue. Some (but not all) FLCS designs automatically augment their pitch commands to account for this term, which helps.

So back to the original question: yes - the goal is a roll around the velocity vector, which requires relatively good sideslip control.

(Oh yeah, and EOM is equations of motion)

Newbie Joined: Nov 07, 2007 Posts: 3 Status: Offline

johnwill wrote:

Roll/pitch coupling is more severe when high roll rate is combined with high angle of attack. The F-16 can roll faster at higher AOA than most other airplanes, so the roll/pitch coupling is simply due to that high performance.

ip is correct that the F-16 rolls around its flight path axis, but his reasons are wrong. To begin with, the difference between the flight path axis and the fuselage axis is the angle of attack. So long as the pilot maintains the same g command during the roll, the AOA with be essentially constant, as will the g. So the airplane wants to roll around the flight path axis naturally. If there is significant pitchup, the g command system will keep the g near-constant. Since the AOA is near constant, it is not converted to sideslip angle,so the ARI is not used to prevent the AOA from converting to sideslip.

The job of the ARI is to maintain the sideslip angle near zero. During a roll, up and down deflection of the ailerons and horizontal tails causes an unwanted yaw moment on the airplane, so the ARI deflects the rudder to balance that moment and keep sideslip near zero.

We are not talking about roll with G's, we are talking about Roll with pitch stick at Zero .

Thanks raptor for the precision about inertia coupling, i was lazy to write the explanation about it Wink

If you are interested in F16 stability issues and FLCS control , i strongly advise you to read this document that explains many mysteries about F16 FLCS , this is particularly oriented toward roll stability and how the FLCS handles it Smile

URL: http://techreports.larc.nasa.gov/ltrs/P ... tp1538.pdf

Senior member Joined: Jul 01, 2007 Posts: 368

Hey Boxer,

For the AA Tacan thing, yeah, the needle will just sit there and spin. But in order to recieve AA Tacan I have to select A-A T/R on the UFC via the Tacan Page and have the correct freq in. For AA Tacan to work, I need to be 63 off of the other guy and both be on X or Y. On the HSI itself, I need to be in just plain Tacan or it won't display. That is just for the Block 30s and below. The new block 40/50 displays the DME for AAT on the lower right of the DED so we use the HSI for other things. As for the tanker, I've never seen one give off bearing info. But that might be different for different jets.

F-16.net

Forum: F-16 Procedures

Ground handling: Braking, NWS and nose gear strut behavior