Flight Director - evil or genius?

[Mikoyan89]

Are you sure about this? From what I've heard in the past, the autopilot is downstream of the controls, and doesn't feedback its inputs to the pilot......

Yes,I don't think there's a helicopter with SAS/AP and the autopilot feedback its inputs to pilot,it should be completely transparent

[Haril]

Surely the easiest way to determine this would be to look at the stick in the cockpit (or the ctrl+enter display) when the autopilot's straining and see if it's in a different position that when it's not? I don't believe the cockpit stick moves at all unless the user moves their controller.

 

It's also known that the collective's altitude hold adjustments aren't fed back to the pilot - the collective is unable to move up or down whilst the brake is engaged, and yet the actual collective still adjusts itself to hold altitude/height.

[AlphaOneSix]

It depends on where the autopilot inputs occur. On some aircraft, the inputs are directly into the hydraulic actuators. The hydraulic actuators are non-reversible, meaning that moving the actuators does not translate back into the flight controls. On other aircraft, autopilot inputs are accomplished through a trim servo, which is attached to the flight controls and usually also contains the springs for the trim feel. In aircraft with trim servos, the autopilot inputs do move the flight controls in the cockpit.

 

A quick example of an aircraft that uses inputs directly into the hydraulic actuators is the Mi-8/17, and in that aircraft, the flight controls do not move in response to autopilot inputs. The AH-64A is similar. Note that both of these aircraft do not have true autopilots, but SAS/SCAS instead (dampening and basic attitude/hover/heading hold functions).

 

My employer operates a version of the Mi-172 that has a much more advanced autopilot, which includes both the original Mi-8/17 autopilot for dampening, as well as a Honeywell autopilot system using trim servos. Inputs from the factory autopilot do not translate back into the flight controls, but inputs from the Honeywell autopilot do.

[Nate--IRL--]

Hmm interesting. Does that mean, in those aircraft with no AP feedback, that in the situation I described above, there would be a sudden loss of 20% control input on pressing the trim button?

 

It just seems an odd design decision if so.

 

Nate

[press]

My employer operates a version of the Mi-172 that has a much more advanced autopilot, which includes both the original Mi-8/17 autopilot for dampening, as well as a Honeywell autopilot system using trim servos. Inputs from the factory autopilot do not translate back into the flight controls, but inputs from the Honeywell autopilot do.

 

I am quite familiar with the Mi-17...being an engineer.....but never heard of that layout. Can you give some more details ?

[SFJackBauer]

FD's main drawback is workload

 

Unless you have mapped most of the critical functions to your HOTAS like I did. I dont need to take my hands off controls much, therefore no need to rely always on autopilot. Perhaps that's my "sin" - in the real thing the pilot would have to use both hands several times to manipulate the avionics during flight?

 

When you are setting up an attack, usualy at range, you need to focus on other things rather than flying the helicopter manually

 

Since im flying HOTAS, I also do not need to concentrate much on flying the chopper, it comes naturally. I can keep scanning outside, check ABRIS/HSI for navigation etc.

 

...

 

Frederf, you made a good post, but you didn't brought anything that could not be done with FD on, or that could be done worse. Except for the fact that in the real chopper the pilot would need to be with his hands off controls during nav (but during engagement, I doubt), no specific, objective downside to using FD has ever been brought. I would be happy to be proven wrong on this point though :)

[Frederf]

A bit of a miscommunication there. I was referring to the pseudo centrepoint the DCS trim system gives not the attitude values given interpreted by the AP.

 

Nate

 

I don't consider the non-FFB centering and FD to be related in any way.

 

You are flying along at 250km/hr, you last trimmed attitude at say 210km/hr and this attitude accelerated you to 250k/h. The attitude hold input into the AP is held even though the aircraft has increased in speed.

 

That's normal. The pitch hold holds pitch just as advertised.

 

The AP is now using almost all of its 20% control authority to maintain the Attitude, in a real KA-50 this is obvious with the stick position, in DCS you have no way of detecting this. What happens when the trim is pressed?

 

The 20% control input from the Autopilot disappears instantly and the Aircraft will now change attitude to reflect this. The result can be extreme. Note this is without moving the cyclic on my desk, just be pressing the trim button we get a 20% change in control input.

 

Ballooning, right? It sounds like you're trimming before the aircraft has stabilized. Saturating a hold will cause it to be missed when it's suspended. Don't saturate the holds and the AP won't be on the brink of control so much.

 

The holds are supposed to be "sticky spots" in attitude. That's the appeal not the drawback.

 

This will not happen in the real Ka-50. The position of the controls in the Real Ka-50 reflect the Centrepoint + autopilot, DCS only relfects the centrepoint.

 

This is not true. The controls in the Ka-50 don't move based on AP input. The channels are tacked on to the pilot input stream before it gets to the actuators. Stick input is stick input; it's a one-way street.

[Nate--IRL--]

I don't consider the non-FFB centering and FD to be related in any way.

 

 

 

That's normal. The pitch hold holds pitch just as advertised.

 

 

 

Ballooning, right? It sounds like you're trimming before the aircraft has stabilized. Saturating a hold will cause it to be missed when it's suspended. Don't saturate the holds and the AP won't be on the brink of control so much.

 

The holds are supposed to be "sticky spots" in attitude. That's the appeal not the drawback.

 

 

 

This is not true. The controls in the Ka-50 don't move based on AP input. The channels are tacked on to the pilot input stream before it gets to the actuators. Stick input is stick input; it's a one-way street.

 

Yes it seems I misunderstood the design logic of the AP system. However there is no indication of AP channel saturation. As I pointed out in the scenario above, simply increasing speed with no other control input can lead to the AP using more authority, with no indication it is doing so. Trimming then removes this authority so where you just wanted to trim heading you now must retrim all controls.

 

The reason I believe FD is popular, is that what you see is what you get. Pressing the trim button does not include any other unwanted control input, with FD on.

 

Nate

[AlphaOneSix]

I am quite familiar with the Mi-17...being an engineer.....but never heard of that layout. Can you give some more details ?

 

I't s a mod from Kelowna Flightcraft. Our aircraft's designation is the Mi-172KF, but I have also seen the designation Mi-17KF for other aircraft. Kelowna Flightcraft is a company in Canada that works directly with Kazan to mod the aircraft with updated instruments, avionics, glass cockpit, etc. during manufacturing.

[AlphaOneSix]

Hmm interesting. Does that mean, in those aircraft with no AP feedback, that in the situation I described above, there would be a sudden loss of 20% control input on pressing the trim button?

 

It just seems an odd design decision if so.

 

Nate

 

Don't know. The Mi-17 AP has 20% authority, but I've never felt it jerk when trimming. When you press the trim button on the Mi-17 (and the AH-64A, incidentally) the AP input is "zeroed", or centered, whatever you want to call it. The AH-64A autopilot (actually for the AH-64A it's called Automatic Stabilization Equipment, not autopilot) is limited to 10% authority, except for the forward pitch channel, which is 20%. On the AS332 Super Puma, it's 5%.

 

On the Mi-17, I don't think I've ever had the AP come anywhere close to using it's full authority. Anytime anyone moves the controls, they always immediately press the trim button. Also, the Mi-17 has feedback transducers that tell the AP when the pilot is moving the controls, specifically so it knows to not "fight" the pilot even if the trim button is not pressed.

[sweinhart3]

I personally like to FD for fixed weapons employment, A2A manuevering, and landing. All other times I use the regular AP channels. I find that even if the aircraft is trimmed well it still takes stick corrections to stay on path in FD mode. I also have not had any problems with the new trimming method or fighting AP authority during flight transition. Its not difficult to keep trim engaged while transitioning and letting go when Im at the attitude or speed I desire.

[Frederf]

A16, about the classic "ballooning" symptom. Is that something that would normally happen on your Mi-17 or is there some sort of counter mechanism? If the aircraft would attain a 5 degree nose down attitude with +0% AP hold input and currently the attitude hold is using -15% authority in pitch to increase the nose down attitude... would suspending that hold cause a pitch up moment?

 

Does moving the stick with the trim button depressed remove the artificial feel? Are there a secondary set of (light) artificial feel springs that provide resistance even when the trim button is held?

[AlphaOneSix]

With regard to your first question, I don't really know. There is no way to tell how much input the autopilot is giving at any point in time. The aircraft has never so much as moved a hair when pressing the trim button, even after moving the controls a bit without pressing the trim button first.

 

Second question, yes, once you press the trim button, all artificial feel is gone. There is still just a slight bit of resistance inherent in the controls, but it's very loose.

[average_pilot]

Anytime anyone moves the controls, they always immediately press the trim button.

That's what it took time to realize when I was novice and tried to fly BS like an airplane. It should be advertised in a big banner at the forums front page :)

[Yurgon]

Far from evil, but not genius either

 

Thank you guys for all the input. From what I've read, the Flight Director discussion can be broken down to this: FD is great for hands-on flying and maneuvering. However, to reduce workload, pilots are discouraged to do hands-on flying all the time, they should instead have the heli assist them whenever possible - which is especially important in combat.

 

I've flown a couple campain mission with very little FD assistance and I have to say, it does make things easier to have the aircraft maintain specific flight parameters so that the Shark can be flown hands-free most of the time.

 

I still think FD is brilliant when a lot of manual input is required - evading enemy fire (Yes, I'm too close. I know. But it happens. A lot, actually... :music_whistling: ) would be a good example and I couldn't imagine aerobatics without FD. IMO, it does come in very handy when needed, especially given its overriding nature, but I think I might actually get used to keep it switched off most of the time.

[GGTharos]

FD is great for hands-on flying and maneuvering. However, to reduce workload, pilots are discouraged to do hands-on flying all the time, they should instead have the heli assist them whenever possible - which is especially important in combat.

 

That is not a correct statement. They are NOT discouraged. They are outright trained and expected to do it this way, and they further state that there's practically no need to use the FD, ever.

 

I still think FD is brilliant when a lot of manual input is required - evading enemy fire (Yes, I'm too close. I know. But it happens. A lot, actually... :music_whistling: ) would be a good example and I couldn't imagine aerobatics without FD. IMO, it does come in very handy when needed, especially given its overriding nature, but I think I might actually get used to keep it switched off most of the time.

 

You don't need FD to evade enemy fire ... besides, it isn't going to let you perform evasive maneuvers any better than just hitting the trim switch right when you need it.

[Yurgon]

That is not a correct statement. They are NOT discouraged. They are outright trained and expected to do it this way, and they further state that there's practically no need to use the FD, ever.

A) You are referring to real pilots. I am referring to a simulation.

B) If there's no need to ever use the FD, why is it there?

 

You don't need FD to evade enemy fire ... besides, it isn't going to let you perform evasive maneuvers any better than just hitting the trim switch right when you need it.

Didn't say "need", said "comes in handy", which I strongly believe it does. With FD on, there's no fighting the Shark, it's more like the Shark helps (me|you|a pilot|whomever) flying exactly where the pilot points the aircraft rather than a battle between pilot input, aircraft input and pilot override.

 

For now, I believe that the Flight Director is there when needed/desired by the pilot. You don't want it? Fine. I do? Fine as well. Seems to me it's quite reasonable to use FD in certain situations depending on pilot favor.

[GGTharos]

A) You are referring to real pilots. I am referring to a simulation.

 

Okay; then we were not on the same page.

 

B) If there's no need to ever use the FD, why is it there?

 

As I said, they said they rarely use it if ever. It's there because it was designed in - I'm not sure how else to answer you that. It's like asking why you have a standby horizon or magnetic compass.

 

Didn't say "need", said "comes in handy", which I strongly believe it does. With FD on, there's no fighting the Shark, it's more like the Shark helps (me|you|a pilot|whomever) flying exactly where the pilot points the aircraft rather than a battle between pilot input, aircraft input and pilot override.

 

Never had a problem with it ... hit trim, turn, you're flying in a new direction, release trim. For an emergency maneuver this is adequate, and anything more is sheer panic IMHO.

 

For now, I believe that the Flight Director is there when needed/desired by the pilot. You don't want it? Fine. I do? Fine as well. Seems to me it's quite reasonable to use FD in certain situations depending on pilot favor.

 

That's what I said; in simulation you get a choice. In RL operations it is simply not used.

[Frederf]

I could see it being very useful for training, FD mode. It shows what "right" is for a level turn and can train the pilot's hand to fly perfectly, sort of an instant grading feedback on textbook maneuvers.

[pakfront]

This may have been mentioned, but:

One thing I missed for a while was that the AP control toggle switch labelled DT/DH (desired track/desired heading) has a middle position as well. This turns off waypoint/steerpoint/ingresspoint input from the navigation system. Instead, whatever heading you released the trim is now your AP's 'destination'.

 

Once I found this mode, i found doing manoeuvres much easier than with FD ON and without unwanted AP inputs.

[Yurgon]

Okay; then we were not on the same page.

Apparently. The whole point of starting this topic was to find out why using the Flight Director in BS was considered bad although to me it felt so comfortable.

 

As I said, they said they rarely use it if ever. It's there because it was designed in - I'm not sure how else to answer you that. It's like asking why you have a standby horizon or magnetic compass.

Concerning standby instruments, redundancy seems to be a very good reason to me: If one instrument fails, it's very good to have a secondary or even tertiary device. I don't see how this relates to the Flight Director, though.

 

I remember a quote from the Falcon 4.0 manual where the radar's Velocity Scan mode was described as being utterly useless and being somewhat of a Westinghouse show-off (IIRC), so in my opinion Kamov surely deserves the right to build in something useless. Except... the reason I started this topic is that in my opinion the Flight Director is far from useless.

 

Never had a problem with it ... hit trim, turn, you're flying in a new direction, release trim. For an emergency maneuver this is adequate, and anything more is sheer panic IMHO.

How about NOE flight? It requires constant adjustments and extreme pilot consciousness. IMO, FD makes this much easier than flying the Shark wíthout FD. For now, I don't think I'll ever do NOE or aerobatics without FD on, because that just seems to add to the workload instead of decreasing it.

 

Same goes for evasive maneuvers: given that the FD overrides pretty much any aircraft input and simply keeps the dampers active, that seems to be extremely helpful for just about any sudden maneuver because there is no struggle between pilot input and aircraft counter-input. Activating FD anytime does override Route Mode and even Auto Hover, doesn't it?

 

Seems to me it's quite reasonable to use FD in certain situations depending on pilot favor.

That's what I said; in simulation you get a choice. In RL operations it is simply not used.

I'm not entirely convinced about RL use of FD mode. Guess we'd have to ask a real Ka-50 pilot. As far as BS is concerned, I see little reason not to use it under the conditions described.

 

I could see it being very useful for training, FD mode. It shows what "right" is for a level turn and can train the pilot's hand to fly perfectly, sort of an instant grading feedback on textbook maneuvers.

Seems logical. I can't say anything about RL pilots, but I would probably recommend that DCS:BS pilots start flying the Shark with FD on because the aircraft responds so much more directly and intuitive. It'll be interesting to see how Miguez is going to teach this in the Leading Edge Trainings (I've learned the entire startup procedure from the first LET video).

 

Coming back to the original question of this topic, I still think that FD is far from evil and does come close to genius once in a while. However, my main misconception was that the Shark had to be flown hands-on 24/7, which is wrong. It's designed so that it can be flown hands-off, and FD counteracts that. However, when hands-on flying is required, IMO FD is the best tool available.

[GGTharos]

How about NOE flight? It requires constant adjustments and extreme pilot consciousness. IMO, FD makes this much easier than flying the Shark wíthout FD. For now, I don't think I'll ever do NOE or aerobatics without FD on, because that just seems to add to the workload instead of decreasing it.

 

It's quite easy when you're doing it at 30KT (about 50-60kph) as SOPs dictate.

 

Same goes for evasive maneuvers: given that the FD overrides pretty much any aircraft input and simply keeps the dampers active, that seems to be extremely helpful for just about any sudden maneuver because there is no struggle between pilot input and aircraft counter-input. Activating FD anytime does override Route Mode and even Auto Hover, doesn't it?

I don't see a need for any violent maneuvers. If in the helicopter you cannot avoid something with a hard turn, you're not going to avoid it.

Most people consider a violent, practically uncontrollable maneuver to be 'sudden'. They are most often entirely unnecessary. If you have an emergency, you hit the emergency AP disengage switch.

 

I'm not entirely convinced about RL use of FD mode. Guess we'd have to ask a real Ka-50 pilot. As far as BS is concerned, I see little reason not to use it under the conditions described.

ED did ask real Ka-50 pilots. They don't use it.

 

Seems logical. I can't say anything about RL pilots, but I would probably recommend that DCS:BS pilots start flying the Shark with FD on because the aircraft responds so much more directly and intuitive. It'll be interesting to see how Miguez is going to teach this in the Leading Edge Trainings (I've learned the entire startup procedure from the first LET video).

It is the correct way to start learning because it does simulate a training helicopter.

 

Coming back to the original question of this topic, I still think that FD is far from evil and does come close to genius once in a while. However, my main misconception was that the Shark had to be flown hands-on 24/7, which is wrong. It's designed so that it can be flown hands-off, and FD counteracts that. However, when hands-on flying is required, IMO FD is the best tool available.

 

That is essentially correct. It should also give you the impression that hands-on flying is seldom required ... even during relatively close combat. It's all trim-move-stick.

[Kaiza]

I would say the FD would only be used as an enroute tool when the pilot wishes to hand fly but wants added SA for inclement weather/night ops. It would be great for VNAV with a low cloud base.

 

I would doubt that it is ever used to teach pilots how to hand fly, as pilots are taught based on outside cues, particularly the horizon. Teaching someone to fly off the FD is instilling bad technique for VFR flight.

Edited June 29, 2010 by Kaiza

[CAT_101st]

All I know and can add to this thread is that I don't like to use FD. But some of the sim pilots in my groop love using it.

[EtherealN]

It's quite easy when you're doing it at 30KT (about 50-60kph) as SOPs dictate.

 

Bah. The best way (well, okey, fun way :D ) to do NOE is 200 km/h and FD absolutely off. I've never felt a need for using FD to somehow make anything easier during NOE and is really confused by the suggestion that it would indeed be easier. As you said GG, you get basically the same function for maneuvers through just pressing trim, but get stability augmentation too when you want it.

 

I've tried flying with FD a couple times, and I really didn't see the point of it. It doesn't give me any capability I didn't already have, but it does remove some very very good capabilities from my list of options.