F-35 FLCS CLAW design

[steve2267]

In the Bee, I have read that in hover / vertical lift mode, the rudder pedals still command yaw (rotation around aircraft Z-axis). Fore/aft stick which would normally control pitch instead controls altitude: push forward to go down, pull back to go up. Left / right stick, instead of commanding yaw commands left / right translation. After reviewing some other threads, it appears that the throttle controls forward / aft movement. Not quite sure how the reverse part works, but one thread mentioned using the throttle to move rearwards.

Now I assume that in conventional flight, the controls operate similarly to other fly-by-wire systems, that is, they command a certain rate in one axis:

• rudder pedals command yaw rate
• stick left/right commands roll rate
• stick fore/aft commands pitch rate

If I have this wrong, please correct me.

So if I command left roll by pulling the stick left, the FLCS will move whatever flight surfaces it deems most appropriate to give me left roll: left flaperon up, right flaperon down, possibly differential movement of the stabs, possibly move the rudders as well. What controls move will depend on my angle of attack / sideslip, speed etc. So far, so good?

Do these "control laws" remain the same even in the very high AOA regime (i.e. >> 25­°): left stick commands left roll (even if it may mean no flaperon/aileron but stabs/rudder only), right rudder commands right yaw (which might command flaperon/stabs, but perhaps no rudder)?

[spazsinbad]

Yes it is difficult to 'imagine' how things work, especially for me when I have never flown a STOVL aircraft or simulator. However in this forum there is plenty of detail in words about the UNIFIED control laws used by the F-35 family with extra bits for STOVL mode. Some say START here: viewtopic.php?f=22&t=24054&p=252995&hilit=Farley+Unified#p252995
OR
viewtopic.php?f=22&t=23199&p=247994&hilit=Farley+Unified#p247994
&
viewtopic.php?f=22&t=20304&p=242760&hilit=Farley+Unified#p242760

Another search: viewtopic.php?f=22&t=20138&p=347365&hilit=Krumenacker#p347365
&
'Krumenacker': viewtopic.php?f=62&t=27677&p=297622&hilit=Krumenacker#p297622

An example from the first HIT above:

Test Flying The Joint Strike Fighter
17 Jun 2011 Graham Tomlinson

"...The STOVL mode control system is derived from ‘Unified’ developed by the ‘RAE’ on the VAAC Harrier. The throttle commands acceleration and deceleration (or thrust on the ground and in the STO mode, and in all conventional modes); in the hover the stick moved backwards/forwards commands upwards/downwards vertical velocity (or pitch rate elsewhere); in the hover the stick moved from side to side commands bank angle (or roll rate elsewhere) and if released returns the aircraft to wings level; in the hover the pedals command yaw rate (or sideslip elsewhere)...."

Source: http://myweb.tiscali.co.uk/hawkerassoci ... ghter.html

[steve2267]

Thanks for the links and pointers Spazs. I didn't see those threads when I searched on +CLAW and +FLCS. (Lots of other reading, though.)

[spazsinbad]

Yes using known search terms is not easy if you are unfamiliar with the forum and content. To ease the pain I think searching 'for all the text' can get better results but that is just my untested theory at this point. Meanwhile attached is a 137 page PDF about a pot pourri of VACC Harrier/F-35/F-35B STOVL specific CLAWS and such including the ENGINE & pilot comments and sheesh ran out of file size space. Online at the 'tinyURLs' below my posts are heaps of stuff - on forum also.

[steve2267]

One of the reasons I started this thread is so in the future if someone searches "FLCS" or "CLAW" they will have a jumping off point to other threads and PDFs like you just linked. Thanks!

FWIW -- your live.com site appears empty. I'll poke around the 2nd tinyurl. Thx.

[steve2267]

Also, the main point I was trying to confirm is that in conventional flight mode, the controls all command a rate around a specific axis and NOT deflection of specific flight control surfaces. So at high alpha, left stick commands a left roll rate which may deflects stabs & rudders but no flaperons (just a guess), whereas in a Cessna 152, left stick (well, yoke) only activates aileron deflection.

I hadn't intended this thread to be dedicated to STOVL Unified control laws etc. But that's ok. Fascinating stuff.

You know... I don't know if this is related or not... My father retired from BAE Control Systems in Binghamton, NY about a decade ago. Before it was BAE, it was Lockheed Martin Control Systems. When visiting once, he got me in to see a demo of a Blackhawk flight control simulator. LMCS had developed a flight control mode & autopilot wherein once activated in the hover, the control stick blended to a new control law:

• left<>right stick .... translate left<>right
• fore<>aft stick .... translate front<>rear (if memory serves)

I think the cyclic controlled up / down.

They (LMCS) were trying to sell it to the military (and others) for search / rescue ops, and probably for fast rope ops too. One feature was that if you engaged the autopilot, and took hands off the controls (stick only, though? maybe had to maintain grip on cyclic) it would hover over that exact spot, crosswinds notwithstanding.

Since they were Lockheed Martin at the time (well, they were sold to BAE, and BAE is a JSF partner) I wonder if what they developed had cross pollinated to JSF in some way. Dunno.

[spazsinbad]

IF you have downloaded the PDF here then the first few pages will make clear where the UNIFIED control laws came. Being UNIFIED the F-35 control laws apply to all variants whilst the F-35B transitions to the extra STOVL unified laws when required. Make sense? So this thread is not just about the STOVL mode - at least the PDF or whatever. As for being empty I have said a zillion times but forgot - potential downloaders from either Microsoft OneDrive or GoogleDrive must register for free FIRST so that they will be able to see the files at each site. No register then emptee. And please follow download advice which is: right click on a file to 'save as' to your computer before opening it with latest Adobe Reader for your OS.

Page 20 of the PDF above starts the JSF Unified CLAW which came from the VACC Harrier Unified Testing beforehand. UNIFIED is the key word. UNIFIED makes it weird for me because pushing forward on stick in STOVL mode is weird to me.

Here is how it starts: viewtopic.php?f=22&t=15079&p=190210&hilit=%2AGJBalas%2A#p190210 Good thread for topic
http://www.nt.ntnu.no/users/skoge/prost ... _ECC03.ppt (5.6Mb)

OLD file formats can be problematic so attached is a PDF made from the complete 71 page PPT mentioned above.

[johnwill]

steve2267, you are generally correct about conventional flight control responses to stick and pedal commands, but not entirely. Lateral stick force does command roll rate, not around the airplane roll axis, but around the airplane flight path axis. The difference between airplane axis and flight path axis is AoA. Longitudinal stick force does command pitch rate under some conditions, but mostly it is g-command. It is called a blended pitch rate / g-command system. Rudder pedals command rudder deflection, but the deflection obtained is variable according to (I'm guessing here) flight condition and yaw axis feedbacks.

[quicksilver]

Steve, another important point to remember about the STOVL machine is that the flight control laws are integrated with the propulsion system. Any time the jet is in Mode 4, "flight control inputs" also "talk" to the propulsion system. In PA, the other variants do so also, but obviously not to the degree that STOVL does.

For example (and as noted in several parts of the link that spaz provided), when the STOVL jet is in jet-borne flight (i.e. sustaining flight almost entirely on propulsion system lift mechanisms), a 'stick' input that would, in conventional flight, command a pitch rate, instead commands aircraft motion in the Z (vertical) axis without changing aircraft attitude. Also, when one makes a stick input that would normally command 'roll', it is not the aerodynamic flight control surfaces that are effecting the rolling moment; it is differential thrust output from the roll-posts.

[steve2267]

For side to side translation, swiveling the 3BSD nozzle slightly to one side will generate translational forces, but it will also produce a moment about the aircraft Z-axis, i.e. a yawing moment. To counteract that yawing moment, it seems you either need to be able to vector the liftfan thrust, OR differentially vector roll post thrust fore/aft on opposite sides. Can the liftfan thrust be vectored via louvres or some similar mechanism? I seem to recall reading about "louvres" somewhere with respect to the liftfan, but I thought that was more with respect to adjusting volume flow rate for purposes of increasing / decreasing thrust.

[steve2267]

steve2267, you are generally correct about conventional flight control responses to stick and pedal commands, but not entirely. Lateral stick force does command roll rate, not around the airplane roll axis, but around the airplane flight path axis. The difference between airplane axis and flight path axis is AoA. Longitudinal stick force does command pitch rate under some conditions, but mostly it is g-command. It is called a blended pitch rate / g-command system. Rudder pedals command rudder deflection, but the deflection obtained is variable according to (I'm guessing here) flight condition and yaw axis feedbacks.

johnwill thank you very much for that description. The lightbulbs are a poppin'. I will have to sacrifice some additional brain cells on this altar.

I am most accustomed to thinking in terms of pitch, yaw, and roll. Is this how fighter pilots think? Or are they thinking in terms of flight path vector and lift vector?

I think I can envision how when performing a high yo-yo if you get alpha up over 30° and roll, you are going to be performing a Herbst maneuver (J-turn). But if you stomp on the rudders at the same time, you are really going to slew that sucker around. I could see where that 28°/sec comes from and really makes the F-35 dangerous close in. I wouldn't want to get slow and stay there, but if you've got the smash to get your speed back... yeah, they might be re-writing the book on knife fighting in a phone book.

[johnwill]

Steve, a comment about your question concerning lateral translation - I have no knowledge how that is done, but from observation it seems to me it is done with a small roll angle, perhaps 1 or 2 degrees. That way both lift vectors have a lateral component without a change in pitch, roll, yaw, or longitudinal vectors. There would be a tiny loss of lift, so maybe a little added thrust world be required.

[steve2267]

Steve, a comment about your question concerning lateral translation - I have no knowledge how that is done, but from observation it seems to me it is done with a small roll angle, perhaps 1 or 2 degrees. That way both lift vectors have a lateral component without a change in pitch, roll, yaw, or longitudinal vectors. There would be a tiny loss of lift, so maybe a little added thrust world be required.

Aah. That makes much more sense. Come to think of it, I recall always seeing in videos of F-35Bs landing aboard ship a very slight roll before / after translation started / stopped.

[spazsinbad]

'steve2267' asked above:

"...Can the liftfan thrust be vectored via louvres or some similar mechanism? I seem to recall reading about "louvres" somewhere with respect to the liftfan, but I thought that was more with respect to adjusting volume flow rate for purposes of increasing / decreasing thrust."

I'll check wot STOVL engine bits are in the 137 page CLAW PDF already downloaded. However there might already be an explanation there.... The F-35B in STOVL Mode 4 can fly backwards at 30 knots, with LiftFan louvres positioned and Aft Engine Exhaust Nozzle 3BSN full forward (at 95 degrees to vertical) I guess. Anyhoo I'll get references..... see page 126.

Powering the Lightning II
April 2012 Chris Kjelgaard

"...…Below the LiftFan, the variable area vane box (VAVB) provides an exit path for the cool air driven downwards vertically by the LiftFan. Rolls-Royce produces the VAVB, which is made of aluminium and contains louvred vane doors. These can be angled all the way from 45° back, through fully vertical to 5° forward to provide variable directionality for the downward cool-air flow from the LiftFan, as commanded by the pilot through the aircraft’s FADEC units.

...When the F-35B hovers, the FADEC commands the 3BSM [3BSN] – which can direct air through a 95-degree range from 5° forward to horizontally back – to swivel downwards to direct hot engine exhaust air in the same direction as the direction of the cool air produced by the LiftFan near the front of the aircraft. The 3BSM [3BSN] can swivel fully from horizontal to vertical orientation in 2.5 seconds…."

Source: http://militaryrussia.ru/forum/download ... p?id=28256 (large PDF)

[quicksilver]

Lateral translation is effected by differential thrust from the roll posts which in-turn creates a slight angle of bank; vector mechanics does the rest. Absent pilot input, the 3BSM on the main engine will maintain aircraft heading during the translation. The lift fan VAVBN vectors only fore and aft (as spaz' links indicate). The CLAW also assumes the pilot wants to maintain altitude during the translation, so the propulsion system will compensate with the appropriate adjustments.