question regarding f 14 spider detent (emergency wing sweep handle)

[Gareth Barry]

Greetings all.

So I have been slowly making progress in designing a full throttle for the f 14. The throttleteck version is a bit pricey for what it is imho, and has certain limitations. My plan is to implement a functioning, servo actuated autothrottle and wing sweep handle, which will talk to the PC via DCS Bios. I also intend to make all of the plans etc freely available so that whoever else is crazy enough to make their own f 14 throttle can do so.

My current question is regarding the 'spider-detent' of the wing sweep handle actuator. My understanding is that there is essentially a 'pin' on the handle that fits into a 'slot' on the actuator, such that when the pilot pulls out/up the wing sweep handle, the handle itself is disengaged from the actuator and is then free to move. The wing sweep position will then be determined by the position of the handle, and not the CADC. However, the CADC will still move the actuator, thus for the pilot to put the system back into auto he/she must first move the handle to match the position of the 'slot' in the actuator (ie the spider detent) and then push the handle back down into it. At this point, the wing sweep handle will then obviously move with the actuator.

Have I got this right? If my understanding is incorrect, please correct me, so that I avoid going down the wrong path in designing a system that imitates this.

Thanks so much.

Regards

Gareth Barry

[sLYFa]

I think you are basically right except that the handle can only be stowed (i.e. pushed in) in the full forward or full back (oversweep) position. The spider detent does not move if you are in emergency mode (i.e. handle pulled out). The way the system is put into normal from emergency is that you move the handle fully forward, stow it, press the master reset button and put the wings in auto mode. This will drive the spider detent from whatever position it is (usually around 55° because the handle is pulled out from the 'bomb' sweep position after landing) to the forward position, pushing it into the handle and latching it.

[Gareth Barry]

Ahh Ok thanks. Hmmm, gonna require some thought as to how to engineer a system that imitates this, but your post is very helpful. 

At the moment i am thinking of 2 'axes' that lock into each other, one for the 'spider detent' and one for the wing sweep handle.

[draconus]

Last time I heard the handle could be moved even in the lowered position.

Here's a manual:

https://www.heatblur.se/F-14Manual/general.html#emergency-mode

[sLYFa]

There is another thing to consider. When the handle is raised from the stowed position, a locking mechanism is engaged which has notches every 4°. So in order to move the handle after putting it out of stow, it needs to be raised another 1inch, moved to the desired position and released to snap into one of the notches (the handle is spring loaded). The rw manual has a description of the mechanism, if you manage to find one (any variant).

[Gareth Barry]

thanks Draconus

Hmmm...sLYFa, reading through the manual seems to suggest I was right regarding being able to put the handle back into the spider detent at any wing sweep position , although with my ADD and general low IQ I could be wrong? From the manual 'To return to the normal mode of operation, the handle should be pushed into the desired position and pressed down and the guard closed. The MASTER RESET button on the fuel management panel should then be depressed and the wing-sweep system set to the same position as the handle. The servo will then drive to the commanded position and re-engage the handle to the spider detent, resuming normal operation.' I could be reading this wrong?

Personally, i have never found the need to open/close the cover mid-flight, so perhaps the procedure of re-engaging the spider detent was specific to the commonly encountered time when this is done, ie before take-off, when then 'desired position' will obviously be full forward sweep?

Draconus, are you saying that opening the hatch, then moving the handle without extending it will also disengage the spider detent and let the wing sweep angle be manually moved- ie that extending the handle itself only serves the purpose of providing extra leverage? If this is the case, it would require some sort of clutch system similar to what I am imagining for both throttle levers, such that the pilot can override the servo without stripping any gears. Also, with the handle out of spider detent, does the spider detent continue to move, controlled by the CADC? Or is the servo that actuates the spider detent thingy also disengaged at some point in all of this? I am guessing the latter?

Just got a bit more drawing to do and then will show all, for anyone's input and criticism.

Thanks so much guys.

sLYFa

gosh, the 4 degree thing is another complication....I can see why none of the manufacturers have given this thing a go...

So does that this mean that in emergency mode, only discrete values of wing sweep are selectable, in increments of 4 degrees?

Thanks again

[draconus]

I'm in no position to argue with SMEs, HB or former maintenance stuff. I can just share what I read here from them or the manuals. Here's recent changelog after they revisited wing sweep functionality.

 

[Gareth Barry]

sLYFa - thanks so much - it turns out I was wrong after going through your recommended reading.... I would suggest Heatblur perhaps update their manual as imho it's description of the mechanism doesn't seem quite right - maybe I was just misreading it. It turns out that the wing can only be returned to the stowed position in the 20 degree and oversweep.

The handle is spring loaded toward the stowed position, but requires depressing the release button on the inboard side of the lever to return it to the stowed position.

Interesting..got some thinking to do...

[sLYFa]

The discription is a bit ambigous in regards to what 're-engaging the servo' means. On the one hand, pulling the handle up does not disconnect the spider detent as you need to pull it back with quite some force to get it out of the detent. On the other hand, pulling the handle up already puts the wings in emergency mode so I guess while you can move the handle back to wherever the spider detent is, the spider detent won't move the handle unless the handle is fully stowed. What is also not quite clear is what is actually commanding the wings to move. The manual says that the handle is directly connected to the hydraulic valves controlling wingsweep, implying that in normal mode, the spider detent just moves the handle thereby controlling wingsweep.However, a fold out suggest that the handle does not control wingsweep in normal mode and instead, the CADC controls the wings electronically. Perhaps some SME can give a more detailed explanation

[RustBelt]

So I had done some looking at this mechanism. The “Spider Detent” is basically a spring loaded clutch like in a drill/driver. It’s coupled under spring tension. Pulling hard on the lever causes it to break away. Re aligning causes it to pop back in. It’s coaxial to the hinge of the lever. 

The Lever controls the wing position, the CADC drives the lever when the clutch is engaged. The pilot drives it when they pull it out of engagement with the CADC coupler.

This causes a feedback mismatch in the CADC causing it to trip out of the loop until you line the lever back up to engage the clutch coupling, and hit the master reset. 
 

The super slick way to mock it would be a Servo motor that will disengage with any load. But still send position data. Otherwise you run a basic motor drive, build the clutch coupling, and have the motor position looped on the lever input. Getting the CADC position out of DCSBios may give you full independent action. And when not using the lever, it has to be ignored otherwise the input and DCS will fight. So a switch on the cover or on lift up is needed to switch it to axis reporting. Or you would need to perfectly match the position movement to keep the sim happy.

 

what’s really needed is a museum to be nice enough to film an actual full actuation of the mechanism. It’s in there just sitting doing nothing in gate guards and hanger displays all over the world. 

Edited May 30, 2023 by RustBelt

[Gareth Barry]

Hi Rustbelt

What you described is basically how I imagined it (the real system) would work. I have been mulling this over, and I THINK I have come up with something that is acceptable. My goal is not to reproduce the system exactly, as in, the exact amount of force required to 'break' the link between the servo and the handle etc. At some point, some compromises have to be made I think. I am progressing on the drawings and 3D models and I think I am getting to the point where I can post something for everyone to comment on, criticise and hopefully improve.

Anyways, here's what I have come up with.

The handle will have another lever to the inside of it, which will be operated by a servo, controlled by DCS bios. On the lever will be an electromagnet, which can be switched on or off. On the handle itself will be a strip of steel for the magnet to grip on to (most all the parts will be laser cut acrylic and 3D printed.) Both the servo actuated lever and the handle will have hall sensors to measure rotation angle. When the pilot shoves on the handle, forcing a misalignment between the handle and the servo actuated lever, a few things will happen;

i. the electromagnet will power down

ii. the arduino will send a message to the PC that the system has now switched to 'emergency wing sweep.'

iii. the arduino will now send the position of the handle to the PC.

electromagnets for arduinos are cheap and readily available.

Also, the '4 degree locks' will be a series of teeth, almost like a section of a gear, scaled such that the notches are in even increments from '20 degrees' to '68 degrees' and then 'oversweep.'

The handle will be spring loaded to the 'down' position.

When it is pulled up, a spring-loaded catch will flick out, over and ultimately into the 'notches' mentioned above. In order to move from one 'notch' to the next '4-degree notch', the handle will have to be pulled out/up some further distance. Pulling the handle up such that that the spring-loaded catch is now engaged with the '4-degree' teeth, a microswitch will trip so that the arduino knows that handle is engaged with the notches.

The notches at the 'oversweep' and '20 degree' position will be somewhat thinner than the rest. When the pilot wants to put the system back into auto, he will push a button on the handle which will retract the spring-loaded catch sufficiently to clear either of these two settings, but not enough to clear any of the others. With the spring- loaded catch retracted and the handle at either the 20-degree or oversweep position, the handle can be pushed down so that the spring loaded catch is now underneath the notches. Pressing 'master reset' will then command the arduino to move the servo operated lever to match the position of the handle, as well as re-energise the electromagnet. When the readings of both hall sensors are aligned, the system is back in 'auto.'

Wow...that was a long description!! Congratulations to anyone who read through that, but honestly, writing it all out help me to organise my thoughts on this. 

[Gareth Barry]

Whilst I'm at it, let me also mention that the autothrottle function I am keen to do similarly, ie with electromagnets to act as a simple 'clutch' between the servo and the handle. The throttle stalks will have hinges to allow sideways rotation, so that they can be moved sideways through the various 'gates'.

[Temetre]

vor 22 Stunden schrieb Gareth Barry:

sLYFa - thanks so much - it turns out I was wrong after going through your recommended reading.... I would suggest Heatblur perhaps update their manual as imho it's description of the mechanism doesn't seem quite right - maybe I was just misreading it. It turns out that the wing can only be returned to the stowed position in the 20 degree and oversweep.

The handle is spring loaded toward the stowed position, but requires depressing the release button on the inboard side of the lever to return it to the stowed position.

Interesting..got some thinking to do...

Oooh, is that why its so difficult to stow the wings? I usually just clicked around on the thing until I somehow could stow it. The manual was a bit confusing to me^^

[Gareth Barry]

Sooooo....

For anyone who is still interested in this, I am posting some of the files of my 3d drawings, the flap handle, as well as both throttle halves.

Still thinking about how to implement the autothrottle and the motorised wing sweep with manual override. 

In another thread, I discussed some ideas about to implement an autothrottle that can easily be disabled/overridden by applying some force to the handle- and was given the excellent suggestion of using a motorised linear potentiometer as used in high-end digital mixing desks.

So then i got to looking at how these motorised linear potentiometers work (ALPS seems to be the common brand) - and it seems to be a simple belt pulley round the motor when then moves the position of the wiper - there don't actually seem to be any complex clutch systems or anything at all between the motor and the potentiometer - I am guessing slipping of the belt is the extent to which there is any clutch mechanism. With that being the case, I am in two minds as to whether to just use a motor with a belt and pulley to move both throttles and the wingsweep handle, with hall sensors on each of the latter so as to measure position, or to buy motorised linear potentiometers and use those. 

In any case, the 4 degree 'notches' can be quite well simulated in that case, by coding the motor to try to 'restore' the position of the wiper to either the 20 degree, 24 degree, 28 degree position etc etc. 

This seems on the whole to match what Rustbelt was suggesting, ie to use motorised potentiometers and clever coding to mimick the system as much as possible.

In any case, here are the files of some of the stuff I have drawn.

f 14 grip lhs cover.FCStd f 14 grip lhs.FCStd f 14 grip rhs cover.FCStd f 14 grip rhs.FCStd f 14 throttle flaps handle.FCStd

[RustBelt]

Electromagnets are an interesting solution. 
 

I’d just do motors and belts. Motorized linear pots are pricey and constrain your form factor.

Edited June 1, 2023 by RustBelt

[Gareth Barry]

Thanks Rustbelt - also for your response in the PM I sent you. I am seeing that the electromagnet solution that I had in order to act as a clutch so as not to strip the gears or burn the armature of the servo is unnecessary. I think I am going to go with a current-load sensor on the servo, and when that exceeds a high value from being stalled due the pilot shoving on the handle, the arduino will switch off the servo motor, simple.

[parkerfly51505]

On 6/1/2023 at 12:29 PM, Gareth Barry said:

Sooooo....

For anyone who is still interested in this, I am posting some of the files of my 3d drawings, the flap handle, as well as both throttle halves.

 

Definitely interested and thanks!  I've been trying to figure out how I want to implement the wing sweep myself.  I have a Virpil CM-3 and just map some of the buttons to the cover and handle for now but that's no fun.    I'm finishing up panels on the full cockpit now but will be anxiously watching your progress.  Those CAD files look great so far.    

[Gareth Barry]

Thanks Parkerfly, it's good to know I'm not the only one crazy enough to want to want an f-14 throttle!

Making steady progress on the CADs. 

However, I have run into something of a dead end in terms of the mechanism of the wing sweep handle. Not so much in terms of the pulling it out of spider detent, as well as the 4 degree notches when in manual - I think I have come up with something that models that. I am trying to model the functioning of the actual yellow handle on the end of the wing sweep handle. Specifically, there is a button on the yellow handle itself which must be pressed to allow the handle to be stowed away. I have no idea how this actual mechanism works. Now, I could design a system that achieves the same thing, and I have tried to make the parts look as authentic as possible - but the way the mechanisms work is not exactly the same. Bare in mind that I want this whole thing to be achievable by someone with access to a laser cutter, 3D printer and not much else in terms of manufacturing - no welding, cnc milling or anything like that.

So, if anyone knows how this part works (the handle itself with the button) please chime in...

[Gareth Barry]

One other part that is giving me a headache- probably due to my lack of modern manufacturing knowledge - is how to go about making the wingsweep cover. For a once off part for myself, I would probably just hand carve a form out of wood and then vacuum form some clear acrylic over that - but again, I would like this to be reproduceable by anyone else if they so wish to do so. So, can 3D printing produce clear parts? And clear enough to be able to see the wing sweep handle underneath? I have my doubts to be honest. Perhaps the side profile could be laser cut from clear acrylic, and then the curved part bent by hand and then glued to the side profile using MEK solvent? That's about the best that I can come up with.

[RustBelt]

FDM won’t make a transparent part. Resin ‘might’ but it may be brittle. One transferrable DIY way to do it would be to 3D print a buck. Then get out the hair dryer (probably more a full power heat gun) and heat mold the clear piece to the buck with some parchment paper in between. Super easy with both Polycarbonate and Acrylic. Heat, drape, cool. Don’t heat on the buck because it will also go soft.

Probably only get one shot at the buck, but you just re-print if you botch it up.

[parkerfly51505]

23 hours ago, RustBelt said:

FDM won’t make a transparent part. Resin ‘might’ but it may be brittle. One transferrable DIY way to do it would be to 3D print a buck. Then get out the hair dryer (probably more a full power heat gun) and heat mold the clear piece to the buck with some parchment paper in between. Super easy with both Polycarbonate and Acrylic. Heat, drape, cool. Don’t heat on the buck because it will also go soft.

Probably only get one shot at the buck, but you just re-print if you botch it up.

I was just watching a YouTube video yesterday on the "art of bending acrylic"!  Never tried it before.  That seems to be the easiest method as long as a decent form can be made.  I have some 3mm clear acrylic on hand but that's possibly too thin to try.  When I get a chance, I'll try it with a heat gun and see how it holds up.

I have no idea how that switch on the yellow handle works either.  So wish we could tear into a real one to see how it works.

[RustBelt]

It’s all about putting the heat where you want it, otherwise it bends where you don’t want it to. 
 

Or just slumps into an unfixable wad. 

[parkerfly51505]

On 6/5/2023 at 2:15 PM, Gareth Barry said:

I am trying to model the functioning of the actual yellow handle on the end of the wing sweep handle. Specifically, there is a button on the yellow handle itself which must be pressed to allow the handle to be stowed away. I have no idea how this actual mechanism works.

You may have already seen this photo but it's a decent closeup of the button mechanism.  Looks like that tab is probably spring loaded and the button somehow pushes it in enough to allow the lever down.   

 

[parkerfly51505]

Found another photo of a pretty ragged quadrant that has a detached side panel letting us see a bit more.  That pin I referenced in the other photo is seen "stowed" due to the missing side plate on the lower part of the wing sweep lever.  

[Gareth Barry]

Thanks so much Parkerfly, those pictures are quite helpful. I may, or may not, decide to try to model something with the same mechanics. As I said, even if I don't those pictures will be very helpful.

At the moment i am leaning towards a simple push-button switch which will let DCS know that the handle is now being pushed in. However, this may change. As I said, getting there. Determined to see this project through, although it may take some time.