Do anyone use a servo failover set up in typical two servos-one elevator applications. (like a Hawk and many others)I mena a mechanical linkage that allows one servo to stall an you end up with half the movement by the other servo. No binding ofcourse. Do you understand what I'm talking about?
What is the best linkage and who sells it?

I have put a picture of exactly that in the copy for next edition of RCJI.

I saw the installation in Paul Dunkley's excellent GosHawk just last Sunday.

The travel remaining if one servo fails will be reduced but may keep the aircraft flyable and it will not cope with ALL modes of servo failure but I believe the most likely failure is probably a burnt out motor which locks up the servo and I believe the system WILL cope in that mode of failure, albeit with reduced travel.

There is a similar system in the artificial feel system of the VC10 airliner (tanker now) which keeps feel the same even in the event of the failure of one system. Just cant remember exactly how it works !

Regards, David Gladwin

A similar system is used on the Skygate Hawk.

Can anyone post pictures?

marcs

Not sure how these other work but here is one. Note it cannot provide reduced movement if one servo drives fully over hard.

http://www.swbmfg.com/self_adjusting.htm

I have made a skecth, a quite bad one I'm afraid but it hopefully will give you an idea of what I mean. It is really quite simple.

Very interesting, Matt, I'll certainly take a closer look at this system, probably order one, as ONE locked up servo on a dual servo Hawk stab. for example, locks up the whole thing, meaning very little redundancy.

The other possibility with a common arm, or with the SWB system, is to have normal stab. or surface travels at low rate and if one servo fails causing reduced travel, then switch to high rate activated by a mixer switch restoring all or most of the pitch response. Just enough to get you home.

And talking of servos, has anyone any experience with the new 6000 series metal cased v. powerful JR servos ?

Regards, David Gladwin.

On a quick glance over the system, seems that you are adding three aditional points of failure:

The two extra critical linkages on the servo bridge and the posibility of one servo failing and going at one end of his travel arc.

As this is not resolving the problem of retaining the same amount of travel even in case of a optimal servo failure, the plane will probably turn incontrolable too... and having a dual rate switch to quickly overcame this will imply in having half the amount of the force applied on the surface on standard operation mode wich is a bad idea. Another built in disadvantage is the greater amount of slop on this system.

As far as I have seen, the best way to give safety redundancy on a single air surface is installing two brutally strong servos on it, wich in case of failure of one of them will still have enough torque to move the surface with all the flight loads and a dead servo thru the full travel.


Back in 1990, I had a giant scale Citabria using this.. but that was because there wasn´t strong servos available at the time!

Does NOT work. Try rotating a servo with a burnt out motor, impossible, they lock SOLID and the torque of another servo will never overcome one. If you use very strong servos with a system such as the SWB or similar geometric system , you can easily use only half the torque, using a dual rate switch uses also half which SHOULD be MORE than adequate, if you choose your servos carefully (JR8711, 6311) and balance the stab. Of course there is no problem in overcoming a passively failed servo.

The other two points of failure are mechanical. 99.999% reliabilty should be eaily obtainable and slop should also be easily eliminated by careful manufacture and tensioning of the cables !

This SWB system is the best I have seen so far for creating a most reasonable degee of servo redundancy on a twin servo stab. Its not perfect but it goes LONG way to solving the problem.

Regards, David Gladwin.

Yes it works, not all the fail modes of a servo includes a locked burnt out motor.

About applying half the torque of a servo with mechanical disadvantage is unfortunately just plain wrong and a safety issue in r/c jet models. In fact, this can led to burning out motors, no matter how carefully you choose them. Not to mention about the pointless extra load in the electrical system.

About the extra mechanical points of failure doesn´t matter how careful you can tension a pulley, pivot, slider, whatever.. it is has two extra rotating points wich multiplies the slop on all the system, that could led to flutter wich is a major safety in r/c jets too.



Kind regards, Enrique


p.s. Here seems that your keyboard have a problem as it activates caps lock in some words.

The SWB system is not the one I'm lookin for. It takes to much space and is to complex. No comment on my sketch? You must click on the picture to see it. I think it is a German company that's making it but I may be wrong.

Ela, I think you system does approximately the same as the SWB version: one locked servo gives half the travel.

I saw your sketch in the last issue of FMT magazine (german). They also explained it works even with one stalled/burnt servo. Burnt servo poses no load on the working servo. You have half of the travel left if it is stalled in the center position.

If otoh a servo breaks at full deflection, you need full deflection from the remaining servo to get back to neutral. Though it is better than no control at all, i doubt that will save your plane if a servo breaks at full down on an elevator...

I like the system though if you have two servos on one control surface. (eg elevator half)