By continuing to use this site, you agree to our use of cookies. Find out more
Forum sponsored by:
Forum sponsored by CML

Failsafe

What should be set?

All Topics | Latest Posts

Search for:  in Thread Title in  
Erfolg08/03/2012 14:44:37
avatar
11863 forum posts
1374 photos

I recently crashed one of my gilders, from a great height.

I am pretty certain that I lost the radio link. I had thought it had gone into neutral control surfaces and low throttle. The crash was an event of a large number of stalls, before reaching the ground.

On inspection later, it was apparent that there was no failsafe to set. The control surface settings being just serendipitous.

I have now replaced the 35 Rx with a 2.4, and I am setting the failsafe. The dilemma is, to what.

If I had set all to neutral and power off, it could easily have gone for miles.

Olly has suggested a rudder or aileron turn. My concern here would be that the model could end up spiralling in, at speed, both potentially wiping out the model and doing some damage.

I then thought perhaps some up elevator as well, hopefully to prevent the spiral dive, perhaps inducing a stall, recovery sequence. The aim to limit the distance travelled and speed achieved.

What do others do? What actually works well.

Simon Chaddock08/03/2012 15:37:54
avatar
5844 forum posts
3100 photos

Erfolg

I too have been pondering this problem with my better gliding planes.

I suppose the ideal solution would be for the failsafe to activate another servo and deploy a CofG mounted parachute like some light planes now have, or release the tailpane to go into a "dethermaliser" mode like the free flight boys used to.

Failing that a stall/recovery mode does work but the cycle needs to be very short to prevent the speed building up during recovery. This usually requires a more powerful elevator than would be required for normal flying.

Only one of my planes (the lightest at 12oz) can really do this and it doesn't go far either as a wing drop at each stall also ensures it keeps changing direction.

I dont know if it would really survive undamaged. I suppose it would depend on the what part of the stall sequence it was in when it hit the ground.

The only fly away I have had was a battery connection failure so the surfaces just stayed where they were but fortunately it was not high and it just glided down and landed on open ground.

Peter Beeney08/03/2012 16:36:37
1595 forum posts
59 photos

Erfolg, I think this is an age old dilemma when it comes to setting failsafes on gliders. (Or on any model, perhaps?). If you set it for a turn the argument is if you are in a thermal it just keeps going up! Which is perfectly plausible, I would think. However, I guess I’d like think that at least I’d like it to come down reasonably close to hand so I think I’d take a chance and try and fix it so that the model went into a turn, and ignore the case of getting into the thermal.

I suspect every model is different, so I’d be tempted to fly the model, move the controls to a position, say up elevator, a bit of rudder, out-spin ailerons if it has ailerons, and see what happens. Then find out the gentlest turn you can and set the failsafe to that. Or, of course, any other setting that just takes your fancy. Plenty of scope to try. You might then be able to ague, that should the model go up in a thermal OOS, wherever it comes down it’s still going to come down in the softest manner possible.

I once did play about with some PCM failsafes, actually in in the air with interference but on powered models, when I didn’t really believe some of the things that were being said about it. Even an i/c powered model can be made to spin down relatively slowly.

In your case, is it not a fact that should the ESC lose signal, or receive interference, then the motor will automatically shut anyway.

Of some slight concern, did you exactly establish why you lost control? If it was a case that you lost receiver power, which sounds possible, then the whole question is academic anyhow, and being on 2.4 wouldn’t make any difference, either. This is a bit of a moot discussion point, the fact that quite a number of failures are power related, which tends to render the failsafe situation somewhat optional, in that the servos will just stay as they are. However, the motor should stop in this event, certainly I’d be proving this, and I said recently in another thread, if this were not the case, I’d not even consider using the ESC.

As Simon says, I’ve considered the idea of a parachute, I once saw a tv programme about an American light plane so equipped years ago, but it requires some thought. It would have to operate on a power failure, or any other failure, and also some positive means of stopping an i/c engine, i.e. a fuel cut off rather than relying on the servo.

This is all probably why it’s still something of an insoluble dilemma……

PB
 

Edited By Peter Beeney on 08/03/2012 16:39:20

Alan Cantwell08/03/2012 16:45:34
3039 forum posts

i must admit, i have never given thought to a glider failsafe, on a power jobby, with an IC engine, its should be to tick over, on electric, totally shut down, the need is to stop the model rising into the airlanes, but a glider? HMM, for me, i think, a gentle turn with up elevator, or even just full up, cant go far with this action, surely?

Engine Doctor08/03/2012 16:48:45
avatar
2711 forum posts
44 photos

Hi the minimum setting for power models is to set throttle to idle .I dont fly gliders but could you set the failsafe so that the model will go into a decending turn ,not a fast spiral but a gentle turn that will loose height without gaining too much speed . You might get lucky and be able to re-gain the link or at least stand a chance of getting the model back in one piece.You would have to experiment with the trim settings and when happy fly the plane down on the sticks and set the failsafe as per the trims.

Former Member08/03/2012 17:02:08
6538 forum posts

[This posting has been removed]

Erfolg08/03/2012 17:09:41
avatar
11863 forum posts
1374 photos

In the case of my crash I do not know the cause. When I got to the model everything seemed to be totally static, no twitching etc.

I left the model switched on, really thinking it was dead. Always keeping the prop pointed away from me.

When I got back to the Tx, it remained the same. That is until I wobbled the sticks to confirm the model was dead. Then I found the control surfaces moved. Further movement confirmed, movement was as usual.

Testing the lipo back home it was still above 4v per cell (3s).

Unlike some, I have found that Futaba Rx,s and my Gold Tx do die. I have changed about 2 now, as they seem to become unreliable. No Rx has died, the range seems to go, with time. I have opened it up, as suggested by a club member to check the aerial. In my opinion, it is difficult to do a continuity check, and it is very securely held.

Is it the ESC, hard to say? It is a Turnigy 40A, Blue series, which I use a lot.

I will go for a turn with the nose held up a little, as it drops the nose in a turn, unless in a thermal.

Peter Beeney08/03/2012 22:40:17
1595 forum posts
59 photos

Erfolg, I think I’d be very inclined to at least try out a full range ground check on your radio anyway. At least that might give you a bit of a clue as where the trouble lay. I’d expect, with the tx standing on something about chair height, the aerial vertical, and carrying the rx, a battery and servo to be able to get at least 500 - 600 meters away with the rx at chest height before you started to seriously start to lose signal.


To do a continuity check, just stick a pin in the end of the aerial wire, make sure you’re touching the inner copper wire and then touch the other end with the other meter lead where it terminates on to the PCB, but I’d say that’s most unlikely, if it it is down on range it’s most likely the crystal, and that’s almost unlikely, too. Or, if you don’t want to take the case apart, use a fine needle as a prod where the aerial wire goes into the case.


We have an ideal site for full range checks, you can either drive or walk, for the exercise, and they invariably pass this test. Air range is going to be considerably more.


If it is the radio that playing silly games then all well and good, but if you are simply changing one good component for another you could get the same situation happening again; and from all the little problems that I’ve seen I think I’d be looking at the power supply first.

Good Luck!

PB

Erfolg09/03/2012 16:07:32
avatar
11863 forum posts
1374 photos

Peter

I will try a range check on the Futaba receiver. It will have to be in something else, as the Sagitta now awaits testing.

FlyinBrian09/03/2012 17:15:59
707 forum posts
4 photos

Firstly,the purpose of "Failsafe" is not to land the model without damage but to prevent the model flying away and becoming a danger to other people or aircraft. This is fairly easy with powered models by setting the throttle to idle and other controls to neutral.

On a glider however that could result in a flyaway so If the model has spoilers or flaps I would set these to be deployed in the event of failsafe occuring.(Ailerons could be used as spoilers if two servos are used)

If that's not possible then I would set the elevator to full up and rudder full over to hopefully spin the model in before it got too far away.

Peter Beeney09/03/2012 21:37:54
1595 forum posts
59 photos

Erfolg, thanks for the reply.

You don’t necessarily need a model, just the rx, battery and a servo. If you get long range with the aerial coiled up a bit it will definitely work with the aerial held out straight. When I do this check, I find usually that at some point the servo is just beginning to ‘twitter’ slightly, the signal is getting weaker and the servo is just beginning to hunt around the centre. At that point there is still full control, but I know I don’t really have to go much further before I start to lose signal at low level, say knee high.
Straight up high in the air is perhaps the best case, actually, I think that Maynard Hill used a standard 27 AM tx for his first model altitude record attempt, at around 14,000 ft, 2.6 miles, but he chose a directional type for later record heights.

I’d fully agree about the failsafe but there are snags. Any failsafe system that the operator can tinker with and adjust is perhaps asking for questionable results in some cases anyway, coupled to a lack of knowledge sometimes about the provision of same, some pilots are unaware of the failsafe facility. The failsafe debate has been anything but failsafe over the years; and I’m sure will continue to be so.

Also as I mentioned before, some of the inline types don’t work correctly anyway; opening the throttle instead of closing it under interference conditions, for instance.
If the definition of a fail-safe mechanism is “designed to return to a safe condition in the event of a failure or malfunction” then a failure of the power supply I consider leaves something to be desired; and not even a system comprising of two batteries and a reputable battery-backer unit is totally immune, as a friend flying his large scale plane found out on one occasion.

With the advent of 2.4 growing on almost a daily basis perhaps, I think the conditions will change slightly. For the foreseeable future perhaps, there will be negligible interference conditions, so we can ignore that. Complete failure of the tx or rx, if the tx failed then the failsafe would operate, but the rx? This might depend on what the failure was, if it were the decoder then the failsafe might not be able to operate, there might not be any pulse to the servo. But again, the track record here must be excellent, too, almost negligible. A servo failing, gears breaking and jamming, say, failsafe will not operate; but does that happen very often, either? Although the incidence rate here seems to be climbing on many of the small park flyer types, the servo seem to be prone to failing.

Not much else left to go wrong, other than the power pack, and at the moment there is little provision for any sort of emergency supply to cover this. If the power does fail the model invariably goes straight in, but there also have definitely been flyaway models. I think personally that the chance of a power failure in the future will outweigh all the other situations put together, so, as always, the integrity of the battery is paramount. And with the right care and attention, I think this can be made to be. As reliable, or more so, than anything else.

So that’s why I’d aways try and establish, whenever possible, the most likely cause of an incident. If it’s my own mistake that’s one thing, but an unexplained disaster is always cause for concern.

PB

Erfolg10/03/2012 10:50:52
avatar
11863 forum posts
1374 photos

Hi Peter

I only intend using a model as a convenient package for the test.

With respect to "failsafes", I do agree and accept the title is misleading. To ensure a model and it's systems "fail safe" will take a lot of effort, equipment and expense, if attainable in an absolute sense.

I have disagreed with many as to what is appropriate. I see a park foam flyer as quite a different issue when compared to a 11 ib., 90 powered, aerobatic machine. Yet as with tie downs?, I believe that common sense as to the risk, consequences, and how to manage the risk in a proportionate way, is the way to go. We modeller need to think, use our imagination, not despair that total safety cannot be achieved. What ever some H&S people will state in training. To many seem to think, because their club insists that a range of safety measures have to adopted, that the indoor electric jobbie should do the same as them.

As to gliders, my own opinion, at the moment, is that it is on balance safer that the model comes down adjacent to the field, rather than travel a long way slowly. Ideally the model should come down slowly, as speed and weight, normally mean a bigger impact. Although i am prepared to change my mind as to what I can ideally can try to achieve.

Alan Cantwell10/03/2012 12:20:31
3039 forum posts

model fails, it crashes safe, whats wrong with that? what title would you give to a reciever that goes to certain settings when it loses contact with the transmitter? reffering to your last four words of your opening post " what actually works well," forgetting my previous softly aproach, the answer to that is simple for a glider, hard over and full up, and the its not the model that dicates if it should have its failsafe set, its the reciever, why they incorparated this system into all 2.4recievers is beyond me,

Erfolg10/03/2012 12:46:52
avatar
11863 forum posts
1374 photos

It is what the model makes contact with matters, in some instances.

The idea that any loss of contact will result in a model automatically crashing seems a little extreme, in my opinion. Would it be considered that this is the optimum solution for a very large or heavy model, as long as it lands in the field?

I much prefer the idea of slow speed, to delay contact with the ground, whilst keeping the model in the immediate vicinity of the field in anticipation that the loss of signal is not permanent.

Although it is only hear say, I have heard that on a typical flight that Spektrum receivers will receive a corrupted signal a number of times per flight. I assume that if true, it also applicable to other manufacturers, other than as other systems do not record these events, we typically have no idea. Anything which automatically results in crashing models, with a temporary or permanent loss of signal, could result with a wreckage strewn field. Would this be desirable, would it necessarily be the best outcome?

Alan Cantwell10/03/2012 12:54:39
3039 forum posts

this is why, on my power models, i set the failsafe to throttle closed to tickover, and all surfaces neutral, it will not go anywhere with no power, and i have a chance to redeem the situation, it satisfies the CAA, it satisfies me, with your glider scenario, with the lift of, say, a slope being the engine, then its hard over and in,

and i use a data logger, and so far, i have not seen any corrupt signals that could cause me any problems, and i fly DX9 and specky recievers in some big planes

Erfolg10/03/2012 14:38:17
avatar
11863 forum posts
1374 photos

Alan smiley

I would not want to justify any incident that resulted in damage, if the failsafe was set to take in any model fast.

As we all know from our school days (applied maths), that KE = mv^2/2. Again the potential to do damage is related directly to V^2, it could be argued, that it would be reckless to maximise speed.

We could do a calc on the velocity using again our basic v bar = delta d/ delta t and the deriatives which show the change in velocity, relative to time, using g for the accelerating force. This could then also be related to our fluid expressions etc. But we do not need to do all that, lets leave that to BEB.

What we all know that the majority of gliders, including the Sagitta, can easily reach 60-100 mph in a prolonged dive. Generally these models accelerate faster (due to their inherently lower drag) and reach a higher speed than the majority of power models (again due to low drag).

The other issue is of the structural integrity of many gliders. many gliders have Glass/Carbon Fibre or similarly strong bodies. Whish do not deform easily. Again taking a BEB approach, the work done in deformation due to all the energy being dissipated in a small distance and area will be high.

From this I would contend that hard over is not a good solution for a glider to be deliberately piled in. I would certainly not want to justify the process.

It would be far easier to justify, a slow speed flight approach and confining the flight pattern as far as it is practical to a known safe area.

I was thinking, this is a typically BEB type scenario for his students, relating many apparently independent relationships to predict an outcome of a system.

Alan Cantwell10/03/2012 14:52:19
3039 forum posts

dont get you, its out of control, and it has to come down, the slope lift is making it go to where it should not go, how the heck would you be able to get it to a know safe area? no amount of fancy, useless calcs will help, in any way, shape or form, you have NO CONTROL, having it go hard over will at least save you a long walk, you could always do what i do, use 35mhz in your gliders

Erfolg10/03/2012 16:37:47
avatar
11863 forum posts
1374 photos

Alan

I am talking about flat field flying. Although the nearest serious housing is a minimum of a mile from our site, there are some farms, roads and lots of cows etc. These are hazards.

I do not fly slope anymore, probably +20 years back, how time does flyfrown. All my experiences on the slope, were even more remote, with respect to location. Generally the most frequent hazard were the sheep, with the occasional hill walker. The other thing with slopes, is that once you are out of the lift area, you are generally coming down, So would still be tempted to go for either straight and level, or possibly a wide circle.

In all cases I still see that excessive speed as being unhelpful and hard to justify.

I do think it is easier to justify full down and pile the model in, with large power models, to keep them within the confine of the field, rather than potentially leaving the safe area, particularly if there were houses near by. Yet I would still hesitate to recommend such a drastic action. Particularly if the model were to actually be just out of the field, it could be disastrous.

I much prefer the idea of low speed impact, endeavouring to keep in what I would consider a safety zone, which could be outside the flying field.

I guess we will not agree, other than we both want to reduce the hazards and consequences.

Alan Cantwell10/03/2012 16:46:10
3039 forum posts

well, now you have given more details, so do i!! power model will not catch a thermal, yours can, so slow over will suffice, as will motor to tickover for me, and all neutral, glider failsafe--interesting, and now, i hope, well sorted

Erfolg10/03/2012 17:31:37
avatar
11863 forum posts
1374 photos

Alan

I started to go over to 2.4 about two years back.but continued with 35, as it worked so well, with no interference or other issues. Both tx's Futaba 6EX models. I hardly used the 2.4 as the 35 models were so satisfactory.

I also bought a 8FG (about 6 months ago) as there not enough memories with the 6EX.

I had no trouble that is until about 6 weeks back, when a gilder (electric conversion) went in about 1/2-1 mile away at +200m high, not apparently responding, having been preceded by an electric sport model, where the radio can almost certainly be said with certainty to have failed. The model piling into a hedge with no control, later inspection showed the radio was not working, at all.

So it is now 2.4 and thoughts about the best failsafe settings, previously just leaving low throttle and neutral. The glider (electric) was at such a height that it could have gone for miles if it had not gone into a series of stalls and recoveries.

All Topics | Latest Posts

Please login to post a reply.

Latest Forum Posts
Support Our Partners
Sussex Model Centre
Slec
CML
Subscribe now
electricwingman 2017
Advertise With Us
Sarik
Latest "For Sale" Ads
NEW POLL - has the pandemic altered your event safety perceptions?
Q: Has the covid pandemic deterred you from attending shows and events in 2021?

 No, I'll be attending just as many as I usually do
 No, but I'll choose my event with greater care
 Yes, I'll attend fewer events going forward
 Yes, I wont attend any where previously I have

Latest Reviews
Digital Back Issues

RCM&E Digital Back Issues

Contact us

Contact us

Magazine Locator

Want the latest issue of RCM&E? Use our magazine locator link to find your nearest stockist!

Find RCM&E!