Air Crash investigation needed

[Paul Williams]

I have just suffered one of them weird crashes that plaque you for months because you just dont know what went wrong. I noticed the plane glitch then recover on a straight and level pass I then tried to bring it round for a landing but it would not respond. The fail safe did not operate the motor was still half throttle, then it turned over and dived leaving me watching helplesly as £700 worth plunged to the deck. The Robarts survived but thats about all. The Spektrum DX7 and AR7000 Tx I later tested in another (cheap) model they showed no faults. The power system still worked and the voltage regulator has a twin feed one to Rx one to CDI so any faults there would have cut the motor. The li-po was 85% charged when tested.
I am stumped and although 2.4 is supposed to be rock solid I cant help but feel interferance or something of that nature caused this accident.

I could not find the black box anywhere LOL.

[Paul Williams]

My point is if the TX or RX is faulty how can I tell?

Edited By Paul Williams on 05/06/2011 19:16:24

[Tim Mackey]

Unfortunately you cant - other than by trying it in another model as you have done.

Sending it back to HH will likely result in either NFF, or replacement - highly unlikely that they will actually say they found a fault.

How was your Tx aerial orientated, and while we are on that subject, how about the rx installation? Did you range check? "Glitching" is VERY unusual, so how can you be 100% sure that the glitch was not a result of a mechanical issue such as a clevis, or servo etc.

[Paul Marsh]

sorry about the loss.

The only thing I can think of is either undervoltage. Did you check your receiver before you powered it down?

If the receiver was flashing, would indicate loss of power. Unless the battery got detached in the crash...

[John Privett]

Just a thought - feel free to dismiss it...

 

You mention the voltage regulator feeds both the Rx and a CDI. I assume that's an electronic ignition unit for a petrol engine. Now I don't have any petrol engines in my IC fleet, but is it good practice to run the CDI and the Rx off the same unit? Shouldn't ignition circuits and the radio electrics be kept as far apart as possible to avoid interference?

Edited By John Privett on 05/06/2011 20:53:55

[Alan Cantwell]

run the battery through a cycler, see if is holding what it shows, this weekend, i have charged up a HUGE battery in my big corsair, it shoed a 105%, it cycled a lot lower, less than half, sorry about the plane, its sometimes a hell of a price we pay in this game

[Alan B]

Hi Paul - sorry to hear about the crash.

 

As Tim has out lined the usual suspects. The favourite is usally a Straight TX Antenna and pointing it directly at the model. Even though you have retested the equipment I have a couple of occasions when I have been ready to fly I have touched the lipo lead and the eletrics basically switched on and off. This was due to a dodgy solder joint at the connector. I now wiggle everything before flying to make sure everything is ok.

 

Now to the conspiracy theory. Is 2.4 Rock Solid? I have 2 incidents recently with loss of signal on 2.4. Over the past year at Barkston we have had just to many incidents where flyers have commented on no response from the model and resulting crashes. The other week Ali from Al's hobbies was visiting and lost a very expensive jet - lack of response. At the warbird meet a £15000 Sea Fury went in, apparently the same problem. Fellow club members have lost a few recently. This has been doing my head in and I'm sure there is some sort of interferance there and I have been making my feelings known. A club mate told me last week that our chairman is speaking to the RAF to find out is some sort of new radio beacon has been installed that we are not aware of something to that effect.

 

In a nutshell there have been just to many incidents to be coincidence. I dont know. Is 2.4 nearly infalable or just recently has something been added to the airwaves that we do not know about. I do not want to start a massive 2.4 debate again with all the technicalities of frequencies etc. Brown outs etc. My own feelings are that for 2-3 yrs I was comfortable with my change over to 2.4. I now feel that I have lost confidence in the system.

 

Just to add that most the incidents occurred on Spektrum - A few JR spektrum module TXs and even on a couple of Futaba systems.

 

Al 

Edited By Alan B on 05/06/2011 21:23:23

[Alan Cantwell]

did the nats suffer from such problems? i was on the stand all weekend, didnt keep pace with the flying, as for the ariel, i was told it does not matter where its pointed, at an angle, or straight out, would love to know which ones right,

[Martin Harris - Moderator]

Posted by Paul Williams on 05/06/2011 19:13:51:

The power system still worked and the voltage regulator has a twin feed one to Rx one to CDI so any faults there would have cut the motor.

 

It's horrible when the cause isn't known. You seem to have covered most things but I'm not sure that you can dismiss the power box though as there might have been a fault on one output. The fact that you saw a glitch before the final lock out would support a possible intermittent power problem.

 

I've never been very comfortable with the fixed nature of the 2 channels pre-selected by DSM2 although informed opinion has it that this shouldn't be a problem - but when you read about video transmission equipment streaming over several adjacent channels then it does get the suspicious grey cells ticking a bit...

 

Let's hope you find the cause for your peace of mind and let me add my commiserations for the loss of your model - we've all been there but it's a little easier to bear if you know it was of your own doing or can pinpoint the cause.

 

 

[Alan B]

Hi Alan

 

I once pointed my TX with a straight ariel when range testing and had no response. I bent the ariel and had imediate response. So that basically gave me my own answer.Try it yourself see what happens.

 

As far as the Nats go. When we fly most weekends I think the Airfield is stil active. When the Nats are on the Airfield is closed and do they they switch everything off - I dont Know

 

Al

Edited By Alan B on 05/06/2011 22:21:04

[Martin Harris - Moderator]

I did some testing with my Jeti a few weeks ago using its telemetry to monitor the signal in real world conditions - there was a very small reduction at extreme range (far far further away than I would normally fly) when pointing the aerial directly at the model but no loss of control and nothing measurable at normal circuit positions.

[Paul Williams]

Thanks for a fantastic response, to try and cover some points;

 

The voltage reg actualy only has one output then splices into two so any problem there would have cut the engine I thought about two systems but I like the idea of the engine cutting should there be a power loss to the Rx.

 

The ariels were in the correct postion the model has flown before without any problems.

 

No I did not range check before taking off some thing I will regret and inprove on.

 

Alan C, the system was Li-PO powered it was a new battery and detatched on inpact so I could not check for flashing lights on the Rx.

 

The Ign unit was on the engine side of the firewall and the voltage reg 4" away from the Rx plus had a ferris ring on the output.

 

I have also range checked with the ariel pointing straight and it failed the range check I binned that set and bought a new one which passed in any postion! I try to remember to angle the ariel but to be honest I might have been flying with it straight. that is some thing else I could experiment with using the same TX & RX and my old hack.

 

I was talking to some club mates who seem to think JR have moved away from Spektrum because of problems which include lockouts. I would lke to find out if any crashes of this nature have happened with the new JR system?

 

The problem is I have not enough understanding of the systems to make any judgement I have seen quite a few glitches at our site when people fly 35 so rushing to change would not solve the dilema.

 

Is 2.4 solid or can the system fail under certain conditions?

What can we do to reduce the risks?

Are different makes more reliable than others?

 

I have one large scale model left in my shed and I dare not fly it, plus I will now have to mope around the house huffing and puffing untill er-indoors says "for gods sake go buy another one. " Its hard work!

 

 

 

 

 

 

 

 

[Tim Mackey]

Posted by Alan C on 05/06/2011 21:20:34:
did the nats suffer from such problems? i was on the stand all weekend, didnt keep pace with the flying, as for the ariel, i was told it does not matter where its pointed, at an angle, or straight out, would love to know which ones right,

 

Definately hold the Tx so that the aerial is NOT pointing at the model.

This is the correct orientation

Fuby....

[Tim Mackey]

In summary, and IMO only......

NO RADIO SYSTEM is 100% certain, and we ALL run a risk of losing ANY model EVERY time we fly. What we ask of our kit is a heck of a lot, and literally millions of little components are all working hard in difficult and hostile surroundings - its little wonder that sometimes they fail. Its always very annoying and frustrating when we get problems for which we cant account, and understand it when people lose faith in their systems.

I have had - probably three that I can recall - unexplained loss of control scenarios with my DX7 Spektrum in over 4 years of continuous and hard use.

I am loath to blame the system per se - simply because I have had THOUSANDS of trouble free flights as well.

Edited By Tim Mackey - Administrator on 05/06/2011 23:26:02

[leccyflyer]

What is the source of the correct transmitter antenna orientation photo above? There's nothing that I can see in the DX-7 manual that instructs users to tilt the antenna so that the tip is vertical.

[PatMc]

Posted by leccyflyer on 06/06/2011 11:34:59:
What is the source of the correct transmitter antenna orientation photo above? There's nothing that I can see in the DX-7 manual that instructs users to tilt the antenna so that the tip is vertical.

 

It's not peculiar to Spektrum/Futaba or even 2.4GHz. What Tim's diagram shows applies to all frequencies. It just that with the shorter aerial it's easier with 2.4 to tilt it significantly. Tilting to the side would be just as effective & possibly slightly better than as shown for the Futaba.

Some of the continental pizza box style 27 & 35MHz Txs had/have the aerials on ball sockets to allow a degree of tilt so that they're near vertical but more importantly never pointing at the model.

[leccyflyer]

Indeed - the majority of Spektrum flyers at the club have their TX antenna orientated sideways, parallel to the ground, so as not to point directly at the model and I believe that information came from one of the early adopters of the technology, who is in the trade.

 

The manual doesn't say anything about the actual orientation, but in an appendix of generic safety information states that the TX antenna shouldn't point at the model.

 

OTOH the Frsky modules state categorically that the TX antenna should be vertical.

[Tim Mackey]

As PatMc says....it was just to illustrate that the aerial should not be "pointed at" the model.

 

[PatMc]

Posted by leccyflyer on 06/06/2011 13:38:42:

 

OTOH the Frsky modules state categorically that the TX antenna should be vertical.

AFAICS that's only stated in the range check procedure instructions. Since the model will be virtually level with the Tx at this time the orientation will be for best signal. There doesn't seem to be any recommendation for when the models in flight.

[malcolm woodcock 1]

I've just managed to stall my WOT4 e in doing some serious damage. I couldn't get the c of g right (70mm from leading edge at root). I had added as much lead as I could to the front end in the cowling but still could not get it further forward than 80mm. Rang Ripmax and they suggested pushing the c of g back to 90mm and adding 90 grams of lead, which is less than I used originally. Can someone explain how moving the c of g back arbitrally is going to work. This is the second shoddy piece of kit I've had from them where I can't make there models work, the first was the Foamy WOT4 which had an esc rated at 30 amps but it drue 33 amps at peak in the system. How do I avoid this happening again.

[Peter Beeney]

 

 

I, too, was very sorry to read about the crash, and particularly the unknown circumstances surrounding it. From the O/P, can I make some assumptions? I’m sure the throttle ‘failsafe’ had tested ok? If so, then the fault would appear to be be related in some way to the power supply, because anything else, such as incorrect aerial orientation or interference, or the tx intermittently ceasing to transmit, (this can happen to the DX5e tx!), in fact any loss of signal for any reason, would result in the throttle closing. Unfortunately it might not have lessened the crash damage that much but at least there would have been a pointer to the fact it wasn’t a power problem.

Stating the obvious, and as in the O/P, the regulator was still working so any fault must have been beyond that. I think the reg. would have supplied the requisite (presumably) 5V, it would be all or nothing, therefore I’d say it’s most unlikely to be a low voltage. So that leaves a disconnection on the rx supply as the fault, or possible power borne interference on the power leads into the rx. I do know this can happen, there is a thread on it, Interference on 2.4 gHz, by Soeren Schmidt, 22 - 06 - 2009. This was a different set of circumstances, mutual inductance on some battery leads, from a CDI unit, but the interference was massive, and entered the rx on the power leads. I think it’s possible that 2.4 is susceptible to this, whereas I’ve never seen any evidence that 35 is. I am by no means saying this is the problem, but because the throttle didn’t shut, very little else seems to fit the bill; but as a minus it appears to be unusual that it had not shown itself before. That first glitch, too, I’m not sure that any rf transmission problems are able to do this, will the microprocessor in the rx pass any spurious information to the servos? Power glitches can cause the servos to go berserk. It would be interesting to measure any voltage excursions on the rx bus-bars, if they are there, it may well be any variations are affecting the servos directly, the glitching is simply the servo trying to continually compensate for the rapidly fluctuating voltage. But why your model’s radio ceased to function entirely, I do not know.
Again I’m assuming that the hack model has a standard battery set up, so the split battery feeds would not have been tested here? Is that possibly another clue?

One alternative to the split supply might be to have two separate batteries but with pair of relay contacts in one leg of the CDI supply; the relay operated by the supply to the rx, and the relay might be a mechanical variety or perhaps a solid state type; if this supply then fails the engine stops. It would be very necessary to make sure there is no interaction, but by keeping all the leads well clear of each other I’m sure it would work ok.

I’ve not been involved with CDI units at all, but from observation I’d say they are not particularly juicy. However, the capacitor charge may be a ‘one-shot’ affair, so that when the current flows it will be a short but hefty pulse, just right for causing a nice inductive wave in any adjacent wires. A strategically placed diode or two and a spot of decoupling capacitance may help to reduce these effects.

Hope you manage to reach some conclusions, if only just to be able to be be able to fly in the future with some confidence.

PB

 

[Former Member]

[This posting has been removed]

[Ben B]

I don't suppose anyone was flying POV nearby? Some POV systems are 2.4Ghz and can cause interference. Fundamentally, no system is foolproof. At our field we have a mobile mast which causes issues at one end and a mobile jammer near the prison which also causes problems. A digital frequency hoping system will help reduce intereference but it can't completely remove it.

[Mark B]

How many people were flying at the time?

I read somewhere that only 12 2.4 txs can fly at one time. I dont know the validity of that statement and it would depend on what brand, but in the search for model justice, it could be a possibility.

 

Having said that - my most expensive model is a 40 size trainer used for just about everything, and its on 40mhz.

 

I'm a student with a lack of money BTW.

I fly many much smaller aerobats and scale models all scratch built, so know what its like to lose a model.

[Pete B]

Hi Mark,

 

2.4Ghz can handle many more Tx's than 12 - I believe it is up to around 100+. That's one reason why Spektrum have introduced the DSMX.

 

I'm a little concerned to see that you are using 40Mhz if you are in the UK. That is for surface models, not aircraft. The problem isn't so much that you will interfere with a surface model, more that your aircraft can be shot down by another 40Mhz Tx being switched on whilst you're flying.....

 

I appreciate you will be on a budget but it's worth thinking about changing to the correct band.

 

Pete