Fault finding after a crash

[Keith Evans 3]

How do you go about fault finding after a crash .

I'm looking for a more in depth response than " check the TX check the RX etc,

Apart from sending the TX to a specialist what can I do as a lay person  .

I currently check for binding ,range check , battery ,aerial security etc . and similarly for RX  servos , on board wiring , switches and battery ,along with clevises push rods , hinge security ,flying surface"s integrity and anything else that may work loose .

It is the TX and RX I'd like more ideas on .

Thanks Keith

[ken anderson.]

hello Keith, hard to give you a definite answer,as only you know if the model was in a fit state to fly before it left the ground...ie,did you change anything about its set up,have you been fiddling with it,put it to one side and forgot you had made any changes,until it left the ground?

 

i would say most of the stuff we use in our hobby is nower days fit for purpose and as such never(rarely) gives cause for concern.one time seeing a model crash was an everyday occurrence and accepted as part of the hobby,now (most) days its rare to see a model go the journey.

 

so if you are regimental in your build/set up and pre/post flight checks, your models should be enjoying longer lives...... the weak part is us,the pilot....dont think that fully answers what you've asked but may throw a bit light on the matter.

 

ken anderson...ne...1..light on the matter dept.

 

PS....i lost one a couple of weeks ago,by getting distracted and choosing the incorrect model...and launching my model into the deck.☠️

[Allan Bennett]

As a lay person, meaning in this context one who has little electronics expertise and/or test instrumentation, there's not a lot else you can do other than rigorous common-sense checking of electronic and physical connectors, range-check, etc.  Another thing is to be aware of any minor changes -- e.g. if you're constantly having to add a bit more trim it usually means something's coming loose.

 

After I lost a couple of models I was saved from losing a third one as my Tx then gave me an 'faulty antenna' alarm.  After speaking to him on the phone the vendor sent me a replacement antenna which I installed myself but, not being 100% sure of my soldering capabilities, I sent the Tx to him for checking.  To cut a long story short it eventually transpired that the Tx had a faulty RF board.  That kind of thing the average modeller can't be expected to discover by "normal" checks.

[Denis Watkins]

I have to quote this one Keith, as I have witnessed pilots berate their radio gear a number of times when this occured.

The Stall.

 

A number of times pilots have claimed they lost control contact with the model when it was either a long way out or coming towards them.

The weakest super sense with our sight, is depth of field and gauging the speed of an object, we are not good at it.

Pilots were unable, due to their viewpoint, to see the model travelling too slowly.

 

Before or at the stall, the surfaces reduce in authority, and it does feel like electronic control has been lost.

As the speed of the model reduces, the sticks of the tranny feel mushy and soft.

 

 

 

[Jon H]

I tend to echo the comments of the Ken and Denis. 99% of model crashes are pilot error, either through stick twiddling issues or insufficient maintenance/poor setup leading to something causing a crash. Real mechanical failures are very rare and, sadly, very hard to diagnose. 

 

In 30 years flying i have suffered only one receiver failure and two servo failures so you have to be pretty unlucky. 

 

I am not trying to suggest it never happens though and realistically the only way to find out is to set the item away for testing. The problem with that is the time for turn around and the chance they may not be able to replicate the fault. 

[David Davis]

I have had a modern transmitter go bad on me twice. In the first intance the screen did not even light up when I switched it on. Fortunately I had a second simpler transmitter by the same manufacturer which i keep as a buddy box. I was able to fly with that. On the second occasion I lost signal in flight. The transmitter was posted off to the agent and a new rf board was installed.

 

Other than that all of my crashes have been pilot error. When I was first learning how to fly over thirty years ago my crashes were caused by inexperience.

 

Having gained some experience I have only crashed twice in the last two years. In the first instance I was bringing my WOT 4 XL into land when suddenly it just spun into the ground. Damage wasn't serious but when I examined the model I noted that the rudder servo output arm was at an acute angle to the servo instead of being at ninety degrees to it. What had happened was that I swapped a Hitec servo with a JR servo without changing the output arm too. Hitec and JR use different splines and the arm had shifted locking the rudder over to the right.

 

On the other occasion I stalled the model on finals. I was flying so slowly that I had no airflow over the elevator so no response to my inputs.The model crashed into the ground from a height of about four feet but being a very lightly built vintage model I've now got quite a big repair job.

 

Pilot error in both cases.

[Denis Watkins]

To buy a range of testers in one go does cost a few bob, but maybe over time, accumulate the basic kit.

An RF output tester is about £20, and will test your aerial.

A multimeter does much testing, for e.g. Continuity, volts, amps.

An ammeter,

A servo tester is invaluable for setting up.

 

Tx gear does self test, with failsafe, and range test, and the monitor shows stick movement and switch selection.

We are not in the dark really.

[Don Fry]

Answering your question, fault finding after a crash, can be difficult. Some things are easy, a battery is flat, ailerons reversed. But go past that point, sorting out faults that existed before impact, and faults caused by impact is the reason the likes of he CAA have forensic crash investigators.

I find telemetry is helpful, as it can give something of a flight data recorder. And it can prevent damage. Last year I was having odd low battery warnings. Investigation revealed a servo that was stiffening up. Digital, it was taking big loads to get its’ instructions done. On the servo tester, it lasted another 10 minutes before failure.  Change servo, off we went, fly. No telemetry, crash, how would I sort out the difference between crash damage, and pre existing damage, without a professional knowledge of servo manufacture, and a fancy forensic laboratory, and staff who can work the stuff in the laboritory

[Dale Bradly]

Completely agree its 99% pilot error, but that extends to errors in setup/preflight, e.g. not charging batteries, or damaged plugs not being resolved.

However  Start the process immediately if possible. Treat it like CAA or the police would. 

Obviously immediate safety must take priority (eg disconnecting and removing damaged lipo), then:

Do NOT turn off your transmitter,  and walk over to the scene and start pulling apart the wreckage. 

Note where servo and/or control surfaces positions are, and recall your control inputs, and compare with what the plane did in its final moments. Is there something there? There is a world of difference between "I have no control" to "I have no elevator"

Check radio responses, does the plane respond correctly? Battery voltages/telemetry indications/range check show anything?

Lipo voltage/Fuel level/carb position??

Only then start pulling things apart, but carefully note damage that doesn't seem to align with the crash. Are plugs in correctly, pushrods still connected?

Probably At home, Test individual components (tx, rx, esc, bec, switch etc) one at a time as part of a complete bench setup, one component might show up faulty 

 

The evidence is often there at the scene, but may have disappeared by the time it's dumped in the back of the car.

 

 

 

[leccyflyer]

Good advice there -especially on not switching off the transmitter. Seeing immediately if there is any loss of radio system function can be a vital part of working out what went wrong. On two separate occasions I've had models flying normally only to flick hard and roll into the ground, to then find at the crash site one aileron servo locked hard over. That's something that you can check at the crash site - how do the controls look and is everything working correctly? Most often it's due to pilot error. but malfunctions do occur on occasion.

 

As an example I read many reports of AR500 brown outs and had never had a single problem, despite using several of their receivers and I didn't have a problem, right up until the flight where we did, resulting in the destruction of our Ripmax Spitfire. Horizon replaced the receiver with the "no fault found, but AR500 replaced with AR600 for customer's peace of mind" comment. In that case the model flying completely normally and then doing something it had never done before, totally unexpectedly was the clue.

 

Likewise for interference -it's often said that interference is widely and wildly exaggerated, but there is little doubt that electric models were more susceptible to internally generated interference on 35mhz. We watched only the other day as a flying pal suffered repeated uncommanded control inputs whilst flying a usually very well behaved model - very clear control glitches, none of which were fatal, but definitely not pilot error or any sort of weather effect.

[Peter Christy]

The actual nature of the crash can often be a clue as to the problem. Putting pilot error to one side, for the moment, the first thing I always ask is 1) Was the fail-safe set? and 2) Did it work?

 

If the failsafe was set correctly, and yet it didn't work, that points to a loss of power to the receiver and servos. If the failsafe did function, then clearly the receiver lost communication with the transmitter. This could be broken aerials (at either end) or a transmitter failure. Electronic failures in receivers that stop it receiving, but allow the failsafe to work are very rare. Equally, outside interference is very rare - but not unheard of - on 2.4 GHz.

 

Other than that, modern radios are extremely reliable, and the usual cause of crashes is pilot error. This is usually a stall caused by slowing up too much on the landing approach, or the dreaded "not flying down-wind fast enough" on a windy day!

 

Each accident is unique, so trying to generalise is difficult, but hopefully the above will help with some pointers to the most common causes.

 

--

Pete

 

[Peter Christy]

8 minutes ago, leccyflyer said:

Likewise for interference -it's often said that interference is widely and wildly exaggerated, but there is little doubt that electric models were more susceptible to internally generated interference on 35mhz. We watched only the other day as a flying pal suffered repeated uncommanded control inputs whilst flying a usually very well behaved model - very clear control glitches, none of which were fatal, but definitely not pilot error or any sort of weather effect.

 

I would be looking for a dodgy connection or switch for that one. If the power was suffering from intermittent losses, that could well cause that kind of scenario...

--

Pete

 

[Chris Walby]

For the crashes other than dumb thumbs I employ the "what its not" approach a technique I picked up from a professional forensic investigator (following two electrical fires at work!). All quite interesting when you are only left with the charred remains of a very complex piece of electrical equipment! Anyway from the last few seconds what did the model do? 

What's it fail safe mode? Did it throttle down, do control surfaces stay on last command or return neutral, what about gear, do they fail safe up or down. When you found the model were the control surfaces in the expected positions? As mentioned above look at the crash site and the state of the model.

 

Over the years I have had servos do wild things due to a poor Y lead aileron connection plug that once unplugged and reinserted never played up again, servos fail due to ingress of water etc. I have a couple of crashes that have been inconclusive like RX loss of signal/ESC overheating which because of the lack of telemetry and repeatability (model damaged beyond repair) I have left the case open as I can't determine which was the root cause.

 

I had a TX and non OEM RX that would bind, range test and work perfectly well pre flight and after crashing (lightly and repairable, just) resulted in me setting a test rig up on the flying field bench (RX battery, RX and some servos) and walking across the flying field....then it would lose signal....in the same arc and the previous 3 crashes - so I stopped using non OEM RX's despite people not having any issues with this combination. The model that crashed was cheap, but the RX was intended for a much larger model and I would have been gutted if I had lost that.

 

PS - Our field has trees, woods and hedgerows all of which do produce some very unexpected turbulence where you would not think it would be, not a problem if you are whizzing about, but if you are close to stall then it can catch you out.

Edited October 12, 2021 by Chris Walby