Spektrum/JR user warning

[Martin McIntosh]

Yesterday I went to the field with my 1:5 Chipmunk which has only had four flights over two years ago. The YS fired up and ran fine but the flaps and ailerons were all over the place; in fact the ailerons were reversed so no flying. This is the third time that this has happened with those brands but the first two models were not so lucky. I googled it and it turns out that it is not uncommon. One cure was to transfer the model to a spare memory then back to the original. This worked but don`t ask me how.

[Brian Cooper]

Thanks for the warning. 

However, I must say this is a new one on me. 

 

The only other make of radio I have encountered which reversed its channels was on Multiplex radio. . It happened twice, with two separate radios and a couple of years apart.

On both occasions I was test flying a model for a club mate.

 

The first one reversed the ailerons just before taking off (they worked properly in the pits).  Fortunately I noticed it before the model left the ground. 

It turned out to be a dry joint somewhere in the Tx. 

 

The second one reversed the ailerons after a couple of minutes in flight (Wot4 XL) . . . That kept me guessing, but I managed to land it without any damage.  I wanted to get it down quickly in case anything else decided to reverse itself. 

We never got to the bottom of this one but we suspect it may have been a mobile phone being too close to the Tx during the flight. 

 

The owner lost faith in the radio and consigned it to the bin. 

 

[Paul De Tourtoulon]

I happened on one of those indoor things Sunday afternoon on the model field, only the brilliant flying of its owner saved it, and yes it was one of 'those' radios Martin.

Edited October 9, 2023 by Paul De Tourtoulon

[Learner]

2 hours ago, Martin McIntosh said:

Yesterday I went to the field with my 1:5 Chipmunk which has only had four flights over two years ago. The YS fired up and ran fine but the flaps and ailerons were all over the place; in fact the ailerons were reversed so no flying. This is the third time that this has happened with those brands but the first two models were not so lucky. I googled it and it turns out that it is not uncommon. One cure was to transfer the model to a spare memory then back to the original. This worked but don`t ask me how.

Which model transmitter?

[Futura57]

Last week I replaced a couple of Assan receivers in well flown models with Spektrum receivers because I am slowing migrating to that brand. The setups were done at home, on my NX8 tx, but when I got to the flying field and did my preflight control checks I realised I needed to reverse the rudder channel. At home the model had been inverted on the table, Duh! Anyway, at the field I also decided to put some exponential on elevator and ailerons. I tested all the controls again and was happy. I then had a successful flight. I then turned to the second model, switched the model memory over on my NX8 (still turned on) and had a couple of successful flights. Super. I went home happy.

 

The next day I went flying with the same two models. This time I flew the second model first because it was already selected on the NX8 at boot-up from the day before. No problem with the controls or flying the model. I then switched the NX8 to the first model memory and to my horror the preflight checks showed that the rudder needed to be reversed, AGAIN! Not only that, I checked the exponential settings I had programmed the day before and they were back to zero defaults! So I repeated the setup again and flew the model no problem.

 

After landing I began to wonder why the settings had to be re-entered. I began to think like the software developer I had been during my entire career, and a simple test case I would write if I were developing the tx code. Surely changes to a model's settings should be permanently stored if I switched to a different model memory rather than switch the Tx off? This turned out to be the problem. 

 

So, if you change a model's settings on a Spektrum tx, consider switching it off then back on to check they have been properly stored before switching to another model memory.

 

Frankly I am tired of the lack of quality in any software controlled devices these days. 

Edited October 9, 2023 by Futura57

[Allan Bennett]

I'm not a Spektrum user, but I did read elsewhere that settings could be altered if one switched off the power before the trannie had completed its boot-up.

[Cuban8]

Twelve years with Spektrum, both DX7 and DX8G2 still both in regular use - never ever had the 'faults' described. Many Spektrum sets in both of my clubs and if anything like phantom servo reversing had been experienced by members, the clubs' grapevines would be alive with it...............so far...........silence.

[Phil Green]

I'm a trusting sort, if Martin, Paul & Futura say it happened then my first assumption is that it happened 🙂

[leccyflyer]

1 hour ago, Cuban8 said:

Twelve years with Spektrum, both DX7 and DX8G2 still both in regular use - never ever had the 'faults' described. Many Spektrum sets in both of my clubs and if anything like phantom servo reversing had been experienced by members, the clubs' grapevines would be alive with it...............so far...........silence.

Hmm, that's sort of what I said about the AR500 problems that some denizens of the internet reported for at least a couple of years, right up until the moment I watched my boy's Ripmax Spitfire half roll and spilt-S into the ground when flying straight and level right down the runway. Horizon checked the receiver out as being faultless, but replaced it with a new AR600 "for customer's peace of mind".

 

Some strange things can occur with computer transmitters and receivers, even when they have been operating trouble free for years. I had a couple of AR6100 receivers that just spontaneously lost their bind and flatly refused to rebind. HH's advice is that there is likely to be a corruption, but that they are not possible to economically repair. Just today, at the field my Ripmax P-51, which has an AR6210 receiver would not initialise. Back on the bench the model rebound successfully, but with the aileron and elevator controls reversed.

[FlyinFlynn]

......never used to happen pre computerised days..... bah humbug!

 

seriously though... loss or corruption of data in electronic memory devices is only a seu away.  There is little protection in R/C transmitters against these events happening,  there is no checksums or backup error checking on model memorys, perhaps there could be, after all, memory is cheap now and the software wouldn't be hard to integrate, it is routinely done in professional applications almost everywhere you look.

[Cuban8]

Perhaps the tone of my comments did suggest that the problems some people have encountered with their gear never happened - that was not my intention.  However, posts seeming to  casually trash the reputation of a major manufacturer with millons of £s in sales and in one of the most litigious markets, i.e the US, where any sniff of an intrinsic issue with any manufacturers' gear would have the lawyers on the warpath, seems odd to me.

Rumour and folklore have been part of R/C equipment stories for decades..............from sunspots to high tension power lines, bluetooth signals from nearby houses, microwave towers and mobile phones, reports of widespread lockouts on the old 35 meg PCM gear (mainly Futaba BTW and as an ex-PCM Futaba flyer there was usually no supporting evidence of an equipment problem per se)......'haunted' parts of airspace at flying sites that attract models to their doom in some weird Bermuda Triangle style - happened  at my two clubs and there's not a shred of evidence to suggest any truth in it - oddly I haven't heard of any airborne 'possessions' for quite a few years.

The trouble is that genuine issues can be lost in the noise and smokescreen of guesswork and speculation.

[Nigel R]

19 minutes ago, FlyinFlynn said:

loss or corruption of data in electronic memory devices is only a seu away

 

Luckily for us, SEU events at ground level happen once-in-a-never.

 

[leccyflyer]

I'd never heard of an SEU event.

 As Rodney would say - Cosmic!

[Martin McIntosh]

Thanks all for reading my warning; it would seem that I am not alone.

A bit fuller version of events just to clarify:

 

When we were given the go ahead to fly electric in the evening at the Nats, I bit the bullet first and attempted a flight with my trusty Cap. 21. Aileron reversal on take off and said Cap. spread over the hallowed tarmac. Radio was a 6ch Spektrum DSM1, previously with no problems.

Next time at my local field it happened with my ancient Limbo Dancer. I mainly use JR DSX9 but the memories were getting full so I bought an identical Tx and transferred some models across using a buddy lead in the usual way. I had done a lot of flying with the newer set by the time I put the LD on it. Obviously checked out OK in the shed but I may not have power cycled it after the transfer. Again reversed ailerons and splat!

Same Tx used on Sun. Other than replacing what proved to be lousy TGY Nanotech LiFe packs with Zippies the model had not been touched for over 2 years, in fact the twin packs had not even had their first charge.

Back to the attempt at flight and I have already told you the rest. I was completely stumped until I googled the problem.

[Nigel R]

If I'm reading this right Martin, your ailerons were ok in the pits when you did your pre-flight check, and wrong when you took off, and all without any power cycles of TX or RX?

 

[leccyflyer]

Spektrum support's position is that receivers cannot "lose their bind", which may just be a semantic or terminological variation on what exactly is occurring. I've certainly known other club members to apparantly lose bind with the receiver/transmitter pair, when a model which worked perfectly on the bench fails to respond at the field. I've experienced that myself. Yesterday's occurrence was one in which the model has not been flown for a number of years and simply did not respond. I do wonder whether it was a result of my having updated the software on my DX9 earlier this year, after having an issue with the model memories being out of synch and have asked the question where it will be seen by the Spektrum gurus who advised that software update. However I'd observe that dozens of other models have not required rebinding since that software update.

 

The symptom of reversed controls is one which other users have reported after having successfully rebound their receiver under such circumstances- essentially the settings in the model memory have not translated through to the correct movement of the controls. It further reinforces the important of a proper pre-flight control check - taking the time to do that, rather than waggle the sticks to see if the control surfaces are moving, taking the time to see if they are moving in the right directions.  

[Cuban8]

I've never experienced unexplained control reversal but I do believe that people weren't just making it up, so obviously something was going on. Considering what would happen in the full size world if a fly by wire system suddenly decided to do its own thing (has been known and one particularly disastrous case in particular is notable) then what as operators of small unmanned aircraft bound by all manner of official regulations,  are our responsibilities in the event of a suspected, potential safety issue with radio gear?

Who would be interested?

I have experienced an issue sometimes with difficulty getting a model to link with my tranny. Not the initial bind. Switching on a model indoors for testing has sometimes been problematic, but I've put that down to a noisy electrical environment that upsets the systems self checking during linking. Switching on near cars (testing) and close to a wire fence has also upset the process. A good thing really because the system will not link until its happy that so many good data packets are received. Keeping clear of noisy electrical environments and metal objects solves the problem. Once linked, the connection remains solid no matter where the model is.

 

 

Edited October 10, 2023 by Cuban8

[leccyflyer]

Andy Kunz at Spektrum believes that the most likely cause of what I experienced is that an incorrect model memory was retained when I updated the software on the DX9 earlier this year, so when I select the Ripmax Mustang in the Select Model Menu, it is not picking up that model. When I rebind to that model memory, those are not the settings for the Mustang at all. He did give a technical explanation on how the bind is encoded in the transmitter, but to be honest that's beyond my ken. The upshot is that the diagnosis is that the model was ultimately bound to the wrong model memory.

[Nigel R]

2 hours ago, leccyflyer said:

Spektrum support's position is that receivers cannot "lose their bind",

 

I imagine they encode a unique TX ID number in their permanent storage.

 

The TX software version may form part of that ID number.

 

That might be deliberate.

 

2 hours ago, leccyflyer said:

It further reinforces the important of a proper pre-flight control check - taking the time to do that, rather than waggle the sticks to see if the control surfaces are moving, taking the time to see if they are moving in the right directions.

 

Couldn't agree more. At the flightline - check all switches / dials in correct position, check controls move as expected, and only then go.

[Denis Watkins]

2 hours ago, leccyflyer said:

It further reinforces the important of a proper pre-flight control check - taking the time to do that, rather than waggle the sticks to see if the control surfaces are moving, taking the time to see if they are moving in the right directions.  

 

 

[Paul De Tourtoulon]

2 hours ago, Nigel R said:

Couldn't agree more. At the flightline - check all switches / dials in correct position, check controls move as expected, and only then go.

Surely you should do this before Every flight on the taxiway.

[GrumpyGnome]

I'm not checking my retracts on the taxiway 😀

[Futura57]

8 hours ago, Cuban8 said:

However, posts seeming to  casually trash the reputation of a major manufacturer with millons of £s in sales and in one of the most litigious markets, i.e the US, where any sniff of an intrinsic issue with any manufacturers' gear would have the lawyers on the warpath, seems odd to me.

Rumour and folklore have been part of R/C equipment stories for decades.............

I've spent the best part of an hour trying to reproduce my problem, but guess what? No joy. But it definitely happened, and I was mildly furious about it at the time.

 

After a lifetime in software development I can attest to the difficulty in recreating a problem sometimes as described by a disgruntled user. It's frustrating for all and sometimes the only way is to instrument the code ( write to a log every time certain functions are executed or certain events occur) and sit and wait for the next occurrence of the problem. Sometimes it doesn't happen again for months or even years, but eventually it does. Then the fun starts by trawling through the log and the realization that, for example, assumptions about the sequence in which certain events are triggered are incorrect. Or how certain seemingly unrelated precursors come into play.

 

In the two and a half years I've had my NX8 I have experienced unbound receivers twice. They worked one day, but not the next. They were genuine Spektrum rxs which I had to rebind.

 

The first day I got my NX8 out of the box I hit a problem. It would not connect to my WIFI router to update to the latest version of the operating system. Thinking there was an issue with my router I messed around with my router until none of the 'smart devices' in my home worked anymore. All had to be reinstalled after resetting the router. After a Google search I discovered many other NX8 users had the same problem. The solution was to connect the NX8 to a PC via USB and upgrade the operating system via that route. So out of the box my NX8 had a bug. Ironically the bug was preventing me updating to eradicate the bug! I find this totally Unacceptable, I don't care who the manufacturer is.

 

[gangster]

I have little sympathy for those who do not do a functional check as the last thing they do before hitting the gas and taking off. It’s your last chance be certain all is well. It takes seconds to slowly cycle Aileron and ele and glance at the ailerons and say to yourself “ Right Up” as you watch the right aileron go up. It ain’t going to reverse itself in the next two seconds. 

[Futura57]

3 minutes ago, gangster said:

I have little sympathy for those who do not do a functional check as the last thing they do before hitting the gas and taking off. It’s your last chance be certain all is well. It takes seconds to slowly cycle Aileron and ele and glance at the ailerons and say to yourself “ Right Up” as you watch the right aileron go up. It ain’t going to reverse itself in the next two seconds. 

I don't disagree. As operators we are prone to error e.g. reversed ailerons. But our equipment should be fit for purpose.