Servos reversed

[Craig Kane]

Whilst carrying out pre-flght checks on my B-25 yesterday I found that all servos (except throttle) had become reversed.

The model had last flown on Wed evening with no issues and I hadn't touched any of the settings on my Tx. The model was still bound to the Tx.

I'm flying with a DX6i and an Orange 6ch Rx

No problems with any of my other aircraft whch use a simiilar setup.

Luckily it was spotted on the ground as it could have been an expensive mistake!

Has anyone else encountered anything lilke this?

[Alwyn Gee]

Hi

Is it possible that you left your trannie switched on and allowed the battery to go flat and then charged it up again before this last session. I had a similar experience a couple of years ago with a DX6i and this was what had happened to me. It seems that if the set runs out of power strange things happen to its memory. You will notice that when switched off manually it comes up on the screen "Saving Settings" before actually going off. It likely can't do this with gradual power loss.

Alwyn

[Craig Kane]

Alwyn,

Basically, no. Everything was shut down normally. I flew the B-25 and then went on to fly other models that evening with no issue.

I had flown other models yesterday with no issue either. One of our members (who just happens to own the local model shop) suggested that it was due to those "dodgy Orange receivers", but I feel he may have more financial motives at work.......

 

It's all very weird. As I said, just thankful that I spotted it pre take-off.

Edited By Craig Kane on 08/07/2013 11:13:46

[Phil Green]

Did you rebind set a failsafe whilst on the wrong model memory maybe?

Orange receivers dont have any reversing facility so its extrememly unlikely to be the rx.

Whilst there are a few unlikely technical possibilities such as flash corruption, the most likely explanation is unintentional misoperation, ie 'human error'!

Cheers

Phil

[Olly P]

Can't be the Rx.

The Rx is essentialy a decoder - if A goes in B comes out - that can't change without a sofware change in the Rx. The servos reversing sounds like a problem with the Tx end, unless the Rx was disconnected and all servos plugged into wrong channels!

[Craig Kane]

I know it sounds like I was fiddling with the setups, but honestly, I flew the models on Wednesday, put them away, took them out on Sunday and flew.

No messing about with binds, model memories, settings whatsoever.

I'm an electronics engineer by trade and I can't offer a rational explanation!!

Going to put it down to "a quirk of the system"!

[RichB]

I had a very similar experience with my DX6i a couple of years ago

I had done all the usual checks and flown one model without problems. I then flew two other models using the same Tx, again without problems. I then came back to the first model. Switched on, waggled the sticks (I had already done my usual detailed checks for the first flight so assumed all would be as before) all the surfaces moved so I launched it, but it didn't seem to respond to the controls and did a wingover into the ground. It wasn't too badly damaged and on checking it I found the ailerons reversed. All other controls were OK except that I'd had a small (8%) rudder/aileron mix active which had now become 100%!

I checked all the other model memories and found just one with only the rudder reversed.

I haven't had any further problems since then but I now do my detailed checks before EVERY flight.

[Chris Bott - Moderator]

Posted by Phil Green on 08/07/2013 11:28:24:

Whilst there are a few unlikely technical possibilities such as flash corruption,

Cheers

Phil

Just clutching at the tiniest of straws - could a flash corruption be in any way shape or form be caused by having a mobile phone in a jacket pocket while holding the Tx close to said phone? I'm sure this idea has been mooted in the distant past for different transmitters, but I don't think any proof has ever been found?

[Peter Beeney]

Craig, - I’d want a lot of convincing that it was anything to do with the receiver, it’s very unlikely that it modifies the servo information in any way, it simply passes it on. Having said that, nothing’s impossible, and many receivers do use a micro processor to organise the servo signals these days…

I experienced something similar very recently, only on the aileron channel, though, but also using a DX6i, the DSMX version. This was a set up used by some youngsters, which they had been using previously, so I can’t say with 100% certainty that it wasn’t ‘got at’, but I’m 99.999% sure. I always have to make any adjustments anyway, even switching on can be fraught where they are concerned, although I do try and persuade them to have a meddle; but I’ve never seen anyone surreptitiously tinkering, and I think that it’s extremely unlikely.

The model is a Riot, and we have another identical combo, 6i and Riot, so I was able to cross-reference, simply because I couldn’t understand what was going on. The channel throws are all normal direction, with exception of the ailerons, which is reversed. So on this tx the aileron channel had definitely reverted to normal direction, for whatever reason, and which I was able to see. I did find this difficult to understand, but now it would seem as though it’s occurred on more than one occasion now, so at least I now know that it’s a possibility.

All our tx’s use primary cells, so there’s no charging to influence anything, and Chris mentioned the mobile phone, but again unlikely in my case, as I don’t have one…

A few years back I had a new student at our club site, with a 6i. We borrowed a DX7 as the slave tx, and after some initial confusion I found that the aileron trims were reversed on the spare channel on the 7 we were using for buddying. I was assured this couldn’t happen, but nevertheless it did. There is no facility for correcting this.
It would seem as though some of this transmitter software is still not entirely bullet proof.

I’ve always been very careful checking control throws, and part of my training regime is gently encouraging a ‘proper’ control throw check before every take off, irrespective, and even if we’ve only just landed, but now like Craig and RichB I’m paying even closer attention, at least for a while.

I wonder just how many more times this might have happened…

PB

[SR 71]

I have lost 9 models over the last 18 months, i use a JR 11 zero and Jr and Spektrum 8 ch receivers, The last four models have been Sprint turbine jet , totaly destroyed in the fire, Fliton inspire F3A destroyed, Revolver 4 off totaly destroyed, Cap 232 180 again totaly destroyed and all of them rolled over and dived into the ground with no control, ie not responding to any input on the transmitter, the only common factor with all of them is the transmitter. i am now waiting for the arrival of a Futaba 14sg. if i have any more problems then i shall find another hobby

[Phil Green]

>>Just clutching at the tiniest of straws - could a flash corruption be in any way shape or form be

>>caused by having a mobile phone in a jacket pocket while holding the Tx close to said phone?

Its possible under certain circumstances Chris. On mine I write-protect the flash after every write, and every write is checked for completion. Theres not much more you can do code wise to ensure the flash is operating properly, following of course the microchip recommended practises. Its usual to include a checksum so that direct flash corruption is detected, but if the corruptin happens in the (pre write) buffer then the mangled data is written with the correct checksum for that data, hence no error is detected.

So a mobile-phone type of corruption would have to happen as the data was actually being written - ie while the write-protect was off - and this happens at different times with different sets. I believe Spek write at the moment of switch off (via a delayed soft switch of course) whereas others write in real-time. My reeds set for example writes whenever a trim is used, every pip is a write. So for this to happen to a Spek is very unlikely - the phone would need to corrupt the write at the very moment the set is being switched off.

Whilst it is possible, I'm inclined to put it down to what we BT engineers used to call "Subs Misop"

Cheers

Phil

[SkippyUK]

Only time I've come across this is on my YAK with unlabelled cheap servos from CG.

Servos work ok but don't have stops at end of travel and if you move either rudder or elevator to max allowable (90 degrees rotation of arm) during handling then it can go round and be 180 degrees out and on these particular servos, they work OK but now 'backwards' and nearly centred as well - all to do with internal gearing.

I now check all arms point upward before powering up.

Don't expect this to be your problem but rough handling can make gears/ arms slip and internal pot be out but..... exactly 180 degrees out is hard to imagine.

On memory corruption, would the corruption have to be the exact reverse bits to be corrupted and no other bits changed???    hard to imagine without some other changes being evident.

Are there any other models that do have the servos reversed that could have over-written the correct model?

Skippy

Edited By SkippyUK on 08/07/2013 21:08:32

[RichB]

My "incident" occurred on the one and only occasion when I had my mobile phone in my pocket. This is something I never normally do and will not do again!

I contacted HH and explained what had happened. They said that it was not possible for the problem to be caused by a mobile phone but that I could send the Tx in for a check if I wished.

[Phil Green]

No electronics tech would use the term 'not possible'. Thats 'office speak'.

I think its unlikely in the very extreme.

[Peter Beeney]

Well there yer go, Skippy, you’ve just consolidated a line of thought that I’d been poggering over for forty odd years… - There might just be a common thread here. (Sorry!)

In my case, it was the aileron servo control that reverted back to normal from reversed within the transmitter, I know this is so because I checked it against an identical set up. All the other channels are in the normal position. A clue might be in Craig’s OP, because he mentioned that the throttle didn’t change.
So could it be that all his servos were coincidentally reversed to get correct model operation, and then they all reverted back in one shot to what might be the default position, normal? The servos would only appear to be reversed if this happened. But with the exception of the throttle, however, because this is usually in the normal mode anyway regarding Spektrum and ESC’s. I’m not sure if this is significant, or helpful even, but as far as I’m concerned, the settings in our transmitter did change.

At the moment I have some DX6i’s that are OOS for a few weeks; and so just as an experiment I will reverse all the servos in one model memory in a couple of them, including the one tx in question, and just switch them on and off randomly for a while; and then check later to see if anything has changed. I’ll be very surprised if it has, I must say, but if I get really lucky, and the channels do revert back to normal all on their own, at least it will be a starting point for someone to investigate this.

In the meantime, I’m going to assume that it’s possible to get an involuntary servo reversal, even if very infrequently, and, as I said, I’ll be very vigilant….

PB