Warning X8R RX - do not use with CDI on IC powered 'planes

[Bob Cotsford]

It's a difficult one, I had a couple of years flying two and four stroke petrol models fitted with X8Rs running the first LBT iteration firmware with no problems whatsoever. Ron has had a very different experience and I can understand his frustration but as the problem isn't repeatable AND FrSky is a Chinese company I think that in making a claim for losses would it would be hard to gain solid proof of fault.

Ron, that's a hard lesson to learn, I'm glad I've gone all-electric now!

[Tim Flyer]

Blimey very sorry to hear that Ron. I thought you had been quiet recently. That’s awful . I must admit I have stuck with glow engines so far, as fo me they are always the simplest easiest and most practical option as i’m unlikely to go bigger than the 180. I’m slowly sticking my Hurricane back together, but while I do that have certainly been enjoying my home baked LA7.

[Ron Gray]

Quiet licking my wounds Tim! To be honest I didn’t see any coincidence in my losses, I just put them down to frank accidents / bad luck or dumb thumbs, mainly because I couldn’t put my finger on something specific. It was only this latest loss which lead me to a conversation with Rich at T9 and he mentioned what happens when the RX locks out that the penny suddenly dropped. The circumstances in each case were identical, all gassers and each time, before the loss I got the ‘telemetry lost’ warning. Looking back at the telemetry logs they too are the same, a flat line for RSSI and battery voltage. I should also have thought more about my own skills, after all I’m not that bad am I? I should also have taken into account that over this time I have flown many other ‘planes, both leccy and glow and apart from 1 switch fault and 1 dumb thumbs there have been no other losses / nasty events.

The trouble is that I do love the larger ‘planes but with them comes £ so I have now started the process of re-equipping my other gassers with 868 MHz gear. I will replace the Sbach as the GA30 was a good match for it (well it appeared to be so in the 2 and a bit flights I had with it) but I’m sure I won’t be able to find such a good deal as the last one!

[Tim Flyer]

I don’t blame you quietly licking wounds that is so bad ...., It’s awful that receivers are sold like that without obvious warnings on them. Thank heavens nobody was hurt either. An out of control model is a nightmare. I know you like your planes as we all do , and as well as the replacement cost in money it’s all the time and effort we put into them when assembling them that is also lost in crashes. I do think we should have an “industry classification” for receivers. There are some horrible ones out there that can easily cause trouble. If range or suitability is limited to specific model types there should be a warning. I cannot understand why some manufacturers make some receivers for outdoor aircraft limited range .

[Frank Skilbeck]

Not sure if changing to 868 MHz will do anything but make you feel better, I don't think the interference was specific to 2.4 GHz but more electromagnetic interference upsetting the electronics from which this particular Rx didn't reboot.

I go to a few aerotows (as a glider guider) and all the tugs are petrol powered an do a lot of flying during the day, different brands of radio used and I've never yet seen one have a radio problem, been on a couple of tows where the engine has conked out (including the maiden of my friends 1/3d scale Minimoa!), but not seen any tug crashes which were due to radio lockout.

I only have one petrol model, but it's a noisy two stroke so only gets flown at sites where noise isn't an issue, so only have very limited personal experience.

[Ron Gray]

@Frank - I know what you mean but the general concensus seams to be that the lower frequencies are less prone to interference, including that from ignition systems. LMA pilots seem to favour this setup as well. I will, however, be carrying out some tests using the new Rcexel ignition tester. During my research of this problem it’s come to light that it is very hit and miss, as I’ve said before, I’ve got other gassers with X8R RXs and there hasn’t been a problem with them. But when I emailed a friend who I know uses FrSky and warned him about the problem he told me that he lost 2 models last year with the exact same symptoms which he put down to other things (cheap gas engine!). He replaced the engine with an OS and hasn’t had a problem since. But he will now be changing RXs.

@Tim - a central register would be a good idea, a bit like the ones that are in existence for servos. The problem I see is that there are other things that can effect the performance of a RX, the location and positioning of aerials being but one. For my part all I can do is to raise it on this forum so that all are aware plus I plan to contact resellers to suggest that they put warnings on their websites. 

Edited By Ron Gray on 04/05/2019 18:23:24

[Frank Skilbeck]

Posted by Ron Gray on 04/05/2019 18:22:34:

@Frank - I know what you mean but the general concensus seams to be that the lower frequencies are less prone to interference, including that from ignition systems. LMA pilots seem to favour this setup as well. I will, however, be carrying out some tests using the new Rcexel ignition tester. During my research of this problem it’s come to light that it is very hit and miss, as I’ve said before, I’ve got other gassers with X8R RXs and there hasn’t been a problem with them. But when I emailed a friend who I know uses FrSky and warned him about the problem he told me that he lost 2 models last year with the exact same symptoms which he put down to other things (cheap gas engine!). He replaced the engine with an OS and hasn’t had a problem since. But he will now be changing RXs.

Edited By Ron Gray on 04/05/2019 18:23:24

I always understood it was the other way round, a spark ignition is creating noise at around 7,000 rpm or 120 Hz and shielding was quite an issue on 35 mhz. Even so your Rx if it saw RF interference should have gone to failsafe not locked up and required a hard reboot.

I'll check on the LMA forum, must admit I've not seen any reports of CDI interference issues with 2.4.

[Stephen Smith 14]

With what this has cost you, you could have gone Jeti. The lost planes would have bought the radio and the horus some receivers.

I bought a Ds24 and was so scared of the amount of money it was costing, took over a year to pluck up courage to spend it but have never once thought after the purchase that was a lot of money, I just think about the amazing quality.

Not to late to buy you won't regret it, and no I'm not connected in anyway just an incredibly satisfied customer. There is also a really good faceache group that can help if your stuck with some very knowledgable people on there even the importer comes on, but if you can get your head round frsky jeti will be easy.

[Ron Gray]

I had thought about Jeti, I really like their kit. However, with over 40 'planes equipped with FrSky the thought of moving them all over scares me! I could take the plunge and buy RXs as needed but even then a DS24 will cost the thick end of £2000 and RXs are twice the cost of FrSky and their telemetry is also far more expensive than FrSky. I have been unlucky in that the circumstances of my losses were unique to the combination of 1 type of RX and some of my CDI units but now that I know the cause I can guard against it for a fraction of the cost of changing to Jeti. Is the Jeti quality 4 x better than my Horus, having actually used a tray DS24 I would say no, it's 'better' but not that much better.

Anyway this isn't about Jeti better than FrSky, it's about making people aware of an issue with a certain type of FrSky RX and when not to use it.

Edited By Ron Gray on 04/05/2019 22:27:29

[Bob Cotsford]

Ron , how about running a redundancy system? Thinking back, mine were all equiped with X8R primary and X6R or X4R backup rx run through an XPS XP10+ sbus/redundancy board which might possibly have masked any problem that did cause the X8R to lock, though I can't say that I ever noticed a model landing with the primary locked out.

[Tony Richardson]

I have to ask is the X8R RX sold in europe programmed differently to here in Canada and North America.

I have been running an X8R for the past 2/3 seasons in a Curtiss P6E with a gas engine and have not had an issue, just wondering if there is a programming difference?

Tony.....

[Bob Cotsford]

I understand that the issue has been discussed on RCGroups and the firmware update was made available in both LBT (Eu) and FCC versions so I'd guess that the issue has been experienced on both sides of the pond.

Edited By Bob Cotsford on 04/05/2019 23:08:22

[Ron Gray]

@Bob - not sure that redundancy back up would have worked as the RX didn’t have loss of signal. But the advantage of the RX9R Pro and the R9M is that they are redundancy RXs so can have another RX connected to them, something I will also be doing on the larger (more expensive) ‘planes. Another tip from the LMA boys!

[Peter Christy]

Reading through this thread, I wonder if we are barking up the wrong tree here, with regards to interference. Firstly, I have to admit that my experience with petrol motors is limited. I only have one, largely because I find them heavy, gutless and noisy compared to their glow equivalents. The only real advantage is cheap fuel! But putting that to one side...

Ignition noise is pretty broadband, but generally, it seems to peak around 600 MHz, tailing off above that. By the time you get to 2.4 GHz, the RF splatter from an ignition system should be pretty low level. Add to that the fact that spread spectrum transmissions are designed to be interference tolerant, and I do wonder if RF interference in the accepted sense is the actual problem.

I would suggest that a more likely scenario is that the impulse interference is directly affecting the microprocessor that is at the heart of any 2.4 GHz system. This processor will operate at a much lower frequency - typically in the MHz region - and could get locked up by a strong enough splat of interference if not adequately shielded.

Those who have been in the hobby a while may recall some fields having problems in the 35 MHz days, eventually being traced to microwave links crossing the field. The answer was to wrap sensitive receivers in silver foil, which usually produced a complete cure.

A big problem with diagnosing this kind of issue is being able to reproduce it "on the bench". Until you can do that, your chances of finding a cure are vanishingly small. My only gasser does not suffer from the problem, so I can't help here. But if anyone is experimenting and *can* reproduce the problem under controlled conditions, try screening the receiver in foil of some kind, and see if that helps. I have a feeling it might!

--

Pete

[Ron Gray]

Peter, I understand where you are coming from but the the evidence, not just from me but from quite a few fliers over the last 2 years, is that an RX lockout occurs with X8R RXs and some CDI units. Some guys have been able to reproduce it using ignition testers which make the CDI unit spark. The problems are two fold, firstly the RX shouldn’t be effected by the interference but more importantly the RX should not lock out. As a result of these issues FrSky brought out a new RX, the RX8R PRO which they in their blurb states this ‘ And a new feature to decrease effect from interference caused by the ignition process’.

As highlighted in previous posts, there are extensive posts about this issue on other forums, unfortunately I didn’t see them, but when you are told by the main UK FrSky retailer that X8R and CDI is a no-no you tend to sit up and take note! I, for one, will not be wrapping my RXs in foil!

[Peter Christy]

I understand, Ron. The point I'm trying to make is that until the root cause of the problem is understood, any "fix" is likely to be guess work, and may or may not be effective.

I have a hunch (and its no more than that, as I am unable to reproduce the fault) that the problem is not at the 2.4 GHz end of the process, but RF pickup direct into the micro-processor itself, without which none of the rest of the system can work. The microprocessor controls everything, from the frequency hopping to the decoding of the signal. It is not itself an RF device, but it does *control* the RF device.

FrSky have, in the past, shown themselves to be very responsive to customer demand. Anything we can do to help them will also help us in the longer term.

Its the same principle as applied in the full size accident investigation world. That is why I ask for anyone able to re-create the problem on the ground to try this simple fix to either confirm or eliminate this diagnosis.

Personally, I would like to go back to the days when ALL receivers were in metal cases, not plastic ones!

Cheers,

--

Pete

[Denis Watkins]

Pete is right

In communications tech, the microprocessors, those little black boxes

Are constantly checked and designed by subjecting them to emf pulses

Microwave, 2.4gig is not susceptible to these with its short 12.2cm wavelength

As suggested, foil screening works on those

Within mobile phones and computers, you will see elaborate perforated push on screens on susceptible chips

Our chip needs screening

[Ron Gray]

If you read through this thread plus associated forums where this topic has been raised (2 years ago) you will see that FrSky were notified of the issue. They released a new version of the firmware but there have still been some (although fewer) reports of RX failure. They then released the new RX.

Putting it bluntly if they can’t guarantee a 100% fix then....... My rationale for going to either the 868Mhz route or the PRO RX is not for the frequency change to fix the problem it is the RX which appear to have interference tolerance, plus, of course if there are any external 2.4 GHz problems the lower frequency should be better.

[Alan Hilton]

My stuff predates lbt I removed the R8X from my cdi model thanks to your post .It was replaced by D series rx which I tested this morning .It passed an engine on range test with no problem and I had 3 successful flights .This is in line with the experience on the US site

Alan

[Paul Marsh]

Yeah, I won't use them for CDI models now, although there are other receivers in the Frsky range.

What are the best receivers to use on Frsky? Any recommendations from members here - tried and tested?

[Peter Christy]

As I say, I've only got one CDI model, and that flies on a Spektrum DSM-2 receiver with a 4-cell NiMh without any problems (yes, really!).

However, I do have quite a few FrSky receivers, both D and X series, and so far, both have proven to be pretty bullet-proof on glow and electric models.

The D series is an older protocol, but has a very good reputation. The fact that others report that they don't suffer from the problem leads me to suspect that they use a different micro-processor. The RF chips must be the same, as X series receivers can be bound to D series transmitters and the transmitters can transmit both D and X protocols.

However, the X protocol (also known as D16) is much more comprehensive, both in the number of channels, and its telemetry abilities. This makes me suspect that it uses a more sophisticated micro-processor, that may in turn be susceptible to a strong electro-magnetic pulse (EMP).

I have no inside knowledge of FrSky systems, so this is pure supposition, but it is a reasonable extrapolation, I think.

Mike Blandford knows much more about these systems, though he is bound by a NDA, I believe, but he may be able to confirm or deny my guess-work.

Since it appears that the X series receivers are prone to this problem, and D series not, I think the recommendation should be pretty clear.

--

Pete

[Ron Gray]

Once again, referring back to other forum threads, there is a strong felling that it is the X8R only that exhibits the problem, neither the X4 or X6 do. I'm slightly sceptical about this as logic would say that there are very few X4 and X6 RXs in use with CDI units. the older D series appear to be OK as do the latest RX8R PRO ones.

FYI here is a video showing the problem being replicated on the bench. Ignore the fact that they have placed the spark plug next to the RX, this is a test! Also have a look at his written comments below the video (unless you speak Italian!)

[Frank Skilbeck]

I alerted a club member who flies petrol with Frsky to this, just a heads up, he told me that there are also some issues with S8Rs randomly locking the rudder over on ignition planes and he has seen this but fortunately on the ground before the model had flown, similar to this

[Ron Gray]

When I saw that video I realised that I had one of those ignition testers! I picked one of my ‘spare’ X8R RXs and placed the spark plug from the CDI unit used in the Sbach loss on top of it, no lock out!. Didn’t have time to investigate further so to be continued!

[Peter Christy]

Couple of points about that video: Firstly, he's using a DFT module in the Tx. That does not transmit D16 (X) modulation, only D8 or V8 (V8 is non-telemetry). Therefore the fault occurred on the receiver when it was working in D8 mode. This may or not be relevant, but I would have preferred to see the test done with the receiver in its "native" mode (D16).

Secondly, there is clearly an edit in the video between the receiver working and not working. I don't know what he did in between times to make the receiver fail.

I'm not doubting that there is an issue here, but that video is a very poor demonstration of the fault for those reasons.

Its interesting that Ron says its only the X8 that is susceptible - not the X6 or X4. I was under the impression that the X6 and X8 were almost identical - certainly they share firmware. The only obvious difference is in the case, which again makes me wonder if this is a screening problem within the receiver. ie a physical rather than electronic issue.

Again, those of us who have been around a while will remember the issues some very well known (and expensive!) transmitters had with locking up when a mobile phone was nearby - hence the advice not to wear a mobile whilst flying. I wonder if this is a similar problem?

Ron: I'm assuming that your ignition system is completely isolated electrically from the radio? No shared ground connections, or anything like that? RC ignition cut-out perhaps?

I'm just trying to get a handle on the mechanism whereby the interference is breaking into the receiver.

--

Pete