What's new
Aloft Forums

Welcome to Aloft Forums. Register a free account today to become a member! Once signed in, you'll be able to participate on this site by adding your own topics and posts, as well as connect with other members through your own private inbox!

Urgent firmware update to all X receivers and RF modules (ACCST D16) just released

Thanks. Yes indeed, some planes are hard to get at and others are just not flown enough or valuable enough to worry about.

But for hand built kits I want to take reasonable measures to protect them.

I have an external module. Can it be updated? Or will it be necessary to buy one that comes updated things are settled.
 
Last edited:
has anyone worked with the spectrum analyzer i am curious will the TX still transmit when in this mode ?

Dennis
In Spectrum analyzer mode the receiver part of the IXJT module is scanning and noting the different signals on the band, While its performing this function the TX part of the IJXT is off, it has to be. You will notice you have a number of controls in spectrum mode including centre frequency and the ability to view parts or the entire RC frequency spectrum
 
Yeah, I suspect Kilrah is correct, your module should continue to work fine. The firmware change should only be for the guts of the FrSky RF boards, nothing else.

Hi Wayne,
well it looks I was a little late for the party ! Last night I upgraded my rx8pro (2)receivers to the new v2.0.1 firmware. I have been going through all my planes this winter
getting ready for the next flying season. After the upgrade the taranis would not rebind to the receivers, so I go back to frsky website to upgrade the internal module,
which I did and of course they now bind.
Today I start doing more research only to find out the firmware frsky (the company) has put out a unstable, and you recommend waiting until the bugs have been worked
out and a stable version comes out before we upgrade.
I`m kinda freaking out, what should us idiots who upgraded before doing the research do ?
1, is there anyway to go back to the old firmware for the receivers/modules ?
2, if no then what do you recommend ? do you think its safe to fly ?
3, is there any time frame on when a stable version will be ready ? I guess the only plus is my flying time is limited till maybe April-ish
but the planes I have the rx8r pros in are a couple of extreme gasers that I really don`t want to lose.
Please help us idiots ! hahahaha Do you think, are they safe to fly or would you put up everything that has a X series receiver in it and
wait for the bugs to be fix ?
 
I have flown the new firmware several times, with different Rxs: X8R, X4RSB, electric to gas on my Horus X10S. The only problem I have had with the new firmware has been a "sensor lost" which occurred once, and I only lost one sensor, a variometer. Everything else (GPS, Airspeed, RSSI, RxBat) still worked.
It is my understanding that the problems with the new firmware are with the telemetry part. The control link is solid. I lost two airframes to the "uncontrolled servo movements", both of them were gas planes, both of which had CDIs that had triggered in the past lockouts (a different story), so I would fly the old firmware.
Wayne may think otherwise, but if telemetry is not critical to your flying, I would keep the updated firmware. Now, as far as your questions
1) Yes you can. You will have to go to the frsky website and download the latest pre V2.0.1 firmware for your receivers. Depending on what radio you have, you may have to dig around to find pre V2.0.1 firmwares for your internal module.
2) See my initial comments.
3) Who knows. The coronavirus epidemic is affecting everything in China.
 
Last edited:
Hi Wayne,
well it looks I was a little late for the party ! Last night I upgraded my rx8pro (2)receivers to the new v2.0.1 firmware. I have been going through all my planes this winter
getting ready for the next flying season. After the upgrade the taranis would not rebind to the receivers, so I go back to frsky website to upgrade the internal module,
which I did and of course they now bind.
Today I start doing more research only to find out the firmware frsky (the company) has put out a unstable, and you recommend waiting until the bugs have been worked
out and a stable version comes out before we upgrade.
I`m kinda freaking out, what should us idiots who upgraded before doing the research do ?
1, is there anyway to go back to the old firmware for the receivers/modules ?
2, if no then what do you recommend ? do you think its safe to fly ?
3, is there any time frame on when a stable version will be ready ? I guess the only plus is my flying time is limited till maybe April-ish
but the planes I have the rx8r pros in are a couple of extreme gasers that I really don`t want to lose.
Please help us idiots ! hahahaha Do you think, are they safe to fly or would you put up everything that has a X series receiver in it and
wait for the bugs to be fix ?
from my reading of the RCG thread, the 2.0.1 firmware fixes the ACCST problem of the rogue servo movement, but at the same time introduces a problem with telemetry so that some sensors do not work. I agree with the above recommendation to avoid the 2.0.1 upgrade, that is what I am doing, and I am continuing to fly as I think the risk in non-EU gear is small enough. I have not had an incident of rogue servo movement, as far as I know, in years of flying with ACCST. IF I had done the 2.0.1 upgrade like Jengel, I would consider (a) the safety is improved, as the telemetry problem is much less dangerous than the servo movement one, and (b) on the other hand v2.0.1 has not been properly tested, maybe there is another issue, worse than the telemetry one, yet to be discovered. Its really a philosophical problem involving human nature, thus humans like to eat snacks, humans dont check things properly, etc, etc, so how do you know everything is 100% safe, maybe you are better off to stay home. Me, I still fly on Boeing planes.

I loved Konrad showing off to Frsky his foamie with Spectrum gear. Yes! Obviously Frsky needs to work out what went wrong and fix its mistakes and management issues, but I will not be shifting back to menu-driven RC unless I am really forced to. I am sticking with Frsky and waiting this crisis out, especially given the lack of competition. I fly slope gliders and only use Frsky gear, but I now understand why some segment of Frsky customers are particularly angry, eg those who fly indoor quads where a clone rx is built into a flight controller, after looking at a video by J. Bardwell

How can Frsky or any company ignore the requirements of what I understand is a big segment of its customers?
 
When I started in aviation we had a saying; If I’m going it's in a Boeing!” This is not so today!

While I still think the Boeing 777 is the epitome of aluminum transport aircraft, the company Boeing is a disgrace.

The last six year of my career was in support of Boeing as a vendor or partner. Much of this time was spent sittting in meetings on mahogany row. It pains me to say that the management of the last three aircraft (737 Max, 747-8 and the 787-8) I was involved with was often rather poor, if not downright horrible! Many of Boeing’s actions with regard to safety actually were made counter to their defined processes. There still are some great engineers at Boeing but since their purchase of McDonald Douglas the "Boeing management" has been focused on monetary issues over sound engineering. This may be an unpleasant reality when needing to make the acquisition of McDonald Douglass look to be working to the Wall Street bean counters.

Having, as an engineer, supplied hardware and services to all 4 major transport aircraft manufactures. I can say without reservation that Boeing is at the bottom of the pile as far as ethics and holding to their stated processes. Can Boeing recover; maybe? But the issues are now systemic in the culture. It will take some very very strong leadership at Boeing to turn a round the commercial division. I know almost nothing about the military side of Boeing. But I have no reason to think it is any different than the commercial side.

Yes, I’d like to see Boeing broken up and other more ethical firms allowed to grow in its place. This is the only way I know to get the real attention of management. And this is coming from a guy that has close to 2 million dollars in Boeing stock. Well, at one time it was close to 2 million dollars! Again piss poor management!

On the subject of piss poor management, here is more proof that I hold to my engineering analysis. Here is a photo after the maiden flights of my foamy jet. You might notice that I'm still using some FrSky products. Please notice the transmitter harness. This has proven to be real comfortable and it interfaces well with other brands of transmitters. Now to the main subject of this thread TX performance, you might notice that the TX is a Spektrum DX-9. It performed without a hitch!

The sad truth is that My FrSky X9D+ has been relegated to only controlling low cost, low emotional value toys like this 64mm Rafale jet. This jet really wasn’t worth the time to figure out how to program a delta wing wing with the pull down menus. FrSky needs to learn that the only reason they are still in the TX race is because of the fact they are running OpenTX. Without OpenTX I’d be done with FrSky TX’s!

BTW; I need to thank Wayne for helping me with an annoying haptic feedback I was getting. It felt like I was being shocked every time I touched the rudder. To be clear the dealers like Aloft Hobbies have been doing an exemplary job picking up the slack, dealing with the inherent ineptitude of FrSky's managemen when it comes to software and customer support! But at some point the core issues need to be addressed at the source of the problem, and that comes from the very top of FrSky’s management, as the problem is FrSky management!


To the last question, that’s much the point, FrSky does not see those using their unlicensed products as customers, but rather leeches. This may be true, but to FrSky’s customer base and future customer base this attitude will cost FrSky more, much more than the clones are costing FrSky.


All the best,
Konrad
Scorpion No FrSky.jpg
FrSky Rafale.jpg
 
Last edited:
I think that FrSky management has done questionable things in the past, but I do not think they handled this problem in a particular bad way. Let me try to summarize the situation as I see it:
1) FrSky develops the new ACCESS protocol. The intent is to extend the number of channels, offer OTA upgrades and remote binding, and at the same time to entice people to buy new hardware. It also offers to provide ACCESS to most of its D16 series receivers and radios (but not all). Sadly D8 hardware, which has proven to be incredibly robust, will be left behind. Can they be blamed for that decision? Not sure, maybe there are technical reasons, maybe it is pure profit driven, but they can't be blamed for developing new products. Spektrum has gone through three iteration of they RF protocol (and the first had its share of problems!): did they offer the ability to upgrade the firmware on their hardware? Maybe now, but I recall that until a few years ago their hardware was not field upgradable.
2) The German distributor discovers a problem with the ACCST D16 protocol. Admittedly, the problem is rare (but it happened to me twice in the last two years!, and I run the FCC fw) and the consequences rather severe (I lost two airframes!). FrSky at this point could say "One more reason to switch to ACCESS!". Instead, I believe correctly, they came up with a patch that by all accounts fixes the original bug. And they did it rather quickly. Could they have done more testing? Yes, but again, the pros and cons need to be weighted: allow people to fly with an equipment that can rekit a plane in a matter of seconds, or push a fix that may interfere with other, less critical part?
I think the mistake here was not to be more upfront with the problem the patch fixed ("uncontrolled servo movement" is rather vague), but at least they offered a solution that potentially undermines future sales of ACCESS. In fact, it appears that ACCESS has been put on ice for now.
3) The timing of the release was bad. First the Chinese New Year, then the coronavirus. The first occurs every year, but the latter is hard to predict.
4) The patch breaks compatibility with third party receivers: Could have been avoided? On RCG, there is a guy who has created a Open Source D16 firmware for receivers that fixes the bug without modifying the original ACCST protocol. So, in hindsight, it appears it could have been done. Why FrSky decided to alter the D16 protocol? Was it to kill two birds with a stone (fix the bug and kill competition from third party receivers and Jumper radios)? Or was motivated by purely technical considerations, as is we really think that this is a better protocol? BTW, I do not doubt that the protocol can be reverse engineered. It will be just a matter of time.
5) I have mixed feelings with FrSky at this point. In the last 3 years I had 5 accidents, of which 4 leading to loss of airframes, because of radio problems. 3 were due to the infamous lockout problem due to CDI ignition noise, the other to the bug. In all cases, the common factor was a noisy RF environment. Note that this coincides with me moving into gas planes. Never had a problem with electric or glow. This makes me wonder if the problem emerged now as more and more people are ditching glow for gas. OTOH, I remember well the problems that Spektrum users had (the brownout issue). Recently, I also discovered that Futaba has its own problems. At least, FrSky has allowed the users to perform firmware upgrades since the D8 series, which the big boys out there have only just started allowing. If you consider that any modern Radio system runs on a firmware, which by definition is bound to have bugs, allowing firmware upgrades should be a default option. But ultimately, I think what makes me stick to FrSky over the years has been openTx, which is truly a great software.

So let people decide.
 
Ululi, I agree with much of what you just posted.

While it is obvious that I’m livid with Fr Sky’s management. I should mention some of the strong points with the radio. First we agree it is OpenTX that really really separates the Fr Sky product from the other major radio providers. (Futaba, Spectrum and Jetti take note!). Fr Sky allows the owner/operator to replace wear items like; broken switches, and worn out gimbals etc.. This is a major draw back of Spektrum. If one breaks a switch during the fly season you are out of luck, as the TX needs to be shipped back to Horizon Hobbies for replacement. This results in unnecessary down time! Fr Sky and Spektrum allow the user to perform firmware updates. It just looks like Spektrum has a more robust processes as they don’t need near as many updates.

To the issue at hand, timing and perception are key issue to any successful marketing plan. In this case Fr Sky has fallen flat on their face! It is known that the current focus of FrSky is to break the clones. This is the true reason for ACCESS, adding other functionality was just an attempt to hide, soften the burden. Having a ACCST “FIX” that bricks much of the legacy products gives the impression that Fr Sky does not give their customer base any consideration. This has only been reinforced when they ignored (discounted) their own distributor/service agents request that they hold off on the implementation, and to have the “FIX” vetted by a third party. (After all they have a long history demonstrating their total lack of ability to validate any software)! Fr Sky needs to put some resources toward software validation. I for one would be willing to pay 10% or more for products that actually work when purchased. (Now that’s a novel idea)!

Your link (sorry I think it was getsuyoubi) earlier in this thread to the “German team” that found the issue, went a long way towards dispelling the perception that the change was driven mainly by marketing concerns. FrSky should have had this information on their site explaining the reason for the firmware update. Again very poor marketing and customer support on Fr Sky’s part!

I have no reason to believe that these failure are very rare. There is no vehicle to report failures! Like you, most have attributed their radio failures to some other outside factors. As Fr Sky didn’t know of the problem how can they make a sound engineering assessment as to the frequency in which it appears? I suspect that the failure mode is much more prominent than reported. I’m sure that many of these failures have been mis diagnosed.

Add this XJT bug to the very poor roll out of ACCESS, one would have thought that FrSky would have stepped back and made a concerted effort to insure that the ACCST D16 2.X.X fix was free of user noticeable faults! Again Fr Sky has demonstrated there utter contempt for their customer base!

I’m of the opinion that the firmware update is very much needed and must be implemented, even if one looses the functionality of some legacy systems. But this update needs to be well vetted (gone through a true software validation process) so as to NOT introduce other failure modes!

Yes, the encryption plan is flawed! Just look at how Apple’s own I-Phone was hacked. Given enough resources all encryption plans can be broken. This is particularly true if amateurs are working the issues, as their time is their own. I think all the RC hacks have come from amateurs, not firms trying to break a competitor’s code. And now that Fr Sky has pissed off a large segment of the hobby there is now a much larger population working on doing just that; breaking Fr Sky’s latest encryption!

Do you have a link to the open source for the D16 firmware? (It is understood that this is an unapproved hack). I’d like to have one of my XJT modules running a more secure link for my legacy Fr Sky systems.

All the best,
Konrad
 
Last edited:
Konrad, I think we agree on most indeed. The Open Source FW applies only to (some) receivers. Here's the link
https://www.rcgroups.com/forums/showthread.php?3391195-FrSky-D16-firmware-for-D8-receivers
It started as a way to turn a D8 receiver into a D16 one (S.Bus) but I think Mike is working on on a version for R8XR.
Mike says that his implementation fixes the CRC issue without breaking the D16 protocol.
At the beginning, it required soldering wires to pads on the receiver board, but now I think flashing can be done via the Tx/Rx ports.
 
Last edited:
I wrote my post above late at night, and on reflection I do not think the error Frsky made in releasing ACCST 2.0.1 with apparently faulty telemetry is nearly as bad as Boeing's recent decisions, as has been documented in a journalist investigation:


What Boeing (and officials from the N.T.S.B. and the Federal Aviation Administration) did remains almost beyond belief. I apologise for this unfair comparison in my post.

Today on my slope near Sydney, there were clear sunny skies, a good S wind, the ocean looked great, and as usual my Frsky-fitted out slopers (running ACCST v1) performed flawlessly. I use a pre-2019 x9d+ and qx7 both with recently upgraded gimbals, and g-rx6 rxs. With products as good as these it seems to me Frsky with OTx are now the dominant players in the industry.

I have never had a rogue servo movement as far as I know, in years of regular flying with ACCST v1. Today I achieved some perspective on the issue. The risk of such an incident may be less than better understood and dangerous problems such as a cracked wire, an elevator horn that is damaged (but the damage is not visible), a faulty/damaged rx or tx battery, severe turbulence, pilot error, etc etc, that I and my slope colleagues live with every day, and manage safely. Btw, does anyone know how I can tell exactly what version of ACCST I am running, all I know is that I have not upgraded to v2.0.1, in fact the only fw upgrades I have ever done are Otx and bootloader; my rxs have been purchased at different times to my tx, is having a different rx and tx ACCST version an issue?

I expect Frsky will soon correct ACCST v2.0.1 and this incident will pass. I hope some solution can be developed for drone pilots with a x9d or qx7, and rxs and modules that cannot work with v2.0.1. An example would be to announce a "legacy" ACCST upgrade version that fixes the rogue servo movement issue but retains compatability with ACCST v1 in older Frsky and 3rd party gear.
 
It is my understanding that the bug in v1 needs a trigger. In other words, in a perfect noise free world, v1 would work just fine. Only when sufficient noise exists (I suspect in my case was
CDI RF noise) packets are (sometimes) not decoded correctly, which now brings the CRC into play. V1 uses a CRC that can fail to flag a corrupted packet as such, which then can cause
the Rx to issue wrong commands to the servos and in the worst case possible, loose sync with the Tx, which then takes about 1 second to be reestablished. I never flew a slope glider, but I suspect that in your case there are very few sources of noise around (no engines, no ignition, just air...)
And yes, I agree that comparing what FrSky did to what Boeing did is extremely unfair to FrSky: after all, they may have botched the response, but at least they addressed the problem instead of trying to burying it. Also, on RCG the latest is that there is a more recent version of V2.X.X that is being tested.
 
Testing is taking place and we hope to have a green light on it soon. I am not applying any pressure for it to be rushed, I much rather have it fully tested before we suggest people use it. When it is time we will email everyone in our list and post up here too. We will explain the issue so that folks can decide what if any action they will take. Some will upgrade, and some will not. I think either option is fine. Eventually we will sell the receivers with the new ACCST as we can not sell something we know has a bug in it.

-Wayne
 
In terms of QC/QA, I think FrSky would to well to consider imprinting the Rx (at least the high end) with a serial number, and having a mechanism to report problems. This is a wild idea that I throw around, but the high end receivers, the ones that people may be flying on expensive hardware, should have a microSD card that by default records each received packet. A veritable black box. In case of accident, the user should be able to upload the data for FrSky to look at. I do not think it would drive the cost
 
So if I understand this, this problem is that a random packet of bad corrupt data causes a servo to fully deflect. What is the typical result. Does the servo return to normal behavior and the event usually go unnoticed? Is the typical crash caused by the momentary and unexpected servo deflection, or because of damage to the servo or controls caused by the deflection? If it is not caused by damage to the servo, would it be unlikely that a single glitch would cause a loss of control of a slow to respond airframe like a 2m RES glider?

It has been said that "noisy environments" seem to be a trigger. Are these glitches very unlikely at a rural flying site with only one person in the air at a time.

I am asking because I built three gliders over the winter. I really hate to do upgrades. Do I need to worry about "this" if flying a glider in a rural location?
 
It looks like you are asking us to do a risk assessment for you. The truth is only you can carry that out.

It is a known issues that the firmware allows corrupted data to pass. This is will happen irrespective of the RF environment.

The killer is that it can take up to 900 milliseconds to "reboot". This is what kills our ships.

Now for some historical perspective most of us have been satisfied with the RF performance of our FrSky XJT radios for over 5 years.
Again most of us have been much happier with FrSky's XJT radio than with the old 72 mHz band radios. Heck, most of use have favored the RF/ packet performance of the FrSky XJT over the Spektrum and just about any other protocol.

What most of us have an issue with is how FrSky has dealt with this issue and a myriad of other issues with products and software that has been released far too early in the development cycle.

Sorry, I can't give you a Go/No Go answer.

All the best,
Konrad

P.S.
I've grounded my high emotional investment models that are running Fr Sky!
 
Last edited:
Well, I was hoping someone would say, yeah, with a big slow plane like a 2m RES glider, holding full elevator or rudder wouldn't do a lot. I guess when Spring comes I could just try that.

I haven't had any crashes I'd blame on my receivers since I started using Frsky stuff in 2012, but I did use mostly D Series receivers for half that time. Still, lots of hours. So, did I not have glitches, or did I have them and not notice them. I will never really know.
 
My fastest and most valuable planes are gliders. Having a glitch on launch is a sure-fire way to loose one!
Even my slow F3-RES 2 meter is now flying Spektrum (I haven't dumped the memory location in my x9D+, in hopes that FrSky can correct their gross errors).

As you don’t think you are going to be flying until spring. Then I don’t think there is an issue. Well, I hope FrSky will have come out with a stable fix by then! ;-0

It is my understanding that the “germans” have come up with a fix for the data packet issue. So that core issue has been addressed. Now unfortunately as is typical of firms that don’t have a proper validation process. FrSky has introduced other issuing into the mix with this fix 2.0.0.1 (or whatever). This is just ridiculous!!!! Like you I have better things to do than chase FW updates. I hate doing updates, and doing them more than once because the management of a firm is totally inept, just pisses me off!

As Is typical for FrSky one needs to wait a few iterations before we get a stable workable software solution!

Since you can wait until spring, I’d suggest you just sit back and watch this 3 ring circus unfold. But by spring do upgrade to the stable firmware before flying your gliders.

Since I fly year round I’ve switch to other RF manufactures until this gets all cleared up. Like I said before, it is OpenTX that is keeping me in the marriage to FrSky.

All the best,
Konrad
 
Back
Top