Possible brown outs ?

I recently installed Deviation 4.0.1 on my Devo 8 TX that I've had for years. Reason being that I wanted to bind to some DSMX recievers that I have.

Anyway, I've now had two different aeroplanes crash in as many days and it seemed to be due to signal loss. For example, I was just flying my P51 an hour ago and was half way through an aileron roll when control was lost and it crashed inverted. The previous crash was very similar but that time I was coming out of a loop when the elevator stopped working and it dived into the ground. Both planes were using Devo RX802's and were not very far away from me, maybe 80m max. I tested both on the bench after the crashes but everything seems to be working properly.

What could be happening here?

I've never had any problems like this before with the stock Devo software.

Thanks.
So there is a known issue with brown outs when telemetry is enabled for Devo?
What is a "nightly build" ?

Hmm, I'm very disappointed that this known issue isn't clearly mentioned somewhere on the forum or in the release notes. The crashes have caused a lot of damage to my planes.

Not sure if I should trust Deviation again even with the telemetry turned off.

Are there any other known issues that I should be aware of that could cause brown outs?

First, let me say I'm sorry you had problems with our software. I understand the pain a crash can cause, especially of a prized model.

Since the release notes are information available at the time of release, I'd be disappointed if they mentioned brownouts, as the release shouldn't have happened in that case. Being a volunteer organization, we don't have to adhere to any kind of release schedule, so there's no excuse for releasing something with known problems that severe.

Searching the forums for "brownout" turned up a lot of hits. Given that we support dozens of protocols with a correspondingly broad range of receivers on Devo transmitters, that's about as good as that's going to get. A better search function would be nice, but see the above comment about volunteer organizations.

To be honest, if what you want is rock solid reliability with as little effort and exposure as possible, deviation isn't for you. Then again, neither is Walkera - you want to be using Spektrum gear (or maybe Futaba; more in a bit). Between being a volunteer organization and the large number of protocols supported, we clearly aren't going to be able to test as thoroughly as a commercial operation supporting one or two protocols and a few receivers.

On the other hand, we do have a publicly accessible issues list (see the Bug Tracker link on the left). Anyone who can post to the forums can report an issue - as well as fixes or workarounds that might help. It is targeted at ongoing development, so finding information about really old releases (at this point, any of them) can be difficult.

But not even the best commercial organization has such a list. At best, they'll bluster and maybe lie a little while checking on things before fixing it in the next production run, and if you're lucky add a tech note under a tab on the product page about things you should avoid that seem reasonable.

HH is very good about helping if one of those problems leads to a crash, and will replace things that broke because of it. We don't have an organization to do that, which is why there are disclaimers on the front page of the main site and in the license file about these things. I've never found a way to contact Walkera support, but their US resellers have been poor at best. Futaba users swear at Spektrum gear as unreliable (and vice versa), but I've not used it, and have no idea if their support is as good as HH. Check RCGroups.com for reports on that, and numerous flame wars between fans of the two.

We also have the forums, where you can come and ask for advice about what to do. If you came and asked which version to run for what you want to do , I would have been told you to run the nightly builds, and I don't know of anyone still recommending 4.0.1. Again, I'm sorry if that's not what happened. If someone recommended you run 4.0.1, you should have been warned about the issues with telemetry. But again, it's a volunteer organization, so you may have missed everyone who would have steered you right.

You can get that kind of support from HH via the phone or their web site, which is good as it's less hit or miss than a public forum, but bad as they repeat the party line, even when it's BS, and you don't get to hear from anyone else. Walkera doesn't have any such thing. And I still don't know about Futuba.

I've had radio issues with deviation - telemetry problems, range issues, responsiveness etc. Never had a crash because of it, though - which is more than I can say for problems with Spektrum gear. But I always treat deviation like any other part of the model I built. I'm less critical of the Spektrum gear, and keep a Spektrum transmitter around just so I can call HH support for help with their aircraft (yes, their support is that good!). This may be why I've had more crashes when their stuff failed than when deviation does.

If you want something you can treat like an RTF aircraft, and just plug it in and go, then deviation is liable to cause problems. If you put the work into it that you do into a kit-built aircraft (well, that I put into a kit-built yacht) and do some research into possible problems, test everything along the way, etc, then you'll generally be fine. But still no guarantees or warranty. For that, go to Spektrum (or maybe Futuba).

shwmae wrote: Not sure if I should trust Deviation again even with the telemetry turned off.

Well,
there are very few known issues in Bug Tracker right now, and none causing a crash in flight.
cmpang e.g. has some thousand flights of experience
The issue vlad is talking about:
- is solved in the nightlies since somewhere around april 2015. See Support for walkera telemetry
- was very rare (maybe once in ~5h) and only with telemetry switched ON
- not all TX have been affected

Thanks for your input, it's appreciated.

Yes, I realise that it's a volunteer organization behind Deviation and they have done a remarkable job and service to the RC community. I also realise that nothing is without risk in this hobby.

However, I have to question the wisdom of leaving 4.0.1 at the top of the download page as that is obviously the one most people like myself would assume to be the latest and most reliable version. I had no idea what "nightly build" meant, I've never heard of that term before so didn't even look at that page.

I've known about Deviation for quite a while now but held off installing it because I knew I'd find the interface a challenge to understand which it has been. I was quite pleased with myself for being able to get to grips with it and program things like expo and 3 way switches etc, it's akin to learning another language!. That's another reason why I was really disappointed to crash two models because it does take time and effort to learn Deviation plus I have no need for telemetry and would have turned it off anyway had I known that it's on by default.

Also, being that we are talking about Walkera Devo transmitters here, I just blindly assumed that running Deviation with the Devo protocol would be stable and reliable. It's the other protocols that I could understand may have some issues but not Devo. Does that make sense?

Could someone tell me where I can find the stock Walkera firmware for the Devo 8 please?
Walkera.com doesn't seem to work anymore.

shwmae wrote: where I can find the stock Walkera firmware?

The best source for Walkera I am aware of is the german walkera-fans.de .
"Sender" is the german word for transmitter

Thanks, they got firmware for the 8S but not 8, weird!

I think the only difference is that the 8s is telemetry capable.
Upgrading the Devo-8 to Devo-8S
I guess that the 8s firmware would work in the 8 as well.
You just can't use telemetry

If you follow the "Links" link in the menu, there's s link to official Walkera software, DFuse and bootloaders, including firmware images for the transmitters. The 8s firmware is there. I believe you'll need to install both the dfu and the library.

The assumption about the Devo protocols is wrong - it's no different from any other protocol, in that we can't use the walkera code, so have to reverse engineer the protocol. But that's also how you get all the old Walkera protocols as well which devention doesn't support.

Unfortunately, the site gets updated even less frequently than the software, as there are fewer people who can do it.

linux-user wrote: The issue vlad is talking about:
- is solved in the nightlies since somewhere around april 2015. See Support for walkera telemetry
- was very rare (maybe once in ~5h) and only with telemetry switched ON
- not all TX have been affected

Having thought about it, it seems a shame to go back to the Walkera firmware now that I've spent hours learning how to use Deviation.

So if you don't mind, I'd like to try and work through the problem with you guys because after reading the above, it seems unlikely that the telemetry being on was the cause of the crashes. Both occurred at around 5 minutes of flying so that surely must rule out the once in ~5h occurrence.

Some facts that may help.
I've flown both planes dozens of times before installing Deviation and never had a problem with either.
I didn't alter anything on these planes, either mechanically or electronically.
I was flying in the same location. I fly fairly low and near to me so they are never more than about 200m away, line of sight.
Both planes have flight stabilization systems installed but from different manufactures. Both were switched on at the time of the crashes.
I set the same amount of servo travel and expo in Deviation as I had done with the stock firmware.
Fixed ID was off.
Power was set to 10mW.
Number of channels was set to 8.
I copied the model setup from the first one to the second and then changed a few settings.
Both planes felt completely different to fly with Deviation compared to before and were erratic at times as if my stick commands were being delayed but that's just a guess. I just know that they weren't flying like before which is a difficult thing to quantify, it's all down to feel and touch.
As I've already said, both planes crashed at around the 5 minutes mark and it seemed to me that I had no elevator control for the first one and no aileron control for the second. It's impossible to say whether I lost all the other channels as well because they were in the ground in a split second.

It's probably going to be impossible to tell me what went wrong but I'm curious to hear some theories anyway.

When you say you set the same amount of servo travel and expo as you did with devention, what exactly do you mean? I've not looked at fixed wing setup on devention, but if you just copied the numbers from devention to deviation on a helicopter, your controls would be very badly broken.

In particular, deviation uses values that go from -100 to +100 for everything, where devention uses values that go from 0 to 100 for some things. If you needed that to translate between ranges and didn't, then you only had access to half the throw on your servos. Also, different transmitters can interpret expo with opposite meaning, so that what's positive expo on one is negative on another. I'm not sure if devention and deviation are different or not.

shwmae wrote: ..................cut
Fixed ID was off.
Power was set to 10mW.
Number of channels was set to 8.
...........cut
.

Just do a ground test with these settings. One servo at throttle channel (full), walk and watch the servo..

www.deviationtx.com/forum/3-feedback-que...er-150mw-100mw#32518

shwmae wrote: it seems unlikely that the telemetry being on was the cause of the crashes. Both occurred at around 5 minutes of flying so that surely must rule out the once in ~5h occurrence.

No, you can't rule out the telemetry issue because you've flown only 5 minutes.
It was very rare, but could have happened any time.
I guess that V120d02s-V2 lost binding mid flight was caused by the telemetry issue.
The only way to avoid this issue is either to use a deviation version newer than ~april 2015 or switch telemetry OFF.

BTW:
To our best knowledge this telemetry issue is hardware related.
Newer deviation versions have a built in workaround:
They reset the CYRF6936 if an error is detected.

mwm wrote: When you say you set the same amount of servo travel and expo as you did with devention, what exactly do you mean?

I meant that I set up the same amount of flight surface deflection as before (measured on the model) and the same amount of positive expo. I don't use dual rates.

Thanks guys, I'll do some range tests and then fly my old battered trainer plane with the telemetry turned off to see what happens.

Seems to me pushing the limits, I'm not a "use the maximum power available always" user, but for big planes and at that distance I'll be using 100mw for sure. I only use 10mw with micro planes, helos and quads.

Couldn't FDR just remove the older versions including 4.01 from the site and point only to nightlies? shwmae is right new people would be attracted to it as the more stable version.

Btw using telemetry on, on a non capable telemetry device as the plain devo 8, is possibly looking for trouble even with the latest nightly.

mwm wrote:
... If you came and asked which version to run for what you want to do , I would have told you to run the nightly builds, and I don't know of anyone still recommending 4.0.1. Again, I'm sorry if that's not what happened. If someone recommended you run 4.0.1, you should have been warned about the issues with telemetry.

The Deviation site is replete with billboard-sized suggestions that 4.0.1 is the place to start.

Go to deviationtx.com (the home page) and start from there. It's very misleading.

The second line of the home page says this:

Version 4.0.0 of Deviation has been released!

I love the software and what it does for my flying models. The contributors are wonderful. But I hate to see others mislead by the poorly constructed and poorly maintained web site.

Deviation is a wonderful, community-based effort, with a wonderful product. But parts of that community are is disarray. The web site in particular. Someone with no knowledge of Deviation should be able to begin at the home page and get started with a minimum number of clicks. Today that's not possible. Days or weeks of study of the site are necessary. And perhaps some luck. The information is there, usually. But the organization is pretty terrible.

Disarray.

(I applaud mwm's tireless efforts to educate and support.)