DX8 Rebooting mid-flight

[agagescu]

Was on my second flight of the day with T500 FBL. Maybe 5 flights on the helicopter, 2nd on a brand new pair of EDGE blades. After what was a pretty normal flight, I came in to land. At about three feet, cyclic went full right with no input and now I need some parts.

I looked down at the transmitter to see the spektrum splash screen followed by the Idle up warning. No shit the idle was up, I was flying what was a nice helicopter before the DX8 did it's thing.

I've read about the other problems; gimbal, unwanted channel input etc., but this is a new one to me.

The battery was fully charged. It was only my second flight.

Has anyone seen / heard about a rebooting issue?

[the collective]

I'd been looking forward to meeting you, but this wasn't what I had in mind... I arrived just after it happened, and it really hurt to see the aftermath.

I have not heard of any spontaneous rebooting behavior before today. I had this sort of thing happen on a dx6i a while back but it turned out to be a bad contact where the batteries go in. I'm pretty sure the DX8 comes with a rechargeable pack instead of a battery "box" with loose "AA" cells so I don't think a bad contact would be the issue... maybe an intermittent plug on the battery?

I'm wondering if you power up the radio with the battery cover off if you can see any change in the voltage reading on the screen if you jiggle the battery plug a bit...

[agagescu]

Good to meet you! Did you ever get that 600 flying?

I'm going to play with it tonight and see if I can replicate that issue. If not, I've got my trusty old futaba and my other 500 for tomorrow if the weather cooperates.

I'll be sending the DX8 to Horizon on Monday and tell them what they can do with it...

Now I'm shopping for either a 11X or 10CAG :)

[the collective]

We got about to the midway point of the BeastX setup, then Al's transmitter battery alarm went off. We cut it short and will resume tomorrow with a fully charged transmitter.

We're thinking 11 or so but I'll drop a text to you when I know for sure if that's okay. I want to get my Rave out... it's been almost a week and I need a flyin fix..

[zer0frequency]

I can related to the tragedy. I just got back from the field after experiencing a similar reboot !!!

Luckily my Trex-500 ESP (with the align MD-500 hughes body) was still on the ground just spooling up when the transmitter started to beep! :roll:

A glance on the screen revealed that the transmitter is warning me of "Throttle high" !! well duh! the heli was spooling up, the throttle stick was up :p Moving the stick down made the transmitter to display the splash screen and it continued to boot.:face

Scary stuff ! Anyone knows if this is a more widespread issue? Has anyone spoken with Horizon about this?

BTW This DX8 just came back after the latest recall and is upto date as far as known firmware issues etc are concerned. The TX and RX Voltages were both fine...

I will try to call them tomorrow. Lets see..

Zer0frequency

[RocketSled]

Was it unusually hot? Had the TX been sitting in the sun? The DX8 is a microcomputer that executes firmware. It's common practice to include a "watchdog timer" in this type of hardware design. Watchdog timers force a reboot if the FW stops running correctly, if for instance it gets in to a tight loop or "goes off in to the weeds". Crappy code might do this sort of thing all the time, but I don't think Horizon's FW is crappy. High temperatures can effect the operation of electronics, changing the timing of the HW and causing the code to execute incorrectly, which could also trigger a watchdog reset.

I couldn't find any temperature specs for the DX8. But assuming it's built with commercial-rated components, they're good to about 70ºC (about 160ºF). The parts generate heat in operation, and might run at 20ºC (about 36ºF) above ambient, or even higher when powered. The TX case has no ventilation, which makes things even worse. Sitting in the sun, on a 90ºF day, I could see the internal temperature easily getting close...

It's just habit for me, but I never leave my TX in the direct sun on a hot day. I see many other fliers who do, though. Leaving the TX on a table in direct sunlight is not a good idea.

[agagescu]

Glad it didn't happen to you when it was in the air. Yours is the only other reboot incident I've heard of.

I spent quite a bit of time trying to recreate that problem unsuccessfully before I sent the radio to Horizon. I couldn't risk another helicopter so most of that time was on Phoenix. That was a poor test since the radio wasn't transmitting during that period.

I did eliminate the battery as an issue b/c it lasted for a lot of hours on the SIM afterwards Plus a couple more tests and the voltage was still pretty good when I boxed it up.

Leaving the TX and that particular RX on for a couple of hours on the bench didn't even drop a frame. After a while the inactivity beeps got annoying. That test wasn't very scientific so I gave up without recreating the problem.

I could have done more testing and spent more time on it but I'm sure they have someone on staff with an EE after their name that should be doing that.

For a while I thought maybe it was just my particular radio. A fluke in the midst of all their other DX8 problems. This certainly isn't as wide spread as their other problems but it's not a singular case.

I told HH over the phone and on the service ticket that I didn't want that particular radio back. They said their technicians would fully test it and make a decision. I insisted and they said that they'd see what they can do and get back to me. They didn't get back to me. Haven't received the radio back but am pretty sure it's going to be the same s/n.

I'd love to know what their response is when you call them. Also, what was your PID and date of purchase? I bought mine directly from HH on 11/26/10. I can post the PID if you're interested but need to dig up my notes.

Picked up a JR11X as my main transmitter. The DX8 will be relegated to foamies and the SIM or sold. I have a lot more faith in JR in spite of their relationship with Spektrum.

Last Spektrum radio I'll ever buy. That's for certain.

-A

I can related to the tragedy. I just got back from the field after experiencing a similar reboot !!!

Luckily my Trex-500 ESP (with the align MD-500 hughes body) was still on the ground just spooling up when the transmitter started to beep! :roll:

A glance on the screen revealed that the transmitter is warning me of "Throttle high" !! well duh! the heli was spooling up, the throttle stick was up :p Moving the stick down made the transmitter to display the splash screen and it continued to boot.:face

Scary stuff ! Anyone knows if this is a more widespread issue? Has anyone spoken with Horizon about this?

I will try to call them tomorrow. Lets see..

Zer0frequency

[agagescu]

It wasn't particularly hot when it happened to me. Ambient was in the mid 80s and I tend to keep my electronics out of the parked car and don't leave it in the sun. More for the screen than anything else.

If they really wanted to t-shoot, they could write events to the SD card. Doubt it's going to happen though. Just knowing whether the last boot was unexpected would go a long way.

Was it unusually hot? Had the TX been sitting in the sun? The DX8 is a microcomputer that executes firmware. It's common practice to include a "watchdog timer" in this type of hardware design. Watchdog timers force a reboot if the FW stops running correctly, if for instance it gets in to a tight loop or "goes off in to the weeds". Crappy code might do this sort of thing all the time, but I don't think Horizon's FW is crappy. High temperatures can effect the operation of electronics, changing the timing of the HW and causing the code to execute incorrectly, which could also trigger a watchdog reset.

I couldn't find any temperature specs for the DX8. But assuming it's built with commercial-rated components, they're good to about 70ºC (about 160ºF). The parts generate heat in operation, and might run at 20ºC (about 36ºF) above ambient, or even higher when powered. The TX case has no ventilation, which makes things even worse. Sitting in the sun, on a 90ºF day, I could see the internal temperature easily getting close...

It's just habit for me, but I never leave my TX in the direct sun on a hot day. I see many other fliers who do, though. Leaving the TX on a table in direct sunlight is not a good idea.

[RocketSled]

You never know, they might capture crash data, a stack dump or something. That's also pretty common practice. Being able to write it to the SD so users could email it in would be cool. But there might be something they can retrieve on a return-for-service radio...

[zer0frequency]

Hi Guys,

The temperature thing makes sense but it wasn't that hot - I think it was around 85~90 F. I had done about 4 flights (with 5~6 minute breaks in between) prior to the "incident" and it was fairly cloudy evening.


I just called HH and the rep I spoke with said that he is not aware of this issue being reported, and suggested I send the radio in for inspection. But given the fact that its not reproducible, I am not sure what the tech would do. It doesn't seem to be very common, right?


One thing is that the RX I was using is DSM2 (AR7000). I dont know if it's related but there have already been a few recalls related to DSM2.

agagescu, what rx were you using?


I am thinking of changing all rx to DSMX ... :dontknow

[agagescu]

Hi Guys,

The temperature thing makes sense but it wasn't that hot - I think it was around 85~90 F. I had done about 4 flights (with 5~6 minute breaks in between) prior to the "incident" and it was fairly cloudy evening.


I just called HH and the rep I spoke with said that he is not aware of this issue being reported, and suggested I send the radio in for inspection. But given the fact that its not reproducible, I am not sure what the tech would do. It doesn't seem to be very common, right?


One thing is that the RX I was using is DSM2 (AR7000). I dont know if it's related but there have already been a few recalls related to DSM2.

agagescu, what rx were you using?


I am thinking of changing all rx to DSMX ... :dontknow

They never heard of it before huh? SR01-4EFC-9238-3340F was my case number. Feel free to reference it when you speak to them.

I was using the AR8000 with one satellite.

I crashed today. And it wasn't radio related. WOO HOO.

[TODDC]

Hi Guys,

The temperature thing makes sense but it wasn't that hot - I think it was around 85~90 F. I had done about 4 flights (with 5~6 minute breaks in between) prior to the "incident" and it was fairly cloudy evening.


I just called HH and the rep I spoke with said that he is not aware of this issue being reported, and suggested I send the radio in for inspection. But given the fact that its not reproducible, I am not sure what the tech would do. It doesn't seem to be very common, right?


One thing is that the RX I was using is DSM2 (AR7000). I dont know if it's related but there have already been a few recalls related to DSM2.

agagescu, what rx were you using?


I am thinking of changing all rx to DSMX ... :dontknow

When I sent my DX8 in for the update. I asked if I could have used it with my 6000 series receivers. I have a 7000 series in my 600. Support told me the DSM2 6000 receivers may have problems and would not recommend it. The 7000 series he said was fine because it could talk to the transmitter fast enough where the DSM2 6000 receivers will not.

[agagescu]

They never heard of it before huh? SR01-4EFC-9238-3340F was my case number. Feel free to reference it when you speak to them.

I was using the AR8000 with one satellite.

I crashed today. And it wasn't radio related. WOO HOO.

The service order was C172126 from 6/28/11 . The SR number above is the service status tracking number.

[agagescu]

When I sent my DX8 in for the update. I asked if I could have used it with my 6000 series receivers. I have a 7000 series in my 600. Support told me the DSM2 6000 receivers may have problems and would not recommend it. The 7000 series he said was fine because it could talk to the transmitter fast enough where the DSM2 6000 receivers will not.

That's an interesting statement: "fast enough". The DX8 should figure out what it's bound to and select the proper frame speed. For the AR500 and AR6**** receivers that should be 22ms. I wonder what the heck that means? But an issue with the receiver or the transmission protocol shouldn't cause the transmitter to reboot.

Either way zer0frequency was using an AR7000, and I had an AR8000 both in DSM2 mode.

There are several bulletins about the AR500 and AR6**** receivers. I normally wouldn't care about those receivers but I do have some BNF stuff I don't feel like upgrading.

The marketing states that DSMX TXs are backwards compatible with DSM2. The service bulletins say not quite. Usually the words compatible and superseded don't go together.

http://www.horizonhobby.com/Support/Product/Bulletins.aspx

There's one statement on http://www.horizonhobby.com/DSMX/

"DSM2 transmitters are compatible with DSMX receivers. Likewise, should you add-on to a DSMX transmitter, the DSM2 receivers you have now will be compatible with it".

Then there are the service bulletins that state:

DSMX transmitters operating in DSM2 mode

"The Spektrum Team has been investigating sporadic reports from DSMX transmitter owners using their products in DSM2 mode. We have discovered that in some rare instances, DSMX equipped transmitters with the Product ID (PID) within this bulletin, may have a backwards compatibility issue that could cause a "hold" with the following superseded DSM2 receivers: AR500, AR6100, AR6100E, AR6110, AR6110E, AR6200 and AR6300."

[zer0frequency]

I am trying to figure out if there is a pattern to this... My AR8000 is from the first batch, so I am thinking that is also DSM-2 and not DSMX.

Is your AR8000 a newer version ? Maybe its all related to DSM2 glitches, some of which they have documented and tried to fix?

[Busher]

I am trying to figure out if there is a pattern to this... My AR8000 is from the first batch, so I am thinking that is also DSM-2 and not DSMX.

Is your AR8000 a newer version ? Maybe its all related to DSM2 glitches, some of which they have documented and tried to fix?

All ar8000 receivers are dsmx, but as the receiver was released before they released DSMX protocol, there are some of the early ones still showing dsm2 on the case. But I have been full assured by Horizon UK that all ar8000 are DSMX. I have recently seen a brand new dx8 set unpacked, and both the receivers ar8000 had a dsm2 sticker on them. Both bind as dsmx.

[agagescu]

I am trying to figure out if there is a pattern to this... My AR8000 is from the first batch, so I am thinking that is also DSM-2 and not DSMX.

Is your AR8000 a newer version ? Maybe its all related to DSM2 glitches, some of which they have documented and tried to fix?

The AR8000 came with the DX8 I bought in Nov, 2010. It would be helpful if they put out more information. Finding a pattern with only two known instances of this exact problem is going to be near impossible.

And if you don't know what the problem is, how are you going to fix it? That was my argument.

[zer0frequency]

Yep! I am in 100% agreement with you. At this point, it seems like we are on our own. They don't recognize this as a wider problem and therefore certainly aren't even investigating. :-(

[agagescu]

Yep! I am in 100% agreement with you. At this point, it seems like we are on our own. They don't recognize this as a wider problem and therefore certainly aren't even investigating. :-(

I got it back yesterday. They upgraded the firmware to 2.01, replaced the antenna whose articulation was bent. Returned it with a note that they can't justify replacing the TX as it is operating correctly. Also, I shouldn't attempt to fly with a broken antenna. The antenna (the actual transmitting element, i.e. wire) was not damaged. The articulation of the plastic antenna housing element was bent. A range test that morning was fine. As were the couple hundred flights with that transmitter prior.

So I was without a receiver for a full month and got an antenna replacement and firmware update which I could have done at home.

They were testing on Model 30, which has a system timer at 19:04 and bound to a DSMX receiver @ 22ms. The firmware update does not reset the system timer. The tracking status of "assigned to technician" was just over two hours according to their ticketing system.

There is no documentation of work being done to the gimbals but a look at the monitor shows travel on all channels going to 99% with the exception of Ail left (100%). Where previously they were +100/-100. No big deal, I know I can re-calibrate, but why not mention that work. I'll open it up to have a look at what was done.

[zer0frequency]

I crashed my scale airwolf (600) yesterday after a wonderful flight, coming in for landing :'( I don't know if the radio went dead (rebooted) or just lost connection but the heli went into fail-safe mode and landed, cracked the body up!!!!!!

Luckily the mechanics and electronics were safe as I was already low altitude trying to land and there wasn't much head speed !

I called horizon yesterday and got into their call back queue but have not heard back from them. :'(

Anyone has any info on this???

[ror1]

A Rx can not make the Tx reboot...or are you saying it is a DSM2 issue and not a DSM2/DSMX issue?

[agagescu]

I crashed my scale airwolf (600) yesterday after a wonderful flight, coming in for landing :'( I don't know if the radio went dead (rebooted) or just lost connection but the heli went into fail-safe mode and landed, cracked the body up!!!!!!

Luckily the mechanics and electronics were safe as I was already low altitude trying to land and there wasn't much head speed !

I called horizon yesterday and got into their call back queue but have not heard back from them. :'(

Anyone has any info on this???

From your description it's hard to tell if you are having the issue I was experiencing. Were you in idle up mode? Was you RX blinking? What firmware are you running? Was your PID affected and if so, has it been in for service?

Sorry to hear about the damage.

[agagescu]

A Rx can not make the Tx reboot...or are you saying it is a DSM2 issue and not a DSM2/DSMX issue?

The discussion around protocol was more headed towards what code the transmitter is running. I.e. Is there a bug in the old DSM2 stack vs. DSMX.

Also, the dx8 transmitter is a receiver as well. Data is being passed bidirectionally.

[desertstalker]

Also, the dx8 transmitter is a receiver as well. Data is being passed bidirectionally.

But not by the RX, the data coming in only comes from the telemetry modules AFAIK. If the RX could send data back you could have telemetry on signal stats from the RX directly.

[agagescu]

You're probably right. The information it would be receiving should be coming from the TM unit. But during the binding process, it is getting information back from the actual transmitter. Am assuming that's how JR do their model match over these protocols. Presumably each RX has UID and framing intended for that particular RX is prefaced with that UID. In an ideal world, that would be hashed.

Wouldn't it be nice if someone developed an open source radio with open source protocols? I can think of about a million things I'd improve on.