Throttle control bug ?

[AndreK]

about 50% of the boot-ups are like this (correct):

I switch on the RC-TX (no throttle) , and plane. the plane changes from failsafe to live control within ~4 seconds, and I have control, the motors beeps, - and I can then apply throttle, and they start.

Another 50% of boot-ups are like this:

I switch on the RC-TX (no throttle) , and plane. the plane changes from failsafe to live control within ~4 seconds, and *starts* the motors at 10-20% throttle, I have control, and I am able to influence the motors RPM, but they won't stop at throttle stick in low position !.

I guess the brushless-motor controllers during boot time , somehow, gets to experience a signal lower than "no throttle" - and calibrate to it.

[Alex Villa]

about 50% of the boot-ups are like this (correct):

I switch on the RC-TX (no throttle) , and plane. the plane changes from failsafe to live control within ~4 seconds, and I have control, the motors beeps, - and I can then apply throttle, and they start.

Another 50% of boot-ups are like this:

I switch on the RC-TX (no throttle) , and plane. the plane changes from failsafe to live control within ~4 seconds, and *starts* the motors at 10-20% throttle, I have control, and I am able to influence the motors RPM, but they won't stop at throttle stick in low position !.

I guess the brushless-motor controllers during boot time , somehow, gets to experience a signal lower than "no throttle" - and calibrate to it.
Do you always turn the RC tx before the RVOSD, RX, ESC?

[AndreK]

Yes, flying Spektrum it makes no sense, but from old days, I have the habit of switching the transmitter on first.
I did check this a few times before reporting this behaviour.

[Alex Villa]

Yes, flying Spektrum it makes no sense, but from old days, I have the habit of switching the transmitter on first.
I did check this a few times before reporting this behaviour.
Did you tried this without the RVOSD? I dont remember seeing this behaviour before.

[AndreK]

No , and I need to inform you that I just tested these speed-controllers - I did not really use them before getting RVOSD.
So , yes, it might be something strange with them..

I will do a lot of testing text time I disassemble the plane, it's really tight built and everything is very well placed/protected.
- in the mean time, we can see if anyone else responds to this thread.

[Alex Villa]

Anyways be sure the failsafes values are properly set, specifically "Set throttle failsafe"

[IanJ]

Hi,

I have a similar bug.......when I power up my ESC will beep continously and I have to power cycle everything until it starts up properly. This only happens with RVOSD connected.

I have tried all combinations of Rx first, Tx first etc etc.

I haven't investigated yet, but I'm thinking RVOSD on power up has some throttle there and which the ESC determines as throttle on at power up.

It must be marhinal though coz sometimes it all powers up fine.

I would expect RVOSD to output everything at zero/minimum at least until the ESC is fully powered up......which can be 5 or 6 secs. And this being nothing to do with failsafes. Just a good P.O.S.T.

Ian.

[Alex Villa]

Hi,

I have a similar bug.......when I power up my ESC will beep continously and I have to power cycle everything until it starts up properly. This only happens with RVOSD connected.

I have tried all combinations of Rx first, Tx first etc etc.

I haven't investigated yet, but I'm thinking RVOSD on power up has some throttle there and which the ESC determines as throttle on at power up.

It must be marhinal though coz sometimes it all powers up fine.

I would expect RVOSD to output everything at zero/minimum at least until the ESC is fully powered up......which can be 5 or 6 secs. And this being nothing to do with failsafes. Just a good P.O.S.T.

Ian.
Hahaha, you programmers are hard to please. I was worried about processor resets, so I choose to give servo pass thru from the very begining, but this might cause some unexpected output at the begining when the servo inputs are not readed yet and the servo output interrupt wants to kick in.
Just guessing here, because I never experienced that on my system.

However continuos beeps are more like the outputs going permantly out of parameters due to no servo pulses or RC link detected at start.

[AndreK]

Hi,
I have a similar bug.......when I power up my ESC will beep continously and I have to power cycle everything until it starts up properly. This only happens with RVOSD connected.


Do you use that ESC's BEC , or do you use a separate BEC ?
I do, the ESC's BEC circuits are poor linear stuff, so I use a standalone switch-mode 6V 5A BEC.

I suspect, that the linear BEC's and the ESC's are starting first, and manages to "understand" some early outputs from the RVOSD while it's booting from the switch-mode BEC.
The second time if powered up within seconds of the first time, everything starts normally
- most likely because some capacitor is charged.

[Alex Villa]

Do you use that ESC's BEC , or do you use a separate BEC ?
I do, the ESC's BEC circuits are poor linear stuff, so I use a standalone switch-mode 6V 5A BEC.

I suspect, that the linear BEC's and the ESC's are starting first, and manages to "understand" some early outputs from the RVOSD while it's booting from the switch-mode BEC.
The second time if powered up within seconds of the first time, everything starts normally
- most likely because some capacitor is charged.
Using ESC-BEC. Ill take a look at RVOSD outputs on startup with the oscilloscope.

[thomas]

the problem is some advanced receivers do NOT output any servo pulses
at all before first lock of good data, due to advanced frequency hopping
and binding codes and such this takes a few sec just after powerup of the
RC RX, I have tried many ESC and found no problems with this,
but how RVOSD read a missing pulse and maybe generate a non valid pulse
to the ESC is unknown to me, people that have problems with this, should try to connect the RX directly to ESC and then see if all is ok,

if a non valid pulse is seen just after powerup some ESC's will
go into dead mode and continue to beep and not recover !

[IanJ]

Hi all,

Did some testing.....still more to come, but here's what I found so far :-

With Thomas's LRS Rx connected, and the RVOSD throttle channel only by-passed the ESC powers up fine every time. No problems.

With the above and the throttle channel now connected via the RVOSD along with all the other channels the system will not startup cleanly everytime. Have to recycle power about 5 times before I get a clean startup.
I also found that once it has started up fine, if I recycle power then it will usually restart fine.
So, if I do a complete cold boot, i.e. disconnect the battery and let all power dissapate, i.e. discharge all capacitors etc, and then power up the chances are it doesn't startup. There starts the 4 or 5 tries till it does.

I should also note that the RVOSD is starting up everytime, I can see that on my LCD.

NB
1. Battery power fine (single 3S battery)
2. No apparent loose connections. No strained connections.
3. Tx is powered up before applying power to the plane.
4. ESC I'm using is ESC-P-30A (advertised for heli, but has worked fine in all my planes)

[thomas]

another trick, you see the isp/updating plugs on the RVOSD ?
find the mclr and gnd pads and simply connect a short circuit wire,
now the RVOSD the dead and you can power it all up, when you have stable servo movements on channels not feed thru the OSD, you release the reset wire, and now you see bootup of the RVOSD with all valid servo signals input, now tell us if it is stable and good with your ESC ?
if that is the case, this can be fiext really easy with a firmware upgrade of the RVOSD.
if the RVOSD still have a glitch out of ESC spec, try to boot up in this order:
RX on, RVOSD ON but reset, release reset, now boot up ESC, and see is fine,
the software engineer of RVOSD needs to know all your findings about all this,
so he can fix it.

[Alex Villa]

another trick, you see the isp/updating plugs on the RVOSD ?
find the mclr and gnd pads and simply connect a short circuit wire,
now the RVOSD the dead and you can power it all up, when you have stable servo movements on channels not feed thru the OSD, you release the reset wire, and now you see bootup of the RVOSD with all valid servo signals input, now tell us if it is stable and good with your ESC ?
if that is the case, this can be fiext really easy with a firmware upgrade of the RVOSD.
if the RVOSD still have a glitch out of ESC spec, try to boot up in this order:
RX on, RVOSD ON but reset, release reset, now boot up ESC, and see is fine,
the software engineer of RVOSD needs to know all your findings about all this,
so he can fix it.
The problem is that if he solder something on mclr, he will not be able to hockup the ICSP pins so easyly again to reprogram.

If RVOSD see any pulse at startup, it will take it, then fixing it within accepted parameters (0,85 to 2.01mS) and will keep sending that untill the Rx send pulses again.

I will think about.

[IanJ]

So, if RVOSD sees 1 single erroneous pulse at startup then it will accept it and re-send it continiously?

If I can find time in the next day or so I will scope the incoming and outgoing pulses (throttle) at startup and see exactly whats happening.

Ian.

[IanJ]

Hi all,

I scoped the output of the LRS Rx. I wanted to make sure to begin with that the Rx was starting up correctly before I moved onto the RVOSD outputs.

I notice that the time it takes for pulses to appear at the Rx output varies after power up.
Most times it takes 3 secs for valid pulses to appear after power up, sometimes 5 or 6 secs. Sometimes longer. I can hear this at the ESC also as it will beep a little longer as it waits for valid pulses.

So now I tried something else. I powered up the Rx with the Tx switched OFF. So, whilst the ESC is beep'ing madly I wait 10 secs THEN turn on the Tx.
The sticks on the Tx are doing nothing, so I scoped the inputs to the RVOSD and the pulse widths are varying as I move the sticks and they seem quite valid. So the Rx looks to be working.
I scoped the outputs from the RVOSD and the pulse widths on all channels seem fixed!, i.e. despite the input varying when I move the sticks the output is just fixed at what looks like minimum pulse width.

So, I think at the moment that the LRS Rx is a bit unpredictable at startup. However, this in itself should not be a proble, afterall we are only powering up the system.
It looks to me that if the RVOSD doesn't receive any valid pulses within a certain period of time at boot up then even if it receives them later it has locked out, i.e. it does start outputting but all outputs are fixed.
I think maybe RVOSD has gone into failsafe mode but nothing on the RVOSD video shows this, the screen is normal as if everything is fine.
Next, I will setup a failsafe of say half-throttle and see if the 'fixed' pulse width problem is indeed that pulse width. I don't know too much about failsafe....does it show anything on-screen?

Boot-up of hardware/software system can be complex, the software engineer has to think of as many situations as possible to allow boot up reliability. Have I found a problem here?

Ideas/suggestions?

PS. I also check my +5vdc supply and it's immediate at power up, no delay, no glitch and no variation.

Ian.

[thomas]

you see the rx powerup time to servo output will offcource vary abit,
the system use wide band frequency hopper, and the rx have no idea
where the tx is on what frequency at powerup,
so it will start to scann the band, when it find data with correct id code
it will hop in sync with the tx since the id codes reveal the hopper system too, I see a 1-2-3 sec delay not more, many other systems have much more delay and much more variations in this. so this must be ok for any ESC just like you have seen with out the RVOSD you have no ESC problems,
I suggest you inform the RVOSD designers about this issue with pulses beeing locked, good with your measurements I think it is now possible for them to recreate this and fix it, only one last thing is you need to inform them about your hardware and software version.

[Alex Villa]

Yes, If RVOSD does not see valid signals at the input at the first 7 seconds it will ignore any further inputs, before you start thinking this is a mistake :D I have to say that this procedure was made to allow to use the OSD without the inputs conected, and not having to see RC link lost indication on the screen all the time.

[IanJ]

Yes, If RVOSD does not see valid signals at the input at the first 7 seconds it will ignore any further inputs, before you start thinking this is a mistake :D I have to say that this procedure was made to allow to use the OSD without the inputs conected, and not having to see RC link lost indication on the screen all the time.

Could you make this timing user adjustable then?
I would gladly help test any updates and prove Thomas's LRS system 100% compatible with RVOSD, and of course other Rx's that RVOSD may not have tested that may fail the 7 sec cut-off.

Ian.

[Alex Villa]

Could you make this timing user adjustable then?
I would gladly help test any updates and prove Thomas's LRS system 100% compatible with RVOSD, and of course other Rx's that RVOSD may not have tested that may fail the 7 sec cut-off.

Ian.
I think maybe the problem is the OSD its detecting RC link lost those seconds the Receiver its starting up. Then after the receiver wake up it change to bypass, and this change of periods at the begining its confusing the ESC. Since I never ever seem any video or report of RVOSD restarting in flight, I can make the startup process to dont send any output pulses untill valid input detected.