Long Range Flying Wing

classic Classic list List threaded Threaded
24 messages Options
12
Reply | Threaded
Open this post in threaded view
|

Re: Long Range Flying Wing

Reto Büttner
Hi Stephen and Chris

Thanks for your explanations and tips.

Our power settings should be correct. We don't use a video link on our
current airframe. Our next try will be:

- Polling mode not acknowledged / no retries (RR = 0)

If that doesn't help we might change the antenna configuration on the
aircraft. Unfortunately on our small airframe with high aerodynamical
goals we are a bit limited on different configurations.

I will let you know our results.

Regards, Reto

2012/7/6 Stephen Dwyer <[hidden email]>:

> Hello,
>
> Regarding range issues:
>
> First thing, is the power setting appropriate on both sides?
>
> As Chris mentioned, definitely possibilities of interference from
> other equipment. This can be very hard to debug. As an example, in the
> past we had horrible problems with GPS reception. After days of
> changing things around, we used a spectrum analyzer to debug, which
> really helped solve the problem. I realize it can be difficult to get
> access to a spectrum analyzer (turned out to be the 500MHz SBC spewing
> junk RFI at 1.5GHz ... a nice harmonic, some careful shielding solved
> the problem).
>
> We had serious asymmetrical link issues for a while as well. After
> some debugging, we discovered that the connector on one of the
> antennas was in poor condition. Switching to a new antenna solved all
> of our problems. However, this asymmetry also showed significantly
> decreased range, even in the better telemetry direction. Again, Chris
> mentioned this: antenna selection, antenna/cable setup and orientation
> are rather important.
>
> Our modems also run warm, but I think this is to be expected with the
> high power. We make sure to allow some air flow around the modems
> (i.e. we don't wrap them in foam).
>
> A while ago I wrote a small module to measure the RSSI on the airborne
> side to help debug range problems. It was quite a while back and I am
> not sure if it is still alive (a quick compile on master failed). I
> will try to fix it in the next couple days if I get a chance, and do
> some proper documentation on the wiki for it. Unfortunately, it is
> currently for LPC21xx only (we use it on a TWOG). Check
> conf/modules/xtend_rssi.xml for more info.
>
> Hope that helps.
>
> Thanks,
> -Stephen Dwyer
>
>
>
>
> On Fri, Jul 6, 2012 at 11:09 AM, Chris <[hidden email]> wrote:
>> Well i have flown to a point where the ground station with its 8db patch
>> could not receive well
>> the planes data but i was still able to give a command to the plane and
>> return it home which indicates
>> good airborne reception.
>> I use as stated the polling mode with no confirmation and retries are set to
>> 0.
>> I have set the addresses somewhere in the top range, FFF0-FFF1 i think and
>> changed the channel
>> to 9 (different hopping frequencies).
>> One day i had a problem communicating with the plane (the downlink was fine)
>> and it proved to be the video tx.
>> Since then i have fitted filters to the video tx (high pass) and the modem
>> (low pass)
>> and all problems were solved.
>> The antenna and the modem have to be in a straight line otherwise SWR will
>> rise,
>> RF will flow through the wires all over the airplane causing havoc and the
>> antenna
>> radiation pattern will be greatly distorted.
>> The same goes for the video tx.
>> Chris
>>
>>
>> Hi Stephen
>>
>> We configured our 9XTends according to your recommendation
>> (Acknowledged Polling Mode):
>>
>> http://paparazzi.enac.fr/wiki/Modems#Configuration
>>
>> As we are sending lots of telemetry and not using datalink joystick
>> control we chose the air module as the base. It gets pretty warm in
>> operation. In the workshop everything works fine.
>>
>> On the field we experience a very asymmetrical link: The telemetry
>> data from the aircraft gets well to the ground station, also at range.
>> But the commands from the ground station only get to the aircraft at
>> very close range. At range we don't get any commands from the ground
>> station to the aircraft anymore.
>>
>> Can you explain this behaviour? How do you achieve a symmetrical data link?
>>
>> We still have a 90 degree bend between the modem body and the antenna
>> on the airplane. On the ground station we have them in a straight
>> line. Could this explain the asymmetrical link?
>>
>> Regards, Reto
>>
>> 2
>>
>>
>>
>> _______________________________________________
>> Paparazzi-devel mailing list
>> [hidden email]
>> https://lists.nongnu.org/mailman/listinfo/paparazzi-devel
>>
>
> _______________________________________________
> Paparazzi-devel mailing list
> [hidden email]
> https://lists.nongnu.org/mailman/listinfo/paparazzi-devel

_______________________________________________
Paparazzi-devel mailing list
[hidden email]
https://lists.nongnu.org/mailman/listinfo/paparazzi-devel
Reply | Threaded
Open this post in threaded view
|

Re: Long Range Flying Wing

agressiva
Stephen, the xbee 900 not have the same parameter than xtend 900 so i cannot make the adjusts.

I only changed the RR =0 but the problem continue.

Today i will try with 19k2 bps

> Date: Sat, 7 Jul 2012 09:56:24 +0200

> From: [hidden email]
> To: [hidden email]
> Subject: Re: [Paparazzi-devel] Long Range Flying Wing
>
> Hi Stephen and Chris
>
> Thanks for your explanations and tips.
>
> Our power settings should be correct. We don't use a video link on our
> current airframe. Our next try will be:
>
> - Polling mode not acknowledged / no retries (RR = 0)
>
> If that doesn't help we might change the antenna configuration on the
> aircraft. Unfortunately on our small airframe with high aerodynamical
> goals we are a bit limited on different configurations.
>
> I will let you know our results.
>
> Regards, Reto
>
> 2012/7/6 Stephen Dwyer <[hidden email]>:
> > Hello,
> >
> > Regarding range issues:
> >
> > First thing, is the power setting appropriate on both sides?
> >
> > As Chris mentioned, definitely possibilities of interference from
> > other equipment. This can be very hard to debug. As an example, in the
> > past we had horrible problems with GPS reception. After days of
> > changing things around, we used a spectrum analyzer to debug, which
> > really helped solve the problem. I realize it can be difficult to get
> > access to a spectrum analyzer (turned out to be the 500MHz SBC spewing
> > junk RFI at 1.5GHz ... a nice harmonic, some careful shielding solved
> > the problem).
> >
> > We had serious asymmetrical link issues for a while as well. After
> > some debugging, we discovered that the connector on one of the
> > antennas was in poor condition. Switching to a new antenna solved all
> > of our problems. However, this asymmetry also showed significantly
> > decreased range, even in the better telemetry direction. Again, Chris
> > mentioned this: antenna selection, antenna/cable setup and orientation
> > are rather important.
> >
> > Our modems also run warm, but I think this is to be expected with the
> > high power. We make sure to allow some air flow around the modems
> > (i.e. we don't wrap them in foam).
> >
> > A while ago I wrote a small module to measure the RSSI on the airborne
> > side to help debug range problems. It was quite a while back and I am
> > not sure if it is still alive (a quick compile on master failed). I
> > will try to fix it in the next couple days if I get a chance, and do
> > some proper documentation on the wiki for it. Unfortunately, it is
> > currently for LPC21xx only (we use it on a TWOG). Check
> > conf/modules/xtend_rssi.xml for more info.
> >
> > Hope that helps.
> >
> > Thanks,
> > -Stephen Dwyer
> >
> >
> >
> >
> > On Fri, Jul 6, 2012 at 11:09 AM, Chris <[hidden email]> wrote:
> >> Well i have flown to a point where the ground station with its 8db patch
> >> could not receive well
> >> the planes data but i was still able to give a command to the plane and
> >> return it home which indicates
> >> good airborne reception.
> >> I use as stated the polling mode with no confirmation and retries are set to
> >> 0.
> >> I have set the addresses somewhere in the top range, FFF0-FFF1 i think and
> >> changed the channel
> >> to 9 (different hopping frequencies).
> >> One day i had a problem communicating with the plane (the downlink was fine)
> >> and it proved to be the video tx.
> >> Since then i have fitted filters to the video tx (high pass) and the modem
> >> (low pass)
> >> and all problems were solved.
> >> The antenna and the modem have to be in a straight line otherwise SWR will
> >> rise,
> >> RF will flow through the wires all over the airplane causing havoc and the
> >> antenna
> >> radiation pattern will be greatly distorted.
> >> The same goes for the video tx.
> >> Chris
> >>
> >>
> >> Hi Stephen
> >>
> >> We configured our 9XTends according to your recommendation
> >> (Acknowledged Polling Mode):
> >>
> >> http://paparazzi.enac.fr/wiki/Modems#Configuration
> >>
> >> As we are sending lots of telemetry and not using datalink joystick
> >> control we chose the air module as the base. It gets pretty warm in
> >> operation. In the workshop everything works fine.
> >>
> >> On the field we experience a very asymmetrical link: The telemetry
> >> data from the aircraft gets well to the ground station, also at range.
> >> But the commands from the ground station only get to the aircraft at
> >> very close range. At range we don't get any commands from the ground
> >> station to the aircraft anymore.
> >>
> >> Can you explain this behaviour? How do you achieve a symmetrical data link?
> >>
> >> We still have a 90 degree bend between the modem body and the antenna
> >> on the airplane. On the ground station we have them in a straight
> >> line. Could this explain the asymmetrical link?
> >>
> >> Regards, Reto
> >>
> >> 2
> >>
> >>
> >>
> >> _______________________________________________
> >> Paparazzi-devel mailing list
> >> [hidden email]
> >> https://lists.nongnu.org/mailman/listinfo/paparazzi-devel
> >>
> >
> > _______________________________________________
> > Paparazzi-devel mailing list
> > [hidden email]
> > https://lists.nongnu.org/mailman/listinfo/paparazzi-devel
>
> _______________________________________________
> Paparazzi-devel mailing list
> [hidden email]
> https://lists.nongnu.org/mailman/listinfo/paparazzi-devel

_______________________________________________
Paparazzi-devel mailing list
[hidden email]
https://lists.nongnu.org/mailman/listinfo/paparazzi-devel
Reply | Threaded
Open this post in threaded view
|

Re: Long Range Flying Wing

AJ Kochevar
In reply to this post by Reto Büttner
Eduardo,
I am using those exact radios on a rotorcraft with the same results you have.  What I have found is that the telemetry going down the link from the aircraft to the groundstation was at such a high rate for 9600 baud it was consuming all of the available timing for transmission.  I tried quite a few settings in the software to try to get them to play correctly, but in the end I went in and adjusted the frequencies of the telemetry.xml file.  Even doubling a couple of them, and making sure to offset the values some can make a world of difference.  What I mean by offsetting is if you want a couple things at 1hz, you will run into less of an issue if you put 1 at 0.9, 1 at 1.0 and 1 at 1.1 vs say puting them all at 1.0.  Give it a try, it certainly worked for me.

AJ

_______________________________________________
Paparazzi-devel mailing list
[hidden email]
https://lists.nongnu.org/mailman/listinfo/paparazzi-devel
Reply | Threaded
Open this post in threaded view
|

Re: Long Range Flying Wing

agressiva
Hello AJ,

here are the settings i am using .
    <mode name="default">
      <message name="AIRSPEED"            period="1.4"/>
      <message name="ALIVE"               period="5"/>
      <message name="GPS"                 period="1.0"/>
      <message name="NAVIGATION"          period="1.1"/>
      <message name="ATTITUDE"            period="0.7"/>
      <message name="ESTIMATOR"           period="1.1"/>
      <message name="ENERGY"              period="2.5"/>
      <message name="WP_MOVED"            period="1.1"/>
      <message name="CIRCLE"              period="1.05"/>
      <message name="DESIRED"             period="1.05"/>
      <message name="BAT"                 period="1.1"/>
      <message name="SEGMENT"             period="1.2"/>
      <message name="CALIBRATION"         period="2.1"/>
      <message name="NAVIGATION_REF"      period="9."/>
      <message name="PPRZ_MODE"           period="5."/>
      <message name="SETTINGS"            period="5."/>
      <message name="STATE_FILTER_STATUS" period="2.2"/>
      <message name="DOWNLINK"            period="5.1"/>
      <message name="DL_VALUE"            period="1.2"/>
      <message name="SURVEY"              period="2.1"/>
      <message name="GPS_SOL"             period="2.0"/>
    </mode>

today i configured my xbee900 to digi mesh with better results on 9600 bps
i set RR and NN as the xtend modem.



Date: Sat, 7 Jul 2012 09:42:26 -0700
From: [hidden email]
To: [hidden email]
Subject: Re: [Paparazzi-devel] Long Range Flying Wing

Eduardo,
I am using those exact radios on a rotorcraft with the same results you have.  What I have found is that the telemetry going down the link from the aircraft to the groundstation was at such a high rate for 9600 baud it was consuming all of the available timing for transmission.  I tried quite a few settings in the software to try to get them to play correctly, but in the end I went in and adjusted the frequencies of the telemetry.xml file.  Even doubling a couple of them, and making sure to offset the values some can make a world of difference.  What I mean by offsetting is if you want a couple things at 1hz, you will run into less of an issue if you put 1 at 0.9, 1 at 1.0 and 1 at 1.1 vs say puting them all at 1.0.  Give it a try, it certainly worked for me.

AJ

_______________________________________________ Paparazzi-devel mailing list [hidden email] https://lists.nongnu.org/mailman/listinfo/paparazzi-devel

_______________________________________________
Paparazzi-devel mailing list
[hidden email]
https://lists.nongnu.org/mailman/listinfo/paparazzi-devel
12