[Canberrauav] MavProxy "Line Noise"

Stephen Gloor stevegloor at gmail.com
Tue Dec 15 10:20:43 AEDT 2015


I get exactly the same thing.  I have tried to eliminate it by connecting
the RFD-900 direct serial to the RasPi and not using a USB FTDI connector
and it made no difference.

I have a mavproxy wifi box with a RasPi running Mavproxy and HostAP
connected to an stock RFD-900 to relay mavproxy out on the wifi to tablets.

I experience really bad link quality when using the Wifi bridge I think due
to these spurious characters.  I have no floating grounds or anything.

Stephen Gloor
WestCoastUAV

On Tue, Dec 15, 2015 at 6:21 AM <canberrauav-request at canberrauav.org.au>
wrote:

> Send Canberrauav mailing list submissions to
>         canberrauav at canberrauav.org.au
>
> To subscribe or unsubscribe via the World Wide Web, visit
>         http://canberrauav.org.au/cgi-bin/mailman/listinfo/canberrauav
> or, via email, send a message with subject or body 'help' to
>         canberrauav-request at canberrauav.org.au
>
> You can reach the person managing the list at
>         canberrauav-owner at canberrauav.org.au
>
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of Canberrauav digest..."
>
>
> Today's Topics:
>
>    1. FirstarEX (Jack Pittar)
>    2. Re: FirstarEX (Adam Kroll)
>    3. MAVProxy "line noise" (Alex Satrapa)
>    4. Re: MAVProxy "line noise" (Hugh Blemings)
>    5. Re: MAVProxy "line noise" (David C)
>    6. Re: MAVProxy "line noise" (Alex Satrapa)
>
>
> ----------------------------------------------------------------------
>
> Message: 1
> Date: Mon, 14 Dec 2015 13:31:27 +1100
> From: "Jack Pittar" <jpittar at bigpond.net.au>
> To: "Canberrauav" <Canberrauav at canberrauav.com>
> Subject: [Canberrauav] FirstarEX
> Message-ID: <ONEALCFBKPLECAJMNIDKAEOBDPAA.jpittar at bigpond.net.au>
> Content-Type: text/plain;       charset="iso-8859-1"
>
> On Sunday I trialled the VoltanexRC FirstarEX 2000 without an autopilot.
>
> Flying: Ray Murray hand launched it and I struggled with the controls for a
> minute or so till I had it trimmed.
> It flew smoothly and glided well, just like the glider it is.
> The all grey colour against a cloudy sky made visibility terrible -
> painting
> the topside of the wing white will be essential.
> After landing, I confirmed that it most likely could be launched from the
> mat without an undercarriage.
> The design of the plane looks just right for hand launching, with a dimple
> each side of the fuselage for the fingers. Ray mentioned it was too
> slippery, so a bit of sandpapering in the right place might be a good idea.
> With the three cell battery there was plenty of power. I do not feel the
> need for a 4 cell battery.
> Transition from powered flight to gliding did not require any trim change.
>
> Hardware description: This is the Plug and Play version. The wings and tail
> feathers are EPO foam with a coat of grey paint, which makes for a fairly
> smooth and firm surface.
> The mainspar is a square section aluminium, joined at the centre with a
> carbon fibre tube fitting into the aluminium spar. There are numerous rods
> and carbon fibre stiffeners in the wing, ailerons, and horizontal
> stabiliser. Despite the stiffeners, the ailerons were bowed like a banana.
> I
> had to add a hinge for the rudder adjacent to the rudder horn because it
> made for a very sloppy actuation without it.
> The fuselage is blow moulded plastic. There is plenty of room and access is
> OK. It feels slippery and glue does not seem to stick to it - super glue
> peels straight off !
> The kit is a step up from the cheap foamy designs. Plastic parts and screws
> are used for assembly such that glue is not necessary, and it should endure
> repeated assembly without deformation of the foam parts. However; the parts
> are not that well aligned so a fair bit of force and fiddling is required.
> The screws supplied with the kit seemed to be a random mix. I had to source
> a few from my stack. There were quite a few screws and a piece of rubber
> left over with no means of discovering where they should have gone.
>
> Assembly: I elected to separate the foam hinges and install hinges cut from
> CA hinge sheet. I made slots slots cut initially with a balsa saw and
> squared up with a thin exacto knife. I used thin foam safe super glue which
> I soaked into the hinge before inserting it. This glue takes some time to
> grab, giving enough time to wiggle and poke and unstick my fingers. I fixed
> the flaps in place because I intend to make this into a quadplane.
> For the centre of gravity I chose 45mm behind the leading edge of the wing.
> The manual, which I read later, suggests 65mm. I was quite pleased with the
> flight, so I will leave it as it is, despite the fact this requires a
> slightly heavier battery of about 380gm.
> The throws I was quite satisfied with were as follows:
> Ailerons inboard end - 18mm up, 10 down.
> Elevator - 15mm up and down.
> Rudder, measured at elevator - 15mm left and right.
>
> Jack.
>
>
>
> ------------------------------
>
> Message: 2
> Date: Mon, 14 Dec 2015 10:24:38 +0800
> From: Adam Kroll <krolladam at hotmail.com>
> To: Jack Pittar <jpittar at bigpond.net.au>, Canberrauav
>         <canberrauav at canberrauav.com>
> Subject: Re: [Canberrauav] FirstarEX
> Message-ID: <COL127-W947258AF358239B1751DFB0ED0 at phx.gbl>
> Content-Type: text/plain; charset="iso-8859-1"
>
> Great, thanks for the write up Jack.  I'll try to maiden mine this weekend.
>
> Cheers,
>
>
>
>
>
> Adam Kroll
>
>
> > From: jpittar at bigpond.net.au
> > To: Canberrauav at canberrauav.com
> > Date: Mon, 14 Dec 2015 13:31:27 +1100
> > Subject: [Canberrauav] FirstarEX
> >
> > On Sunday I trialled the VoltanexRC FirstarEX 2000 without an autopilot.
> >
> > Flying: Ray Murray hand launched it and I struggled with the controls
> for a
> > minute or so till I had it trimmed.
> > It flew smoothly and glided well, just like the glider it is.
> > The all grey colour against a cloudy sky made visibility terrible -
> painting
> > the topside of the wing white will be essential.
> > After landing, I confirmed that it most likely could be launched from the
> > mat without an undercarriage.
> > The design of the plane looks just right for hand launching, with a
> dimple
> > each side of the fuselage for the fingers. Ray mentioned it was too
> > slippery, so a bit of sandpapering in the right place might be a good
> idea.
> > With the three cell battery there was plenty of power. I do not feel the
> > need for a 4 cell battery.
> > Transition from powered flight to gliding did not require any trim
> change.
> >
> > Hardware description: This is the Plug and Play version. The wings and
> tail
> > feathers are EPO foam with a coat of grey paint, which makes for a fairly
> > smooth and firm surface.
> > The mainspar is a square section aluminium, joined at the centre with a
> > carbon fibre tube fitting into the aluminium spar. There are numerous
> rods
> > and carbon fibre stiffeners in the wing, ailerons, and horizontal
> > stabiliser. Despite the stiffeners, the ailerons were bowed like a
> banana. I
> > had to add a hinge for the rudder adjacent to the rudder horn because it
> > made for a very sloppy actuation without it.
> > The fuselage is blow moulded plastic. There is plenty of room and access
> is
> > OK. It feels slippery and glue does not seem to stick to it - super glue
> > peels straight off !
> > The kit is a step up from the cheap foamy designs. Plastic parts and
> screws
> > are used for assembly such that glue is not necessary, and it should
> endure
> > repeated assembly without deformation of the foam parts. However; the
> parts
> > are not that well aligned so a fair bit of force and fiddling is
> required.
> > The screws supplied with the kit seemed to be a random mix. I had to
> source
> > a few from my stack. There were quite a few screws and a piece of rubber
> > left over with no means of discovering where they should have gone.
> >
> > Assembly: I elected to separate the foam hinges and install hinges cut
> from
> > CA hinge sheet. I made slots slots cut initially with a balsa saw and
> > squared up with a thin exacto knife. I used thin foam safe super glue
> which
> > I soaked into the hinge before inserting it. This glue takes some time to
> > grab, giving enough time to wiggle and poke and unstick my fingers. I
> fixed
> > the flaps in place because I intend to make this into a quadplane.
> > For the centre of gravity I chose 45mm behind the leading edge of the
> wing.
> > The manual, which I read later, suggests 65mm. I was quite pleased with
> the
> > flight, so I will leave it as it is, despite the fact this requires a
> > slightly heavier battery of about 380gm.
> > The throws I was quite satisfied with were as follows:
> > Ailerons inboard end - 18mm up, 10 down.
> > Elevator - 15mm up and down.
> > Rudder, measured at elevator - 15mm left and right.
> >
> > Jack.
> >
>
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL: <
> http://canberrauav.org.au/pipermail/canberrauav/attachments/20151214/1217df66/attachment-0001.html
> >
>
> ------------------------------
>
> Message: 3
> Date: Tue, 15 Dec 2015 00:22:58 +1100
> From: Alex Satrapa <grail at goldweb.com.au>
> To: canberrauav at canberrauav.org.au
> Subject: [Canberrauav] MAVProxy "line noise"
> Message-ID: <D6E388F8-8621-49C7-9934-C26EFB979117 at goldweb.com.au>
> Content-Type: text/plain; charset=utf-8
>
> I’m getting spurious output in MAVProxy which looks like line noise, and
> causes APM Planner 2.0 to decide that it has lost contact with the
> autopilot.
>
> My setup at present is quite simple: a laptop (in my case, 2010 11”
> MacBook Air running 10.10 El Capitan, connected to (Peter Barker’s split
> dipole, so he remembers where it is) telemetry radio via USB cable. I run
> MAVProxy from the command line like so:
>
> (MAVProxy)18:43 MAVProxy ➤ mavproxy.py
> --master=/dev/cu.SLAB_USBtoUART,57600 --quadcopter --aircraft=X525 --speech
> --out=udpin:0.0.0.0:14550
> No speech available
> Loaded module speech
> Connect /dev/cu.SLAB_USBtoUART,57600 source_system=255
> no script X525/mavinit.scr
> Log Directory: X525/logs/2015-12-14/flight2
> Telemetry log: X525/logs/2015-12-14/flight2/flight.tlog
> Waiting for heartbeat from /dev/cu.SLAB_USBtoUART
> MAV> online system 1
> STABILIZE> Mode STABILIZE
> fence breach
> GPS lock at 0 meters
> APM: APM:Copter V3.3.2 (7f16e4d6)
> APM: PX4: 34e1d543 NuttX: 7c5ef883
> APM: Frame: QUAD
> APM: PX4v2 002B0023 3333470D 31313533
> Flight battery 100 percent
> Received 505 parameters
> Saved 505 parameters to X525/logs/2015-12-14/flight2/mav.parm
> /btG at d^Q4;=LOITER> Mode LOITER
> POSHOLD> Mode POSHOLD
> LOITER> Mode LOITER
> STABILIZE> Mode STABILIZE
> 3DmdWz#CIggYd
>  .X3&;=ddLT3KC<(>6dx^e3q<<=SzI>d!p3o#;=b_d3n;>F3DmdeUmD(GC^d3<=Id!@3bg;_>[Q:}di.3Q;=Pdj'3;n:=3DmdMYfUmDQQX>dmA37;<=dN20;L=ddY(o2<=
>
> The mode changes are me flipping the mode switches on the 9XR Pro.
>
> I have to disconnect the UDP stream and then reconnect in order for APM
> Planner 2.0 to be happy again … but then more spurious output arrives and
> APM Planner 2.0 dies again.
>
> APM Planner 2.0 works just fine when it’s connected directly to the
> telemetry radio, so I’m guessing that this “garbage” is actually MAVLink
> data coming through that MAVProxy is not correctly parsing (as opposed to
> the telemetry being corrupted on its own). Does that sound right?
>
> Where would I start looking in the code to find out what’s going on?
>
> Is there a way to get a parsed output of the MAVLink data so I can see
> what’s going on either side of the bogus stuff?
>
> Also, sorry if this is the wrong place to bring this issue up. Should I
> create an issue in the MAVLink GitHub?
>
> Alex Satrapa
>
> PS: MAVProxy reports version 1.4.38:
>
> (MAVProxy) 0:13 mavproxy ➤ mavproxy.py --version
>     1 Δmaster
> MAVProxy is a modular ground station using the mavlink protocol
> MAVProxy Version: 1.4.38
>
>
>
>
>
>
> ------------------------------
>
> Message: 4
> Date: Tue, 15 Dec 2015 00:56:07 +1100
> From: Hugh Blemings <hugh at blemings.org>
> To: Alex Satrapa <grail at goldweb.com.au>,
>         canberrauav at canberrauav.org.au
> Subject: Re: [Canberrauav] MAVProxy "line noise"
> Message-ID: <566ECA77.5050006 at blemings.org>
> Content-Type: text/plain; charset=utf-8; format=flowed
>
> Hi Alex,
>
> I'm going out on a bit of a limb here but there is (to my tired and late
> night eyes) something vaguely deterministic looking about the line noise
> - certainly the 3Dmd pattern repeats - presumably this is a sequence
> that gets legitimately sent in the datastream ?
>
> Is it possible you've got a floating/missing ground or 3.3V logic trying
> to talk to a 5V TTL input or some such ?  It does indeed look like it
> could be actual line noise ?
>
> Cheers,
> Hugh
>
>
>
>
>
> ------------------------------
>
> Message: 5
> Date: Tue, 15 Dec 2015 07:38:06 +1100
> From: David C <cottrill.david at gmail.com>
> To: Hugh Blemings <hugh at blemings.org>
> Cc: canberrauav at canberrauav.org.au
> Subject: Re: [Canberrauav] MAVProxy "line noise"
> Message-ID:
>         <
> CAAiewESZiu20yQH3E53VPJ2c7WUKH9RdQW9W2FtkpYed0SfdwQ at mail.gmail.com>
> Content-Type: text/plain; charset="utf-8"
>
> I have consistently had the same problem. The only thing I've noticed is
> that there is often snippets of raw mavlink data in the 'noise' , and that
> the quantity of noise is related to data throughput. On the up side, it
> doesn't seem to affect operations. Multiple software versions for radio /
> pixhawk / mavlink had no noticeable impact.
>
> Alex, can you fire it up with a direct USB connection? At the very least it
> will eliminate a few variables. I don't have gear of my own to test with
> right now. I'm guessing (a very poorly educated guess) that the source of
> the noise is either noise on the transmission picked up from the unshielded
> serial connections around the pixhawk, or it's a radio bug.
> Exactly which telemetry radios are you using and where did you get them?
> Mine were 3DR v2 433Mhz from eBay.
>
> Thanks,
> David
> On 15 Dec 2015 01:03, "Hugh Blemings" <hugh at blemings.org> wrote:
>
> > Hi Alex,
> >
> > I'm going out on a bit of a limb here but there is (to my tired and late
> > night eyes) something vaguely deterministic looking about the line noise
> -
> > certainly the 3Dmd pattern repeats - presumably this is a sequence that
> > gets legitimately sent in the datastream ?
> >
> > Is it possible you've got a floating/missing ground or 3.3V logic trying
> > to talk to a 5V TTL input or some such ?  It does indeed look like it
> could
> > be actual line noise ?
> >
> > Cheers,
> > Hugh
> >
> >
> >
> >
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL: <
> http://canberrauav.org.au/pipermail/canberrauav/attachments/20151215/3427c420/attachment-0001.html
> >
>
> ------------------------------
>
> Message: 6
> Date: Tue, 15 Dec 2015 09:20:59 +1100
> From: Alex Satrapa <grail at goldweb.com.au>
> To: canberrauav at canberrauav.org.au
> Subject: Re: [Canberrauav] MAVProxy "line noise"
> Message-ID: <24AB1D8F-73E4-476D-A77A-B16CDF60C993 at goldweb.com.au>
> Content-Type: text/plain; charset="utf-8"
>
> On 15 Dec 2015, at 00:56, Hugh Blemings <hugh at blemings.org> wrote:
> > Is it possible you've got a floating/missing ground or 3.3V logic trying
> to talk to a 5V TTL input or some such ?  It does indeed look like it could
> be actual line noise ?
>
> I’m not sure how I’d check for floating/missing ground or incompatible
> logic types. What I have is a PixHawk (PX4v2, Copter V3.3.2) attached to a
> uBlox M8N GPS Module with Compass (connected to GPS and I2C ports). At the
> time of writing, the quad is stationary and disarmed. The lines to the
> telemetry radio are a bunch of single wires, they’re not shielded or
> twisted, so I guess some noise could be present from crosstalk with the GPS
> serial line or even just environmental noise. It will be emissions from the
> PX4 itself no doubt, since I was getting this line noise while out in the
> middle of a sports field too!
>
> On 15 Dec 2015, at 07:38, David C <cottrill.david at gmail.com> wrote:
> > I have consistently had the same problem. The only thing I've noticed is
> that there is often snippets of raw mavlink data in the 'noise' , and that
> the quantity of noise is related to data throughput. On the up side, it
> doesn't seem to affect operations. Multiple software versions for radio /
> pixhawk / mavlink had no noticeable impact.
>
> > Alex, can you fire it up with a direct USB connection? At the very least
> it will eliminate a few variables. I don't have gear of my own to test with
> right now. I'm guessing (a very poorly educated guess) that the source of
> the noise is either noise on the transmission picked up from the unshielded
> serial connections around the pixhawk, or it's a radio bug.
>
> Just confirming that there are no “line noise” characters in the USB
> output, though I’m having trouble reconnecting to USB once the PixHawk is
> booted. I’m clearly doing something wrong since MAVProxy just sits there
> waiting for heartbeat (PX4 is detected if I invoke “mavproxy.py
> —master=/dev/cu.usbmodem1” but not if I invoke “mavproxy.py
> —master=/dev/cu.usbmodem1 —quadcopter”).
>
> > Exactly which telemetry radios are you using and where did you get them?
> Mine were 3DR v2 433Mhz from eBay
>
> The radios appear to be a Hobbyking clone 915MHz on the drone, and one
> I’ve borrowed from Peter Barker (or James Pattison?) during the UAV course*
> which is a naked PCB with an artisan† dipole antenna attached.
>
> For the moment I’ll just accept the “line noise” as actual line noise and
> expected/normal. I’m still curious as to why it’s presented: perhaps as a
> safety mechanism so we can see if there are MAVLink packets arriving that
> MAVProxy isn’t handling?
>
> Alex
>
> *  The order for autopilot and accessories didn’t arrive in time for the
> course, so we’ve cobbled together our drones using things old, borrowed and
> blue.
> † artisan: hand crafted, of unknown quality but possessing rustic charm.
>
> -------------- next part --------------
> A non-text attachment was scrubbed...
> Name: signature.asc
> Type: application/pgp-signature
> Size: 481 bytes
> Desc: Message signed with OpenPGP using GPGMail
> URL: <
> http://canberrauav.org.au/pipermail/canberrauav/attachments/20151215/6e41da9d/attachment.sig
> >
>
> End of Canberrauav Digest, Vol 60, Issue 9
> ******************************************
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://canberrauav.org.au/pipermail/canberrauav/attachments/20151214/9b1f1f81/attachment-0001.html>


More information about the Canberrauav mailing list