“Waiting for Device” in SD, intermittent.

[grahambaker]

I'm an admin in a team of same on a large flying forum (>60,000) users.

One of the things we have a lot of problems with is people using data compression or saving in browsers on mobile devices. We end up with failed logins, odd invalid session errors, or constant connection/disconnection issues, with need to re-log in.

If using an mobile device, it may be worth disabling data saving/compression. Or, if you feel you need that in your main browser, download and install a different one for PAW. Googling will show you how to turn the feature on/off for testing...

But SkyDemon isn't a browser app.

[Admin]

Like this?
$PALOG,20200526,142305,THROTTLE,1,UV,0,FC,1,TH

There are a couple like that every minute in the log.

And a lot of these:

$PALOG,20200526,105117,WIFI,UDP,192.168.1.27:4000,DISCONNECTED
$PALOG,20200526,105117,WIFI,UDP,192.168.1.27:2000,ATTEMPT
$PALOG,20200526,105117,DEBUG,MODEC-FILTER-PASS,7380C8,206,SIG,200,MAX,170,FILTER,16198,VDIFF

Unfortunately I didn't check the home page during flight.  I can fire it up now if that would help, but so far it has always been fine on the ground.

I do have a confession though - while looking for the track logs, I noticed that I had loaded an old aircraft configuration (flown before lockdown) and forgot to switch back to my usual. The old config had connection type set to AUTO not UDP.  So I have loaded the correct config now, but don't know if that is actually relevant.

Thanks

yep thats it - smoking gun 
1,UV - means UnderVoltage (true)

Sorry if asked before, USB lead type and battery type ?

Thx
Lee

[exfirepro]

I do have a confession though - while looking for the track logs, I noticed that I had loaded an old aircraft configuration (flown before lockdown) and forgot to switch back to my usual. The old config had connection type set to AUTO not UDP.  So I have loaded the correct config now, but don't know if that is actually relevant.

Thanks

Very relevant - that could certainly explain the iPad disconnects.

WRT the Throttling,

In your original post you stated (with reference to earlier disconnects) that...

'This has been happening for some time, and previously very helpful PaW Support have soldered the power cable in [so presumably standard Rosetta one] (to eliminate loose usb connection as cause).'

The cable could of course still have developed a fault internally or where attached to the connectors at either end through repeated flexing. You can easily check this by powering up the unit then flexing the cable while watching the normally solid red 'Power LED' (in the corner next to the longer PAW antenna socket on the Rosetta), for any sign of flickering or going out which will indicate disconnection.

I also would be interested to hear what power supply arrangements (i.e. what make/model of socket/ power pack or whatever and cabling thereto) you are using.

Best Regards

Peter

[Uptimist]

Like this?
$PALOG,20200526,142305,THROTTLE,1,UV,0,FC,1,TH

There are a couple like that every minute in the log.

And a lot of these:

$PALOG,20200526,105117,WIFI,UDP,192.168.1.27:4000,DISCONNECTED
$PALOG,20200526,105117,WIFI,UDP,192.168.1.27:2000,ATTEMPT
$PALOG,20200526,105117,DEBUG,MODEC-FILTER-PASS,7380C8,206,SIG,200,MAX,170,FILTER,16198,VDIFF

Unfortunately I didn't check the home page during flight.  I can fire it up now if that would help, but so far it has always been fine on the ground.

I do have a confession though - while looking for the track logs, I noticed that I had loaded an old aircraft configuration (flown before lockdown) and forgot to switch back to my usual. The old config had connection type set to AUTO not UDP.  So I have loaded the correct config now, but don't know if that is actually relevant.

Thanks

yep thats it - smoking gun 
1,UV - means UnderVoltage (true)

Sorry if asked before, USB lead type and battery type ?

Thx
Lee

Thanks Lee and Peter.  Answering both your questions together:

The USB lead type is the Juicy Bitz lead which came with the unit when I purchased it (mid 2018?).  Last summer Keith V kindly soldered the PAW end to the board to try to resolve issues I was having then when I suspected a loose connector. 

The battery is an Anker 20100 unit, fully charge (four blue dots).

The cable (0.5m long maybe?) from the PAW is plugged directly into the Anker 20100 unit.

Based on everything said so far, it seems like the cable has to be the prime suspect.  Perhaps always working OK at home as it's stationary, and intermittent in aircraft because of vibration.  Maybe something to do with heat too - it can get quite warm in the a/c.

I'll try flexing it around at home to see if I can replicate.

If I need to order a new cable, can you recommend a specific one (e.g. Amazon link)?

Many thanks.

[steveu]

I'm an admin in a team of same on a large flying forum (>60,000) users.

One of the things we have a lot of problems with is people using data compression or saving in browsers on mobile devices. We end up with failed logins, odd invalid session errors, or constant connection/disconnection issues, with need to re-log in.

If using an mobile device, it may be worth disabling data saving/compression. Or, if you feel you need that in your main browser, download and install a different one for PAW. Googling will show you how to turn the feature on/off for testing...

But SkyDemon isn't a browser app.

Sorry, agreed, but I was looking at Lee's post where he said:

there seems to be a few anecdotal stories on Android web browsers
Which web browser are you using on android ?

Can you confirm that your android device has ALL battery saving options disabled
This is known to mess up the WiFi connection

thx
Lee

[Uptimist]

So I did a highly scientific field test to see if moving cable and unit around would reveal anything.  Results were puzzling.

The first time I powered up, the home page showed Voltage warning and Throttle from the get-go.  I moved cable around, shook unit, sat it in the sun etc.  Nothing changed.

So I unplugged the battery, left it for a minute, then powered up again.  This time, voltage fine, no throttling, everything OK.  I shook it around, sat it in the sun, flexed the cable all over the place.  Nothing changed, no throttling, no voltage warning.

I left it running for about 90mins, gave it another few shakes during that time and still no change.

I inspected the log afterwards, I don't know how to read it, but the only thing which maybe looked odd was a bunch of these which started showing up in high volume after about an hour.

$PALOG,20200529,113658,DEBUG,MODEC-FILTER-PASS,4073FD,223,SIG,200,MAX,170,FILTER,1730,VDIFF
$PALOG,20200529,113659,DEBUG,MODEC-FILTER-PASS,4073FD,223,SIG,200,MAX,170,FILTER,1730,VDIFF
$PALOG,20200529,113659,DEBUG,MODEC-FILTER-PASS,4073FD,223,SIG,200,MAX,170,FILTER,1729,VDIFF
$PALOG,20200529,113659,DEBUG,MODEC-FILTER-PASS,4073FD,223,SIG,200,MAX,170,FILTER,1729,VDIFF
$PALOG,20200529,113659,DEBUG,MODEC-FILTER-PASS,4073FD,223,SIG,200,MAX,170,FILTER,1729,VDIFF

And a number of these randomly interspersed about 5 mins apart:

$PALOG,20200529,105334,PAWRT,400BC1,TIMEOUT


So I'm not sure what to conclude from that - perhaps a poor/dirty USB connector (cable to battery), or maybe I got unlucky with the battery and it's a faulty one.  I can try a different battery.  Changing the cable is more of a hassle.

The other thing I have observed is that when running SD on the iPad, if you put SD to background (e.g. to check another app), and ipad goes to sleep mode, then you return to SD, the dreaded "Waiting for Device" appears, and it doesn't automatically reconnect.  Is that expected behaviour?

Thanks again.

[Admin]

So I did a highly scientific field test to see if moving cable and unit around would reveal anything.  Results were puzzling.

The first time I powered up, the home page showed Voltage warning and Throttle from the get-go.  I moved cable around, shook unit, sat it in the sun etc.  Nothing changed.

These values are latched, also called 'sticky bits' - once set they do not clear, otherwise there is no way to tell an event occured.

So I unplugged the battery, left it for a minute, then powered up again.  This time, voltage fine, no throttling, everything OK.  I shook it around, sat it in the sun, flexed the cable all over the place.  Nothing changed, no throttling, no voltage warning.

OK

I left it running for about 90mins, gave it another few shakes during that time and still no change.

I inspected the log afterwards, I don't know how to read it, but the only thing which maybe looked odd was a bunch of these which started showing up in high volume after about an hour.

$PALOG,20200529,113658,DEBUG,MODEC-FILTER-PASS,4073FD,223,SIG,200,MAX,170,FILTER,1730,VDIFF
$PALOG,20200529,113659,DEBUG,MODEC-FILTER-PASS,4073FD,223,SIG,200,MAX,170,FILTER,1730,VDIFF
$PALOG,20200529,113659,DEBUG,MODEC-FILTER-PASS,4073FD,223,SIG,200,MAX,170,FILTER,1729,VDIFF
$PALOG,20200529,113659,DEBUG,MODEC-FILTER-PASS,4073FD,223,SIG,200,MAX,170,FILTER,1729,VDIFF
$PALOG,20200529,113659,DEBUG,MODEC-FILTER-PASS,4073FD,223,SIG,200,MAX,170,FILTER,1729,VDIFF

I presume you have the mode-C  filter enabled - this is showing the state of the dynamic filter

And a number of these randomly interspersed about 5 mins apart:

$PALOG,20200529,105334,PAWRT,400BC1,TIMEOUT

This indicates an A/C is no longer being received, so either now out of range, or no longer emitting.

So I'm not sure what to conclude from that - perhaps a poor/dirty USB connector (cable to battery), or maybe I got unlucky with the battery and it's a faulty one.  I can try a different battery.  Changing the cable is more of a hassle.

Either is possible.

The other thing I have observed is that when running SD on the iPad, if you put SD to background (e.g. to check another app), and ipad goes to sleep mode, then you return to SD, the dreaded "Waiting for Device" appears, and it doesn't automatically reconnect.  Is that expected behaviour?

yes it was asleep, and so not listening, and has to start listening again

Thx
Lee

[Uptimist]

Ok thanks Lee.  I’ll try a different battery next time I’m up and will report back. At least I know what to look for now.  It’s frustrating to be having the problem when I am using all the recommended bits.