PostFrontal Forum
PostFrontal Forum
PF Home | International_Forum | Profile | Register | Active Topics | Members | Search | FAQ
Username:
Password:
Save Password
Forgot your Password?


 All Forums
 LK8000 International Support
 Hardware
 BlueFlyVario_TTL_GPS_v12 and LK8000 on kobo
 New Topic  Reply to Topic
Next Page
Author Previous Topic Topic Next Topic
Page: of 2

bambuko
Falchetto

United Kingdom
140 Posts

Posted - 10/08/2018 :  12:20:07  Show Profile  Reply with Quote
Is anybody else using this latest Bluefly vario with LK8000?
I have connected it to two different kobos and in both cases kobo suddenly shuts down in flight.

One of these kobos worked fine with v11 version of Bluefly vario for a long time, so I am suspecting it might be something to do with v12 and LK8000?

v12 apparently works fine with Xcsoar, but has not been tested before with LK8000

Have you fitted BlueFlyVario_TTL_GPS_v12 and is it working OK?

Alistair, is very kindly trying to help me to debug the issue from the other side of the world, but it's not an easy job.

It is also possible, I am doing something wrong?

so... if you had succes with v12 and LK8000 on kobo please let us know

Aleksandar
Falchetto

Yugoslavia
147 Posts

Posted - 13/08/2018 :  09:50:47  Show Profile  Visit Aleksandar's Homepage  Reply with Quote
I am interested for that solution, please any info if BFV-V12 work with LK8000 :)

Best regards,
Aleksandar Cirkovic
Go to Top of Page

AlphaLima
Moderator

Germany
1544 Posts

Posted - 13/08/2018 :  13:13:43  Show Profile  Visit AlphaLima's Homepage  Reply with Quote
quote:
Originally posted by bambuko

One of these kobos worked fine with v11 version of Bluefly vario for a long time, so I am suspecting it might be something to do with v12 and LK8000?




So the question is, what are the changes between v11 and v12.
Go to Top of Page

bambuko
Falchetto

United Kingdom
140 Posts

Posted - 13/08/2018 :  14:11:21  Show Profile  Reply with Quote
The changes that have caused the problem seem to be the baud rate between GPS and Bluefly.
V11 used 9600 and V12 used (a default for the new GPS+Glonass) 115200.
BTW (for clarity) - baud rate between GPS and Bluefly (i.e. uart1BRG) seem to be the source of problem, not the rate between Bluefly and LK (i.e. uart2BRG).
See:
http://blueflyvario.blogspot.com/2018/07/blueflyvariottlgpsv12-released.html
This in turn caused problem for LK (but not for XCSoar apparently?) - something to do with the way LK deals (or does not deal) with a malformed GPS sentences?
Solution was to reduce the baud rate between the Bluefly and the GPS to reduce the chance of random error.
Assisted by Alistair, I was able to modify hardware settings on V12 version.
Since then I have been test running it continously (on the ground) without any crashes.
Alistair will run a vario for a few hours to try to capture what kind of errors might be being induced.
Test flying to follow.


Edited by - bambuko on 13/08/2018 14:30:17
Go to Top of Page

AlphaLima
Moderator

Germany
1544 Posts

Posted - 13/08/2018 :  14:40:44  Show Profile  Visit AlphaLima's Homepage  Reply with Quote
However, LK should not crash with corrupte data, did you have NMEA Checksum enabled?
Can you make a NMEA log of the "corrupt" data stream?
and what was the buadrate in LK? downmixing from high to low baudrates is allways a problem and needs enough holdoff to be implemented.

BTW.: 115200Bd for a GPS source does not really make sense in my eyes.

Edited by - AlphaLima on 13/08/2018 14:53:18
Go to Top of Page

bambuko
Falchetto

United Kingdom
140 Posts

Posted - 13/08/2018 :  15:28:48  Show Profile  Reply with Quote
quote:
...However, LK should not crash with corrupte data,...

Agree, however, as I have said two different models of Kobo kept crashing

quote:
... did you have NMEA Checksum enabled?...

Can't see this option in Device config?

I have seen it before... but not on this version of LK?

later edit (from v6.0 manual):
NMEA checksum previously configurable in Com port setup is now part of engineering menu, and by default it is enabled. It is a very bad idea to disable NMEA checksum while using a real RS232 serial line, which is not granting data integrity during transmission.


so the answer to your question is yes - NMEA checksum is enabled


quote:
...Can you make a NMEA log of the "corrupt" data stream?...

Have a copy of NMEA log from one of the runs that ended up with white screen event.
Happy to email it to you, although Alistair couldn't see anything obvious when I shared it with him?

quote:
...and what was the buadrate in LK? downmixing from high to low baudrates is allways a problem and needs enough holdoff to be implemented...

Originally it was 115200 between GPS and Bluefly and 115200 between Bluefly and LK, so I am guessing this was not downmixing?

Now it is 9600 between GPS and Bluefly and 115200 between Bluefly and LK

quote:
...BTW.: 115200Bd for a GPS source does not really make sense in my eyes...

I am not an expert, just a user being assisted by experts...


Edited by - bambuko on 13/08/2018 19:17:19
Go to Top of Page

AlphaLima
Moderator

Germany
1544 Posts

Posted - 13/08/2018 :  20:30:53  Show Profile  Visit AlphaLima's Homepage  Reply with Quote
Please send the NMEA Log to AlphaLima (at) lk8000 (dot) it
Go to Top of Page

bambuko
Falchetto

United Kingdom
140 Posts

Posted - 13/08/2018 :  20:42:32  Show Profile  Reply with Quote
Posted
Go to Top of Page

Aleksandar
Falchetto

Yugoslavia
147 Posts

Posted - 21/08/2018 :  16:18:01  Show Profile  Visit Aleksandar's Homepage  Reply with Quote
Any new information, is it V12 work stabile?

Best regards,
Aleksandar Cirkovic
Go to Top of Page

bambuko
Falchetto

United Kingdom
140 Posts

Posted - 21/08/2018 :  18:38:36  Show Profile  Reply with Quote
It is as far as I am concerned - has been working OK, since the changes described here.
Didn't get any comments from AlphaLima yet, so don't know if there is anything obvious in LK8000 NMEA log?
I will be using it (as soon as the weather gets better) and obviously will report if there are any problems...
Go to Top of Page

AlphaLima
Moderator

Germany
1544 Posts

Posted - 22/08/2018 :  12:19:37  Show Profile  Visit AlphaLima's Homepage  Reply with Quote
You wrote you sent the wrong file, but did not send the correct one, so far!
Go to Top of Page

bambuko
Falchetto

United Kingdom
140 Posts

Posted - 22/08/2018 :  16:58:31  Show Profile  Reply with Quote
quote:
Originally posted by AlphaLima

You wrote you sent the wrong file...


Sounds like we misunderstood each other...
After sending you the log,I have noticed, that LK actually produces two logs one dated 2080 and another dated 2018.
For example (from the latest test I have run):
NMEA_2018-08-13-19-02-55.txt
and
NMEA_2080-01-05-23-59-52.txt

I don't know or understand the significance of having two logs? rather than just one (perhaps you can explain?)

I only have one log (i.e. 2080) and that is the one I have sent to you.
I didn't say this was the wrong one, just that it was one of the two.

The best way to do this job would be to have spare BlueFly vario to experiment with, but the one I have, is soldered to the kobo and in the glider and now works fine (with different internal baud rate), so clearly not keen to take it all apart again.

I have spare kobo to play with, but no spare Bluefly vario, so I guess, unless you find anything interesting in the log I have sent you, or until we have the hardware to play with, the question will remain un-resolved....
Go to Top of Page

AlphaLima
Moderator

Germany
1544 Posts

Posted - 22/08/2018 :  19:17:12  Show Profile  Visit AlphaLima's Homepage  Reply with Quote
So you created a log on 1st of May ???
Looks like your date setting was wrong.
Go to Top of Page

bambuko
Falchetto

United Kingdom
140 Posts

Posted - 22/08/2018 :  20:03:47  Show Profile  Reply with Quote
No, I didn't create the log in May.
The only testing I have done was in August.
These two log files were created in one run on the August 13th (i.e. LK creates two logs, one with correct date and another with the weird 2080 date).

I still don't understand why?
Unless it is something to do with it being GPS/Glonass, rather than (as previously) only GPS???
Don't know?

quote:
...Looks like your date setting was wrong...

are you saying that I did something wrong?
or are you saying that something else is wrong?
date is set by gps?
Go to Top of Page

AlphaLima
Moderator

Germany
1544 Posts

Posted - 25/08/2018 :  23:40:16  Show Profile  Visit AlphaLima's Homepage  Reply with Quote
I'm not saying you did something wrong.
All I can say the the logfile has a timestamp of a far future.
This was probably caused by a wrong/corrupt NMEA stream.
Yes the date is set by the GPS. Did you have a GPS fix?

Go to Top of Page

bambuko
Falchetto

United Kingdom
140 Posts

Posted - 26/08/2018 :  10:20:27  Show Profile  Reply with Quote
Yes, I did have GPS fix.

I can experiment with a GPS module and kobo.
I have VK2828U7G5LF (GPS/GLONASS) which afaik is similar to the one used on BlueFly vario V12.
Let's see whether I get similar logs and date problems?
Go to Top of Page
Page: of 2 Previous Topic Topic Next Topic  
Next Page
 New Topic  Reply to Topic
Jump To:
PostFrontal Forum © PostFrontal - La community del Volo a Vela Go To Top Of Page
This page was generated in 0.18 seconds. Snitz Forums 2000

Suggerimenti dello staff:

               

Note legali (italiano):
Privacy Policy Cookie Policy
Legal notice (english):
Privacy Policy Cookie Policy

PostFrontal S.A.S. di Giuliano Golfieri & c. - P.I. 05264240960