About this post: viewtopic.php?f=21&t=6504
It is not a real bug or error but it is a problem by design.
The connection with the SLI handling software will be terminated after 5 sec of inactivity.
This problem has been addressed a long time ago.
The problem is not only present with SLI's but also with I2C connections.
Very logic because the I2C module is also a kind of SLI.
The 5 sec is also visible in the log screen on the TXT (Putty ssh connection with screen -f)
There has been already published also a workaround for it, see:
viewtopic.php?f=8&t=6176&p=45797&hilit=SLI+5sec#p45797
It is also addressed in on of the examples:
https://github.com/fischertechnik/txt_d ... /README.md
Put a stayAwake function in a SLI and call this every 3-5 sec.
Don't put the wait time in the SLI function itself because the function is blocking.
For the function of the Lock see: https://github.com/fischertechnik/txt_d ... e-property
If there is no next SLI element to lock on, the executing can be blocked.
==============================================================================
Zu diesem Beitrag: viewtopic.php?f=21&t=6504
Es ist kein wirklicher Fehler oder Irrtum, sondern ein Problem.
Die Verbindung mit der SLI-Handhabungssoftware wird nach 5 Sekunden Inaktivität beendet.
Dieses Problem wurde vor langer Zeit behoben.
Das Problem tritt nicht nur bei SLIs auf, sondern auch bei I2C-Verbindungen.
Sehr logisch, da das I2C-Modul auch eine Art SLI ist.
Die 5 Sekunden sind auch im Protokollbildschirm des TXT sichtbar (Putty ssh-Verbindung mit Bildschirm -f).
Es wurde bereits eine Problemumgehung dafür veröffentlicht, siehe:
viewtopic.php?f=8&t=6176&p=45797&hilit=SLI+5sec#p45797
Fügen Sie eine StayAwake-Funktion in einen SLI ein und rufen Sie diese alle 3-5 Sekunden auf.
Geben Sie die Wartezeit nicht in die SLI-Funktion selbst ein, da die Funktion blockiert.
About the 5sec I2C/SLI problem in "Kontakt mit fischertechnik"
Forumsregeln
Bitte beachte die Forumsregeln!
Bitte beachte die Forumsregeln!
Re: About the 5sec I2C/SLI problem in "Kontakt mit fischertechnik"
vleeuwen hat geschrieben: ↑22 Dez 2020, 19:46About this post: viewtopic.php?f=21&t=6504
It is not a real bug or error but it is a problem by design.
What I found in my motor pwm test is the following:
If you take 1024 x the TXT pwm period you get exactly that trigger the 5 sec problem. It is remarkable that it would correspond to a 10 bit software or hardware limit. I measure a pwm period of 4785usec. I suspect that this is a link to the 5sec error.
Frans.
Re: About the 5sec I2C/SLI problem in "Kontakt mit fischertechnik"
@fotoopa
If you take a look into the the log on the TXT (putty, screen -f) you will see that it is a simple time out in case of in-activity in the I2C/SLI manager which is running on the primary processor (ARM Cortex A8 ).
The motor/IO control software is running on the secondary processor (ARM Cortex M3).
If you take a look into the the log on the TXT (putty, screen -f) you will see that it is a simple time out in case of in-activity in the I2C/SLI manager which is running on the primary processor (ARM Cortex A8 ).
The motor/IO control software is running on the secondary processor (ARM Cortex M3).