Sorry,
What with timezones and busy today, I have not be able to test.
Do you want test14 tested, or wait for 15?
Cheers,
Yak.
Servo's go crazy when saving unit
Thank you very much!
With this information it seems that the problem is in very different part so I know nearly exactly what is wrong.
Data are processed as valid even they are invalid, there is some coincidence so it seems as it is OK for the unit.
It is very good it is fixable after the event, I thought it is'nt and memory is corrupted somewhere, but thankfully it is not.
Simply data stream is shifted somehow, so there are rather random values in the output.
In the latest test15 there is little better mechanism for checking frames.
http://spirit-system.com/dl/fw/test/spi ... test15.bin
With this information it seems that the problem is in very different part so I know nearly exactly what is wrong.
Data are processed as valid even they are invalid, there is some coincidence so it seems as it is OK for the unit.
It is very good it is fixable after the event, I thought it is'nt and memory is corrupted somewhere, but thankfully it is not.
Simply data stream is shifted somehow, so there are rather random values in the output.
In the latest test15 there is little better mechanism for checking frames.
http://spirit-system.com/dl/fw/test/spi ... test15.bin
Spirit System developer
Looks like you might be onto a winner.
I have installed test15 and have repeated the previous tests.
As previously I:
Powered on TX & Heli
Updated firmware to test15
Power cycled Heli.
Checked the sticks in the diag screen
Powered off TX
Checked failsafe values (correct)
Clicked on log.
Tail did not glitch
Powered on TX, values went back to TX sending "normal" & Heli is responding as expected.
Powered off TX
Clicked on log.
Tail did not glitch
Powered off Heli
Powered on Heli
Restarted software
Checked the sticks in the diag screen
Powered off TX
Checked failsafe values (correct)
Clicked on log.
Tail did not glitch
Powered on TX, values went back to TX sending "normal"
Powered off TX
Clicked on log.
Tail did not glitch
I have done this about 10-15 times, all were uneventful & turned off/on the TX about 40 times.
While the TX was off, the tail/swash was responding to movements of the heli.
After the TX was turned back on, heli was responding as expected.
There was one exception where nothing was responding, but it may have been me turning the TX off/on too quick.. A power cycle brought it back.
Check with JMann, with his save testing, but test15 is looking good!
Cheers,
Yak.
I have installed test15 and have repeated the previous tests.
As previously I:
Powered on TX & Heli
Updated firmware to test15
Power cycled Heli.
Checked the sticks in the diag screen
Powered off TX
Checked failsafe values (correct)
Clicked on log.
Tail did not glitch
Powered on TX, values went back to TX sending "normal" & Heli is responding as expected.
Powered off TX
Clicked on log.
Tail did not glitch
Powered off Heli
Powered on Heli
Restarted software
Checked the sticks in the diag screen
Powered off TX
Checked failsafe values (correct)
Clicked on log.
Tail did not glitch
Powered on TX, values went back to TX sending "normal"
Powered off TX
Clicked on log.
Tail did not glitch
I have done this about 10-15 times, all were uneventful & turned off/on the TX about 40 times.
While the TX was off, the tail/swash was responding to movements of the heli.
After the TX was turned back on, heli was responding as expected.
There was one exception where nothing was responding, but it may have been me turning the TX off/on too quick.. A power cycle brought it back.
Check with JMann, with his save testing, but test15 is looking good!
Cheers,
Yak.
Awesome, thank you very much again!
If mentioned "hang" can be reproduced too, I guess I can fix it too.
Now I know where the problem was exactly
If mentioned "hang" can be reproduced too, I guess I can fix it too.
Now I know where the problem was exactly
Spirit System developer
It happened once, and after about 10 power cycles & 20 - 30 tx power on/offs.ZeXx86 wrote:Awesome, thank you very much again!
If mentioned "hang" can be reproduced too, I guess I can fix it too.
Now I know where the problem was exactly
I was doing it fairly quickly, so may have been my fault not waiting for the tx to be off completely before trying to turn it on again.
My TX may not have reconnected with the RX properly.
I was not able to repeat it, but was going more slowly than before.
Glad to hear you might have got the tail of it.
The fun of troubleshooting is that you do not always know what the problem is until it is fixed.
The skill is being able to predict the fix..
Will you be cleaning test15 up, & releasing an interim release or should I leave test15 installed?
Let me know if I can help anymore!
Cheers,
Yak.
P.S.
I have mentioned this elsewhere, but thanks for you help with the stablization settings.
10 Stick priority /5 Direction Control Rate helped alot!
Last edited by Yak on Sun 09. Nov 2014 12:27:24, edited 1 time in total.
Great, I am very glad it was sorted out.
Here is probably final version (we have to test it for a while before release).
http://spirit-system.com/dl/fw/test/spi ... test16.bin
There is little modification so it will be good to test it too.
Here is probably final version (we have to test it for a while before release).
http://spirit-system.com/dl/fw/test/spi ... test16.bin
There is little modification so it will be good to test it too.
Spirit System developer
Ok,
Have finished testing with test16 using the same methodology as before (log files)
Power cycled heli about 10 times, TX about 20-25 times.
No issues, everything performed as expected.
No repeat of the previous "Hang" which may have been my fault.
Confirm with JMann, (saves) but it is looking good if the cause was common to both it's looking good from here.
Time for some zzz (11:00pm local)
Cheers,
Yak.
Have finished testing with test16 using the same methodology as before (log files)
Power cycled heli about 10 times, TX about 20-25 times.
No issues, everything performed as expected.
No repeat of the previous "Hang" which may have been my fault.
Confirm with JMann, (saves) but it is looking good if the cause was common to both it's looking good from here.
Time for some zzz (11:00pm local)
Cheers,
Yak.
Perfect, good night.
Here it is time for lunch
Here it is time for lunch
Spirit System developer