oXs vario 2.5 issue

Development & General Chat for the superb openxvario project.

Moderator: rainer

User avatar
kalle123
Posts: 905
Joined: Sat Mar 29, 2014 10:59 am
Country: -
Location: Moenchengladbach

oXs vario 2.5 issue

Post by kalle123 »

Hi all.

Yesterday evening was sitting at my desk and assembling another vario. (arduino pro mini + GY-63). Nothing special so.
Configuration with oXs conf 2.1, software oXs 2.5. TX Taranis, RX X6R (1.7., not 1.8 and opentx 2.0.8).

ALT and VERT SPEED both 0 at no movement! RESET on the pro mini and values moved :shock:

So, took the vario apart and tested the GY-63 separate with uno and MS5611 sketch. OK.

Put oXS firmware on that combo and ... same! No value movement. RESET and its fine :?

Changed the GY-63 to a DROTEK MS5611 shield. ... same! No value movement. RESET and its fine :?

Had a bad night thinking what could be the cause but was going to put things aside.

This morning in GERMAN forum http://fpv-community.de/showthread.php? ... sor/page20

I did not trust my eyes. Another RESET case!!!

Therefore I am posting here now.

Attached debug screen and config.h

br KH
Attachments
oXs_config.h.zip
(1.96 KiB) Downloaded 223 times
Bildschirmfoto55.jpeg

mstrens
Posts: 1435
Joined: Fri Dec 27, 2013 7:49 pm
Country: -

Re: oXs vario 2.5 issue

Post by mstrens »

In your debug screen, you can see a line Sport protocol = 0.
This means that oXs considers that the receiver is a D serie (and not a X serie). Therefore it activates the Hub protocol instead of the Sport protocol.

In order to detect the kind of protocol, oXs checks if the voltage level changes on the pin connected to Rx (Hub/sport).
Normally, a X serie rx send a polling code every 12 ms and oXs wait for 20 ms for a change of voltage.

So the issue could have 2 reasons:
- or the Rx is not yet 100% running ( does not yet send a pooling code on the Sport bus) when oXs checks the change of voltage. In this case we could delay the check performed by oXs
- or the Rx is not sending anymore a pooling code every 12 ms. In this case we could increase the delay of 20 ms.

If the issue happens at each startup, I think that the first reason is the rigth one.

Could you test following change:
in file oXs_out_Frsky.cpp, you have those lines

Code: Select all

    PCIFR = (1<<PCIF2) ;	// clear pending interrupt

// to see if SPORT is active, we have to wait at least 12 msec and check bit PCIF2 from PCIFR; if bit is set, it is SPORT
    delay(20) ;

You could insert a line "delay(3000) ; " before like this

Code: Select all

     delay(3000) ;    
     PCIFR = (1<<PCIF2) ;	// clear pending interrupt

// to see if SPORT is active, we have to wait at least 12 msec and check bit PCIF2 from PCIFR; if bit is set, it is SPORT
    delay(20) ;

3000 means that the set up of oXs will be delayed by 3000 millis sec (= 3 sec).

If this does not work, you can increase the delay but it make no sense to go above 10 sec.

If this still does not work, it could be that the second reason is the rigth one. In this case, the issue should be solved if you increase the existing 20 msec delay. So you could try replacing the line delay(20) ; by delay(100) ;

Perhaps, that this issue is the result of a change in the Frsky firmware on Rx level. Do you use a new Rx (a 2015 model) or did you upgrade the firmare to be EU compliant?
User avatar
kalle123
Posts: 905
Joined: Sat Mar 29, 2014 10:59 am
Country: -
Location: Moenchengladbach

Re: oXs vario 2.5 issue

Post by kalle123 »

Hi mstrens.

Helpful as ever. Will check and report back asap.

Did post a link in German forum fpv-community to your advises.

br KH
User avatar
kalle123
Posts: 905
Joined: Sat Mar 29, 2014 10:59 am
Country: -
Location: Moenchengladbach

Re: oXs vario 2.5 issue

Post by kalle123 »

mstrens wrote: You could insert a line "delay(3000) ; " before like this

Code: Select all

     delay(3000) ;    
     PCIFR = (1<<PCIF2) ;	// clear pending interrupt

// to see if SPORT is active, we have to wait at least 12 msec and check bit PCIF2 from PCIFR; if bit is set, it is SPORT
    delay(20) ;

That did it!! With both of us here in Germany.

I am still on pre EU conform firmware. As I learned from the other case here, a delay of 500 works fine, so maybe in a future version of oXs that delay should be added in file oXs_out_Frsky.cpp. But still curious, why that happened with that arduino - sensor combination ....

br KH
User avatar
kalle123
Posts: 905
Joined: Sat Mar 29, 2014 10:59 am
Country: -
Location: Moenchengladbach

Re: oXs vario 2.5 issue

Post by kalle123 »

Think, it is not yet over :?

Did the test with delay(3000) with a arduino uno + GY-63 combo. That is OK.
DSCF3780.JPG
Then substituted the uno with the pro mini*, I used on sunday.
DSCF3786.JPG
Frozen zeros and after RESET values ALT and VSpd moving, but ALT!!??
DSCF3785.JPG
Same sketch on uno and pro mini. At the moment don't have a spare pro mini, but supply is on the way ...

* pro mini is of new design/layout with additional A6/A7 ports.

br KH

mstrens
Posts: 1435
Joined: Fri Dec 27, 2013 7:49 pm
Country: -

Re: oXs vario 2.5 issue

Post by mstrens »

@Kalle123,
It is very strange.
I do not think that new design arduino has an impact.

First of all, I presume that you apply the power supply exactly at the same time on Rx and on Arduino and that Arduino is already connected to Rx Sport bus.
Please note that if you apply power supply first on Arduino and later on on Rx, then it is "normal" that Arduino does not recognise that it is connected to an Sport receiver.
The check on a change on the level voltage on SPORT is performed only when Arduino start (in fact after that Arduino runs the bootloader).

As further test, you could try :
- to increase the delay 3000
- to check if the delay of the polling is wel the expected one.
Normally the Rx send a polling on SPORT bus every 12ms.
In order order to know if this is really the case, it would be good to put an oscilloscope on the SPORT signal generated by the Rx when there is no oXs (and no other sensor) connected.
You should see some blocks of data (when Rx send a polling byte on the Sport bus). The idea is to measure the enlapsed time between the start of 2 blocks.
Do you have an oscilloscope (or a logical analyser) to perform this test?
Otherwise, you can try changing the other delay. I refer here to the line "delay(20)".
I suggest that you change the value "20" by a value "40" or even "100".

I have no explanation for the strange Alt that you get.
If you perform the same test again, do you get the same strange Alt?
it could be that there is an error in oXs.
Which vesion of IDE are you using for compiling? Is it 1.0.5? if not could you try with this version?
User avatar
kalle123
Posts: 905
Joined: Sat Mar 29, 2014 10:59 am
Country: -
Location: Moenchengladbach

Re: oXs vario 2.5 issue

Post by kalle123 »

Hi mstrens. Please give me some time to do the testing, you recommended.

My pc ist non op, mobo faulty and replacement is coming. Have to post this on an old machine with an LXDE desktop. :(

I have a rigol oszi and a saleae clone here. So that is not the problem. But I have to get my pc running ...
.
.
.
.
First of all, I presume that you apply the power supply exactly at the same time on Rx and on Arduino and that Arduino is already connected to Rx Sport bus. ---- YES, fixed connection like that.

- to increase the delay 3000
---- it is set to 3000, did not play with that value

Which vesion of IDE are you using for compiling? Is it 1.0.5? ---- I am using 1.0.6

br KH
User avatar
jhsa
Posts: 19480
Joined: Tue Dec 27, 2011 5:13 pm
Country: Germany

Re: oXs vario 2.5 issue

Post by jhsa »

kalle123 wrote:
My pc ist non op, mobo faulty and replacement is coming. Have to post this on an old machine with an LXDE desktop. :(
I know the feeling. Had it last year. With fireworks and all that. :(
But it is a good excuse to get a new much faster PC :mrgreen:

João
My er9x/Ersky9x/eepskye Video Tutorials
https://www.youtube.com/playlist?list=PL5uJhoD7sAKidZmkhMpYpp_qcuIqJXhb9

Donate to Er9x/Ersky9x:
https://www.paypal.com/cgi-bin/webscr?cmd=_s-xclick&hosted_button_id=YHX43JR3J7XGW
mstrens
Posts: 1435
Joined: Fri Dec 27, 2013 7:49 pm
Country: -

Re: oXs vario 2.5 issue

Post by mstrens »

@kalle123,
Take time to do the testing.

If you want, you could do some tests with the version in attachment.
This is the latest version I just made.
Changes are:
- oXs can be used as light controller (see description file)
- oXs allows to add a parameter in order to try to reduce drift of altitude generated by change of temperature
- oXs have new parameters in order to force SPORT protocol or HUB protocol to be used; automatic detection is still possible but I do not know (yet) why it does not work in all cases (I added the line for "delay(500)" but did not change the "delay(20)")
Attachments
openXsensor_2015_03_10.rar
(82.13 KiB) Downloaded 226 times
User avatar
kalle123
Posts: 905
Joined: Sat Mar 29, 2014 10:59 am
Country: -
Location: Moenchengladbach

Re: oXs vario 2.5 issue

Post by kalle123 »

@João. You know that feeling :D

But after going through INTEL SX386, DX2/66, P75 .... it is now the first time, I switched to AMD and an ASUS board with UEFI and secure boot. And ordered a 8core cpu with 125 watt of power loss. Either I'll need a air conditioning here and/or wear ear protection. I ordered the upgrade kit on Sunday evening and on Monday late at night I found the cause of all my trouble. A BAD CABLE. But now the kit is on the way and I'll go through that ....

@mstrens. Could do a short test. Connected my saleae clone, started recording on the s_port channel of that X6R and then put power on the rx. Please see the three screenshots attached here ...
Hope this helps a bit. 1.pic power ON, 2. and 3. pic first two signals on s_port zoomed

Going to do a test with your newest software tomorrow. But now I have to meditate, how to get this AMD thing into my good old INTEL machine................... :mrgreen:

br KH
Bildschirmfoto7.jpeg
Bildschirmfoto8.jpeg
Bildschirmfoto9.jpeg
mstrens
Posts: 1435
Joined: Fri Dec 27, 2013 7:49 pm
Country: -

Re: oXs vario 2.5 issue

Post by mstrens »

@kalle123,
Thanks for the test with X6R.

It shows that polling occurs as expected every 12 msec.
So the line with "delay(20)" should be OK because there should be at least one level change within 20 msec.

Perhaps good to make a test increasing the value in the line delay(3000) (e.g. trying 5000) and activating the debug option in order to see on the PC terminal if sport is activated or not.

Take time to make new tests; this is for sure not high priority. :)
User avatar
kalle123
Posts: 905
Joined: Sat Mar 29, 2014 10:59 am
Country: -
Location: Moenchengladbach

Re: oXs vario 2.5 issue

Post by kalle123 »

@mstrens.
Hi, just thought about it.
Would it be helpful to record with saleae the data between arduino and X6R s_port at power on (Taranis TX ON)

- oXs ver.2.5 delay(3000) or (5000) with arduino uno + GY-63

- oXs ver.2.5 delay(3000) or (5000) with arduino pro mini + GY-63 ?

And attach the recordings here? Can you visualize those on your pc?

br KH
mstrens
Posts: 1435
Joined: Fri Dec 27, 2013 7:49 pm
Country: -

Re: oXs vario 2.5 issue

Post by mstrens »

@kalle123
It could be very useful.
I have a saleae clone too. So I expect I could read your log.
Ideally, it would require to capture not only the signal on SPORT but also some signal from Arduino in order to know how far is the Arduino start up process when Rx polling occurs.
In order to do so, I should add some code to oXs.

I am busy the whole day. So I could to this change only this evening.
User avatar
kalle123
Posts: 905
Joined: Sat Mar 29, 2014 10:59 am
Country: -
Location: Moenchengladbach

Re: oXs vario 2.5 issue

Post by kalle123 »

@mstrens.

OK. No hurry. And please give your instructions for monitoring.

br KH
mstrens
Posts: 1435
Joined: Fri Dec 27, 2013 7:49 pm
Country: -

Re: oXs vario 2.5 issue

Post by mstrens »

@kalle123,

I came back home later than expected.
I will not be able to code a version of oXS with extra synchronisation signal before to morrow.
I will give instructions when the version is ready.
mstrens
Posts: 1435
Joined: Fri Dec 27, 2013 7:49 pm
Country: -

Re: oXs vario 2.5 issue

Post by mstrens »

@kalle123,

Here the new version that should allow a better analysing of the signals.

In this version, there are 2 new parameters in file oxs_out_frsky.h
#define DEBUG_SETUP_PIN 5 // allows to send a pulse on an output pin during the enlapsed time that Arduino runs the set up
#define DEBUG_SPORT_PIN 6 // allows to send a pulse on an output pin during the enlapsed time that Arduino checks for SPORT

You could change the value 5 and 6 if you want to use other pins.

The idea is the folowing:
When Arduino starts, it runs first an init() function and then enter the setup() function before entering in a loop().
The setup function runs only once at the beginning. This function calls many other functions, one containing the code that will detect SPORT protocol
I have added some code in order
- to generate a level HIGH on the pin defined by DEBUG_SETUP_PIN (only while Arduino is running the function setup)
- to generate a level HIGH on the pin defined by DEBUG_SPORT_PIN (only while Arduino is running the code to detect SPORT protocol, so during 20 msec).

So, you should connect you logic analyser to arduino with 4 leads.
- Gnd
- SPORT signal (like you already did)
- pin defined by DEBUG_SETUP_PIN
- pin defined by DEBUG_SPORT_PIN
Triggering the logic analyser could be done on a rising pulse on DEBUG_SETUP_PIN
Sample rate could be 1mhz or more.
Recording time could be about 3 sec.
In this test, Arduino should be connected to Rx (via the SPORT)

You can start recording before powering on.
Normally recording should start automatically when Arduino enter the setup function (so a few seconds after power on because it first runs some bootloader code).
If this does not work as expected, perhaps that a resistor (any value between 1k and 10k should be ok) should be added between DEBUG_SETUP_PIN and GND (because before oXs start up, the pin is in high impedance and so the level can fluctuate).
At the end of the test, you should check if the Tx receives Alt & Vspeed (so checking if SPORT protocol has been recognised or not).
I some way, I hope that it will not be recognised otherwise the test does not bring additional because the issue is not present.
This test can be done with the option DEBUG activated or not activated. It should normally not make a difference except that setup function should take a longer time with DEBUG activated.

Analysing the log, we should normally see a pulse on DEBUG_SETUP_PIN and somewhere (nearly the end of this pulse), there should be another pulse on the DEBUG_SPORT_PIN (for about 20msec).
If SPORT is not recognised (so if the issue occurs), I expect that there should be no level change on the SPORT bus during the 20 msec of the "DEBUG_SPORT" pulse.
It should be good then to look further in the log when levels change on the SPORT bus.
This is the most logical explanation I have for the issue.

Note: if there are some level changes on SPORT bus during the 20 msec then the bug should be in oXs.
Attachments
openXsensor_2015_03_12.rar
(82.51 KiB) Downloaded 208 times
User avatar
kalle123
Posts: 905
Joined: Sat Mar 29, 2014 10:59 am
Country: -
Location: Moenchengladbach

Re: oXs vario 2.5 issue

Post by kalle123 »

@mstrens.

You send me a oXs ver. 3.1

I did one adjustment in oXs_config.h //Hope, this was correct!?
File is attached here.

// ***** 1.2 - FrSky protocol and device ID (required when Sport protocol is used) *****
#define FRSKY_TYPE_SPORT <-----
//#define FRSKY_TYPE_HUB

Compiled with arduino IDE 1.0.6 and flashed on both uno and pro-mini (both used before in test)

No issue!

Saleae on 2 MHz and 5 sec. Monitoring 4, 5 and 6

1.Taranis ON.

2. Power ON RX+arduino

3. REC

BOTH combos working directly when moving GY-63 :o

Leave the test like it is at the moment, so no problem to do a further test.

br KH
oXs_config.h.tar.gz
(2.49 KiB) Downloaded 283 times
Bildschirmfoto uno.jpeg
2 MHz, 10 M Samples uno-working.logicdata.tar.gz
(28.38 KiB) Downloaded 305 times
Bildschirmfoto pro mini.jpeg
mstrens
Posts: 1435
Joined: Fri Dec 27, 2013 7:49 pm
Country: -

Re: oXs vario 2.5 issue

Post by mstrens »

@kal123,
Thanks for the test.
Sorry if I did not specify to keep the config file as originally made.
In fact, if you uncomment the line #define FRSKY_TYPE_SPORT, then it means that oXs does not have to check for SPORT protocol (and so it selects it directly).

Best would be to do the test again but having this line as comment (so adding "//")

I also see that level on pins 5 and 6 is not exactly as expected. I think that when Vcc is on but pin are still in high pimpedance, then they are logged as HIGH.
Could you at least on pin 5 add a resistor to Gnd in order to force a low level when the pin is still in high impedance.

Thanks.
mstrens
Posts: 1435
Joined: Fri Dec 27, 2013 7:49 pm
Country: -

Re: oXs vario 2.5 issue

Post by mstrens »

@kal123,
looking further to your screen shot, there is some thing strange:
I presume that Pin 6 becomes definitively low when setup function starts (when there is no resistor to gnd).
It seems that with Pro mini, SPORT polling exists when setup function starts.
This is not the case with Uno.

So, I would expect that automatic SPORT detection would be OK with the pro mini and not with the uno.
Still, this is just the opposite of what you have noticed in your previous tests (one or 2 days ago).
So I do not understand but let wait for the new tests with a resitor.
Or perhaps did your invert the 2 logs in your post?
mstrens
Posts: 1435
Joined: Fri Dec 27, 2013 7:49 pm
Country: -

Re: oXs vario 2.5 issue

Post by mstrens »

@kal123,
looking further to your screen shot, there is some thing strange:
I presume that Pin 6 becomes definitively low when setup function starts (when there is no resistor to gnd).
It seems that with Pro mini, SPORT polling exists when setup function starts.
This is not the case with Uno.

So, I would expect that automatic SPORT detection would be OK with the pro mini and not with the uno.
Still, this is just the opposite of what you have noticed in your previous tests (one or 2 days ago).
So I do not understand but let wait for the new tests with a resitor.
Or perhaps did your invert the 2 logs in your post?
User avatar
kalle123
Posts: 905
Joined: Sat Mar 29, 2014 10:59 am
Country: -
Location: Moenchengladbach

Re: oXs vario 2.5 issue

Post by kalle123 »

@mstrens.

So next steps on my side.

Comment this line OUT //#define FRSKY_TYPE_SPORT

Put an resistor between pin 5 and GND (any value between 1k and 10k)

... and do the tests again!

Come back with results asap

Will double check "Or perhaps did your invert the 2 logs in your post?"

br KH
User avatar
kalle123
Posts: 905
Joined: Sat Mar 29, 2014 10:59 am
Country: -
Location: Moenchengladbach

Re: oXs vario 2.5 issue

Post by kalle123 »

@mstrens.

Next tests.

Uno + GY-63 and pro mini + GY-63 working!

//#define FRSKY_TYPE_SPORT
oXs_config.h.tar.gz
(2.49 KiB) Downloaded 241 times
3.3k between GND and pin 5
Bildschirmfoto uno1.jpeg
2 MHz, 10 M Samples uno-working1.logicdata.tar.gz
(26.51 KiB) Downloaded 278 times
Bildschirmfoto pro mini1.jpeg
cu KH
mstrens
Posts: 1435
Joined: Fri Dec 27, 2013 7:49 pm
Country: -

Re: oXs vario 2.5 issue

Post by mstrens »

@kal123,
Thanks for the tests.
So, we are a little unlucky because oXs now works with both Arduino.
Still I see on your test with the uno that the polling starts only a few millisecond before oXs check the SPORT bus. So, if Arduino should start just a little slowlier, the SPORT would not have been detected.
This means that the check on the type of protocol could be delayed; so the line 96 (in oxs_out_frsky.cpp) with delay(500) could be changed in delay(1500) in order to be more secure.

For the test with the pro mini, it is not possible to know if it is the same because the log starts recording later.
In fact, I think that you set up your logic analyser in order to start recording when there has been an rising edge on EACH of the 3 pins and so it starts only when pin 6 goes up at the end of the set up.
In order to know when the polling starts compared to the start of the set up, I think that the log should be triggered only by one rising edge (or on SPORT signal or on pin 5).
note: I do not know why the test with the uno starts recording earlier.

Perhaps good to repeat those tests a few time to see if the SPORT polling occurs always at the same time compared to the start of setup function (pulse on pin 5)
User avatar
kalle123
Posts: 905
Joined: Sat Mar 29, 2014 10:59 am
Country: -
Location: Moenchengladbach

Re: oXs vario 2.5 issue

Post by kalle123 »

@mstrens, I'll repeat the tests ;)

I started the recording manually, but I will put a trigger on all 3 channels. There is the possibility to put in a pre trigger ....
shot.png
(You are talking here to a retired mechanical engineer, who does not know anything about electronics :mrgreen: I got that seleae clone two weeks ago and that is the first time, I am doing something with it!)

Is it enough for you to see a screenshot of start sequence or shall I attach the recordings here?

Shall I flash both arduinos with that setting before doing those tests?

This means that the check on the type of protocol could be delayed; so the line 96 (in oxs_out_frsky.cpp) with delay(500) could be changed in delay(1500) in order to be more secure.

cu KH

PS. What is with the 3.3k resistor. Leave that in or take it out?
mstrens
Posts: 1435
Joined: Fri Dec 27, 2013 7:49 pm
Country: -

Re: oXs vario 2.5 issue

Post by mstrens »

@kal123

FYI I am also a retired mechanical engineer. Still I played with electrinics and IT since quite many years. :mrgreen:

I have no deep knowledge about seleae clone. I had an older version and I just downloaded the latest version in order to be able to read your attachement.

About trigerring, I am not 100% sure how it works.
Here my understanding (but I am perhaps wrong).
When you press the start button, the analyser look at the signals but do not log immediately the data.
It starts recording when the trigger condition(s) is (are) OK.
The trigger conditions are to set up for each channel based on the icons displayed to the right of the channel name.
In order to trigger only on a rising edge on e.g. SPORT signal, the set up should show
- a rising edge for the first channel (SPORT)
- a "X" for the 2 other channels

In order to analyse the data, you can do it the way you want (screen shot and/or attach the recording). I can now read your recording. So do it the easier way for you.
If you put the screen shot, the most important is that the pulses on the 2 last channels are present and that there are some SPORT level change (left and right if they exist).

Best is to flash both arduino with the same settings (e.g. with delay(1500). Normally the results should be the same.

Please leave the resistor on pin 5 because it allows to get a nice pulse on this channel.
In fact you could even add a resistor on pin 6 to get only a nice pulse but it is not so important because normally I just look at the pulse on pin 6 when the pulse on pin 5 occurs.

Thanks for your collaboration.
User avatar
kalle123
Posts: 905
Joined: Sat Mar 29, 2014 10:59 am
Country: -
Location: Moenchengladbach

Re: oXs vario 2.5 issue

Post by kalle123 »

@mstrens. WILCO ;)

Post results asap.

br and a nice evening - KH

... and thank YOU for your work!
User avatar
kalle123
Posts: 905
Joined: Sat Mar 29, 2014 10:59 am
Country: -
Location: Moenchengladbach

Re: oXs vario 2.5 issue

Post by kalle123 »

@mstrens.

So, found time (retired people never do have enough time :D ) to do 3 subsequent test with both combos.

TX ON, saleae START, POWER on arduino + RX ON.

PIN 5 with 3.3k to GND. Delay in out_frsky set to 1500.

Please find all info in attached files.

br KH
screenshot_start.tar.gz
(292.76 KiB) Downloaded 271 times
logicdata.tar.gz
(191.13 KiB) Downloaded 231 times
oXs_config_out_frsky.tar.gz
(12.21 KiB) Downloaded 276 times
mstrens
Posts: 1435
Joined: Fri Dec 27, 2013 7:49 pm
Country: -

Re: oXs vario 2.5 issue

Post by mstrens »

@kal123,
As far as I can see, in the 6 tests, the SPORT polling is already available when oXs tries to detect SPORT protocol.
So in all cases, the data should be directly displayed on the Tx and there should be no issue.

The screenshots look very similar for the pro mini on one side and for the uno on the other side (tests are repetitive).
The only difference between pro mini and uno is that it seems that pro mini starts slowlier (probably a different bootloader).
With the pro mini, the SPORT polling is already available nearly 1 sec before oXs enters the setup function while with the Uno, the SPORT polling is not yet present when oXs enters the setup.
So the instruction delay(500) (and even safier 1500) is required for Uno but it is not really required for pro mini.

This result is strange because it is in contradiction with:
- your first post (saying that pro mini did not work); The version 2.5 used had no instruction delay(500) (or more) but this does not seem really required if I look at the tests done today.
- your post on 10 march at 6:07 saying that it was not OK with pro mini but OK with UNO. For some value of delay it would normally be the opposite. Perhaps I misunderstood this post: I thought that both arduino were running the same sketch and so it was strange that the pro mini did not work. But in fact, probably that you did not yet add the delay(3000) for the pro mini and so it was just a confirmation of the first post result.

Conclusion:
The new version is OK but we do not really know why the version 2.5 was NOK in your case (and for one other german user).
I propose to add the delay (probably with a value 1000 which seems enough in all cases) in next release of oXs because it helps in some cases.
As already coded, this release will also allow to force the protocol to be or SPORT or HUB for the user that have only one type of receiver and do not need automatic detection of protocol.

Thanks again for the tests.
Do you have any other suggestion?
User avatar
kalle123
Posts: 905
Joined: Sat Mar 29, 2014 10:59 am
Country: -
Location: Moenchengladbach

Re: oXs vario 2.5 issue

Post by kalle123 »

@mstrens

Errors do happen, but I am 99% sure, that firmware in all tests, I did, was identical.

Good to see, that the issue seems to be solved. Awaiting your next version of oXs and maybe a new version from davx for the configurator ...

br KH
User avatar
jhsa
Posts: 19480
Joined: Tue Dec 27, 2011 5:13 pm
Country: Germany

Re: oXs vario 2.5 issue

Post by jhsa »

That configurator is a veeeeery nice tool ;) :)

João
My er9x/Ersky9x/eepskye Video Tutorials
https://www.youtube.com/playlist?list=PL5uJhoD7sAKidZmkhMpYpp_qcuIqJXhb9

Donate to Er9x/Ersky9x:
https://www.paypal.com/cgi-bin/webscr?cmd=_s-xclick&hosted_button_id=YHX43JR3J7XGW

Post Reply

Return to “OpenXVario - an open source vario supported by the open source firmwares!!”