New beta version of openXsensor (=openXvario)

Development & General Chat for the superb openxvario project.

Moderator: rainer

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

Re: New beta version of openXsensor (=openXvario)

Post by mstrens »

Here a new version I made to day.
In the previous version, before calculating the total energy every 2O ms, I made a first smoothing on Altitude (every 20 ms) and a first smoothing on Airspeed (every 5 ms about). This could create some phase shift.

In this version, I calculate the total energy every 20 ms using:
- the raw altitude returned by the MS5611 every 20 ms
- a moving average presure returned by 4525DO. The moving average is calculated every 20 ms based on the 4 last for values from 4525DO (a value is returned around every 5 msec).

Then I apply a differential smoothing on this total energy (in order to get a raw dTE (delta total energy ) and finally I apply an additional smoothing on the raw dTE and the hysteresis; this logic was already the one used for the uncompensated VSpeed.

This algorithme sould avoid phase shift and need only one smoothing instead of 3.

Perhpas it will be better (it depends if both sensors have similar noise level).

Let test it.

Edit :
in this version:
- Temp1 is the rawcompensation based on the differential pressure returned by 4525DO (it is equivalent to an altitude in cm)
- Temp2 is the rawTotalEnergy (= Raw altitude + compensation) (in cm)
Attachments
openxsensor_2014_09_06_V1.rar
(67.17 KiB) Downloaded 294 times

Carbo
Posts: 467
Joined: Fri Aug 02, 2013 6:55 pm
Country: Germany
Location: Freinsheim RP

Re: New beta version of openXsensor (=openXvario)

Post by Carbo »

The artifact issue is completely gone! Perfect sound, no more inappropriate beeps, very good feeling for the air. One small issue, it is a little bit undercompensated. We had the possibility, to adjust compensation via ppm, is that still alive? If not, would you please reactivate it, this is the only thing i miss so far. TEK probe starts already trembling, it knows, the end is near.
2014_09_06.png
Stratos_E-2014-09-06.zip
(185.76 KiB) Downloaded 265 times
mstrens
Posts: 1435
Joined: Fri Dec 27, 2013 7:49 pm
Country: -

Re: New beta version of openXsensor (=openXvario)

Post by mstrens »

Carbo wrote:The artifact issue is completely gone! Perfect sound, no more inappropriate beeps, very good feeling for the air. One small issue, it is a little bit undercompensated. We had the possibility, to adjust compensation via ppm, is that still alive? If not, would you please reactivate it, this is the only thing i miss so far.
Fine.
This first version had already control on sensitivity via ppm, dynamic smoothing and hysteresis.
This first version had no compensation adjustment via ppm. I added it this new version.
It is strange that it is under compensated because it just calculated based on airspeed (or I made some mistake in the calculations)
Attachments
openxsensor_2014_09_06_V2.rar
(67.17 KiB) Downloaded 277 times
mstrens
Posts: 1435
Joined: Fri Dec 27, 2013 7:49 pm
Country: -

Re: New beta version of openXsensor (=openXvario)

Post by mstrens »

@Carbo,
In your log, I always get the Airspeed in knots.
As said this morning, it seems that there was some discreapancy with the value OXS generates.

I made a test to check it with a fix dummy value (1000) being sent as Airspeed and Temp1.
On my Tx, I get Temp1 = 1000 (ok) and Airspeed 185 which mean 185 km/h. This seems OK because 1000 (in 1/10 of knot/h) = 100 knot/h = 185,2 km/h .
But more strange: in the log I get a header whith km/h ( and not knot/h).
So it seems that my version of openTX perform the conversion in the log too.
I think I am using 2.0.5 version of openTX.
Which version are you using?
Carbo
Posts: 467
Joined: Fri Aug 02, 2013 6:55 pm
Country: Germany
Location: Freinsheim RP

Re: New beta version of openXsensor (=openXvario)

Post by Carbo »

Version 2.08 and i have knots in all logs but km/h in display. Units in TX is metric. Do you have the header in km/h and the corresponding km/h value in the data fields?

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

Re: New beta version of openXsensor (=openXvario)

Post by mstrens »

Carbo wrote:Version 2.08 and i have knots in all logs but km/h in display. Units in TX is metric. Do you have the header in km/h and the corresponding km/h value in the data fields?
Yes I have the header in km/h and the corresponding km/h value in the data fields.
Here a log from my SD card.
I did not modified the CSV file.
As said in previous post,for this log, I made a special version of OXS where I filled airspeed and Temp1 with "1000".

Perhaps there is some explanation to be found on the forum.
Note: I am using a french version (via companion)
mstrens
Posts: 1435
Joined: Fri Dec 27, 2013 7:49 pm
Country: -

Re: New beta version of openXsensor (=openXvario)

Post by mstrens »

mstrens wrote:
Carbo wrote:Version 2.08 and i have knots in all logs but km/h in display. Units in TX is metric. Do you have the header in km/h and the corresponding km/h value in the data fields?
Yes I have the header in km/h and the corresponding km/h value in the data fields.
Here a log from my SD card.
I did not modified the CSV file.
As said in previous post,for this log, I made a special version of OXS where I filled airspeed and Temp1 with "1000".

Perhaps there is some explanation to be found on the forum.
Note: I am using a french version (via companion)
HAve a look at issue #1362 on openTX.
I presume they made some changes between 2.0.5 and 2.0.8
Carbo
Posts: 467
Joined: Fri Aug 02, 2013 6:55 pm
Country: Germany
Location: Freinsheim RP

Re: New beta version of openXsensor (=openXvario)

Post by Carbo »

mstrens wrote: HAve a look at issue #1362 on openTX.
I presume they made some changes between 2.0.5 and 2.0.8
Interesting, someone developed a commercial version. I look forward eagerly to the comparison. The announced total energy compensation has not been realised, although it is such an easy thing :-)
Carbo
Posts: 467
Joined: Fri Aug 02, 2013 6:55 pm
Country: Germany
Location: Freinsheim RP

Re: New beta version of openXsensor (=openXvario)

Post by Carbo »

Had to make a small change in openxsensor.ino, to get compensation via PPM to work (mstrens is offline for a few days). A compensation of 116% sounded good while flying, in the log dTE is still a little bit undercompensated (like TEK). PPM from +60% to + 90% allows adjustment of compensation from 80-140%, ratio 30:60 (1:2), so easy to convert. Decoded PPM is transmitted back by OXS in the Temp1 field in this version.
Compensated.png
The trend of TEK and dTE are nearly congruent, the noise in both signals is the airnoise, (microthermals). In the evening with calm air i will check again.

Tried the original version once again, it worked, as foreseen - there must have happened something, when transferring to arduino. Here is a log from mstrens original version.
Stratos_E-2014-09-11.zip
(55.71 KiB) Downloaded 270 times
Edit: removed the pointless modded version
Last edited by Carbo on Thu Sep 11, 2014 5:23 pm, edited 1 time in total.
Carbo
Posts: 467
Joined: Fri Aug 02, 2013 6:55 pm
Country: Germany
Location: Freinsheim RP

Re: New beta version of openXsensor (=openXvario)

Post by Carbo »

Here the flight in calm conditions:
Stratos_E-2014-09-07.zip
(70.49 KiB) Downloaded 255 times
Here is a chart of the complete flight, good to recognize, that in fast sections the energy reduces, in the climb afterwards the energy not rises. Even the added electric energy in the climbs is clearly visible.
4all.png
Here a detail additionally with the uncompensated vario (AccelX)
Flug4.png
sfabris
Posts: 8
Joined: Sat Jan 04, 2014 12:28 am
Country: Italy
Location: Vicenza

Re: New beta version of openXsensor (=openXvario)

Post by sfabris »

Hi,

If I understand correctly, this beta code has not been checked in google code, is'nt it?
Would be possible to have the beta code on google code repository, or even better on a github repo?

Thanks,

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

Re: New beta version of openXsensor (=openXvario)

Post by mstrens »

sfabris wrote:Hi,

If I understand correctly, this beta code has not been checked in google code, is'nt it?
Would be possible to have the beta code on google code repository, or even better on a github repo?

Thanks,

Simone
Most of the code has been tested (at least to support airspeed, dTE vario and 2 baro sensors) by one user.
It seems to work quite wel.
The config file has been modified too (see the file inside the version with the explanation).
Please note that the configurator is not yet updated. So you have to edit the config.h file manually.

I also made some changes in order to support the 2 Frsky (SPORT and HUB) protocol simultanously. At power on, OXS check if it is connected to a X serie receiver. If yes, it activates automatically the SPORT protocol and otherwise the HUB protocol. So you can use the same OXS for both receivers without having to reload another firmware.
I also made some clean up of the code.

I uploaded this new version (named OpenXsensor_v2.0) on github.
You can find it on the new web site for openxsensor under downloads and tags at this link:
https://code.google.com/p/openxsensor/source/browse


This version contains quite many changes and I can't garantee that it is bug free. So if you get some issues, let me know and I will look for a solution.
sfabris
Posts: 8
Joined: Sat Jan 04, 2014 12:28 am
Country: Italy
Location: Vicenza

Re: New beta version of openXsensor (=openXvario)

Post by sfabris »

mstrens wrote: I uploaded this new version (named OpenXsensor_v2.0) on github.
You can find it on the new web site for openxsensor under downloads and tags at this link:
https://code.google.com/p/openxsensor/source/browse
Thanks a lot!

Simone
rolfz
Posts: 13
Joined: Sat Oct 04, 2014 3:08 pm
Country: Switzerland
Contact:

Re: New beta version of openXsensor (=openXvario)

Post by rolfz »

Hello
I am new in the forum, trying to make OXS work in Hub mode with Taranis!
With the standard setup, I can currently detect the following packets
BaroB, BaroA, VSPeed, ACCX, ACCZ, TEMP1, TEMP2 and FUEL

So missing Airspeed and ACCY.
Can someone help me to know hot to make this word ?
Also I let my Taranis speak out the ACCZ and I get " G xxx". Is there a way to make this m/s or similar ?

Big thanks to the creators, you do a great job with your code!!!
Rolf
Kind regards
Rolf
rolfz
Posts: 13
Joined: Sat Oct 04, 2014 3:08 pm
Country: Switzerland
Contact:

Re: New beta version of openXsensor (=openXvario)

Post by rolfz »

Has the Hub version (oxs 2.1) been tested ?
I am running it on a Arduino Mini 328 3.3v version running at 8Mhz

I can see the following being sent to the RX UART, so some fields are missing
BaroB, BaroA, Vspd, Accx, Accz, T1, T2, Fuel (so Airspeed and AccY are missing)

Of course, only the transmitted values are visible on Taranis, no Airspeed and no AccY value
Kind regards
Rolf
mstrens
Posts: 1435
Joined: Fri Dec 27, 2013 7:49 pm
Country: -

Re: New beta version of openXsensor (=openXvario)

Post by mstrens »

rolfz wrote:Has the Hub version (oxs 2.1) been tested ?
I am running it on a Arduino Mini 328 3.3v version running at 8Mhz

I can see the following being sent to the RX UART, so some fields are missing
BaroB, BaroA, Vspd, Accx, Accz, T1, T2, Fuel (so Airspeed and AccY are missing)

Of course, only the transmitted values are visible on Taranis, no Airspeed and no AccY value
I made a few tests with a D serie receiver (= Hub protocol) and it seems to works in general but not really with the config.h file provided on ggogle site.

Here an extract of the config.h file on google.
#define SETUP_DATA_TO_SEND \
DEFAULTFIELD , ALTIMETER , 1 , 1 , 0 ,\
VSpd , PPM_VSPEED , 1 , 1 ,0 , \
DEFAULTFIELD , AIR_SPEED , 1 , 1 ,0 , \
AccX , VERTICAL_SPEED , 1 , 1 ,0 , \
AccY , VERTICAL_SPEED_2 , 1 , 1 ,0 , \
AccZ , PRANDTL_DTE , 1 , 1 ,0 ,\
T1 , TEST1 , 1 , 1 ,0 , \
T2 , TEST2 , 1, 1, 0 , \
Fuel , SENSITIVITY , 1, 1, 0

This config was used by a member that used an OXS with :
2 baro sensors (MS5611) , the second one was connected to a TEK probe in order to get a pneumatically compensated vario
1 4525DO sensor that provided the airspeed and an electronically compensated vario (using also vertical speed from the first baro sensor).
This config was used to test the firmware and to compare the pneumatically compensated vario with the electronic.
Probably this config is not a normal one because I presume that it does not make a lot of sense having as well a pneumatically and an electronically compensated vario. Normally if should be the one or the other.
This member uses this config with an X serie Receiver (so using the SPORT and not the Hub protocol).

Now, some comments on your 2 issues:
When using the Hub protocol, it is "normal" that airspeed is missing. The reason is that Frsky did not assign a code for the hub protocol to this kind information because they have no airspeed sensor.
A code has been assigned for airspeed by the OpenTx developper but only for the SPORT protocol.
So in Hub protocol, sending the airspeed in the "defaultfield" has no effect.
If you have a 4525DO sensor measuring the airspeed, you could change the config in order to let OXS send the airspeed in another field (like temp1, temp2, ...).
Another solution should be to send the airspeed in the GPS speed but this requires some changes in the OXS firmware.

In the here above config, AccY is supposed to be filled by the vertical speed calculated by the second baro sensor.
Do you have a second baro sensor?
Do you change the config file in order to put the line
#define VARIO2 // set as comment if there is no second vario
as comment.
If this line is put in comment, then OXS does not try to transmit the vertical speed 2 in AccY because it can detect that it does not exist.
If the line is not in comment, then OXS should add in the frame sent to the RX a code 0x25 followed by the vertical speed (could be 0 if sensor does not exist).

Can you explain me how do you proceed in order to check the content of the frame sent to RX. This could help me looking for a bug.

Another reason that could explain why AccY is not present in the frame is the fact that in the hub protocol FrSky put a limit to the max number of data allowed in one frame.
Perhaps that we reach this limit with all the data to be sent. Still I do not think that this is the reason.
rolfz
Posts: 13
Joined: Sat Oct 04, 2014 3:08 pm
Country: Switzerland
Contact:

Re: New beta version of openXsensor (=openXvario)

Post by rolfz »

Thank you for your answer, very helpful.

To check the message sent, I use a small LogicAnalyser from IKALogic, that one is connected on the TX line of the Mini-Arduino output.
Once the module is in place, I can run the logic analyser software on the PC and see block by bloc what is sent. See picture below.
OpenXsensor_Hub.jpg
So after reading your message, and if I enable again Vario2, I effectively get the package with code 0x25, so question answered and issue solved.
Concerning the horizontal speed, I exactly did what you suggeted, I put the AirSpeed onto the AccY, and I now get the horizontal speed in replacement to the vario2 which I do not use.

Here my current setup working fine with my old D8R-II receiver, seems to work fine :D

#define SETUP_DATA_TO_SEND \
DEFAULTFIELD , ALTIMETER , 1 , 1 , 0 ,\
VSpd , PPM_VSPEED , 1 , 1 ,0 , \
AccX , VERTICAL_SPEED , 1 , 1 ,0 , \
AccY , AIR_SPEED , 1 , 1 ,0 , \
AccZ , PRANDTL_DTE , 1 , 1 ,0 ,\
T1 , TEST1 , 1 , 1 ,0 , \
T2 , TEST2 , 1, 1, 0 , \
Fuel , SENSITIVITY , 1, 1, 0

I will just need to check if I can replace AccY on the Taranis with HSpeed :-)
Kind regards
Rolf
mstrens
Posts: 1435
Joined: Fri Dec 27, 2013 7:49 pm
Country: -

Re: New beta version of openXsensor (=openXvario)

Post by mstrens »

rolfz wrote:Thank you for your answer, very helpful.

To check the message sent, I use a small LogicAnalyser from IKALogic, that one is connected on the TX line of the Mini-Arduino output.
Once the module is in place, I can run the logic analyser software on the PC and see block by bloc what is sent. See picture below.
OpenXsensor_Hub.jpg
So after reading your message, and if I enable again Vario2, I effectively get the package with code 0x25, so question answered and issue solved.
Concerning the horizontal speed, I exactly did what you suggeted, I put the AirSpeed onto the AccY, and I now get the horizontal speed in replacement to the vario2 which I do not use.

Here my current setup working fine with my old D8R-II receiver, seems to work fine :D

#define SETUP_DATA_TO_SEND \
DEFAULTFIELD , ALTIMETER , 1 , 1 , 0 ,\
VSpd , PPM_VSPEED , 1 , 1 ,0 , \
AccX , VERTICAL_SPEED , 1 , 1 ,0 , \
AccY , AIR_SPEED , 1 , 1 ,0 , \
AccZ , PRANDTL_DTE , 1 , 1 ,0 ,\
T1 , TEST1 , 1 , 1 ,0 , \
T2 , TEST2 , 1, 1, 0 , \
Fuel , SENSITIVITY , 1, 1, 0

I will just need to check if I can replace AccY on the Taranis with HSpeed :-)
In order to avoid having to modify the Taranis, do you want that I try to allow to send the airspeed in the fields foreseen for GPS speed?
rolfz
Posts: 13
Joined: Sat Oct 04, 2014 3:08 pm
Country: Switzerland
Contact:

Re: New beta version of openXsensor (=openXvario)

Post by rolfz »

yes, that would be very cool, thanks in advance...
Kind regards
Rolf
mstrens
Posts: 1435
Joined: Fri Dec 27, 2013 7:49 pm
Country: -

Re: New beta version of openXsensor (=openXvario)

Post by mstrens »

rolfz wrote:yes, that would be very cool, thanks in advance...
I put on google a new version of 2 files (oxs_out_frsky.cpp and oxs_out_frsky.h).
Those files are in the folder tags\OpenXsensor_v2.0 but not yet included in the downloads.

I did not test the change but I hope it will work.
At least it compiles without errors.

With this version, when airspeed is sent as DEFAULTFIELD, OXS will put the airspeed in 2 fields normally used for GPS speed.
Field GPS_SPEED_B should contains the airspeed in knots without decimal.
Field GPS_SPEED_A should contains the decimal part of airspeed.
Please note that I do not know how many decimal are foreseen. Currently I sent only one decimal because I know that in SPORT protocol the airspeed is sent in 1/10 of knots.

Please test this version and let me know if it is ok.
If it is ok, I will put a version 2.2 in the dowloads folder.
rolfz
Posts: 13
Joined: Sat Oct 04, 2014 3:08 pm
Country: Switzerland
Contact:

Re: New beta version of openXsensor (=openXvario)

Post by rolfz »

That's just perfect, GPS speed is in the message (code 11 and 19).
header 11 and 19 can be seen in the message
header 11 and 19 can be seen in the message
I just checked with my Taranis, and I cannot find the GPS-Speed on it, does anyone know where to find it ? I cannot beleive the setup of the Taranis is different from SPORT to HUB protocol ??
Kind regards
Rolf
mstrens
Posts: 1435
Joined: Fri Dec 27, 2013 7:49 pm
Country: -

Re: New beta version of openXsensor (=openXvario)

Post by mstrens »

rolfz wrote: I just checked with my Taranis, and I cannot find the GPS-Speed on it, does anyone know where to find it ? I cannot beleive the setup of the Taranis is different from SPORT to HUB protocol ??
On Taranis, there are 3 telemetry screens where you can select the fields you want to display.
For each of the 3 screens, you can put 12 fields (4 rows of 3 fields)
For each field, you can select in a drop list.
I think that the GPS speed is named Spd. It is between fields named T2 and Dist.
rolfz
Posts: 13
Joined: Sat Oct 04, 2014 3:08 pm
Country: Switzerland
Contact:

Re: New beta version of openXsensor (=openXvario)

Post by rolfz »

Unfortunately I do not get anything on Spd /Vit in french, I tried to change the formating code to get the data after T1/T2 but there is no data on it.
I get the vario data and PRANDTL_DTE on the screen, working fine, but no speed data, which I can see sent over the TX line.
So the issue is not the OpenXsensor code, but something wrong inside the Taranis firmware.
Kind regards
Rolf
mstrens
Posts: 1435
Joined: Fri Dec 27, 2013 7:49 pm
Country: -

Re: New beta version of openXsensor (=openXvario)

Post by mstrens »

rolfz wrote:Unfortunately I do not get anything on Spd /Vit in french, I tried to change the formating code to get the data after T1/T2 but there is no data on it.
I get the vario data and PRANDTL_DTE on the screen, working fine, but no speed data, which I can see sent over the TX line.
So the issue is not the OpenXsensor code, but something wrong inside the Taranis firmware.
Perhaps not.

I think at 2 items that could explain this issue.
1) OXS allows already (in hub mode) to send Gps Alt (in 2 fields ALT_A and ALT_B). It is this part of code that I copied and modified in order to send the airspeed as Gps speed.
When I copied this code I saw folowing lines:

// need to send a gps fix before openTX displays this field....
SendValue(FRSKY_USERDATA_GPS_LONG_A, 1);
SendValue(FRSKY_USERDATA_GPS_LONG_B, 1);
SendValue(FRSKY_USERDATA_GPS_LAT_A, 1);
SendValue(FRSKY_USERDATA_GPS_LAT_B, 1);
// now send the data

I did not expected that this was valid for gps speed too and so I put those lines as comments in the code used for airspeed.
Perhaps that activating those lines will be enough to let openTX display the Gps speed.
So you could make folowing test:
In file oxs_out_frsky.cpp, edit the 4 lines 1408 up to 1411 : you must just remove the "//" at the begining of the lines.
So e.g. line 1408 which is
//SendValue(FRSKY_USERDATA_GPS_LONG_A, 1);
should become
SendValue(FRSKY_USERDATA_GPS_LONG_A, 1);
This is n easy test.
Please note that doing so, OXS will add 4 more fileds to be sent. It could be that we reach the maximum number of field that can be put in one frame and that the last fields would be discarded.

2) Normally, in the Hub protocol, the GPS data are part of a different frame. Perhaps that the Rx (or openTX) does not accept to get GPS data inside the same frame.
In this case, I would have to put the GPS data in a different frame but this is more complex.


I propose that you first test the solution 1 (uncommenting 4 lines) before implementing the solution 2.
rolfz
Posts: 13
Joined: Sat Oct 04, 2014 3:08 pm
Country: Switzerland
Contact:

Re: New beta version of openXsensor (=openXvario)

Post by rolfz »

Thank you for the details,
What I have seen is that thoes lines where not commented on my copy. But maybe I also need to configure them into the stream that is sent to the receiver.
I did not check at the right place, I will do this later today. Do I also need to configure oxs_config.h ?

Btw, I already removed the 2 fields Test1 and Test2 as I was expecting some limites. The message is still sent with this two lines removed.

I will need to check more carefully later today and let you know if I made some progress.
Kind regards
Rolf
mstrens
Posts: 1435
Joined: Fri Dec 27, 2013 7:49 pm
Country: -

Re: New beta version of openXsensor (=openXvario)

Post by mstrens »

rolfz wrote:Thank you for the details,

What I have seen is that thoes lines where not commented on my copy. But maybe I also need to configure them into the stream that is sent to the receiver.
This inside oxs_config.h.

Btw, I already removed the 2 fields Test1 and Test2 as I was expecting some limites. The message is still sent with this two lines removed.

I will need to check more carefully later today and let you know if I made some progress.
The latest version of OXS that I posted yesterday at this link
https://code.google.com/p/openxsensor/s ... _frsky.cpp
has wel the 4 lines as comments inside the function SendGPSSpeed.
Those lines are from 1408 up to 1411.
It is those lines that you should uncomment.

Please note that those 4 lines exist also a few lines before in the code for the function SendGPSAlt. Still this function is not called with the config.h file that you are using.
rolfz
Posts: 13
Joined: Sat Oct 04, 2014 3:08 pm
Country: Switzerland
Contact:

Re: New beta version of openXsensor (=openXvario)

Post by rolfz »

Fantastic, mstrens !!!!

After removing the // as you indicated, I now get

id codes .....0x1A, 0x12, 0X1B, 0x13, 0x11, 0x19....
This are full GPS fields + Speed on the message sent, and the Taranis display shows speed as expected.
Many thanks, this is great!!!!

So ill try the Vario with speaking altitude and speed when good weather is back.

Wonder if GPS data and Accellerometer can be sent as well ;-) but I will check that myself!

I just mesured the full frame length and it takes approx. 60ms to send, so with HUB protocol at 9600baud, we may reach the limit...
Kind regards
Rolf
rolfz
Posts: 13
Joined: Sat Oct 04, 2014 3:08 pm
Country: Switzerland
Contact:

Re: New beta version of openXsensor (=openXvario)

Post by rolfz »

I finally could make a practical test with my car.

! configured vario switch with 3positions: Silent, Speaks altitude, Plays Vario sound
! second button switch on/off speaking speed.
So all data could be switched on/off during flight, but still be visible on the lcd telemetry screen.

Results
+ Vario data and Speed seems to be realistic. That part is ready for a real flight
- Altitude seems to get locked on max altitude, does not get lower when comming back to same altitude.
- I did not yet manage to use PPM input, is there anything to add from the default values to get this working ?

So far so good, results are very motivating :-)
Kind regards
Rolf
mstrens
Posts: 1435
Joined: Fri Dec 27, 2013 7:49 pm
Country: -

Re: New beta version of openXsensor (=openXvario)

Post by mstrens »

rolfz wrote:I finally could make a practical test with my car.

! configured vario switch with 3positions: Silent, Speaks altitude, Plays Vario sound
! second button switch on/off speaking speed.
So all data could be switched on/off during flight, but still be visible on the lcd telemetry screen.

Results
+ Vario data and Speed seems to be realistic. That part is ready for a real flight
- Altitude seems to get locked on max altitude, does not get lower when comming back to same altitude.
- I did not yet manage to use PPM input, is there anything to add from the default values to get this working ?

So far so good, results are very motivating :-)
About altitude:
I never got a comment that Altitude was not working with OXS.
Please take care that Taranis has several fields about Altitude. There are I think (not sure about the codes) Alt, Alt+ and Alt-.
Alt+ is the maximum altitude (since power on of Taranis) and is calculated by openTx based on the Altitude sent over telemetry (e.g. by OXS)
Please check that you did not use Alt+ instead of Alt in your setup on Taranis.

About PPM:
Explanations are given in the file oxs_config_description.h
Please read it.
In the config file on google , PPM is configured.
The main lines are
#define PIN_PPM 2 // default is 2 but my own device use 3
#define PPM_MIN_100 980 // 1500 - 512 ; // pulse width (usec) when TX send a channel = -100
#define PPM_PLUS_100 1990 // 1500 + 512 ; // pulse width (usec) when TX send a channel = +100

Then there are several other lines to configure in order to adjust:
- vario sensitivity
- vario compensation (for compensated vario using airspeed sensor)
- selecting the vario (uncompensated, pneumatically compensated, electronically compensated) that sent the vertical speed to TX
- resetting airspeed
Do not hesitate to ask for explanation if the oxs_config_description.h file is not clear enough.
rolfz
Posts: 13
Joined: Sat Oct 04, 2014 3:08 pm
Country: Switzerland
Contact:

Re: New beta version of openXsensor (=openXvario)

Post by rolfz »

Thank you, ALT+ was my mistake, I did not realize it was max-altitude, and ALT was not just next to it in the menu.
Ill go a bit more into details of the description.h, this will probably explain why I do not get feebback on Sensivity field.

I would understant I can trim, the vario sensivity with a Pot-Button, and the Vario would send back the value with the Sensivity variable if I define it in the protocol like you did it, ie: Fuel , SENSITIVITY , 1, 1, 0
Kind regards
Rolf

Post Reply

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