[kangchen06]

Hello all,

I have a Revolution with a barosensor MS5607-02BA03 rather than the classical MS5611-01BA03.
Values returned by the controler are:

Altitude 5490 m
Temp 38,51 °C
Pressure 50581 kPa

Temperature is correct. It's a good point.
The problem is more the altitude because my real altitude is 43 m (and not 5490m without oxygen !)

Other error: 5490 m = 50,581 kPa not 50581 kPa ! (If I'm not wrong)

Is there something to set in the code of the Revo firmware to take into account the fact the barosensor is not the official MS5611-01BA03 but a MS5607-02BA03 and then return good results ?

PS: Altitude variations seem to return consistent results.
 
Regards

[kangchen06]

I can answer partially to my post. To earn money, they have replaced the MS5611 with a MS5607. The precison is the half of the original (20cm rather than 10cm).

Of course formulas to calculate the pressure are a little bit different but modify the code accordingly seems to be not too hard: next step.

Eric

[TheOtherCliff]

So it sounds like each increment is now 20cm instead of 10cm.

I wonder if the code can tell the difference in the two versions, if not we could add a factor for the user to set.  Actually, if you build your own code, you could change one line in sensors.c

    float altitude = 44330.0f * (1.0f - powf((sample) / PIOS_CONST_MKS_STD_ATMOSPHERE_F, (1.0f / 5.255f)));

probably just change ?
    sample
to
    sample*2.0f

[jtrout19]

why not just purchase the correct baro and replace it?

[Mateusz]

IMO Adding this as feature would probably mean that it us ok to use cheaper baro with worst precision. I would return board and request money back asap. But if it's too late then change in the code. I would not attempt to replace baro myself as it is difficult and it's delicate component.

[jtrout19]

I have replaced smaller things than that on boards before. Just a little hot air and some steady hands.

[DocHardinger]

I just build the firmware with adjusted sensors.c file and it seems to work then.
Just as info ;-)
Just the firmware building seems to work with windows, but not the GCS software.
Am i right that i need to build GCS with Linux in order to get the controller work correct with GCS???
By now im using the 15.09 GCS windows version, so I think that does not work correct because the firmware is newer now...

[Mateusz]

@DocHardinger Yes

Both GCS and Firmware must talk using the same UAVTalk version, otherwise wired things can happen. You should always use GCS that is matching firmware version.

P.S there is also Android App (LP2Go) designed for complement GCS in the field. It's not meant for setting up air-craft, but rather to allow changing settings in the field, or everything that might be useful to do outdoors. I really think it's worth setting up Bluetooth module !

[DocHardinger]

Thx for your reply!

its working now with debian, the new 16.09. version and the edited sensors.c file.
I hope dev team will modify configuration in system in GCS to adjust this factor.
I know its a hardware fault but i think its hard to get a good Revolution controller :-)

And i hope they will fix the windows compile problem, but maybe its my fault. I will try to install everything again and try to compile it again.

Im also trying to get the Revolution board working without an external MAG. Trying to put the board further away from all these high currents. Will see if its still working fine if its 80mm above the main frame ;-), but i think thats another topic...

So thx again

[kangchen06]

Solution:
Modify formulas to calculate preassure in file pios_ms5611.c / function PIOS_MS5611_ReadAD (according to MS5607-02 datasheet).
Then upload the new firmware and all is working fine.

Thank you all for your comments and advises !

Eric