barometer testing + PilotAware graphic

[neilmurg]

Is there a copy of the PilotAware graphic in high definition? I want to create labels for my PAW which include the MAC address and plane ID + hex code.

I have 2 PAW units with barometers. I noticed when I look at the 192.168.1.1 page it reports 1013 hPa until I connect Skydemon, then the page shows what I assume is it's calculated QNH.
Today Heathrow (closest airfield with weather) reported QNH 1025
PAW 4F reported 1032 with my altitude from phone at 260 feet (pretty close)
PAW 78 reported 1026 with my altitude from phone at 260 feet

both units are on the latest release 20151114, they have identical baro types on Harkwood shields

I took out the GPS as in the house it doesn't get a lock and spams the logging screen.

To test I:
  Powered the unit and waited for it to generate a wifi;
  connected the phone to 192.168.1.1;
  checked the phone had a valid GPS position;
  cold started PilotAwareAndroid
  checked the 192.168.1.1 page - baro Auto, 1013
  cold started Skydemon and pressed 'Go Flying', made sure it had a GPS lock (yellow plane appeared)
  went back to 192.168.1.1 and checked QNH and height

then repeated the above with the second unit after cold starting the phone

I will need to check at other QNH values to see if this is just an offset problem or if 1 of the baro's is faulty.
Is there a way to get a baro reading without starting Skydemon?
Is my test sequence OK, have I missed something, is there an easier way?

[Admin]

Is this what you want ?

PilotAware-Logo-Proof.pdf

[neilmurg]

That's great, thankyou.

Further to my other point about baro accuracy, I'm doing further testing on my 4 baro's. I hope the results will be useful. And to answer my own question about if there's an easier way, I guess I could do a Noobs install and read the baro chips directly rather than go though the PAW.

I read the spec sheet http://cache.freescale.com/files/sensors/doc/app_note/AN4528.pdf and it says in there that an offset can be applied to increase accuracy, the offset is stored in NV memory I think, so needs to be re-applied on startup.

Quote from the above doc:
The absolute measurement method is affected by several offset errors resulting in Poff. Sources of offset errors could be due to device-offset variation (trim errors), mechanical stresses (mounting stresses), shifts due to temperature and aging. The MPL3115A2 is factory calibrated for sensitivity and offset for both temperature and pressure measurements and the default values are used on power up. Using internal-offset correction registers, users can adjust the offset values to easily remove these errors after power up for the highest possible accuracy. Further calibration overrides the factory settings and does not affect the factory-calibration values. Users can directly write the correction values of pressure, altitude and temperature into respective user-offset registers (OFF_P, OFF_H and OFF_T). These values will be used by the internal algorithm during the output calculation.

[IainM]

Looks like the 6mBar spread between the two units is within the absolute accuracy range quoted in the spec sheet?  (+/- 4mBar) Thank for the link, lots of interesting info in the data sheet.

Possibly an optional calibration routine to store the pressure offset based on manually entering QFE before takeoff could be useful, but in the context of collision avoidance is an error of +/-120 feet worth bothering about?  We'll be wanting a traffic warning long before that level of accuracy is needed, although obviously the more accurate it can be the better.