So I was super excited about receiving my BMP280 sensors to build an oled custom display altimeter for off roading and general use. After uploading the code and adjusting my float level of my local hPa pressures accordingly my altitude was calibrated to my exact elevation.
Everything was good until the next day when the reading was about 100 meters higher than the day before and the sensor had not changed location. I know that these sensors use baro readings to calculate altitude so how on earth is it ever going to be real-world accurate if the pressure is always changing???? You would have to run your arduino to the pc every hour or so and recalibrate for it to actually be accurate to real world stats. Will still work for off roading as the accuracy of incline/decline (comparable to hPA float inserted in code) will be +/- .3 meters but will never be a true altitude reading.
Any input here would be appreiciated . Also if someone knows of an altimeter project that wil self calibrate and be correct all the time that would be great.
Its good for short term relative altitude readings. It is not a long term (hours) solution. Typically, as in aircraft, you dial in the current baro before takeoff for the same reason.
2 Likes
Concratulations. You have just discovered why pilots ask their flight controllers for the local air pressure at sea level (QNH) when the fly into a certain area.
https://pilotclimb.com/qnh-qfe-and-qne-altitude-height-flight-level/
1 Like
Thanks, seems so weird I searched for someone with the same problem dealing with the BMP280 but could not find it.
So what you're saying is , we have the technological abilitly to ram 16 billion transistors into a cellphone and have every one work flawlessly but we still don't have a way of reading altitude that doesn't constantly need calbration???? 
You can use a GPS receiver to calibrate your baro altitude reading, however a baro altitude reading has more relative accuracy than a 3D GPS reading. Which is one of the reasons why aviation still relies on calibrating the quite accurate altimeter before and during each flight.
Nice, makes sense now. Thanks for the replies all.
Or turn the altimeter to the height above sea level of the airport.... All due to the whether, high pressure, storm and low pressure....
RTK GPS measurements are accurate to cm, but a moving receiver requires a correction data stream from an accurately located base station.
Barometric altitude measurements are quite accurate, as long as you keep track of the local atmospheric pressure changes.
For use in a motor vehicle, you can calibrate your altimeter by consulting a map with altitude contour lines. When you cross a known altitude line, set the device to match. Our home is at 2800 ft., if I wanted to use such an altimeter, I would set it to 2800 ft., for whatever the pressure happened to be and It would be accurate for quite a time/distance.
Sometimes a good clue is not enough to overcome the lack of a good search engine. 
I've deleted some replies, please keep this civil. A short rest from the forum awaits the next person who posts a rude reply. The duration of the rest will depend on my mood at the time.
Enjoy 
2 Likes
I am working on a similar vehicle altimeter using the BMP280, initial idea was to have the system calibrate itself on startup and then display altitude relative to starting location.
In case you are still working on this scootermcgavin I hope these observations are useful,
To set the current local pressure I take 10 pressure readings at 100 millisecond intervals then average them to find the current local pressure. Then use that as the new datum pressure in the Adafruit library code instead of the 1013.25. The relative altitude returned is then around zero but varies between readings much more than I think it should do. I can check the actual local pressure from a METAR airport a few miles away, at the time of writing that is 993, my location is slightly elevated (54m) my measured pressure is 987.8, I feel that is good enough as an absolute measure, but I find the drift strange. After about 10 minutes the altitude can be showing 3 metres even though the test rig has not moved - see photo.
Powering the test system from a battery altitude changes by 2 metres between floors.
The test has run now for over an hour, during which time the pressure reading has remained within a narrow band, but the calculated altitude has changed by over 6 metres.
I am interested to know if you found a solution.
.
Hello @rac-47uk ,
I am not clear if your reply is intended to be a helpful comment for the OP or a new question. If a new question then it should be a separate topic. Please clarify what you want, if a new question then I will split your reply to a new topic (please don't start another, I can split this for you).
Thanks
Apologies for the confusion, I intended a comment to the OP. I will edit and make that clear, especially as I have found further information whilst researching this topic and run another test.
1 Like
Are you testing indoors, while some wind is blowing around your house? That usually creates pressure fluctuations that the sensor picks up. E.g. opening and closing the door of the room you are in, will be visible in the sensor readings.
Yes, testing indoors, at this time there is a wind blowing, not very strong but blowing. I note the comment about the door too. I will try moving outside and see how it affects things. The moving vehicle will affect pressure too, that's something I had not though of.
Thanks for your interest.
Why is the BMP280/BME280 made by Bosch? For an engine is the real height not important. The air pressure make sense. And that is what the sensor do. That the air pressure fall with the height is correct. But at the same time the barometer can rise. You go higher and your meter say that you go down.
I have a analog barometer/altitude meter over 30 years for walking in the mountains. The manual say that you must recalibrate the meter every time you cross a place with altitude designation. The instruments gives then the approx height and by fast air pressure fall that there is bad weather coming.
Fuel/air ratio computation.
1 Like
