Temperature Adjustment of Barometric Altitude

[ihbarddx]

I recently purchased a Blue Ravin 6 DOF flight computer, and launched it four times. Thought the results were good illustrations of the effect of temperature correction. Here are the altitude-time curves for 6DOF inertial altitude (Inert), raw barometric altitude (Raw), and temperature-corrected barometric altitude (Correct). Note that the 6 DOF curves are subject to my limited calibration skills (the best-developed of which is swearing), and to drift. It ain't perfect, but it's an independent measurement.

The flights took place on hot, summery days in West Palm Beach FLA, so the raw barometric altitudes are low. Temperature correction yields barometric curves closer to the inertial curve. Since Inertial and barometric curves have nothing to do with each other, apart from their measurement of true altitude, this is a good sign.

BTW, the fact that we tend to launch in warm weather is ONE reason that accelerometer altitudes tend to exceed barometric altitudes. That is, the discrepancy is not entirely the fault of the accelerometer, even when single-axis instruments are involved.

 

[cerving]

What are you using to correct the temperature in flight? Very few thermistors have enough sensitivity and a high enough slew rate to accurately measure temperature changes during a typical hobby rocket flight... and most definitely not the ones built into the baro sensors. They're fine for ground-level temperature-pressure correction, or for balloons... but not so much for rockets. In most flight data that you'll see, the temperature changes little if at all, especially on a flight of "only" 800'/8 seconds

 

[ihbarddx]

What are you using to correct the temperature in flight? Very few thermistors have enough sensitivity and a high enough slew rate to accurately measure temperature changes during a typical hobby rocket flight... and most definitely not the ones built into the baro sensors. They're fine for ground-level temperature-pressure correction, or for balloons... but not so much for rockets. In most flight data that you'll see, the temperature changes little if at all, especially on a flight of "only" 800'/8 seconds

I'm using the temperature from the thermometer on the field, and the standard temperature lapse with altitude (6.5 Kelvins/km) built into the Standard Atmosphere formula. Balloon data can be used for a more accurate lapse rate, but the linear decline in temperature with altitude in the troposphere comes from the ideal gas model. The onboard sensor measures instrument compartment temperature, which is a poor measure of ambient temperature on hot days.

So, basically, I'm substituting the true absolute ground temperature for 288.15 in the formula (which corresponds to 15 degrees Celsius). Tell ya the truth, the lapse rate don' make that much difference between the ground and 850 feet. (EDIT: Sorry. I see that you anticipated this.)

Point being, not that barometric altimeters are perfect, but that temperature correction improves their accuracy. That's why pilots temperature-correct their altimeters.

 

[ihbarddx]

What are you using to correct the temperature in flight? Very few thermistors have enough sensitivity and a high enough slew rate to accurately measure temperature changes during a typical hobby rocket flight... and most definitely not the ones built into the baro sensors. They're fine for ground-level temperature-pressure correction, or for balloons... but not so much for rockets. In most flight data that you'll see, the temperature changes little if at all, especially on a flight of "only" 800'/8 seconds

Because it only just now comes to mind...

In 2005, I collected data on temperature lapse with altitude. I was in a large airliner, flying from Beijing to Newark. The plane had a GPS display, which listed GPS altitude and temperature. Having nothing better to do for 16 hours, I collected data on takeoff in Beijing and landing in Newark. Note that the lapse rate isn't 6.5 k/km, but the data are very nearly linear. (I grant you some horizontal distance was covered, but... they're still linear.) FWIW, theory says that perfectly dry air would yield a lapse rate of something over 9 k/km. Cloud cover (phase change) would likely have messed the data up a bit.