Flight Computers and Temperature

As a side note, my main business deals with accurately measuring pressures in the petroleum industry. You would be amazed (or perhaps not) at how much a temperature difference can change the pressure reading from a given sensor. In other words, a pressure sensor at low temperature will give a different reading than the same sensor at the same pressure at a higher temperature. In my line of work, any pressure reading has to be coordinated with its associated temperature reading, then an algorithm is applied to the two numbers to arrive at the 'corrected' pressure (and the individual calibration is unique to each sensor). Again, this interaction can be pretty significant.

While this is overkill for current rocketry applications, it may well be the way the things are headed, so the ability to (as an initial step) store a temperature reading with the pressure reading is probably a good thing...

David Erbas-White

If someone wants to bring altimeter to my lab some Saturday, I can do the test in a NIST traceable environment.

Doc

I agree with you, David, and I don't think it's overkill at all. Even in a tropospheric flight, ground temperature alone can make a remarkable difference.

The altitude reading is proportional to the absolute base altitude temperature. For planes, base altitude is normally sea level. For HPR, it is normally ground level. The assumed temperature in the U.S. standard atmosphere is 15 degrees C or 59 degrees F. The correction would therefore be actual absolute temperature over assumed absolute temperature:

CorrectedAltitude = AltitudeReading*(273.15+TempC)/288.15

On hot days, the altimeter underestimates; on cold days it overestimates. At 12 degrees F (-10 degrees C), an altimeter reading of 1000 feet would therefore denote an corrected altitude of 905 feet.

There is also an assumed temperature lapse rate of 6.5 degrees per km, but it turns out that errors in this rate do not have huge effects.

Bottom line: For most flights, ground temperature alone would be a big help.

Regards,

-Larry Curcio

I think you may be missing part of my point -- while the temperature has a bearing on the pressure, which will cause an error in a pressure altimeter reading, there is ALSO sensor error that is related to temperature. The sensor itself will respond differently, based on temperature (its response characteristics). So, not only must one take into account the temperature/pressure correction, but the temperature IN AND OF ITSELF is important in interpreting the pressure data.

While the sensor response is quasi-linear, it's off enough that a curve fit is necessary in order to get accurate PRESSURE data. From the accurate PRESSURE data, one can then convert to a better fit for the ALTITUDE data.

David Erbas-White

David,

Sorry I was off the page.

Regards,

-Larry C.

There is a distinct problem in measuring temperature very fast. There is a time constant associated with a temperature sensor. If you have ever taken a thermometer outside in winter (or brought it inside) you will find that the reading does not change instantly. The thermometer has to cool down or heat up before it can measure the temperature.

You can minimize the "time constant" of the temperature sensor by making it very small - giving it a small "thermal mass". It would also help if you can get the sensor on the outer surface of the rocket.

Ideally, you measure the temperature of the pressure sensor itself, because as others have pointed out, the temperature of the sensor will mke a difference as well. Now, is there a baro altimeter that has the corrections programmed in? Probably not for a reasonable, hobbyist price.

FWIW.

I think the first error you mention isn't real. Thermal effects influence sensor error (the second error you point out), but absolute pressure is absolute pressure regardless if you're at 300K or 3000K.

1psi acting on a surface at 300K will exert the same force as 1psi at 3000K. If you were to make the sensor adiabatic (its temperature didn't change), then it doesn't matter how cold or hot the fluid you're measuring is.

Dave

Nevermind, I misunderstood what you were trying to say... basically that pressure as a function of altitude varies day-to-day, and temperature is one of the factors influencing that... I gotcha now, and you're right.

Dave

That is for absolute readings. But altimeters report a difference between one reading and another.

There is one, but I do not know what the thermal mass of it is. It is used by the GPS flight group. It has a temp sensor onboard with formulas to compensate the pressure. But you are still stuck with pressure changes even if you can accuratly record pressure. Robert

That's true. I thought it worth mentioning, though. If you can correct that much error with one reading...

Regards,

-Larry C.

overestimates.

Right, the GPSFlight units have a baro add on board that is temperature compensated, and the temperature data is included in the telemetry stream along with pressure data. We're currently testing a new unit that is accurate to 100k feet. Fun stuff.

most of the commercial (resistive) temperature sensors I found had a Tc of 5 to 15 seconds. obviously not fast enough to measure during ascent but the measurements could be OK for descent - under parachute - do the math, not fast enough for tumbling or streaming rockets.

I thought this thread started out about measuring temperature changes that would effect the pressure sensor. If the temp sensor is inside the pressure sensor then that's all that matters.

For faster response you would need to use a thermocouple, they can be very tiny.

RDH8