Altimeter sampling rate

The Rocketry Forum

Help Support The Rocketry Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

shockie

High Plains Rocketeer
Joined
Jun 2, 2009
Messages
1,184
Reaction score
625
Location
My Old Kentucky Home
What determines the sampling rate of an altimeter?. For example, the Raven 4 states it has a 400 Hz accelerometer. Does that mean it samples 400 times a second? ELI5.
 
Yes. That’s is exactly what it means. You will get all 3 axes sampled 400 times per second.

Lots of things can determine the sampling rate: processor speed, processor load, sensor limits, noise, etc. Usually there is some sort of trade off happening, i.e. lower sample rate to reduce noise.
 
Yes. That’s is exactly what it means. You will get all 3 axes sampled 400 times per second.

Lots of things can determine the sampling rate: processor speed, processor load, sensor limits, noise, etc. Usually there is some sort of trade off happening, i.e. lower sample rate to reduce noise.

Higher sampling also usually means more storage used, like mentioned it's a trade-off.
 
The Bosch sensors we use also have the ability to average multiple samples between measurements but you also need a temperature sample to do the calibration. You will get less noise letting the sensor avg 16 pressure samples and 1 temp vs running it faster and using 1/1. It’s just more data. Eventually though the trade off is small and it’s better to do the filtering in the firmware where you have more control. One thing to note, if you are interested in velocity, you will get a much better result the longer the time period is.
 
Does higher sample rates equate to higher resolution.

For example, compare and contrast 400 vs 50 for a short time frame say 0.3 seconds
 
Does higher sample rates equate to higher resolution.

For example, compare and contrast 400 vs 50 for a short time frame say 0.3 seconds


Temporal resolution yes, you would get 120 measurements vs 15 respectively. Depending on the sensor, this may give you a lower pressure resolution though. That is the balance, really you want a useful altitude, not just pressure points.
 
Temporal resolution yes, you would get 120 measurements vs 15 respectively. Depending on the sensor, this may give you a lower pressure resolution though. That is the balance, really you want a useful altitude, not just pressure points.
ok Russ you are going to have to show me the math on how you got 120 and 15 from 400 aND 50...eli5
 
Yes. That’s is exactly what it means. You will get all 3 axes sampled 400 times per second.

Lots of things can determine the sampling rate: processor speed, processor load, sensor limits, noise, etc. Usually there is some sort of trade off happening, i.e. lower sample rate to reduce noise.
Hmm, looking at the Featherweight website:
  • 400 Hz axial Accelerometer, +/- 105 Gs
  • 200 Hz lateral Accel., +/- 105 Gs
  • 20 Hz Baro data, +/- 0.3% accuracy!
  • 20 Hz voltage on each of 4 outputs
  • 40 Hz output current
  • 20 Hz high-precision temperature sensor
  • 20 Hz for All flight events used for deployment logic.
So there are three different sample rates in play, and only 2 axes are measured - vertical and lateral. Only the axial accelerometer is measured at 400 Hz, all others are slower, the baro at only 5% of that.

Not a knock on the Ravens, I think they are great altimeters. Just FYI.

And yes, higher sample rates generally equate to higher resolution, but whether or not that is useful depends on many factors – the how, what, and why you are measuring something. Will changing the baro sample rate from 20 to 200 give me a more accurate altitude reading? Probably not. But will it let me more accurately 'see' a mach shockwave passing over the static ports? Probably. (Actually I have no idea on the last one but it sounds good.)


Tony
 
Last edited:
Will changing the baro sample rate from 20 to 200 give me a more accurate altitude reading?

No. That’s not how barometric pressure sensors work. Increased sample rate won’t necessarily increase your accuracy, but it sure will increase the noise. At least in my experience.

Additionally, you will definitely see the shockwave in the data at 20Hz. In fact, you will probably see a difference through the entire supersonic flight region, but it might depend.
 
No. That’s not how barometric pressure sensors work. Increased sample rate won’t necessarily increase your accuracy, but it sure will increase the noise. At least in my experience.

Additionally, you will definitely see the shockwave in the data at 20Hz. In fact, you will probably see a difference through the entire supersonic flight region, but it might depend.
Not to nitpick, but you took my question out of context. It was purely rhetorical, I wasn't looking for an answer, I was framing it from the perspective of the OP. You'll notice I answered my own question to try and provide prospective on sample rates to the OP.

I've flown through Mach a fair number of times and I'm very familiar with how altimeters work. And noise isn't always a bad thing, sometimes it's necessary. Just ask UhClem about noise and Kalman filters.


Tony

PS: I try and always notate that I have snipped or abbreviated a post that I quote, I think it's good way to avoid misunderstandings
 
Last edited:
Not intentional to take it out of context. Some of the quotes get lengthy so I try to trim just the relevant part to keep things short and readable. Also, I tend to read things literally, so I often miss the subtleties.
 
Most altimeters on the market have a baro sampling rate of about 20 samples/sec, that's perfectly adequate for detemining altitude and getting a reasonable time vs altitude/velocity curve. With an accelerometer, you may benefit from much higher sampling rates particularly if you're trying to look at an EX motor's performance for possible chuffing or pulsing effects that might be aliased out with a lower rate.
 
Back
Top