Variometer/Altimeter

Go To Last Post
52 posts / 0 new

Pages

Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

With apologies if this is the wrong place to put this... but does anyone have any experience building such an animal? The search engine produces a few references, but nothing helpful, and while google produces a lot of good information, there's little recent or directly equivalent.

It's essentially an integrating altimeter which logs and displays altitude (referenced to take-off, mean sea level, or a reference sea level pressure) and also indicates short term (1/4 second) and average (30 second) changes in altitude. This latter option is accompanied by an audible warning; beeps or continuous tone indicating lifting and falling and the frequency of the tone indicating the rate of change.

As a bonus, USB connection to a PC should allow the copying of logged data, and of course, it should do all this in a minimum weight, minimum power sort of way :)

My first tentative sketches suggest something along the lines of the motorola MPX4115 pressure sensor which produces a temperature compensated linear 0.6v-4.6v output; some analogue processing to amplify, filter, and shift the output Ievel to 0-5v for the ADV input (for better noise/range) followed by oversampling to improve the dynamic range.

Conversion of pressure to height probably via a look-up table; a real-time clock on TWI, a nice fat serial eeprom for the logging data, and a piezo sounder.

Display on a serial LCD - the Lascar SP 5-GFX1 looks rather good - and either a joystick switch or a rotary/push coder to control it. USB from the FTDI 245 which looks rather interesting and reasonably inexpensive, and an M32 to tie it all together.

The expensive bit is likely to be the pressure sensor, but this should all go together for under fifty quid for a proto, I think.

Any thoughts?

Neil 8)

p.s. I'm a paraglider pilot, in case you were wondering.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Sounds OK to me Neil. And I agree its the sensor which will cost.

Gonna need the abilty to calibrate the unit? Eeprom stored calibration values.

Depending on how many samples you wish to log you may need external flash or eeprom. I SUPPOSE its possible to use unused program memory flash locations for logging.

Yes, the FT245BM is cool. I've used it and would reccomend it. The price is right as well.

DFR

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:

a nice fat serial eeprom for the logging data,

You might consider an MMC card as well. Pre-format on a PC or other device. Then you can either upload the data as you described, or pop the card for desk analysis. Can then swap cards for another "run" and still have all the data for analysis.

Lee

You can put lipstick on a pig, but it is still a pig.

I've never met a pig I didn't like, as long as you have some salt and pepper.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Thanks guys. Nice to know I'm not attempting anything too far off the wall...

Theusch, I want a self-contained and *cheap* unit; nothing to fall off or break, and it has to be drivable in gloves. A 32kB eeprom is about three quid, and as I need only store the sixteen bits of pressure data (I can derive everything else from that off-line for display/download) I can get over two hours sampling every second, of over ten sampling every five. Should be enough to keep me going...

Dana, yes, it will need calibrating and the built in eeprom is ideal, though I can set the DC conditions with a meter I'll need a low pressure chamber to set the range. Maybe I should just locate the pressure sensor with the nipple on...

Cheers,

Neil

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Neil,

This activeflightsystems.com is what I have been working on for the last 3 years ;-)

The pressure sensor is an MS5534 which is pre-calibrated and temperature compensated (and pretty expensive too!). It's a digital device and the datasheets explain how to get from pressure to altitude and they even offer a solution tailored to low power 8-bit micros.

We use Atmel dataflash for the logging memory with a commercial FS over it.

Cheers,

Tim.

------------------------------------------------------------------
www.activeflightsystems.com

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

A variometer is a "differentiating" altimeter. Historically, though, they are more than this. They are often used in gliders and sailplanes where a simple "rate of climb" instrument is not very adequate.

Historically, there is a reservior (think "thermos bottle"). Then, you measure airflow, in or out, often with a venturi. The flow results from difference between the air pressure in the reservior and the local atmosphere. The simple vario used in gliders has a u-tube with some "balls" in it and the u-tube is hooked to the venturi in such a way that it senses not only rate but direction sense (up and down). In gliders there are issues with significant changes in air temperature (due to flying into or out of thermals) showing false indictions of rate of climb and this system is an attempt to correct for that. I am not altogether confident that simple rate of change of atmospheric pressure, as sensed by an electrical pressure sensor, won't be subject to this problem.

Jim

 

Until Black Lives Matter, we do not have "All Lives Matter"!

 

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Jim,

It will be subject to that effect. There are 'total energy' varios which compensate using a venturii, but they're difficult to mount on a paraglider. Most paragliders though care mostly about immediate up and down - am I in the thermal or not? However, the sensor I have in mind is temperature compensated.

Though this is experimental; sooner or later I'll fly it with a professional instrument and see how it compares...

Fathert, that looks like a nice thing... though it does a damn sight more than mine will. I like... btw - (if I could work out how to do it I could set my profile to show that I'm in the UK) I assume you fly too - where? Green Dragons in Surrey for me, and then Dunstable when I have a few more hours.

Cheers,

Neil

Edited to say - MS5534 - now that looks nice. If only I can find a way to persuade Olimex to drill a square hole... gets rid of all the analogue stuff and calibration, and as cheap as the Motorola part.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

If you're just interested in "simple" up or down, why not just use an accelerometer and skip all the pressure calibration problems ?
Also, you may want more than 5 hours of logging time. 10 will cover even most 500 and 1k triangles, but 5 may be a bit too small.

/Jesper
http://www.yampp.com
The quick black AVR jumped over the lazy PIC.
What boots up, must come down.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Inclinometer is subject to acceleration errors! Further, in lift, you can climb with no inclination with respect to the horizon.

Jim

 

Until Black Lives Matter, we do not have "All Lives Matter"!

 

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I also want some warning that I'm getting near my legal flight levels, or that hypoxia is likely :D

Neil

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

For that, you really only need altitude sense, without the vario.

The oxygen requirement threshold in the US used to be 10,000' or 12,000', don't remember which. It may have changed. Flying in Colorado put you a lot closer to that threshold before you even start. Can you get that high in a paraglider?

Jim

 

Until Black Lives Matter, we do not have "All Lives Matter"!

 

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

A fair point ;)

I believe that in the states paragliders are permitted 15,000 feet free flight and 18,000 with permission.

I'm in the UK where the airspace is very crowded; my normal site tops out at 1800 feet before I bump into controlled air space. But I start a lot nearer the sea :)

Neil

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The Motorola Freescale sensors are $18 and change from Digikey. Is that what you intend to use? They have both differential and gauge versions. Instead of measuring flow in/out of a reservior you can use a sealed reservior and measure the differential to local atmosphere. Half atmospheric pressure is at 10,000 feet so a 15 PSI range is adequate, maybe less for better resolution?
The 4.6 volt range is not a problem. The analog power only has to be within Vcc - 0.3V. Drop Vcc a bit. Even the higher clock speed versions work down to 4.5V.
There is an application note on digital filtering, you can measure the sensor directly and internally run it through in the code.I doubt noise will be a problem at the low sampling rate you are using. You measure static pressure so gusts should not factor in.
Most of the A/D's are 10 bit, not 16. You could pack the data and expand it later to reduce memory requirements. Would resolution of 10 feet (10,000/1024) be enough? Possibly not for variometer work.
Most chips have USART built in and for the amount of data it would only take a couple of seconds to send across. No extra parts like required for USB.
Standard day temperature is linear with altitude to 36089 feet. You can add a temperature sensor and correct the measured pressure back to standard day. The chip would have no problems handling the pressure-altitude equation calculation at your data rate. I am using a m8 at 8 mHz and doing floating point calculations for ignition timing. The calculation loop only takes 270 microseconds.
You can use one of the built-in timers to generate the climb/descend tones. You will need a driver for the speaker though. Alternatively, use a 555 timer to drive the speaker and change the control voltage for pitch changes.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hi Neil,

For the MS5534 you don't need a square hole, the part can be mounted either way up so you can mount it with the circular bit poking though a round hole in the PCB. I can't remember what we pay for them but it's just plug and play, Intersema even supply C source code for the drivers so we had it up and running very quickly. The only thing to watch it that the part is extrememly sensitive to noise and to get 1m accuracy you need to get an error of less that +/- 0.1mb.

Yes, I fly, but only with an engine strapped to my back - the lazy way ;-) Been off for six months due to knee ligament ligament surgery though but I usually fly round the Guildford area.

Cheers,

Tim.

------------------------------------------------------------------
www.activeflightsystems.com

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I have used MS5534 aswell.
An accuracy of 5m is easily possible and the device is pre-calibrated.
For re-calibration, just buy a new sensor as it's the cheapest option.

The serial interface is not very nice, but it can be connected through the SPI (requires some software tricks). You don't need a look up table, as you have the ICAO (International Civil Aviation Organization) equation for standard atmosphere as quoted by Intersema.

I achieved vertical speed measurements by numerical integration with 8 points (sample frequency of 8 Hz) but found the output to be very noisey which (not that suprising). I suggest 32 points minimum.

However, if just want to build a variometer/bleeper, then absolute measurement accuracy is not important. You could use a really cheap sensor and the AVR ADC intead.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

GPS wouldn't be a goo aproach?

Guillem.

Guillem.
"Common sense is the least common of the senses" Anonymous.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

It's apparently not sufficiently accurate on height... besides, then I'd want coloured displays with terrain maps...

Neil

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

There are GPS modules that output position in NMEA format through serial port (or even UART port at 3v3) for a cheap prize. Of course, acuracy is an issue.

Guillem.

Guillem.
"Common sense is the least common of the senses" Anonymous.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

GPS units are not generally very good on altitude. It improves as the number of received sats increases, but it is genrally a magnitude worse than surface coordinates.

Jim

 

Until Black Lives Matter, we do not have "All Lives Matter"!

 

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

....

JChristoff
Illinois

Last Edited: Thu. Jan 6, 2011 - 06:47 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Couple years ago we built a general aviation lcd graphics instrument that showed altitude, airspeed and vertical velocity on scrolling tapes. We used a sensym 15psi abs sensor for altitude and an LT inst amp and external LT1290 12 bit a/d on the spi. One lsb was about 20 ft at 30,000 ft. Used a sensym 1.5 psi differential sensor for airspeed. I calculated vertical veloctity by measuring delta feet / deltat seconds. I was getting 6 or 8 updates a sec, which was too fast to see on the lcd almost. All sw fp on a 16Mhz HC12. The flight instructor that flew it commented on how sensitive and responsive the VVI was... unlike the pneumatic ones that always had a lot of lag.... he said he could just tap the yoke and see the VVI bump, so I think all you need is a 15psi sensor and a 12 bit a/d and you're golden.

Imagecraft compiler user

Last Edited: Tue. Jan 24, 2006 - 12:37 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

for one

JChristoff
Illinois

Last Edited: Thu. Jan 6, 2011 - 06:45 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Thanks Bob.

JChristoff
Illinois

Last Edited: Thu. Jan 6, 2011 - 06:46 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

We (Aviation Instrument Technologies www.aircraftinstruments.com) are an FAA certified repair station for pressure instruments. We have a mercury barometer that is accurate to about 10 ft if you can eyeball the bottom of the meniscus against the scale with no parallax. The elevation outside on the taxiway at ZPH is 90 feet, so if I go crank 2992 into one of the altimeters it should read about 90 ft, if the conditions are close to 'standard'. A big 1 inch test tube 30 inches tall full of mercury. Cool stuff.

Imagecraft compiler user

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

When I change to expand the same amount?

JChristoff
Illinois

Last Edited: Thu. Jan 6, 2011 - 06:46 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

That's easy to calculate. Just use the pressure to altidude equation for ICAO standard atmosphere as I mentioned earlier. Can't find the equation??? see Intersema application notes.

You will not achieve the accuracy that you want using the MPX4115.
Each sensor will be different as the tolerance is quite high at +/-1.5%.
You would need a very well designed (instrument level) analogue circuit, and in any case you don't have the calibration equipment which costs $10000's.

So easiest thing for you to do, is to buy a digital senor pre-calibrated. Choose the Intersema sensor and you will be impressed!. No calibration to do...no analogue circuits. Hook it up to your SPI and your done!
You can get 1ft resolution (just).

In reality your accuracy will be limited by local fluctuations in air pressure and fluctations in QFE. ). There is no point in having a resolution of 0.58 ft.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Okay, the design progresses with the Sema chip, a real-time clock which also provides the 32k reference to the sema, a 512kbit eeprom for logging, and a mega32 simply for the code space. I'll run that on internal clock, and put a time crystal on the rtc.

All this will run on 3v, but the lcd wants five (a serial lascar 128*64, to be used without the backlight), which is rather a shame... I think the easiest approach is to use a level converter between the sema and everything else, but that of course leaves the problem of power supplies.

I'm thinking that a battery (2*AA or AAA cell) will provide the 3v side - thus giving the cleanest supply to the pressure sensor - and a switch-mode step up to get the 5v side.

Seem reasonable?

Neil

p.s. thinks - I wonder if the LCD will work at 3v? Hmmm

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Sounds reasonable.
Perhaps step up, then regulate down for AVR (3 volts) and MS5534.
You need to calculate which is the most efficient.

You also need to ensure absolute minimum ripple voltage to the MS5534 in order to achieve the best results. Could use an LC filter and/or a large low ESR caspacitor such as Sanyo.

Aso consider that 2*AA batteries will not give you 3 volts for very long. Their voltage will drop over the life cycle.

MS5534 does not have to be 32KHZ, you can drive it to the nearest (lower) frequency using the output of a timer.

Do pay carfull attention to the timing diagrams for MS5534 data aquisition sequences Which is a potential falling point

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Powertip manufactures LCD's that work at 3V. Take a look at PE12864LRU-004.

Guillem.

Guillem.
"Common sense is the least common of the senses" Anonymous.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

ka7ehk wrote:
GPS units are not generally very good on altitude. It improves as the number of received sats increases, but it is genrally a magnitude worse than surface coordinates.

Jim


Why is that?
My understanding was that you need 3 sats if youa are at a known altitude, and 4 sats otherwise. As long as you have found 4 I can't see why altitude should present any problems.

Four legs good, two legs bad, three legs stable.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Be careful, the MS5534 is very sensitive to noise on its VCC, if you're using a step up DC-DC converter then you may have problems filtering out the ripples. Our device uses a separate micro reg. for the barometer and we no longer use a step up DC-DC.

GPS needs 4 satellites to get a 2D fix, the fourth one is required to synchronise the receiver clock with the satellites' clocks. I never did understand how they did that until I read this excellent tutorial on the Trimble website http://www.trimble.com/gps/why.html!

Cheers,

Tim.

------------------------------------------------------------------
www.activeflightsystems.com

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:
GPS needs 4 satellites to get a 2D fix, the fourth one is required to synchronise the receiver clock with the satellites' clocks.

I beg to differ. At sea level, i.e. a known altitude, three satellites are sufficient for a fix.

Four legs good, two legs bad, three legs stable.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Ok you're right, if the altitude is precisely known but the receiver doesn't know its altitude. Here I'm at 70m above sea level so my 2D fix would be wrong if the GPS assumed I was at sea level.

Having said that, apparently (in other words, I've just read...;-) ) most receivers will fix using 3 satellites by making assumptions about the altitude but if I understand correctly they need either 4 satellites or to know their altitude accurately to get a precise fix.

...I see you're in Surrey, are you now going to tell me you're working on the Gallileo system? If so I'll shut up right now ;-)!!

Cheers,

Tim.

------------------------------------------------------------------
www.activeflightsystems.com

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Ok, so we've gone bit off topic with the GPS thing, but I'll add my 2 cents worth:

The first satellite is used only as a time referenence, since GPS receivers do not have an atomic clock.

AFAIK. You receive a signal from all satellites simultaniously as all satellites use the same frequency). The received pseudo random noise signal is sampled and logged by the receiver.

Each satellite transmits a pseudo random noise (PRN) signal pattern, periodically (timed by an atomic clock). The time difference between the beggining of the logged signal and the received signal for any particular satelite is determined by a time shift correlation of the PRN reference pattern for that satellite.

So, 2 satellites give you one time difference measurement and 1 hyperboloid.
3 satellites give you two time difference measurements and 2 hyperboloids.
4 satellites give you three time difference measurements and 3 hyperboloids.

To determine position in 3 dimensions you need a minimum of three hyperboloids....i.e. 4 satellites, but you can substitute the curvature of the earth and use it for the third hyperboloid.

Anyway, my point is that you will always need one more satellite than perhaps you might imagine. Hence 3 satellites are required (+ the earth hyperboloid) for position and 4 satellites for position and altitude.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

looks good

JChristoff
Illinois

Last Edited: Fri. Jan 7, 2011 - 05:06 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hi Jeff,

We use the MS5534 in a 3.3v system anyway, in fact it's connected to an ARM not an AVR at all, although we use an AVR in our sensor box which transmits data to our display unit where the MS5534 is located.

I would guess you'll have to use a level shifter (MAX3370 for example) but I'm not best qualified to answer this as I'm not really an electronics person!

Cheers,

Tim.

------------------------------------------------------------------
www.activeflightsystems.com

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

My tentative circuit diagram contains an MC14050 buffer which can take inputs over its positive rail. Three sixths of it between the CPU and the sensor chip, and one more between a DS1307 real-time clock chip and the sample clock input. All the rest at five volts.

I'm still torn between switch-mode step up from three volts - better power density and efficiency than linear regulators from say 6 or 9 volts, but the potential for noise on the battery line (even though it should be low output impedance.)

Neil

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Maybe if I can find a diode with a 1.5V~2.2V forward drop I could use it and a resistor for each of the three lines

JChristoff
Illinois

Last Edited: Thu. Jan 6, 2011 - 06:44 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Well, the good news is that my prototype shows signs of life, but I'm having problems with the pressure sensor MS5534...

Having hacked the code that Sema provide, I get the four calibration words and the two pressure sensor words (resetting before each conversion) and I'm clocking the 5534 with a 32K square wave from OC2...

Here's what I get when the data words are translated into the six calibration values:

c1 = 10418
c2 = 1078
c3 = 460
c4 = 381
c5 = 882
c6 = 31

and the pressure and temperature readings d1 and d2 are:

d1 = 14965
d2 = 26474

Unfortunately, when I plug those into the magic formula that sema provides, I get:

calibration temp UT1 = 27280
dt = -806
temp = 136 <-- 13.6 degrees, probably about right
offset = 4337
sensitivity = 34631
x =12143
pressure = 62946 <-- 629.46mb - not unless I've moved to the top of the alps...

I'm totally confused. When I plug in the sample numbers Sema provide, it all works beautifully; but this looks like the pressure sensor's confused...

I'll try my spare sensor, but... anyone got any ideas?

Neil

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hey Neil,

Sorry I can't help with this device but keep us posted on how well it works, I'm realy interested.
Is it sensitive to vibration or light?

JChristoff
Illinois

Last Edited: Thu. Jan 6, 2011 - 06:29 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

It's light sensitive, but it only seems to affect the temperature reading, not the pressure reading. A dark red translucent case seems to make the temperature behave... the data sheet warns of this.

Neil

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Would anyone let me know where I can buy this sensor (5534) please? I'm in the US and would like a place where I can order online. I don't see order page from either Servoflo or Intersema

Thanks in advance

Thanh

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

try nickbelsondesign.co.uk

cheers,
gussy

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Thanks very much for the info Gussy

-Thanh

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

There's a place in MA that handles US sales: http://www.intersema.ch/site/comp_contact.php

Neil

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Thanks Neil. I saw that from Intersema website but they didn't offer web ordering. I just called them up and they offered phone order.

Was the problem with sensor reading caused solely by light interference?

Thanks

Thanh

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Light interference give an offset error; Sema warn of noise of about .4 millibar, which needs averaging. I'm seeing a little more noise than that, but I think it's due to a noisy switch-mode power supply putting glitches on the 3v rail. I haven't yet had time to confirm that.

Neil

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hi,

Accelerometer prob not a good way to go since you'd need a 3 axis device to detect ALL altitude changes (unless you use gimbal to ensure the vertical axis always good) also

Quote:
If you're just interested in "simple" up or down, why not just use an accelerometer and skip all the pressure calibration problems ?
- Jesper

surely ther'd be calibration for accelerometer also.

I reckon pressure is the way to go.

IRO pressure-altitude conversion --> afaik the relationship is exponential/logarithmic but i don't see why it can't be linearised if you don't need a lot of range (maybe a quadratic fit would be good also)

Here's something tho: surely as humidity increases the mass of the air would increase, resulting in increased pressure and a falsely increased altitude? hmm.. i wonder how strong the effect is (and if my logic is correct :P )
Also you'd prob have to sheidl the sensor from wind

Sahil

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

There is a complete solution available,
http://www.aerialrobotics.eu/alt...
Solving all the issues with calibration and temperature drift is far fromtrivial.
You can contact me for details:
kbosak@box43.pl

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I produce two altimeters for use with high powered rockets.

The Freescale MPX6115 (current version of the 4115) would easily fit your requirements. If you limit your altitude (for instance at 10000 ft) you can offset and amplify the MPX6115 output so 0-5v scales to 0 to 10000 ft and get about 7ft resolution with a 10 bit ADC. With a little noise on the output and oversampling you can get this down to 2-3 ft resolution.

The MPX6115 is temperature compensated for pressure. When you calculate altitude from the standard atmospheric model there may be an error if the actual air temperature is different than what is assumed in the model. For that reason you should also include a temperature sensor so that you can better calculate altitude. If you are more concerned about short term altitude changes then this temperature correction is not as important.

Sorry about the long post, bottom line is that the MPX6115 sensor and the AVR 10bit ADC can meet your needs IMO.

--john

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Here are the three digital barometric sensors I know of (I2C/SPI)

SCP1000
MS5534
MPL115A

I wonder if these or an MPX4115->opamp->ADC approach would be more accurate? I guess it comes down to which has less noise, a discrete parts chain (sensor/opamp/ADC) or the whole chain built into the die of the digital chip.

Pages