Which Sensor-Chip(s) are built-in in the Pyra Mobile edition?

[tarator]

@slaeshjag gave me the link to some Sensor-datasheets which were discussed on the Kernel-Mailinglist, for the Compass, humidity, pressure and temperature sensors
Which of these chips are finally soldered on the pyra?

 

[levi]

Depends on whether you went for the mobile unit or not I think. The SE units have:
Wifi, microphone, touchscreen, keyboard.

The mobile unit has that and adds:
GSM, GPS, barometer, hygrometer, accelerometer, gyro.

I think that's right and about the scope of it.

Edit: Probably also compass

 

[tarator]

Yes, I got preordered the mobile edition. But which of these chips are soldered on the board:

From http://p.rdw.se/pyra-sensor-datasheets.txt:

On Sun, May 29, 2016 at 07:33:07AM +0200, H. Nikolaus Schaller wrote:
> I think for the sensors they are even public:
>
> https://ae-bst.resource.bosch.com/media/_tech/media/datasheets/BST-BMC150-DS000-04.pdf
> https://ae-bst.resource.bosch.com/media/_tech/media/datasheets/BST-BMG160-DS000-09.pdf
> https://ae-bst.resource.bosch.com/media/_tech/media/datasheets/BST-BME280_DS001-11.pdf
> https://ae-bst.resource.bosch.com/media/_tech/media/datasheets/BST_BNO055_DS000_13.pdf
>
> There is a user-space driver for the BNO055 but some iio kernel driver would be better:
> https://github.com/BoschSensortec/BNO055_driver

 

[tarator]

Regarding the Pyra Wiki [1] there are three sensors soldered on the board.
Is there a reason, why there are two pressure sensors?

* The BME280[2] can measure: relative humidity, temperature and pressure
* The BMP180[3] can measure: pressure
* The BO055[4] is for orientation purposes


The wiki also states, that there is a gas sensor, but none of the three mentioned sensors in the wiki is capable of doing this.
Is this simply an error in the wiki?

[1] https://pyra-handheld.com/wiki/index.php?title=Sensors
[2] https://www.bosch-sensortec.com/bst/products/all_products/bme280
[3] https://www.bosch-sensortec.com/bst/products/all_products/bmp180
[4] https://www.bosch-sensortec.com/bst/products/all_products/bno055

 

[Pocak]

edit: I guess I should be clearer on this: the Pyra won't ship with redundant sensors. It's just that the PCB has some extra "slots," so that if the preferred chips become unavailable, production can continue with an alternative without needing a change to the PCB.

The wiki is just out of date. We now have an additional data point: the schematic drawing.
It shows that the main PCB has place for the following sensors:

• BMG160 (measures rotation)
• BMC150 (acceleration & magnetic field)
• BMP180 (pressure & temperature)
• BMP280 (pressure & temperature)

• BNO055 (rotation & acceleration & magnetic field)
• BME280 (pressure & humidity & temperature)

Note that some of these might alias physically or in I2C address, eg. it may not be possible to populate both the BMP180 and the BMP280, but then there's no point to that anyway. I guess having alternatives is a hedge against potential issues like lack of drivers, malfunction, or some chips becoming unobtainable. So a future board could have a BNO055 instead of a BMG160+BMC150, and userspace applications would hardly notice the difference.

According to the schematic, the mobile edition is slated to have a BMG160, a BMC150, and a BME280, but I think we should treat that as preliminary. What I think is going to happen is that when it's time to order these parts for mass production, ED is going look at prices, any problems that came up in testing, etc., then decide which chips to get.

As for gas sensing, I guess that's outdated info. Or maybe it's a vocabulary thing and someone thought it's the same as pressure sensing. The about page lists it, too, but then that was written a long time ago. Bosch has the BME680 for volatile gas sensing, but it's larger than any of the above, so it would take a change to the PCB to include it.

 

[tarator]

@pocak thanks for your response. I will update the wiki page regarding the information which is stated here: https://pyra-handheld.com/wiki/index.php?title=Mainboard

 

[Binky]

As for gas sensing, I guess that's outdated info. Or maybe it's a vocabulary thing and someone thought it's the same as pressure sensing.

I don't remember anyone suggesting the Pyra needed to be able identify specific gases in the atmosphere (and besides, the case isn't very well ventilated, so it wouldn't really work anyway).
My bet is that one of our esteemed polyglots (or even google translate, for that matter), got confused by the definition/description of a barometer.

I'm not entirely sure what the use of a barometer is in this case anyway. Estimating altitude? Surely it would be more accurate to use GPS and look up altitude on a topographical map? (GPS altitude readings tend to have a disproportionate uncertainty margin compared to their horizontal resolution). Still, never hurts to have too many sensors...

 

[tarator]

@Binky the barometric sensor could be used for indoor navigation (e.g. change of floor level). I think it can also be used to get a faster GPS fix. (Not sure about that though)
Another use-case would be the utilization of the pyra for a weather station. (Which is awesome in my opinion).

 

[Binky]

By my googling, air pressure variations due to weather conditions are in the order of ±1kPa
air pressure variations due to altitude are in the order of 10Pa/m

Given that tall buildings often have forced ventilation or air-conditioning of some sort, it sounds unlikely that you could reliably work out which floor you're on by pressure alone.

Perhaps it would work if you had a radio beacon at ground level broadcasting a reference pressure. Maybe?

 

[tarator]

@Binky good point. Maybe I'll try it once I have a pyra in my hand.

 

[Silent-Hunter]

I just want an app that displays the old "rain, fair, change" dial.

 

[Binky]

...or how about a virtual one of these:

(Although I think they might hygrometers, some of them)

 

[FBnil]

..or how about a virtual one of these:

Good idea, but I would replace these with a dragon with an umbrella and a mummy in the desert.

like this:

do'h typed in dessert instead of desert...

 

[Pocak]

I'm not entirely sure what the use of a barometer is in this case anyway. Estimating altitude? Surely it would be more accurate to use GPS and look up altitude on a topographical map? (GPS altitude readings tend to have a disproportionate uncertainty margin compared to their horizontal resolution). Still, never hurts to have too many sensors...

Well, the often repeated theory goes like this: barometers are used to measure changes in altitude. As you say, GPS is not very precise vertically, but if you average its readings over an hour, it can give a decent approximation of your absolute altitude. Meanwhile, a barometer can't tell you your absolute altitude, nor long term changes in altitude, as pressure varies with weather. But over a few minutes, you can assume weather to be constant, so a drop in pressure usually corresponds to walking/riding uphills. So GPS-es and barometers compensate nicely each other's weakness, and by carefully combining them you get a better altitude reading.

Personally, I see the barometer most useful for outdoor sports tracking. It will count meters of ascent/descent much more precisely than a topological map or GPS could.

 

[levi]

Given that tall buildings often have forced ventilation or air-conditioning of some sort, it sounds unlikely that you could reliably work out which floor you're on by pressure alone.

You seem to have a better understanding of pressure than I do, but I personally doubt many buildings are airtight enough to resist outside air pressure, or the weight of air between floors which causes lower pressures as you climb. If it were possible to pressurise a building, it might make opening the fire escape quite tricky.

 

[Binky]

You couldn't easily seal a building to maintain a pressure difference with the outside world - but you don't have to. If there's a flow of air (perhaps driven by fans in a climate-control system) then it follows that there must be a pressure difference, much as an electric current implies a potential difference. Lifts (elevators) also force air around buildings like pistons. Even people slamming doors can create sudden gusts in the right conditions, as might wind blowing past open windows, by the venturi effect.

None of these things creates a very big pressure variation, but it does introduce noise to the altitude "signal". If your building has a floor every 3m, and pressure varies with height at 10Pa/m, you only need 15Pa fluctuations to start getting incorrect results. 15Pa isn't very much.

...and that's all assuming you can compensate for the weather. Without compensating for weather, your altimeter would have an uncertainty margin of several tens of floors!

Edit: As pocak says, you might well have better luck measuring changes than absolute values.
[doublepost=1496694503,1496694029][/doublepost]

Personally, I see the barometer most useful for outdoor sports tracking. It will count meters of ascent/descent much more precisely than a topological map or GPS could.

OpenCycleMap (e.g. here: http://opencyclemap.org/?zoom=15&lat=54.60964&lon=-3.22303&layers=B0000) has contour lines every 10m - which seems pretty good.

Edit again: Wind can do odd things in mountainous areas. Is there a meteorologist in the house?
Too many edits: OpenCycleMap gets its topography data from NASA radar surveys, which are public-domain.

 

[Pocak]

Wow, that's pretty good indeed. I was thinking of the crude heightmaps in offline car navigation apps (Sygic and the built-in stuff in cars) when I wrote that; I take my comment back.

BTW, do you know an app that combines horizontal GPS coordinates with topography data?

I figured the Pyra's case would protect the sensor from wind, and a pocket or backpack would help further, but given the tiny pressure differences involved in altitude tracking, I wonder if that's enough. I haven't had a chance to play around with this stuff, as neither of my tablets has a barometer, so I'm just spewing armchair expert stuff here as usual.