Path: ...!eternal-september.org!feeder3.eternal-september.org!news.eternal-september.org!.POSTED!not-for-mail
From: John B. <slocombjb@gmail.com>
Newsgroups: rec.bicycles.tech
Subject: Re: Garmin altitude problems
Date: Sat, 10 Aug 2024 07:02:48 +0700
Organization: A noiseless patient Spider
Lines: 161
Message-ID: <abbdbjhcae4uf1fqgd7df8ioof8rdd0ssu@4ax.com>
References: <HAOsO.12050$MJMb.1498@fx43.iad> <6he7bjp0qtkrif3t3n72q46tk0peot758q@4ax.com> <3Y0tO.41962$nLcd.12608@fx09.ams4> <gbn9bj9g6lg7bmfjcnr0pheah6b8nsjog5@4ax.com> <3aqabj58cc9e2fkintbin8c2v4kqqo0fhp@4ax.com> <d5icbjpe7men8nrimh37lu9cd3hnsvh7d8@4ax.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 8bit
Injection-Date: Sat, 10 Aug 2024 02:02:49 +0200 (CEST)
Injection-Info: dont-email.me; posting-host="6b484413bd7a29261a57ae7481e6e511";
	logging-data="269277"; mail-complaints-to="abuse@eternal-september.org";	posting-account="U2FsdGVkX19RaGIpvZpACeAYw21apRjTUyRdUeBWFVE="
User-Agent: ForteAgent/7.10.32.1212
Cancel-Lock: sha1:/6IM4330wgjXQCpm15zZA9Fpib4=
Bytes: 8449

On Fri, 09 Aug 2024 10:44:09 -0700, Jeff Liebermann <jeffl@cruzio.com>
wrote:

>On Fri, 09 Aug 2024 08:14:59 +0700, John B. <slocombjb@gmail.com>
>wrote:
>
>>On Thu, 08 Aug 2024 08:22:35 -0700, Jeff Liebermann <jeffl@cruzio.com>
>>wrote:
>>
>>>On Thu, 08 Aug 2024 10:18:39 GMT, Roger Merriman <roger@sarlet.com>
>>>wrote:
>>>
>>>>Jeff Liebermann <jeffl@cruzio.com> wrote:
>>>>> On Wed, 07 Aug 2024 17:57:59 GMT, Tom Kunich <cyclintom@yahoo.com>
>>>>> wrote:
>>>>> 
>>>>>> The formula for change in altitude with air pressure in the lower
>>>>>> atmosphere (about to 35,000 feet) is:
>>>>> 
>>>>> Nope.  From what's left of your formula, I'll assume you're trying to
>>>>> calculate the air pressure at a given altitude.  There is no *change*
>>>>> in altitude involved.  Looks like you lost all the Greek letter
>>>>> symbols and formula symbols.  
>>>>> 
>>>>>> P=Pb[1
>>>>>> Where:
>>>>>> 
>>>>>> 
>>>>>> Pb
>>>>>> TM,b
>>>>>> LM,b
>>>>>> h
>>>>>> hb
>>>>>> R
>>>>>> g0
>>>>>> M
>>>>> 
>>>>> That unreadable mess should look something like these:
>>>>> 
>>>>> "Atmospheric Pressure Calculator"
>>>>> <https://calculator.academy/atmospheric-pressure-calculator/>
>>>>> 
>>>>> "Air Pressure at Altitude Calculator"
>>>>> <https://www.omnicalculator.com/physics/air-pressure-at-altitude>
>>>>> 
>>>>> Note that you need to know the barometric base pressure (Pb) at sea
>>>>> level for such a calculation to work.  It's NOT going to work if your
>>>>> riding up or down a hill where the barometric pressure might change.
>>> 
>>>>In fairness the 830 recalculates on start up every so often particularly if
>>>>it’s a changed location or it may well do so every time but don’t watch it
>>>>start up in general.
>>>>
>>>>Roger Merriman
>>>
>>>Garmin works in mysterious ways.  Garmin's self calibration algorithm
>>>is a bit stranger than one might expect.  For example:
>>><https://forums.garmin.com/sports-fitness/cycling/f/edge-830/174850/elevation-auto-calibration>
>>>"Every time I start to record an activity on my Edge 830 I get this
>>>message "Elevation calibrated to location" and immediately the
>>>elevation number goes up by +/-12m out of the blue with me standing
>>>still in the same spot.
>>>The problem is that the elevation shown right before I press the
>>>record button (before the "auto-calibration) is more accurate than the
>>>one modified."
>>>"The value is coming from the saved location that is near your
>>>starting point."
>>>
>>>I seem to recall reading somewhere that Garmin recommends calibration
>>>before starting on a ride.  I'm too lazy to find where Garmin said
>>>that.
>>>
>>>However, that message is over 5 years old and the problem might have
>>>been fixed long ago with a firmware update.
>>>
>>>The barometric altimeter accuracy in the Garmin Edge 830 is specified
>>>at +/-50 ft accuracy (which is better than the +/-400 ft accuracy for
>>>the GPS).  There's also a +/-10 ft error for reasons unknown:
>>><https://support.garmin.com/en-US/?faq=sFMkjQFdnZ99DcunfAue66>
>>>"Elevation accuracy of +/-10 meters is for any given reading during an
>>>activity, not the total elevation gain/loss at the end of an
>>>activity."
>>>
>>>Very mysterious methinks.
>>
>>Just how do these garmin gizmos work? Back when I worked on airplanes
>>the Altimeter measured the altitude above the home field. During
>>preflight, before engine start, the altimeter was set to "zero" and
>>during that flight measured height above the location where it was
>>set.
>>
>>I believe in more modern times one calls the "Tower" and they give you
>>the altitude above sea level corrected by current temperature.
>>
>>Where does the Garmin gets it base altitude from or is "0" simply the
>>altitude, and temperature,  when/where you turn it on?
>
>A barometric altimeter works in a similar, but not identical way.  For
>Garmin, you calibrate the altimeter from a known altitude location
>from a designated reference.  If your starting location is known to be
>100 ft AMSL (above mean sea level), you set your Garmin for 100ft and
>you're done.  If you don't have an accurate map, using the GPS
>altitude, averaged over a long period will suffice, but you need to
>enter the datum used to define zero altitude used by the GPS.  Notice
>that you don't enter the barometric pressure to make this work.  For
>the Garmin 830, in manual mode, it's simply set and ride:
><https://www8.garmin.com/manuals/webhelp/edge830/EN-US/GUID-1833C7CC-667E-48C7-B7D2-7A6041470478.html>
>
>On later devices, Garmin uses the GPS altitude and DEM (digital
>elevation model) to calibrate the barometer.  This is for the Garmin
>Forerunner 965 watch:
><https://www8.garmin.com/manuals/webhelp/GUID-0221611A-992D-495E-8DED-1DD448F7A066/EN-US/GUID-BC734846-01A7-4F33-86D4-DFBDBC06CDB4.html>
>Again, notice that barometric pressure is not involved in establishing
>a reference altitude.
>
>Garmin also has an "auto cal" mode, which does all this automagically.
>I couldn't find an explanation on how it works from Garmin.  My
>guess(tm) is that it uses the nearest GPS reference available where
>the bicycle was not moving and the DOP (dilution of precision) is
>sufficiently low for the GPS altitude to be considered accurate.
><https://en.wikipedia.org/wiki/Dilution_of_precision_(navigation)>
>Again, notice that barometric pressure is not involved in establishing
>a reference altitude.
>
>Once a reference altitude is established, the barometric pressure
>becomes involved.  The math is easy enough:
>"Atmospheric Pressure Calculator"
><https://calculator.academy/atmospheric-pressure-calculator/>
>"Air Pressure at Altitude Calculator"
><https://www.omnicalculator.com/physics/air-pressure-at-altitude>
>For faster response, Garmin might use a lookup table to convert air
>pressure to altitude.
>
>However, barometric pressure change with weather, location and
>environment.  The MEMS pressure sensor might also drift.  Therefore,
>the altitude must be recalibrated occasionally.  My guess(tm) is that
>Garmin uses the nearest known GPS location, with the lowest DOP, to
>establish a new reference.  If the new reference altitude is not quite
>perfect, the rider sees a "glitch" or discontinuity in the elevation
>profile graph when the reference is reset.  Garmin probably tries to
>"smooth" this transition, making the graph look more reasonable.
>
>I was wondering why Garmin elevation accuracy specification for the
>Edge 830 was +/-400 ft.  GPS elevation accuracy should be about +/-120
>ft for most GPS's.  Some possible causes are:
>1.  The EDGE 830 was turned off at night and not given sufficient time
>to download a few days worth of ephemeris data.  Garmin advised to put
>the GPS somewhere with a good view of the sky and let it download for
>about 15(?) mins.
>2.  The GPS antenna on the 830 is probably tiny.  That will reduce
>signal strength and therefore accuracy.
>3.  The Sony GPS receiver on the 830 uses GPS and *EITHER* Glonass or
>Galileo.  It really should be using all three to see more satellites.

The basic system sounds much like my old airplane altimeter. But how
is atmospheric pressure due to temperature handled?
-- 
Cheers,

John B.