Bicycle Computer Altimeter Accuracy

[Artoi]

In article
<tedbennett-F2FCF5.12520517092007@earthlink.vsrv-sjc.supernews.net>,
Ted Bennett <[email protected]> wrote:

> In my experience those altimeters are fairly useless. Since they figure
> altitude by measuring air pressure fluctuations, which changes according
> to temperature as well as altitude, there is built-in inaccuracy. Add
> to that the cheapness of the instrument and you're bound to get the sort
> of "measurements" you did.
>
> If you want more accuracy, get a GPS which measures your altitude by
> triangulation from satellites rather than air pressure. Usually within
> one meter.

The truth is that the barometric altimeters are more sensitive and is
good for small and rapid changes in elevation. While GPS is good for the
X/Y coordinate but less precise in the vertical plane. Those Garmin
units eg. Edge 305, uses the barometric meter for their altitude data
but they constantly get calibrated by the data stream from the GPS
system. In other words, the GPS data significantly reduces the inherent
errors in a barometric unit.
--

 

[Mark]

KF wrote:
> On Sep 17, 3:01 pm, "Clive George" <[email protected]> wrote:
>> T'other way round - the barometric is more accurate. At a guess, a plane
>> doesn't really care that much about a 10m altitude difference, until it gets
>> close enough to the ground that the pilot can work that bit out. (fx: reads
>> Garmin website. It appears they don't use GPS for altitude unless the main
>> barometric one has died).
>
> Not necessarily so, but let's at least define what we mean by
> "accurate."
> Take a GPS altimeter reading. With a decent "view" of the sky, it's
> almost always within 20 feet or so of the correct altitude and NEVER
> varies as the barometer fluctuates. So if you don't know where you are
> and don't know what the barometric trend has been, the GPS
> gives far more reliable elevation information.
>
> On the other hand if you have just calibrated your barometric
> altimeter and you climb a small hill (say, 200 feet high) the
> barometric altimeter wil be able to tell you quite accurately how high
> the top of the hill is, and even without calibration, can tell you how
> much you have gained (the difference in elevation), probably to within
> a few feet (although not all have that kind of fine resolution). But
> try to measure the top of the hill again tomorrow without
> recalibrating and you could easily be off by a couple hundred feet
> just due to normal atmospheric changes. The GPS will always be
> subject to the +/- 20 feet at any one location but will very rarely be
> off more than that. So while the barometric altimeter under ideal
> conditions may tell you that the hill is 200 +/- 5 feet high, under
> unknown barometric changes it may really only tell you that the hill
> is 200 +/- 200 feet high.
> And if we change the hill to 2000 feet and still use the same +/-
> figures, suddenly the +/-20 feet of the GPS starts to look VERY
> accurate.
>
> My experience from a number of years of use of both types of
> instruments is that I'll take the overall consistency/accuracy of the
> GPS over the barometric altimeter any day, in fact the GPS is about
> the only way to conveniently reset a baromeric altimeter after unknown
> changes in elevation/barometric pressure.

Heh, heh, my barometric altimeter watch is a really cheap one. I'm
almost certain I see abrupt changes in the reported altitude due to the
pressure wave of passing trucks. It's still a fun toy, considering it
was cheap, and it's usually not off by too much over the course of a day.

Mark J.

 

[[email protected]]

On Sep 17, 3:41 pm, "[email protected]" <[email protected]>
wrote:

> In the OP's case, the readings might be different if e.g. the weather
> changed, or he started off very early in the morning when it was
> cold at the base, or something like that. It's easy to check a
> single climb against a topo map, but harder if the computer's
> number is "total climbing" from cumulating all the small ups and
> downs.

In certain circumstances, a bicycle power meter has greater resolution
in measuring elevation change than either the consumer level
altimeters or GPS receivers that one finds on bicycles.

 

[Michael Press]

In article <[email protected]>,
"David L. Johnson" <[email protected]> wrote:

> Clive George wrote:
> > "Ted Bennett" <[email protected]> wrote in message
>
> >> If you want more accuracy, get a GPS which measures your altitude by
> >> triangulation from satellites rather than air pressure. Usually within
> >> one meter.
> >
> > That would be why the GPSs which record height have a barometric
> > altimeter on them as well?
>
> A GPS uses the barometer primarily, with the satellites used as a
> periodic check, since the satellite data is not all that accurate for
> altitude, but the barometer is subject to drift due to atmospheric
> conditions.

Strictly speaking, drift in a measuring instrument is a
change in its ability to make the primary measurement
caused by changes in the instrument itself. With
barometric altimeters it is interpretation of the
measurement that is the problem, not drift in their
measurement of ambient air pressure.

--
Michael Press

 

[[email protected]]

On Sep 17, 10:40 pm, [email protected] wrote:
> On Sep 17, 3:41 pm, "[email protected]" <[email protected]>
> wrote:
>
> > In the OP's case, the readings might be different if e.g. the weather
> > changed, or he started off very early in the morning when it was
> > cold at the base, or something like that. It's easy to check a
> > single climb against a topo map, but harder if the computer's
> > number is "total climbing" from cumulating all the small ups and
> > downs.
>
> In certain circumstances, a bicycle power meter has greater resolution
> in measuring elevation change than either the consumer level
> altimeters or GPS receivers that one finds on bicycles.

Yesbut, resolution is not the same as accuracy, as you
know. It may be accurate under certain circumstances,
but in the mountains, there is often a wind blowing.

For real accuracy, any metrology should be calibrated
against a standard. Topo maps are pretty good for this
application.

Ben
measure once, publish twice

 

[Ron Hardin]

!Jones wrote:
>
> On Mon, 17 Sep 2007 15:34:42 -0400, in rec.bicycles.tech Ron Hardin
> <[email protected]> wrote:
>
> >You'd like the up and down totals to agree, though.
>
> Personally, I prefer the downhill... I'm not into deferred
> gratification.
>
> Jones

Actually I prefer uphills. That's the change that 10 speed gearing brought
years ago.

From a strict physics point of view, you waste the least energy on uphills;
any work you do winds up saved as potential energy.

Then you fritter it all away on air turbulence on downhills, which are the most
wasteful of energy.
--
[email protected]

On the internet, nobody knows you're a jerk.

 

[Ron Hardin]

Barometric is subject to wind effects ; if you open the window of a light
airplane, the altimeter jumps a hundred feet.

The same thing happens without you noticing on the ground - whether you're
in the shelter of the mountain and so out of the wind (but therefore at
a lower than natural atmospheric pressure) or on the upwind side and
therefore in the wind, the wind produces local highs and lows as it goes
around natural obstructions, large and small.

If there's no wind at all, you can think of the entire atmospheric systems
as the same effect, except you get a correction to set for those, over the
radio or something. But wind is what creates and sustains pressure changes,
be they continential-large or tree trunk small.
--
[email protected]

On the internet, nobody knows you're a jerk.

 

[_]

On Mon, 17 Sep 2007 12:52:05 -0700, Ted Bennett wrote:

>
> In my experience those altimeters are fairly useless. Since they figure
> altitude by measuring air pressure fluctuations, which changes according
> to temperature as well as altitude, there is built-in inaccuracy. Add
> to that the cheapness of the instrument and you're bound to get the sort
> of "measurements" you did.
>

I seem to remember JB (the one who bikes, not the one who blusters) posting
about the now-defunct Avocet gadget that had an altimiter in it; the units
were designed with a one-time programmable adjustment so that they could be
tweaked for accuracy after manufacture.

 

[Clive George]

<[email protected]> wrote in message
news:[email protected]...

> For real accuracy, any metrology should be calibrated
> against a standard. Topo maps are pretty good for this
> application.

Heh - which reminds me of of the bemusement I had when I first heard the
phrase "Topo map" used. Having been brought up with OS maps, I thought all
sensible maps would have contours. Little did I know how spoilt I was...

cheers,
clive

 

[Dorfus Dippintush]

Ron Hardin wrote:
> Barometric is subject to wind effects ; if you open the window of a light
> airplane, the altimeter jumps a hundred feet.
>
> The same thing happens without you noticing on the ground - whether you're
> in the shelter of the mountain and so out of the wind (but therefore at
> a lower than natural atmospheric pressure) or on the upwind side and
> therefore in the wind, the wind produces local highs and lows as it goes
> around natural obstructions, large and small.
>
> If there's no wind at all, you can think of the entire atmospheric systems
> as the same effect, except you get a correction to set for those, over the
> radio or something. But wind is what creates and sustains pressure changes,
> be they continential-large or tree trunk small.


On a bike it doesn't make much difference but it does if I'm moving in
my car. Winding down the window of my car makes a change to my GPS
indicated altitude, by a few metres.

Dorfus

 

[Booker Bense]

In article <[email protected]>,
David L. Johnson <[email protected]> wrote:
>Ted Bennett wrote:
>> "David L. Johnson" <[email protected]> wrote:
>
>>> A GPS uses the barometer primarily, with the satellites used as a
>>> periodic check, since the satellite data is not all that accurate for
>>> altitude, but the barometer is subject to drift due to atmospheric
>>> conditions.
>>
>> Hmm. That differs from my understanding, which results from my
>> conversations with pilots. But I'm willing to be corrected, as pilots
>> are likely not to be GPS experts. Can anyone else who is knowledgeable
>> chime in?
>>
>I got this from Garmin, IIRC.
>

The garmin units definite use a barometer for relative altitude
and use the GPS to calibrate. Same with the etrex summit, I think
there is a big difference between GPS in a plane where you pretty
much have clear signal all the time and GPS 4 feet from the
ground were the signal can be blocked and lost quite often.
This leads to a lot of jitter in the signal and makes GPS
attitude cumlative measurements less accurate. Basically the
combo of the two improves the accuracy of both.

The Garmin Edge also uses a traditional bike speedo for cadence
and speed/distance correction for exactly the same reason.

I have noticed that the altitude measurement gets more accurate
and consistant as the software of the unit gets upgraded.

_ Booker C. Bense

 

[Lou Holtman]

Ted Bennett wrote:
> "Clive George" <[email protected]> wrote:
>
>> "Ted Bennett" <[email protected]> wrote
>>
>>>> That would be why the GPSs which record height have a barometric
>>>> altimeter on them as well?
>>> Some do it by barometric pressure measurement, actually most do.
>>> However units used in aviation use the satellites for better accuracy.
>>> Hardly practical for bicycle use, I would agree.
>> T'other way round - the barometric is more accurate. At a guess, a plane
>> doesn't really care that much about a 10m altitude difference, until it gets
>> close enough to the ground that the pilot can work that bit out. (fx: reads
>> Garmin website. It appears they don't use GPS for altitude unless the main
>> barometric one has died).
>>
>> cheers,
>> clive
>
> This may be nearly unheard of on this group: I was wrong.
>
> Shows where repeating hearsay can lead one astray. Thanks.
>
> Ted
>


Anyone who recorded and displayed an altitude graph using a GPS in a
fast and curvy descent could tell you.

Lou
--
Posted by news://news.nb.nu (http://www.nb.nu)

 

[Doc O'Leary]

In article <[email protected]>,
Colin Campbell <[email protected]> wrote:

> Yesterday, I did one of Southern California's classic mountain rides -
> Glendora Mountain Road. According to my bike computer, I did 1298
> meters of climbing, and 1541 meters of descending (ending up exactly
> where I started).
>
> Is there any way of guessing which of these numbers might be more accurate?

Flip a coin if you want to guess. If you want to *know*, go an online
map (and/or Google Earth) and measure the route.

--
My personal UDP list: 127.0.0.1, 4ax.com, buzzardnews.com, googlegroups.com,
heapnode.com, localhost, teranews.com, x-privat.org

 

[Booker Bense]

In article <droleary.usenet-BAC011.13471918092007@sn-ip.vsrv-sjc.supernews.net>,
Doc O'Leary <[email protected]> wrote:
>In article <[email protected]>,
> Colin Campbell <[email protected]> wrote:
>
>> Yesterday, I did one of Southern California's classic mountain rides -
>> Glendora Mountain Road. According to my bike computer, I did 1298
>> meters of climbing, and 1541 meters of descending (ending up exactly
>> where I started).
>>
>> Is there any way of guessing which of these numbers might be more accurate?
>
>Flip a coin if you want to guess. If you want to *know*, go an online
>map (and/or Google Earth) and measure the route.
>

Most topo programs provide even less accurate results that even
the worst bike altimeter.

_ Booker C. Bense

 

[[email protected]]

On Sep 17, 11:30 pm, "[email protected]" <[email protected]>
wrote:

> > In certain circumstances, a bicycle power meter has greater resolution
> > in measuring elevation change than either the consumer level
> > altimeters or GPS receivers that one finds on bicycles.
>
> Yesbut, resolution is not the same as accuracy, as you
> know. It may be accurate under certain circumstances,
> but in the mountains, there is often a wind blowing.

Right, but the reason why most of us want to know the elevation gain
is because elevation gain has come to be a proxy measure for effort.

FWIW, take a look at this:
http://anonymous.coward.free.fr/wattage/cda/indirect-cda.pdf
It's an old version and I've done a bit more with it since then, but
notice that I picked out a drainage grate on the edge of a "flat"
road. I'd appreciate your comments.

 

[[email protected]]

On Sep 18, 12:04 pm, [email protected] (Booker Bense)
wrote:
> Doc O'Leary <[email protected]> wrote:

> >Flip a coin if you want to guess. If you want to *know*, go an online
> >map (and/or Google Earth) and measure the route.
>
> Most topo programs provide even less accurate results that even
> the worst bike altimeter.

IMO, topo maps provide very accurate measurements of
start and finish altitudes, but topo programs that compute
cumulative elevation gain are quite unreliable. This is
likely because they model both the road and the elevation
contours as a bunch of line segments and introduce noise
into the elevation profile. When I look at a profile from one
of these programs, it's often much jaggier than any real
road would be.

Ben

 

[Arthur Shapiro]

In article <[email protected]>, Colin Campbell <[email protected]> wrote:
>Yesterday, I did one of Southern California's classic mountain rides -
>Glendora Mountain Road. According to my bike computer, I did 1298
>meters of climbing, and 1541 meters of descending (ending up exactly
>where I started).

For what it's worth, when I do GMR starting at Boulder Springs Road (the one
stop light, a convenient place to park), my Cateye ranges between 4860 and
5240 feet of climbing for the ordinary back-and-forth to Mt. Baldy Village. I
wish the results were more consistent.

Art

 

[[email protected]]

someone writes:

>>> In certain circumstances, a bicycle power meter has greater
>>> resolution in measuring elevation change than either the consumer
>>> level altimeters or GPS receivers that one finds on bicycles.

>> Yesbut, resolution is not the same as accuracy, as you know. It
>> may be accurate under certain circumstances, but in the mountains,
>> there is often a wind blowing.

> Right, but the reason why most of us want to know the elevation gain
> is because elevation gain has come to be a proxy measure for effort.

> FWIW, take a look at this:

http://anonymous.coward.free.fr/wattage/cda/indirect-cda.pdf

> It's an old version and I've done a bit more with it since then, but
> notice that I picked out a drainage grate on the edge of a "flat"
> road. I'd appreciate your comments.

One of the most important features in recording altitude gain is
hysteresis in the accumulator. That is to say, if a bicycle is ridden
on a country road that undulates +- up to ten feet in long waves, an
instrument with no hysteresis would accumulate great climbs even
though the rider didn't notice any or gain any elevation.

Realizing this, I submitted a patent on the instrument on which I
worked for Avocet. Besides ignoring atmospheric pressure undulations,
I also wanted RR underpasses to not appear as altitude gains, gains
for which the average bicyclist does not shift into lower gears as one
does to climb mountain roads.

Jobst Brandt

 

[Colin Campbell]

Arthur Shapiro wrote:
> In article <[email protected]>, Colin Campbell <[email protected]> wrote:
>> Yesterday, I did one of Southern California's classic mountain rides -
>> Glendora Mountain Road. According to my bike computer, I did 1298
>> meters of climbing, and 1541 meters of descending (ending up exactly
>> where I started).
>
> For what it's worth, when I do GMR starting at Boulder Springs Road (the one
> stop light, a convenient place to park), my Cateye ranges between 4860 and
> 5240 feet of climbing for the ordinary back-and-forth to Mt. Baldy Village. I
> wish the results were more consistent.
>
> Art
>
Art,
I don't find a Boulder Springs Road using Google Maps. Where is that?

I usually start from the 210 Freeway & Grand in Glendora. North on
Grand to Sierra Madre, east on Sierra Madre to GMR. Occasionally, we'll
start from Costco (W Foothill & Todd in Azusa). This past Sunday, we
started from a fellow club member's place in La Verne (I was mostly lost
at the start).
Colin

 

[[email protected]]

On Tue, 18 Sep 2007 17:46:05 -0700, Colin Campbell
<[email protected]> wrote:

>Arthur Shapiro wrote:
>> In article <[email protected]>, Colin Campbell <[email protected]> wrote:
>>> Yesterday, I did one of Southern California's classic mountain rides -
>>> Glendora Mountain Road. According to my bike computer, I did 1298
>>> meters of climbing, and 1541 meters of descending (ending up exactly
>>> where I started).
>>
>> For what it's worth, when I do GMR starting at Boulder Springs Road (the one
>> stop light, a convenient place to park), my Cateye ranges between 4860 and
>> 5240 feet of climbing for the ordinary back-and-forth to Mt. Baldy Village. I
>> wish the results were more consistent.
>>
>> Art
>>
>Art,
>I don't find a Boulder Springs Road using Google Maps. Where is that?
>
>I usually start from the 210 Freeway & Grand in Glendora. North on
>Grand to Sierra Madre, east on Sierra Madre to GMR. Occasionally, we'll
>start from Costco (W Foothill & Todd in Azusa). This past Sunday, we
>started from a fellow club member's place in La Verne (I was mostly lost
>at the start).
>Colin

Dear Colin,

http://yellowpages.superpages.com/d...lay_mode=map&route_type=fastest&ROUTE_UNITS=2

Cheers,

Carl Fogel