Results 1 to 11 of 11
  1. #1
    mudnthebloodnthebeer
    Reputation: quaffimodo's Avatar
    Join Date
    May 2004
    Posts
    587

    305 Altimiter issues

    Any experienced Garmin users know what's going on here? Here's two plots of the same ride. The first is taken directly from the GarminTC software, the second is MotionBased output. The ride began and ended at the same point, but as you can see from the first plot, the altitude measurement was way off. MotionBased got the altitude correct, but consistently credits me for roughly twice the climbing and descending I've actually done. I'd be grateful for any insight that you'd be willing to share.
    Attached Images Attached Images
    Also, the streets are full of horizontal dropouts...

    BSNYC

  2. #2
    Full of holes
    Reputation: SwissBuster's Avatar
    Join Date
    Jan 2004
    Posts
    337
    Quote Originally Posted by quaffimodo
    Any experienced Garmin users know what's going on here?
    At risk of stating the obvious, the Garmin software is rounding the data. Motionbased is measuring the raw data as provided by the GPS. Your GPS altimeter readings are fluctuating, and this is most pronounced on flat sections (like the start of your ride). I've notice this on my Forerunner, and I expect would be the same with any GPS (except perhaps one with a barometric altimeter).

    I use a small program called Trackan which works with OziExplorer tracks. It allows you to put a minimum height gain threshold on your tracks to smooth the line. For example, I usually tell it to ignore any height gain <25m when summing. In your case I'd go with the Garmin total which looks more realistic.

  3. #3
    On MTBR hiatus :(
    Reputation: Speedub.Nate's Avatar
    Join Date
    Dec 2003
    Posts
    9,145
    My guess is that the altimeter never recalibrated at the start of the ride. Did you happen to check your elevation display before starting out?

    The altimeter plot shows you starting at ~1200', but in reality should have been closer to ~700'.

    I assume you climbed to the same peak three times. It looks like the GPS corrected the altimeter once on your way up the first time, and then made a second correction during the first big descent. After that, it tracks seem to line up.

    I doubt that over the course of your 19 mile ride the atmospheric conditions changed significantly enough to cause the initial discrepancy, but it's possible that if your unit was hot from spending a day in the car, that may have accounted for some error.

    I disagree with SwissBuster's opinion that Garmin is rounding the elevation plot. Yeah, to a certain extent, they are rounding (filtering) some noise, but the reason for the barometric altimeter in the 305 is for the smooth, accurate contours like the one displayed in the first image.

    The second image is typical of a GPS-only elevation plot, but it's also typical of a plot you'd receive when charting a route over an electronic topo map (plot your route on a Google Earth-based service, such as MapMyRide.com, and you'll see a very similar -- and inaccurate -- elevation profile). I don't know if Motion Based uses topo points or GPS points to generate the plot you posted, but that "noise" is what is responsible for the exaggerated MB total elevation figures.
    speedub.nate
    MTBR Hiatus UFN

  4. #4
    mudnthebloodnthebeer
    Reputation: quaffimodo's Avatar
    Join Date
    May 2004
    Posts
    587
    Quote Originally Posted by SwissBuster
    At risk of stating the obvious, the Garmin software is rounding the data. Motionbased is measuring the raw data as provided by the GPS. Your GPS altimeter readings are fluctuating, and this is most pronounced on flat sections (like the start of your ride). I've notice this on my Forerunner, and I expect would be the same with any GPS (except perhaps one with a barometric altimeter).

    I use a small program called Trackan which works with OziExplorer tracks. It allows you to put a minimum height gain threshold on your tracks to smooth the line. For example, I usually tell it to ignore any height gain <25m when summing. In your case I'd go with the Garmin total which looks more realistic.
    Thanks. I agree that the Garmin track is more accurate, but my real concern is the fact that the altimiter initially reported the starting altitude at least 500' too high. The atmospheric pressure that day was very close to standard, and had not changed appreciably in the 24 hours prior. I was hoping that there was some way to take the barometric altimiter out of the loop initially. It's really a shame that it can't be adjusted for pressure change-that would have been a simple bit of functionality to add.
    Also, the streets are full of horizontal dropouts...

    BSNYC

  5. #5
    mudnthebloodnthebeer
    Reputation: quaffimodo's Avatar
    Join Date
    May 2004
    Posts
    587
    The elevation displayed on the unit at the beginning of the ride was reflected in the track. It wasn't reading accurately until about halfway down the first descent. I ride that route a lot, so I'm familiar with the elevation of all the major landmarks. The unit did spend some time in the car prior to the ride, but not in direct sunlight-that's a reasonable theory though. It accurately displayed the elevation of my home this morning after spending the night on my desk, so I'll be curious to see how it behaves this afternoon on my ride. Stand by for an update, and thanks to everyone who responded.
    Also, the streets are full of horizontal dropouts...

    BSNYC

  6. #6
    On MTBR hiatus :(
    Reputation: Speedub.Nate's Avatar
    Join Date
    Dec 2003
    Posts
    9,145
    Quote Originally Posted by quaffimodo
    The unit did spend some time in the car prior to the ride, but not in direct sunlight-that's a reasonable theory though.
    Please report back how that plays out.

    These baro units have a temp sensor internally to detect / correct for high internal temps, not necessarily ambient. They cook in a vehicle and retain heat very well.
    speedub.nate
    MTBR Hiatus UFN

  7. #7
    trail rat
    Reputation: slocaus's Avatar
    Join Date
    Jul 2005
    Posts
    7,707
    In your MB account, have you checked to make sure that the MB Gravity feature is on? It is turned off by default. I've found the same thing on some rides uploaded to MB, and turning Gravity on solves that for me.

    Also my experience supports Nate's comment about the barometric altimeter getting calibrated at the start. I live about 12 miles inland from the Pacific Ocean in central CA. We get higher humidity, "marine layer" in the mornings, and then it drops quite a bit as temperature rises. I know the actual elevation of my front yard, thanks to my landlord having some surveys done, and it can read 150 feet off when first powered on in the morning, after being turned off on a warm afternoon.

    I've found that turning my 305 on at least 15 minutes before I begin my ride, so it can calibrate, makes a huge difference in accuracy. It seems to "remember" the readings when it was turned off, including the poor satellite reception in my house (I live in a freakin' cave..... ), and takes a little longer to acquire satellites than if it is powered off in the yard, and on the next day outside the house, uh, cave.
    CCCMB web - Facebook
    SLO trail maps - conditions

    "The physician heals, Nature makes well" - real fortune cookie

  8. #8
    Full of holes
    Reputation: SwissBuster's Avatar
    Join Date
    Jan 2004
    Posts
    337
    Quote Originally Posted by Speedub.Nate
    I disagree with SwissBuster's opinion that Garmin is rounding the elevation plot... The reason for the barometric altimeter in the 305 is for the smooth, accurate contours like the one displayed in the first image.
    I missed the main point in the OP's question; that the starting elevation was out and stayed out for a couple of miles. I also missed the fact that the 305 has a barometric altimeter. It seems logical to me that the higher temps in the car could produce this. I have never seen this issue with my 205.

    Although it would not help in this case, in OziExplorer you can edit single data points so where you get an errant reading, you can delete it from the profile. This sometime happens to me when starting up the device or if it loses the connection.

    To Nate's point - I guess it depends if MB is reading and displaying the same data as the Garmin is. Perhaps the Garmin reads the barometrically smoothed data and MotionBased is uncorrected satelite-derived elevation data only, or perhaps simple topo data (the fact that it was at the right elevation for the first couple of miles suggests it reads topo data). Does anyone know?

  9. #9
    On MTBR hiatus :(
    Reputation: Speedub.Nate's Avatar
    Join Date
    Dec 2003
    Posts
    9,145
    And two other questions:

    (1) How quickly does the unit sync baro elevation to GPS elevation?

    (2) When does / doesn't it re-sync? Will it do it while in motion? While climbing / descending?
    speedub.nate
    MTBR Hiatus UFN

  10. #10
    trail rat
    Reputation: slocaus's Avatar
    Join Date
    Jul 2005
    Posts
    7,707
    Quote Originally Posted by Speedub.Nate
    And two other questions:

    (1) How quickly does the unit sync baro elevation to GPS elevation?

    (2) When does / doesn't it re-sync? Will it do it while in motion? While climbing / descending?
    In my experience, after power up, it takes 10-15 minutes for the 305 to adjust to correct elevation in my front yard (give or take 20 feet, the screen stated accuracy). If you manually switch to the satellites page, it will show you the accuracy there, and I have elevation + accuracy as the two things that display on the map page. I tested this at least a dozen times when the 305 was new to me, and occasionally over the 6-8 months that I have had it.

    #2, I don't know, but have been googling for that, plus searching on MB and GroundSpeak forums. So far no info. Of course, if it does, the barometric pressure varies normally, so it may bet better or worse. I'm hoping it continually compares satellite and barometric readins, but who knows.

    One other point to the OP is that Garmin TC has a smoothing feature that you can turn on or off on the graph screen, but it does not appear to affect the readings on the totals page. Of course the TC documention is sooooo detailed!
    CCCMB web - Facebook
    SLO trail maps - conditions

    "The physician heals, Nature makes well" - real fortune cookie

  11. #11
    mudnthebloodnthebeer
    Reputation: quaffimodo's Avatar
    Join Date
    May 2004
    Posts
    587
    Quote Originally Posted by Speedub.Nate
    And two other questions:

    (1) How quickly does the unit sync baro elevation to GPS elevation?

    (2) When does / doesn't it re-sync? Will it do it while in motion? While climbing / descending?

    Here's what I've learned to date. When I turned the unit on just now, it initially displayed an incorrect altitude, but one that was consistent with the atmospheric pressure change that's occurred since it was last turned off yesterday and reading accurately. It displayed the correct altitude within 5 minutes. I get good satellite reception here at my desk.

    I was reviewing some tracks that I had downloaded but not looked at closely and found another one that showed an incorrect starting altitude for the same ride, roughly 250 feet high. It didn't sync properly with know altitudes until over an hour into the ride. I can't recall where the unit was prior to the ride, but I'm pretty certain that it wasn't in the car. As a side note, the parking area at the trailhead is surrounded by high terrain and trees on three sides and there are places in it where I can't get the unit to initialize.
    Attached Images Attached Images
    Also, the streets are full of horizontal dropouts...

    BSNYC

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •