Related
This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

645 M - Barometric Altimeter fundamental issue??

Ingo Schmitz
over 4 years ago

I talked my brother into a 645M after having lots of fun with my 945. His barometric altitude data doesn't make sense. I am wondering if there is a systematic issue:

The height data isn't reacting correctly to terrain changes during biking. It's most notable when remaining stationary after completing a downhill section. While stationary the height data continues to change to lower values over the next three minutes until it reaches the correct value. A height over distance plot looks like he took a 100 ft plunge. The same effect (only opposite) happens after an uphill section. In more technical terms there seems to be an abnormal long time constant so the rate of change of the barometric sensor is way to slow.

When comparing the height profile to data recorded with an older GPS-based Garmin cycling computer it becomes blatantly evident that the entire height profile from the 645M is simply wrong and not tracking changes. 

He has tried resetting the 645, calibrating its barometric sensor (only affects offset) to no avail. His data over the last couple of month shows the watch did this from new. So a clogged port addressed by soaking in warm soapy water isn't going to fix that. He tried anyway and it made no difference. Eventually he returned the watch and got a brand-new replacement under warranty. It shows exactly the same erratic behavior.

Below is an example where the issue becomes very obvious. During a one-minute downhill the height dropped only 11m (36ft). In the subsequent 3 minute break from 1:10:00 - 1:13:00 the height data dropped another 31m (102ft) while he is stationary. That makes no sense!!

  

Thoughts? Is this "works as designed" or a defect?

  • 0 over 3 years ago

    I can only confirm this. The altimeter is too slow for bike rides or skiing. Works fine for hikes for example because your pace is slow and the altimeter can keep up. I already had one unit replaced under warranty, but this one is the same. I'd like to know if the sensor is the same they use on the 945 or Fenix series or indeed it is a different one, where would I get that info?

  • 0 over 3 years ago

    I can also confirm this behavior. But this comes from SW- Update with version 7.0.0 or an Update before. I know that the altimeter was working correctly / better on version 6.2. <irony on> So garmin decided to corrupt the altimeter, as you should buy a new watch please <irony off>. The Fenix series 5 or 6 doesent have this problem. Best regards!

  • Former Member
    0 Former Member over 3 years ago in reply to Keffan
    Keffan said:
    this comes from SW- Update with version 7.0.0 or an Update before. I know that the altimeter was working correctly on version 6.2. So garmin decided to corrupt the altimeter

    So, there's no chance minerals from water or sweat have built-up in the barometer port causing this issue, or your sensor might be failing for one reason or another?  The "logical" explanation is "Garmin decided to corrupt the altimeter" with an update, but you're not exactly sure which version caused it?  Pure genius!  We need more brilliant and insightful people like you to assist on these forums. 

  • 0 over 3 years ago in reply to Former Member

    Please calm down! What exactly did you just contribute to the forum? Yes, the article was to be understood with a bit of irony, but you are all smart guys, aren't you? The reason why I am sure that it is a software problem is as follows: I have already sent 2 watches FR645M to Garmin and received always a refurbished replacement device. All had the software version 7.0.0. On my test track they all showed the above problem after 10 minutes of driving on the first descent. The altimeter responded slowly downwards, even so slowly that it counted down as I drove up the hill again. The meters in altitude just weren't right anymore.
    But now I have received a device from someone, but it has never been updated. It was version 6.2. on it and reacted much faster to the changes in altitude on the test track. Yes, it could all be coincidence, but I am sure it is a software problem.
    Incidentally, Garmin has also changed the behavior of the calibration with the software updates. Since version 7.0.0 or earlier, the calibration behaves differently than with version 6.2. If you enter an altitude during the activity, it is no longer accepted immediately but is reached slowly. Just as slowly as when I'm driving down the hill.

    And no, I cannot say which update "improved" this feature, as this watch automaticlly Updated today to 7.0.0 while I was connecting to Garmin Express. I had set the option "do not update" but it did it over night, after connecting to garmin express.

    As you can see, I've dealt with the topic for some time and also reported it to Garmin, but they probably won't solve it anymore. Unfortunately!

    Regards!

  • Former Member
    0 Former Member over 3 years ago in reply to Keffan
    Keffan said:
    Incidentally, Garmin has also changed the behavior of the calibration with the software updates

    There's quite a difference between Garmin changing the behavior in an update then Garmin deciding to corrupt the altimeter as you said.  One is a change which you don't care for and the other is an accusation that a manufacturer intentionally broke a function on their product.

    Keffan said:
    Yes, the article was to be understood with a bit of irony, but you are all smart guys, aren't you?

    No, it's not understood.  You're a stranger posting on a forum where people from all around the world come to use. No one could possibly know what you now claim you meant. We can only reply to what you actually wrote. 

  • 0 over 3 years ago in reply to Former Member
    No, it's not understood.  You're a stranger posting on a forum where people from all around the world come to use. No one could possibly know what you now claim you meant. We can only reply to what you actually wrote. 

    This is true. I want it to be public that I'm very shure there is a software problem here. Presumably it has to do with the altitude calibration. And yes, I'm already frustrated how Garmin exaggerates its update policy. It is not the first time that devices have been updated and proven functions have been disrupted as a result.

    There's quite a difference between Garmin changing the behavior in an update then Garmin deciding to corrupt the altimeter as you said.  One is a change which you don't care for and the other is an accusation that a manufacturer intentionally broke a function on their product.

    I think this is the point. Garmin knows about this issue from my side since 03/21  and they reported me that the refurbished device is also defect? This is exactly what I meant but yes, you are right we can assume it wasn't there intention to implement this but obviously it is their intention not to fix it.

    (post from Info Schmitz is from 10/20 and I assume that  some of garmins developer read here)

  • 0 over 3 years ago in reply to Keffan

    My brother went through one warranty replacement - same issue and since moved on to a 945 and its height data is spot-on same as mine. I did buy a 645M for my wife (super-deal on Prime day) and it showed the same problems with slow height data from day one. So I am convinced it’s a systemic issue and very likely due to excessive filtering of the barometer sensor data. She doesn’t care much about it so I let it slide.

    I agree with others that while Garmin didn’t do this intentionally it’s still disappointing that they don’t address it in their SW roadmap. My guess: they conclude working on newer features is more profitable even with the opportunity cost (lost customer satisfaction). Welcome to corporate decision making 

  • 0 over 3 years ago in reply to Ingo Schmitz

    I experienced the altimeter being slow to react to altitude changes since the beginning, I bought the 645 a few months after its release. That unit was replaced within 1 year due to the altimeter being actually broken (by swimming with it I guess) but the new one is the same, I do not think it is related to any software update but to the sensor used, probably a cheaper one compared to the one in the 945.

  • 0 over 3 years ago in reply to juza

    Just to clearify: My experience is that the altimeter is reacting slow quiet after starting the activity if you do not wait 30sec for calibrating or do the calibration yourself. If I run or cycle with calibrated altimeter, it was reacting fast. This is also what garmin wrotes on the online manual to fix altitude problems.

    But, if I'm cycling or running for about 10mins it swtiches again to "that uncalibrated mode," you see that the altitude is filtered extremly again, like it was uncalibrated at start. Yesterday I found out that I could break this issue if I calibrate the altimeter again, but this is not what I want to do every activity. And this behaviour has changed since some SW-update  or has been at least extended or is done more often that I recognized it.

  • 0 over 2 years ago

    Hello. Barometric altimeter in FR645 works well, and is reactive. The elevation issue in this watch is due to automatic calibrating of altimeter by gps. When this function is active during activity it slows down the elevation datas saving.

    You must regularly manually calibrate altimeter in order to deactivate automatic calibration during activity.

    Unfortunately, even if you deactivate it during activity, this automatic calibration is active at every activity start. As long as this GPS altimeter calibration is not successful, it will continue periodically. If it is successful, you could run or bike for many hours without elevation issue.

    It would be nice if Garmin turned this function off completely when we manually calibrate the altimeter.