Altimeter Fenix 5

"The Auto Mode can only be considered buggy if we see it making choices which would never make sense in that situation."
In Auto Mode I saw my watch indicating rising and dropping elevation ranging 125m-350m in few hours while on my kitchen table - my house is at 60m elevation. This is a bug to me.

"Just making the wrong guess based on a sound decision algorithm is not enough to consider the Auto Mode buggy."
To me a routine yielding erroneous results out of reasonable error range is buggy.

Please keep posting. This is really interesting. If it turns out that you can affect the ambient pressure reading by installing and uninstalling watch faces, it might explain why some users have had the same problems on several watches.

I can't remember if it was already posted in this thread, but some F5 users have found out that static electricity from their clothing could affect the pressure reading. So you should be careful with that while testing.

With respect to static electricity I don't think that was my problem yesterday. I think that I've been wearing the same combination of clothing earlier without any problem with the watch.

Now to the most interesting bit... After removing the watch faces yesterday evening the watch has been behaving normally... :D
At 23:37 it showed:
Ambient pr., Baro, MSL, Temp
989.1, 998.7, 105, 31.3
then at 07:51
Ambient pr., Baro, MSL, Temp
995.6, 1006, 90, 32.7

My external barometer went from 1001 to 1006 in the same period.

I even went for a 15k run this morning, that is quite a bit up and down with a total elevation difference of about 200m. When returning home the watch was 20m of, which is in the ball park of what I expect from it.

I'm not going to push my luck by installing the watch faces again, but I think there might be something there that messes up the altimeter function in the watch.

"The Auto Mode can only be considered buggy if we see it making choices which would never make sense in that situation."
In Auto Mode I saw my watch indicating rising and dropping elevation ranging 125m-350m in few hours while on my kitchen table - my house is at 60m elevation. This is a bug to me.

"Just making the wrong guess based on a sound decision algorithm is not enough to consider the Auto Mode buggy."

To me a routine yielding erroneous results out of reasonable error range is buggy.

A reasonable error range on a barometric altimeter is approx. 400 meters. That is equal to an ambient pressure change of 40-50 millibar which is perfectly normal.

Example: You calibrate the altitude to 60 m on a day where a 1030 millibar high pressure is right on top of you. Some days later, the high pressure has been replaced with a low pressure of 980 millibar. If the watch still has the same calibration, it will now show somewhere around 500 meter altitude.

That is how a barometric altimeter works. It is not buggy hardware. It is not buggy software. It is the surrounding conditions changing.

If you want more precise measurements, you need to make sure that the altimeter is correctly calibrated everytime you use it. You can't use yesterday's calibration if the ambient pressure has changed in the meantime.

At 23:37 it showed:
Ambient pr., Baro, MSL, Temp
989.1, 998.7, 105, 31.3

then at 07:51
Ambient pr., Baro, MSL, Temp
995.6, 1006, 90, 32.7

My external barometer went from 1001 to 1006 in the same period.

The ambient pressure looks fine to me. There is a small difference in the ambient pressure change measured by the watch (6.4 millibar) and the change measured by the external barometer (5 millibar), but most of that could be explained by rounding errors.

And as discussed before, the watch has had to guess what caused this change. In this case, the watch has guessed that a little less than 2 millibar out of the 6.4 millibar ambient pressure increase was caused by you moving to a lower altitude. In your case the guess was wrong - you did not move at all - but that is the nature of guesswork.

Now to the puzzling part:
The barometric pressure shown by the watch has not behaved in accordance with this. The ambient pressure increased by 6.4 bar, and a little less than 2 millibar of this increase was used for a 15 meter altitude decrease in the altimeter. The remaining 4.5 millibars ambient pressure increase should then have been used for increasing the barometric pressure. But the watch increased the barometric pressure by at least 6.8 millibar (taking rounding errors into consideration). Which is actually more than the registered change in ambient pressure. This seems wrong.

I wonder if there is a typo in your first dataset from 23:37. Near sea level, each millibar difference between Ambient Pressure and Barometric Pressure should be equal to an altitude difference of 8-9 meters. (I live 40 meter above sea level, and when I calculate the watch readings backwards, I get 8.6 meter/millibar. The exact figure will change somewhat with the pressure, but that is insignificant with these small pressure variations). However, when I calculate your watch reading from the first dataset backwards, I get 11 meter/millibar. This seems very wrong.

If I do the same calculation for your second dataset, I get 8.6 meter/millibar, which is exactly the same as I get for my own watch.

Hello, I would like some advice for my watch. I have been in contact with customer support regarding the altimeter function. I understand the inherent error involved in how it calculates elevation and barometric pressure. However, my elevation over the course of four hours while in my home changed from -20,000 meters to +12000 meters. This has to be a problem with the hardware and/or software. The ambient pressure on the watch at this time was showing around 300 or 350 mbar and the temperature was reading -60 °C.

Since that time after I performed a calibration and it appears to be closer. However I believe that the temperature is used to calculate and because the watch was so low on my temperature it was behaving wildly. Do you think that my wild fluctuations were because it needed to be manually calibrated or is this something I should ask for an RMA for in case it is defective?

This is an interesting thread. My fenix 5 has had the altimeter issue traveling thousands of feet both up and down while sitting in my house. The warm water soap bath seems to stop it and it’s ok for a couple days.

I hunt and and want to use the altimeter for fairly accurate readings while in the field, what would be the most effective way of getting accurate readings while not having internet access for multiple days.

I am thinking turning the watch mode to altimeter and then calibrating every morning to my known altitude from my map. Would I be correct? And will the barometer reading be useless while in altimeter mode?
If that is wrong and there is a better way to use it I’m all ears!

I love the watch and I also want to use it for my purposes. I’d rather learn how to use it with its limitations effectively than just assume it’s defective and get an inferior piece of gear.

I hunt and and want to use the altimeter for fairly accurate readings while in the field, what would be the most effective way of getting accurate readings while not having internet access for multiple days.

I am thinking turning the watch mode to altimeter and then calibrating every morning to my known altitude from my map. Would I be correct? And will the barometer reading be useless while in altimeter mode?

When you are in altimeter mode, the barometer reading will be fixed. (At least should be - on my watch I don't have the luxury of switching manually between these modes so I can't do any tests to verify this behaviour.)

So if barometric pressure around you changes, and your watch is in altimeter mode, the barometer reading on the watch will not change. And even worse, since it is altiude you are looking for: The altitude displayed will now be wrong, because the watch will be using the an outdated barometric pressure as reference for the altitude measurement.

It is very likely that you would get a better result in Auto mode. Or not. Your usage case is probably the worst possible for the Auto algorithm. I don't know if anyone except the developers knows the finer detail of the Auto algorithm, but it is clear that the speed of ambient pressure change plays a big role:

• If the ambient pressure changes quickly, the watch will assume that you moved to another altitude, so it will keep the barometric pressure and update your altitude.
  
• If the ambient pressure changes slowly (and constantly over a long time?), the watch will assume that you are staying at a fixed altitude and the barometric pressure around you have changed. So it will keep the altitude and update the barometric pressure.
  

This algorithm works nicely for person like me who travel from home to work, stays there for 8 hours, travels from work to home, stays there for some hours, goes for a run, comes home etc. So most of my hours are spent at an unchanged altitude, and all the altitude changes during the day happen quite fast.

But in your case, I assume that you will be constantly and slowly moving for a lot of hours every day, and you will have some slow altitude variations during the day. If you for example spend 8 hours moving to another location at a 35 meter higher altitude, the altitude variation will only change your ambient pressure with 0.5 millibar/hour. Such a slow ambient pressure variation could easily be caused by changes in barometric pressure too.

So the Auto algorithm will be seriously challenged and probably make a lot of wrong guesses. But you will have to test for yourself to find out whether those guesses will be worse than the errors coming from leaving the watch in Altimeter mode.

Under all circumstances, frequent calibration will be the key to success. If you have known altitudes from a map, that is good. Otherwise you could watch your GPS altitude over some time, for example 10 minutes, work out the average reading and use that for calibration.

If you are frequently passing through the same spots during the hunt, you can also somewhat automate the calibration. See the link in my post #71 in this thread.

I wonder if there is a typo in your first dataset from 23:37. Near sea level, each millibar difference between Ambient Pressure and Barometric Pressure should be equal to an altitude difference of 8-9 meters. (I live 40 meter above sea level, and when I calculate the watch readings backwards, I get 8.6 meter/millibar. The exact figure will change somewhat with the pressure, but that is insignificant with these small pressure variations). However, when I calculate your watch reading from the first dataset backwards, I get 11 meter/millibar. This seems very wrong.

It wasn't a typo, but you are probably right that it was wrong. To get ambient pressure, barometric pressure, altitude and temperature I made a screen on one of the activities showing all four. Unfortunately I have set it to auto calibrate the altitude when starting an activity, and so it calibrates the altitude according to the GPS, but when it calibrates the altitude it does not recalculate barometric pressure. I think only use it as a kind of offset value for that spesific activity. So even if the activity screen shows e.g. 105MSL, when I quit the activity and goes to the ABC widget it will show a different altitude that corresponds to ambient and barometric pressure.

When you are in altimeter mode, the barometer reading will be fixed. (At least should be - on my watch I don't have the luxury of switching manually between these modes so I can't do any tests to verify this behaviour.)

So if barometric pressure around you changes, and your watch is in altimeter mode, the barometer reading on the watch will not change. And even worse, since it is altiude you are looking for: The altitude displayed will now be wrong, because the watch will be using the an outdated barometric pressure as reference for the altitude measurement.

It is very likely that you would get a better result in Auto mode. Or not. Your usage case is probably the worst possible for the Auto algorithm. I don't know if anyone except the developers knows the finer detail of the Auto algorithm, but it is clear that the speed of ambient pressure change plays a big role:

• If the ambient pressure changes quickly, the watch will assume that you moved to another altitude, so it will keep the barometric pressure and update your altitude.
  
• If the ambient pressure changes slowly (and constantly over a long time?), the watch will assume that you are staying at a fixed altitude and the barometric pressure around you have changed. So it will keep the altitude and update the barometric pressure.
  

This algorithm works nicely for person like me who travel from home to work, stays there for 8 hours, travels from work to home, stays there for some hours, goes for a run, comes home etc. So most of my hours are spent at an unchanged altitude, and all the altitude changes during the day happen quite fast.

But in your case, I assume that you will be constantly and slowly moving for a lot of hours every day, and you will have some slow altitude variations during the day. If you for example spend 8 hours moving to another location at a 35 meter higher altitude, the altitude variation will only change your ambient pressure with 0.5 millibar/hour. Such a slow ambient pressure variation could easily be caused by changes in barometric pressure too.

So the Auto algorithm will be seriously challenged and probably make a lot of wrong guesses. But you will have to test for yourself to find out whether those guesses will be worse than the errors coming from leaving the watch in Altimeter mode.

Under all circumstances, frequent calibration will be the key to success. If you have known altitudes from a map, that is good. Otherwise you could watch your GPS altitude over some time, for example 10 minutes, work out the average reading and use that for calibration.

If you are frequently passing through the same spots during the hunt, you can also somewhat automate the calibration. See the link in my post #71 in this thread.

Thanks for you’re detailed reply, I appreciat it! ill have to see how it performs I guess.

On the the issue of the erratic altimeter issue that myself and others experience... the warm water soap bath would stop the erratic behavior and the watch would act normally for a couple of days before acting up again.
I read (possibly in this thread) someone said that they uninstalled all of their downloaded watch faces and since then they had no problems with their watch. I tried this and I have not had any issues for about four days. My watch has actually been very accurate since I deleted the watch faces.

May be be worth a try for anyone experiencing issues. So far so good. I deleted the watch faces “actiface” and the “kuiu” watch face. Hopefully I don’t have anymore issues... it might be interesting for people having altimeter issues to list the watch faces they have installed and possibly find a connection?

It seems very strange that a watch face provoque such a problem, but the fact is that since last post I do not have any problems (no baths needed) and coincidence or not I was using actiface and mywatch during the problem and now I'm using big-time and garmin analog.