Baro sensor drifting badly - APM 2.6

[Doug]

I have been trying to help a local user who was complaining about poor auto-land performance and found that his baro sensor was drifting badly, like ten meters over an hour.   The drift goes at a  relatively constant rate through the log, so it didn't look temperature related. I told him he probably just had a bad sensor.  He emailed be today saying he got a new board and is seeing the same behavior.

I have never seen this behavior before.  I attached a .tlog form his old board - haven't seen one yet from the new board but he describes it as the the same.  Have others seen this behavior?

-Doug

[Andrew Tridgell]

Hi Doug,

> I have never seen this behavior before. I attached a .tlog form his old
> board - haven't seen one yet from the new board but he describes it as the
> the same. Have others seen this behavior?

yes, I see it regularly. I sometimes check the local aviation weather
report and see what the rate of change in atmospheric pressure was for
that time of the day, and I tend to find that it does approximately
match the change I see in the barometer.

When I'm flying at my local field 10 meters per hour is common.

Cheers, Tridge

[Jesus Alvarez]

Dough

My two APM do the same.
Bark drifts at a constant but low rate. So it does not sound extrange to me.

I thought that bark drift is something normal...

This is why I think sonar is a must if you want auto landings.

[Henri Karrour]

You are not alone! I do have 4 APM and they all behave very badly.

To start with:

- They are very sensitive to: Light, temperature and vibrations!

- I did power the APM through a LDO linear regulator to insure that no electrical noise is present on the supply.

  I did check with an oscilloscope that no noise was to be seen on the 5 volts rail.

I do have very different results across the 4 APM.

One like to go with negative values! Other gain constantly altitude (APM on the ground outside)!

Another like to start with negative values and then comes back above ground after about an hour!!

There is no wind (enclosed in a black plastic box) no light and motors aren't armed!

The best of the lot sort of work to almost real values as follows:

Power it up and wait about 1/2 hour. Then switch off main power for 10 seconds and re apply power...

At all times motors are not turning and Aircraft is on the ground.

Using Mission planner 1.3.7 uild 1.1.5307.24027. 

This is a really bad way to measure altitude...

Henri

[Jonathan Challinger]

We have the best barometer available on the market and you do have to wait for the temperature to settle (around 10 minutes) in order for it to stabilize after you power PIXHAWK. Other than that, it will drift with atmospheric pressure.

However, one thing I would strongly advocate for is changing the sensor to one that allows us to connect a probe. Then we can start working on probe designs and placement that minimizes disturbances with speed.

--
You received this message because you are subscribed to the Google Groups "drones-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to drones-discus...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Meier Lorenz]

Hi Jonathan,

You might want to look at this variant of the 4525DO
http://www.digikey.com/product-detail/en/4525DO-SS5AI015AP/223-1125-ND/3569054

For applications where vertical speed has an effect on the fuselage.

-Lorenz

On 22 Jul 2014, at 11:48, Jonathan Challinger <mr.cha...@gmail.com<mailto:mr.cha...@gmail.com>> wrote:

We have the best barometer available on the market and you do have to wait for the temperature to settle (around 10 minutes) in order for it to stabilize after you power PIXHAWK. Other than that, it will drift with atmospheric pressure.

However, one thing I would strongly advocate for is changing the sensor to one that allows us to connect a probe. Then we can start working on probe designs and placement that minimizes disturbances with speed.

On Mon, Jul 21, 2014 at 10:12 PM, Henri Karrour <henr...@iinet.net.au<mailto:henr...@iinet.net.au>> wrote:
You are not alone! I do have 4 APM and they all behave very badly.
To start with:
- They are very sensitive to: Light, temperature and vibrations!
- I did power the APM through a LDO linear regulator to insure that no electrical noise is present on the supply.
I did check with an oscilloscope that no noise was to be seen on the 5 volts rail.
I do have very different results across the 4 APM.
One like to go with negative values! Other gain constantly altitude (APM on the ground outside)!
Another like to start with negative values and then comes back above ground after about an hour!!

There is no wind (enclosed in a black plastic box) no light and motors aren't armed!

The best of the lot sort of work to almost real values as follows:
Power it up and wait about 1/2 hour. Then switch off main power for 10 seconds and re apply power...

At all times motors are not turning and Aircraft is on the ground.

Using Mission planner 1.3.7 uild 1.1.5307.24027.

This is a really bad way to measure altitude...
Henri



On Thursday, April 17, 2014 9:25:15 AM UTC+10, Doug wrote:
I have been trying to help a local user who was complaining about poor auto-land performance and found that his baro sensor was drifting badly, like ten meters over an hour. The drift goes at a relatively constant rate through the log, so it didn't look temperature related. I told him he probably just had a bad sensor. He emailed be today saying he got a new board and is seeing the same behavior.

I have never seen this behavior before. I attached a .tlog form his old board - haven't seen one yet from the new board but he describes it as the the same. Have others seen this behavior?

-Doug



--
You received this message because you are subscribed to the Google Groups "drones-discuss" group.

To unsubscribe from this group and stop receiving emails from it, send an email to drones-discus...@googlegroups.com<mailto:drones-discus...@googlegroups.com>.

For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups "drones-discuss" group.

To unsubscribe from this group and stop receiving emails from it, send an email to drones-discus...@googlegroups.com<mailto:drones-discus...@googlegroups.com>.

[Meier Lorenz]

On re-reading this: I meant horizontal speed has an effect on the pressure within the fuselage 8). Sorry for the confusion.

> To unsubscribe from this group and stop receiving emails from it, send an email to drones-discus...@googlegroups.com.

[Anthonie Michael Muller]

Hi,

I have had bad baro performance on apm boards , espesially the first apm 2 dual layer boards when the telemetry tx power is full, problem whent away when reducing to min tx power and a capacitor on telemetry power also hellped!

was the test performed wit the telemetry connected?

[Tom Pittenger]

This is standard baro drift error and all barometic sensors suffer from it, even ones in large sscale aircraft. It is due to atmospheric changes and can not be avoided without using additional ground-based information like a differential baro or a range finder. Best solution is to use sonar or lidar.