Temperature effects on electronics
I notice from the published pictures on this site that many models have
the electronics boards strapped to the outside of the model, maybe only
protected by heatshrink, or similar.
When you use high precision measurements for GPS, IMU and air pressure, I
think you will suffer very badly from temperature and temp gradient
effects on those boards.
We got caught with this in evaluating a GPS board. We put the bare
board on a bench out in the open, and logged the position and velocity.
Both drifted badly, about 20 times worse than the published data, and
lead us to a long argument with the manufacturer until we ( somewhat by
accident ) shielded the board from air drifts. The drift dropped
immediately to an acceptable level. Yes I know GPS is all 'digital',
but in fact there is a very critical component, the TXCO oscillator,
which must remain exquisitely stable for the system to work, and temp
drifts over a few seconds were badly upsetting it.
Precision analog components like gyros, accelerometers and pressure
sensors will also suffer from these effects. Think about what happens
as you fly a model through air currents, thermals, etc, all at different
temperatures.
It would not surprise me if some of the bad GPS and IMU data I see
reported are due to the same problems, especially on those bare boards.
My recommendation would be to put all boards in an aluminum box ( even
foil would help ), and then put that in an insulating container or foam
wrap. This would cut down the rate of temperature change, and probably more
important, the temperature gradient, across those sensitive components.
the electronics boards strapped to the outside of the model, maybe only
protected by heatshrink, or similar.
When you use high precision measurements for GPS, IMU and air pressure, I
think you will suffer very badly from temperature and temp gradient
effects on those boards.
We got caught with this in evaluating a GPS board. We put the bare
board on a bench out in the open, and logged the position and velocity.
Both drifted badly, about 20 times worse than the published data, and
lead us to a long argument with the manufacturer until we ( somewhat by
accident ) shielded the board from air drifts. The drift dropped
immediately to an acceptable level. Yes I know GPS is all 'digital',
but in fact there is a very critical component, the TXCO oscillator,
which must remain exquisitely stable for the system to work, and temp
drifts over a few seconds were badly upsetting it.
Precision analog components like gyros, accelerometers and pressure
sensors will also suffer from these effects. Think about what happens
as you fly a model through air currents, thermals, etc, all at different
temperatures.
It would not surprise me if some of the bad GPS and IMU data I see
reported are due to the same problems, especially on those bare boards.
My recommendation would be to put all boards in an aluminum box ( even
foil would help ), and then put that in an insulating container or foam
wrap. This would cut down the rate of temperature change, and probably more
important, the temperature gradient, across those sensitive components.
Replies to This Discussion
-
Permalink Reply by Ken on -
"I notice from the published pictures on this site that many models have
the electronics boards strapped to the outside of the model, maybe only
protected by heatshrink, or similar."
Many models? Really? I doubt that. Aside from Marcy, can you provide links to these GPSes and IMUs in heat shrink (or not) on the outside of the airframe?
If one is using the DIYdrones recommended platform, the autopilot electronics are inside the foam cockpit, the servos are exposed. The thermopiles, if used, need to be exterior to function. ESC's, often but depending on loading, need airflow for cooling so some people stick them outside. They are often wrapped in heat shrink. Are these what you are referring to?
I think if you read carefully, the majority of people having GPS and IMU issues haven't yet gotten off the assembly bench.
-
ModeratorPermalink Reply by Sgt Ric on -
You really should spend some time reviewing this site (more than the 5 minutes you seem to have).
As Ken mentioned, the external shrink-wrapped component is the ESC, as well as systems using the XY and Z sensors, which must mount externally.
On this site you will see people flying through inclement weather, sometimes at amazing speed ranges, who do not seem to suffer from your perceived equipt vulnerabilities.
Again, most of the problem reports you mentioned can be traced to bad soldering or assembly errors.
I am sure we can assume that was the case with your project, but thanks for the suggestions anyway.
-
ModeratorPermalink Reply by Sgt Ric on -
I can just imagine the problems I would have figuring out the proper CG for a flying picnic cooler!
-
Permalink Reply by Adrian Jansen on -
Here is one pic I saw: http://diydrones.com/profiles/blogs/my-first-arducopter
I do not intend to 'poke holes' in any particular design, just point out some stuff I learned from experience in 20+ years of dealing with high precision electronics and sensors. If you are aware of the problems, and have taken steps to mitigate them to your satisfaction, well and good. If not, then you are free to do what you like with my advice.
-
ModeratorPermalink Reply by Sgt Ric on -
My remarks were very smug, sorry for that...I must have had a hard day.
-
Permalink Reply by Adrian Jansen on
-
Apology accepted, we all have bad days.
-
Permalink Reply by Andrew Robinson on
-
I know well of shielding. I had a gas sensor PCB board from another manufacturer that would dual purpose as an excellent close range cellular signal strength indicator and other sensors with precision heaters required to operate within spec for accurate measurements. A gust of wind could indeed ruin your day measurement wire.
My stuff is exposed right now, but I also am just doing initial testing.
For the final unit I have a nice thick plastic enclosure lined up with EMI shielding paint. Gyros and Accelerometers are usually temperature compensated and the way the DCM drift-correction algorithms are setup that kind of offset is generally slowly corrected for.
Most people's "alpha build .01" is unshielded out in the open, but I don't think a lot of people post photos after that point.
Social Networking
Top Content
Contests
Season Two of the Trust Time Trial (T3) Contest has now begun. The fourth round is an accuracy round for multicopters, which requires contestants to fly a cube. The deadline is April 14th.A list of all T3 contests is here.
Groups
-
Documentation Editors
51 members
-
ArduPlane User Group
671 members
-
ArduCopter User Group
1280 members
-
Taiwan ROC ArduPilot Gro…
8 members
-
Andropilot Users Group
182 members
Advertisement
© 2013 Created by Chris Anderson.
Powered by
