Replies to This Discussion

DeveloperPermalink Reply by R_Lefebvre on June 20, 2012 at 6:04pm

How does MuMetal compare to Permalloy?  Permalloy is easy to get.

I think a sandwich of aluminum and MuMetal would make a good shield.  The aluminum on the bottom would block some of the rotating magnetic field, and then let the MuMetal take the static field.

• ▶ Reply

Permalink Reply by Dave C on June 21, 2012 at 12:29am

I've got a great source for mumetal - they are sending me loads of sample thicknesses this week :)

I'll report back with results. 

• ▶ Reply

Wiki NinjaPermalink Reply by Gary McCray on June 21, 2012 at 11:14am

Just did a little research on MuMetal and Permalloy.

The biggest problem is that they will in fact significantly reduce the signal strength of the Earths magnetic field at the magnetometer (and are often used to produce just that effect.)

This means they would need to be far enough away from the magnetometer for this effect to be minimal.

From what I can see, keeping the magnetometer an adequate distance away from significant static magnetic fields is the best solution to that problem.

As for the moving magnetic fields or RF interference, both Aluminum and more conventional Faraday style shielding (Steel or copper) provide potential for benefit without negative side effects.

Aluminum has the advantages that it dampens slow moving electro-magnetic fields but is itself non magnetic and it reflects higher frequency RF. Increased thickness somewhat increases effectiveness on electro-magnetic fields.

Basically Aluminum can be an excellent shield for the GPS and is at least somewhat useful for the magnetometer.

DeveloperPermalink Reply by R_Lefebvre on June 21, 2012 at 11:25am

M

Wiki NinjaPermalink Reply by Gary McCray on June 21, 2012 at 10:12am

Totally agree,

Thanks for all the great input regarding this.

In the case of GPS flakiness, a microwave level interference problem, the aluminum seems like it is pretty much the solution because of the way it attenuates, but for the magnetometer in particular, Mu Metal should bock any stray local static magnetic field.

In fact, Mu Metal is so effective I am a bit concerned that if it was too close to the magnetometer it might interfere with its own detection of the earths magnetic field.

I do not know if this is a problem, but it probably needs to be clarified.

Easy to do with a test, so:

I am going to get some MuMetal and give it a try to see what happens and will look into "Permalloy".

I like the sandwich idea with the aluminum.

• ▶ Reply

Permalink Reply by Dave C on June 21, 2012 at 3:44pm

Yup, I'm going to test this all out, samples on their way. My initial plan is to encase or at least shield the esc's, the motor wires and the battery, and leave the APM in open air, all bar the tupperware... We are looking to block interference from the power circuits here, not stifle the APM.

DeveloperPermalink Reply by R_Lefebvre on June 21, 2012 at 3:53pm

Another reply deleted... is it just me?

I'm using Chrome, I wonder if that's the problem?

Anyway, my thinking is that a flat sheet of MuMetal, >1" below the APM will block the interference from all the power bits, and probably the vertical component of the mag field, while leaving the XY component of the earth's mag field unchanged?  I guess we'll see.

DeveloperPermalink Reply by R_Lefebvre on June 21, 2012 at 4:06pm

Dave, about suspending it on a stick, I've had good luck putting it on the tail in the heli.  I'm getting really good mag numbers now.  Putting it up on a stick is kind of an inelegant thing to do for a quad though.  I wish there was  better way.

Permalink Reply by Marc Puig Pasarrius on June 22, 2012 at 7:52am

I've bought mumetal, and first impressions are: don't use it, it's worse than nothing! With the mumetal under APM I get almost plain mag heading:

I recalibrated the compass after putting the mumetal sheet (only static calibration, I don't have telemetry). It's not grounded. I have automatic compass declination enabled.

I think the problem is as Gary commented that my compass is too close to the mumetal shield.

I really think the best approach is the used by dji: keep compass and GPS away from noise. Maybe I'll use the naza GPS mount and I'll try to put the APM1 compass and GPS there.

Permalink Reply by Michael Müller on July 6, 2012 at 3:21am

Hi Gustav, I have the same GPS-Lock issue because of interferences and I want to put a shield (aluminium or copper) between APM and power board.

My question:

Would you please explain how to ground the shield? I think connecting a cable to some ground, but which one? 12V from LiPo or 5V from ESC/BEC or from the UBEC which is powering the APM?

Thank you.

• ▶ Reply

Permalink Reply by Gustav Kuhn on July 6, 2012 at 4:04am

Hi Micheal,

What airframe? Pictures?

To point out, I work in the RF field, and manage Radio and TV transmitters from 5W to 25 KW.

We are also running some sensitive low power receiving type equipment, on the same sites!

And we found there is no one solution to fit all :-(

On to RC and APM's

Bottom line is, make sure your components like ESC's, UBEC's have the proper toroids, as close to the actual device, as possible.

Try to twist all wires carrying DC.

Try to separate these devices as far as possible from stuff like GPS antennas.

On my fixed wing drone, I extended the wiring, the GPS is way behind the wing now.

I have a stock jDrones quad, followed the recommended build, no problems with GPS, am running Spektrum telemetry, and xBee.

There is no easy way to just slide a piece of metal in between stuff, ground it (preferably the ground point on the Power Distribution Board), and make it work.

If the grounding wires isn't less than about an inch in length, it is pointless at radio frequencies.

Butt even after 32 years in the field, we still call RF "Black Magic", so you might be lucky with just a sliding some foil between the different devices, and connect that to the PDB negative :-)

• ▶ Reply

Permalink Reply by John Campen on June 20, 2012 at 3:58pm

any photos of the board and setup? You must make sure wires have chokes and shields and well away form the mag. Also the Xbee needs to be well away form this as well because ti will send the mag haywire

• ▶ Reply

• ‹ Previous
• 1
• …
• 25
• 26
• 27
• 28
• 29
• …
• 68
• Next ›
• Page