What's new
Van's Air Force

Don't miss anything! Register now for full access to the definitive RV support community.

Pitot Tube wiring past Magnetometer

Brent 801

Member
I'm preparing to run the wires from my Garmin GAP 26 heated pitot inboard to the fuse. Is there any concern about running the pitot heat wiring close to the Garmin Magnetometer which is installed in the left wing where a Dynon ADAHRS would by located if using Dynon?
 
Pitot heaters are heavy current users, this means serious potential for magnetic interference. Best practice would be to run a twisted pair (pwr/grd) from the tube all the way to the switch and ground block and keep the wiring as far as practical away from the magnetometer, then keep your fingers crossed.
 
Last edited:
My Experience

I have mine quite close to the pitot, and am using Dynon's temperature regulating probe which I imagine is noisier than just an on/off spike since it has to regulate.

First time I tested it was close enough to fail the self-test. In my case I was able to reroute the wires so that they were 6 inches further away, at which point I no longer saw any impact when I turned it on or off during the self test.

Distance really does matter a lot, and 3 or 4 inches is too close.

(In my case not a twisted pair install)


Derek
YMMV
 
Yes, pitot heat power will interfere with the Garmin magnetometer mounted on the ADAHRS shelf. I ran a power wire in the "bundle" in the wing with a ground to a rib by the pitot - failed miserably. Then I ran a ground wire (not twisted at this point as Walt has recommended) - better, but still failed. Then I rerouted the ground farther from the magnetometer, twisted it around the power wires(s) near the magnetometer and got better results. It still fails Garmin's interference test for about 3 seconds during power up and power down, but I called it a victory.
 
I know there are differences between the 9 and 14, so is there a reason the magnetometer located in the wing rather than the back tail deck area? Im not flying yet so I don't know how mine will work, but there isn't nearly as many wires back there as by the pitot. There must be a reason. Thanks
 
Yes, pitot heat power will interfere with the Garmin magnetometer mounted on the ADAHRS shelf. I ran a power wire in the "bundle" in the wing with a ground to a rib by the pitot - failed miserably. Then I ran a ground wire (not twisted at this point as Walt has recommended) - better, but still failed. Then I rerouted the ground farther from the magnetometer, twisted it around the power wires(s) near the magnetometer and got better results. It still fails Garmin's interference test for about 3 seconds during power up and power down, but I called it a victory.

Is there any reason not to run coax for pitot heat? I'm guessing it would be much better than twisted pair.
 
The physics of twisted pair is better than coax for this application. So coax isn't better, and it is bulkier to boot.

In hindsight I wish I had installed the pitot to the right wing; that would have been better overall. But my Gretz mast is firmly installed on the left and won't get moved. The shelf is in the left wing too - also not going to move. One thing that may make a difference for me is I have the Dynon pitot heat and not the Garmin, with the rest of my kit Garmin. Maybe it will act a little different current-wise.

Its interesting because for sure with pitot heat off it will all work fine. Its when I turn it on that it will be wonky - maybe. But as the pitot heats up, the current variations will diminish, and the magnetometer should settle down albeit with some new fixed error. I wonder what that will be.

If the magnetometer in the wing is a total loser I suppose I can at anytime make a shelf in the tail section and put it off to the right side up there.

I think I'd rather have a card-compass on the dash than a second magnetometer. What happens in the G3X system if you don't install a magnetometer? Can that be ignored?
 
Is there any reason not to run coax for pitot heat? I'm guessing it would be much better than twisted pair.

Other than cost and weight, no - but make sure the center conductor and shield can handle 10 amps dc without significant loss.
We're concerned with a steady state magnetic field here, not RF. A tightly twisted pair should be nearly as good as a coaxial arrangement. But (in either case) this only works if the ground return is not connected to the pitot tube or its mounting harware. You need the return current thru one wire to equal the feed current thru the other.
 
But (in either case) this only works if the ground return is not connected to the pitot tube or its mounting harware. You need the return current thru one wire to equal the feed current thru the other.

I'm a few months from needing to worry about this and haven't studied it yet but this confuses me a bit. You DON'T want the ground connected to either the pitot tube or its mount?!? :confused:
 
I'm a few months from needing to worry about this and haven't studied it yet but this confuses me a bit. You DON'T want the ground connected to either the pitot tube or its mount?!? :confused:

You want the 'return line' (often called the ground wire) from the heater to go back along side of, or better twisted with, the wire which supplies the current to the 'hot' or input side of the heater. Ground it on the firewall, or much better, at a universal grounding point/block. The return wire should not be connected to the metal tube or any metal structure, which would allow the return current to flow thru the metal structure. You want the only possible path to be thru the return wire, so its current is equal (but opposite) to the input wire, so their magnetic fields will cancel out. If your pitot tube has its return line connected internally to the metal structure, you will not be able to easily cancel out the input wire's magnetic field.
 
I'm a few months from needing to worry about this and haven't studied it yet but this confuses me a bit. You DON'T want the ground connected to either the pitot tube or its mount?!? :confused:

Correct, wire the ground all the way back to the bus with same size wire as the power wire.

This applies more generally: all your high-amp circuits should go into a "star" where all the return current comes back to a single bolt on the bus--your reference ground. If you return the ten amps through aircraft skin, then connect another sensitive instrument's ground to the skin halfway to where the high-amp current originates, you will see that ground voltage fluctuate from reference zero as the skin has a small but finite resistance. If the high-amp current is intermittent, your readings on the other instrument end up jumping all over the place.
 
The problem with the pitot heat controller is that the chassis itself is grounded, and this is the same ground as the line that normally bolts to the wing structure. This is done for a few reasons, and is part of the way the MOSFET in the system is grounded to carry the heater current and the heat sink design to the case. So I don't think that it is possible to completely return the ground current via the ground lead. Most might go that way but a very significant portion will return via the chassis mount. The controller case should not be isolated from the mount with fiber washers; it is designed to be grounded. The ground return line exists to ensure there is a good ground return separate from the chassis mounting, and because the resistance to ground via the case is sometimes a little higher because the mounting hardware isn't as reliable for electrical connections. But I haven't checked the differences because I haven't cracked the case to see. A lot of circuits like this have a ground lead on the same connection inside that bolts to the case - the screw that holds the MOSFET down.

I think that putting a larger diameter wire cuts down on impedance as well as reducing overall resistance but is probably a trivial difference for the magnetic field of the power feed.

If it wasn't for the proximity to the magnetometer it would be moot. Nuckoll's book says wire to the chassis don't worry about it and that is correct in general.
 
Last edited:
For JDA's case I would try to mount the box to a heat sink, and keep the heat sink electrically insulated from the airframe. Any decent electronics store sells very thin mica sheets, along with dielectric grease. These are electrically insulating but provide pretty good heat transfer.
 
I think I was a bit sleepy when I wrote my response last night. It registers with me now that connecting the ground to the mounting hardware or the tube would route ground through the airframe. I guess I must have been thinking "where is ground coming FROM?," if that makes sense. But, at the same time, I was thinking that the tube was grounded through the chassis as Dudley (JDA) says. I can see mounting with mica and dielectric grease if that provides an adequate heat sink but I'm wondering if the mounting bolts wouldn't still ground to the wing. I guess you would also use the dielectric grease in the bolt holes and use a fiber washer but then I wonder if you wouldn't end up with some grease on the bolt/nut interface.
 
Has anyone installed the pitot in the right wing just to remove any doubt of interference with the magnetometer? I?ve heard both sides of the coin. Some have done shielded and/or twisted pair with both good and bad results. Doesn?t look like it would make any difference as far as location on the plans or any other conflict.
 
I put my pitot in the right wing. When I called Van?s about it they said ?well we have just always put it in the left? no issue to put it in right. I?m not flying yet but I have only the lights and magnatrometer in the left wing tip, everything else is in the right. I?m hoping this is a simple solution to this issue. Also I ran a shielded 12 ga wire to the pitot.
 
What about running a single 12ga shielded wire and using the shield for the ground return? This is how lots of LED lights are wired.
 
Probably can be done on some stuff but my avionics guy said with the higher amperage of the pitot the ground return needs to be the 12 or 14 gauge as well. Going to do a 2 wire shielded 12 gauge.
 
Another data point from a flying 14... my installation is all Garmin with an unregulated GAP26 on a Gretz mast, magnetometer on the bracket in the left wing, Vans (SteinAir) standard wiring plus an 18 ga ground wire for the probe running all the way to the firewall (so no twisted pair). Turning the pitot heat on barely registers on the interference test. Moving the ailerons gives a higher spike oddly enough. I ran the wires exactly per the plans and, as Walt mentioned, crossed my fingers. Maybe that?s the secret but my interference test passes easily.

I have read of others having interference when using the regulated probe in a similar installation.
 
Back
Top