What's new
Van's Air Force

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

Garmin GMU11 alignment question ?

Larry DeCamp

Well Known Member
Garmin told me they just do electronics," ask the RV community". So,the issue is the magnetometer should align within .5 degree with longitudinal axis of the fuselage. OK, I can define the axis with plumb bobs and string, but defining a line of any significance on the GMU11 ( little box) seems impractical. My solution is to measure the fuselage axis heading (string on the ground) with iphone compass and then adjust the GMU11 box so the compass readout agrees. Any suggestions appreciated.
 
Wouldn?t aligning the mounting holes on the GMU do the trick.
I might misunderstand the challenge, but I would:

1. Select location (that is important as well).
2. Draw a line parallel to the longeron
3. Offset that line to where the GMU mounting holes will land
4. Mark the holes location using the GMU
 
Bjorn response

The longerons are not parallel,but a fuse center line vector could be established within the dimension of a string. The issue I perceive to be unrealistic is the fact that .5 degree over the length or width of the GMU 11 is pretty small dimension. Squaring or paralleling that with the fuse reference leaves a lot of opportunity to drift ( rulers,strings, plumb lines, calipers, pencil lines etc.) I am probably making too much of this. Pitch and roll axis have EFIS correction options within limits, but I don't read this one of those?
 
Of course, I didn't catch "fuse center line vector" . I was in 2D pitch mode :p.

Where to you plan to positioning it ? It could be necessary to build a platform. That platform can be positioned using combination of level for pitch and an angle secured to bulkheads which is aligned with a center string. This could give you a two-axis alignment.

I also heard that a simple boy scout compass can be used to test for magnetic interference, just in case there is some magnetized steel in the mounting area.
 
I think I'd take that .5 with a grain of salt. I've just looked at the G3X installation manual to refresh my memory. The installation does indeed call for .5 degrees. But then I look at the configuration (sections 34 or 35 depending on your display) and the pitch calibration is made in reference to the AHRS, with instructions to re-calibrate if the magnetometer is moved or changed. The heading calibration is made by taxiing to a compass rose and aligning to magnetic North within 5 (not .5) degrees and then allowing the software to guide you through a 360 degree pirouette. Given that, I think your plan for alignment is probably going to be more than adequate.

When you add to that the difficulty in trying to determine the centerline of your aircraft, I believe you should aim for 'good enough' rather than 'perfect'. I mention the centerline, because there may be slight variations if you measure between the longerons, along the bottom of the fuselage, rudder hinge to spinner tip (there is a slight engine offset), etc., so the reference line may not be exactly accurate. I think the software correctly establishes 'heading' during the calibration, which should be accurate enough since track is normally determined by GPS and only by wind/heading calculations in case of GPS failure.
 
Thanks for the help.

I did not catch in the installation instruction that it will auto calibrate. My MGL did exactly that and works well. Thanks again...Larry
 
Hello Larry,

It is true that you don't have to accurately align the aircraft with magnetic north at the beginning of the magnetometer calibration. +/- 5 degs is fine.

You don't even have to perform this calibration on a compass rose, just a location where the earth's magnetic field is reasonably free of disturbances.

What is not true is that it doesn't matter how accurately you align the magnetometer with the centerline of the aircraft. This is one of the few "knowns" the calibration algorithm relies upon.

If you install the GMU 11 with an alignment error, that error will always be reflected in the heading displayed by the system.

A 0.5 deg error is insignificant, because heading is a relatively inaccurate measurement, and the TSO requirements for heading accuracy reflect that.

Even if you had a 1 degree alignment error, it is unlikely you would see that in flight, but anything more than a couple of degrees is likely to be detected at some point, especially if you visit a marked compass rose.

When we were first developing and flight testing the GMU 11, I installed one in the left wing of my homebuilt against the main spar believing this orientation would be perpendicular with the aircraft centerline. I built this plane 20 years ago, and had forgotten that the outer section of the main spar is swept back a few degrees on each side.

Each time we calibrated the GMU 11, it consistently reported a heading a few degrees left of the heading reported by the GMU 22 mounted in a different location in this plane. After looking at the raw data, we came to the conclusion that the unit had to be misaligned by the difference in headings.

Sure enough, when I referenced the main spar plans, I was reminded of this swept angle which was very well known to me at this point since it was within 1 deg. of the heading difference between the two magnetometers.

I built a wedge of the correct angle needed to correct for the sweep angle and aligned the GMU 11 perpendicular to the centerline. No more heading disagreement after that.

Thanks,
Steve
 
Thanks, Steve, for increasing MY fund of knowledge.

Hi Patrick,

You are welcome.

Magnetometer calibration is tricky business when there is no requirement to have to tell the system when you are at each point on the compass rose, and instead make it watch the earth's magnetic field and figure it out for itself given just a starting point that is roughly magnetic North.

All of our certified and non-certified systems have always worked this way to remove the need to have a compass rose on each field where calibration is to be performed. It is good to use a compass rose when you have it because these sites are generally magnetically "clean", but not required.

Thanks,
Steve
 
Back
Top