rjcthree
Well Known Member
I had a problem with my vertical card compass swinging with the combined current load of nav lights, landing lights and strobe light power supply. With all three on, the current is about 15A in one direction from the breakers to the switch, running along three closely packed wires of about 20” length left to right, about six inches under the compass. The compass swing was about 25 degrees from N-S with switching of the load, and would hitch or hang up at anything more than 45 degrees from N-S. Moving the compass was not an easy option, moving the wires was not an option.
I tried the 5”x12” mag shielding from ACS bent in a U shape with about 2” of clearance between the compass and the mu metal. It came up most of the right side and about 2” up the left of the compass. The front, rear and top were open. Initial results were positive until I moved from my hanger, where it became obvious that the shielding was also shielding the earth’s magnetic field. N-S it was ok, but get beyond 30 deg of N-S, and the compass would freeze. The response to current on-off was about 5 degrees, a significant but useless ‘improvement’.
I bought more shielding and consulted with a university contact who is doing high voltage/high frequency power supply research and a few things got cleared up. My first try created a shielded tunnel. Remember, when you use shielding, everything gets shielded, not just the bad stuff– or in the case of mu-metal, directed through the metal. Duh!
The second shield was simply a horizontal flat plate, 5”x8”, like a shelf, about three inches under the compass. It’s offset due to the IP bracket (tip-up), three inches to the right and five inches to the left of the CL of the compass. It’s contact cemented to a 0.020” aluminum backplane and velcroed in place.
The result is very positive. When cycling power on to power off the compass swing is about 3 degrees now (not much more than a twitch), well within compensation range. I’m changing out my nav bulbs for LEDs; that should further reduce the swing. The one thing I can say about my setup is that the things that made it swing were essentially ‘always on in flight’ (for me – I fly with LED landing lights on usually), so I can reliably dial it out. I’ll probably only ever use it the compass if both the Dynon and Garmin give it up, or when I’m in a compass/stopwatch mood, but still nice to have.
Other thoughts: Had I ran power to the switch and back the same way in a twisted pair, I likely would have had no problem. Likewise, a shielded (by mu-metal) conduit would have worked, but would have required significant rewiring. Mu metal isn’t a damper or absorbant, it just redirects the magnetic field – the field would rather go through the metal than air, it’s sort of like a magnetic field detour in non-scientific terms (hey, I’m a BSME, this sparky stuff is weird!). In the end, the energy is going go in and come out the other end. The mu-metal is a pain to work, and it is very sharp, and you get one bend before it gets brittle. If you use a pop rivet around the compass, be aware that the mandrel may get magnetized by the stretching and breaking process (like in constructing a map box, radio stack, or tip-up frame to deck rivets). I was pleased to find my radios SL30 & GTX327 were not part of the problem, nor was the engine state.
As always, YMMV.
I tried the 5”x12” mag shielding from ACS bent in a U shape with about 2” of clearance between the compass and the mu metal. It came up most of the right side and about 2” up the left of the compass. The front, rear and top were open. Initial results were positive until I moved from my hanger, where it became obvious that the shielding was also shielding the earth’s magnetic field. N-S it was ok, but get beyond 30 deg of N-S, and the compass would freeze. The response to current on-off was about 5 degrees, a significant but useless ‘improvement’.
I bought more shielding and consulted with a university contact who is doing high voltage/high frequency power supply research and a few things got cleared up. My first try created a shielded tunnel. Remember, when you use shielding, everything gets shielded, not just the bad stuff– or in the case of mu-metal, directed through the metal. Duh!
The second shield was simply a horizontal flat plate, 5”x8”, like a shelf, about three inches under the compass. It’s offset due to the IP bracket (tip-up), three inches to the right and five inches to the left of the CL of the compass. It’s contact cemented to a 0.020” aluminum backplane and velcroed in place.
The result is very positive. When cycling power on to power off the compass swing is about 3 degrees now (not much more than a twitch), well within compensation range. I’m changing out my nav bulbs for LEDs; that should further reduce the swing. The one thing I can say about my setup is that the things that made it swing were essentially ‘always on in flight’ (for me – I fly with LED landing lights on usually), so I can reliably dial it out. I’ll probably only ever use it the compass if both the Dynon and Garmin give it up, or when I’m in a compass/stopwatch mood, but still nice to have.
Other thoughts: Had I ran power to the switch and back the same way in a twisted pair, I likely would have had no problem. Likewise, a shielded (by mu-metal) conduit would have worked, but would have required significant rewiring. Mu metal isn’t a damper or absorbant, it just redirects the magnetic field – the field would rather go through the metal than air, it’s sort of like a magnetic field detour in non-scientific terms (hey, I’m a BSME, this sparky stuff is weird!). In the end, the energy is going go in and come out the other end. The mu-metal is a pain to work, and it is very sharp, and you get one bend before it gets brittle. If you use a pop rivet around the compass, be aware that the mandrel may get magnetized by the stretching and breaking process (like in constructing a map box, radio stack, or tip-up frame to deck rivets). I was pleased to find my radios SL30 & GTX327 were not part of the problem, nor was the engine state.
As always, YMMV.
Last edited:
