I have had the same problem many others have had with my Raymarine autopilot: transmit at high power (100 watts) on the HF band (SSB), and the autopilot reboots, locks up, or just goes goofey.
After extensive stress testing on every band from 160 to 10 meters, I've successfully stopped that behavior with a ferrite toroid choke on the autopilot power leads, and a snap-on ferrite on the autopilot data leads (seatalk from my wind instrument and NMEA 0183 serial data from my chartplotter).
Radio frequency interference (RFI) is tricky: its effects are episodic and suppressing it is full of variables, so "your actual mileage may vary" but, the solution I found is cheap and shouldn't cause any harm. I determined by process of elimination that RF on the autopilot power supply leads is the primary source of problems, and that a snap-on ferrite on the power leads isn't sufficient.
The ferrite toroid I chose for the power supply leads can be purchased from Digikey: http://www.digikey.com/product-detail/en/28B2400-000/240-2120-ND/571928. It'll cost about $10 with shipping. I did all the usual math on its magnetic properties, which you probably don't care about, and it's core material has the best compromise for 2 to 30 MHz. The toroid (donut shaped) core is 35 mm in diameter. Place it as close as possible to the autopilot connector. Pass BOTH the positive and negative leads through the toroid and make at least 8 turns around it while taking care not to flip the positive/negative wires over as they wrap around. Positive and negative should remain parallel all the way around the toroid. I was able to accomplish that with 14 AWG wire. I used zip ties to hold the turns in place.
It is essential that BOTH the positive and negative supply wires are wrapped in parallel around the toroid because, otherwise, the supply current to the autopilot will saturate the core and it will provide no RFI benefit. With the positive and negative wires running parallel, the DC current in the coil's turns has an algebraic sum of zero. This is called a "common-mode configuration" which prevents core saturation from the high DC currents drawn by the autopilot. Once a magnetic material saturates, it is no more effective than air.
For the data leads (NMEA and Seatalk) I used snap-on ferrite suppressors from DX Engineering: http://www.dxengineering.com/search...on-ferrite-beads?autoview=SKU&keyword=ferrite. Those too should be installed as close as possible to the autopilot connector. Be sure to use ferrites that have an inner core diameter that is just large enough to fit over the wires, i.e., a tight fit. This is to get the best coupling between the wires and the core material ("inductive coupling is reduced by the square of the distance"). Now I know Raymarine says not to install any ferrites they don't supply. But I believe my results are better than the outcome of asking Raymarine to: "please redesign your product to work on my boat."
Good luck!
If you're interested in the physics of what's happening inside a toroid, I suggest reading this article: http://hyperphysics.phy-astr.gsu.edu/hbase/magnetic/toroid.html
After extensive stress testing on every band from 160 to 10 meters, I've successfully stopped that behavior with a ferrite toroid choke on the autopilot power leads, and a snap-on ferrite on the autopilot data leads (seatalk from my wind instrument and NMEA 0183 serial data from my chartplotter).
Radio frequency interference (RFI) is tricky: its effects are episodic and suppressing it is full of variables, so "your actual mileage may vary" but, the solution I found is cheap and shouldn't cause any harm. I determined by process of elimination that RF on the autopilot power supply leads is the primary source of problems, and that a snap-on ferrite on the power leads isn't sufficient.
The ferrite toroid I chose for the power supply leads can be purchased from Digikey: http://www.digikey.com/product-detail/en/28B2400-000/240-2120-ND/571928. It'll cost about $10 with shipping. I did all the usual math on its magnetic properties, which you probably don't care about, and it's core material has the best compromise for 2 to 30 MHz. The toroid (donut shaped) core is 35 mm in diameter. Place it as close as possible to the autopilot connector. Pass BOTH the positive and negative leads through the toroid and make at least 8 turns around it while taking care not to flip the positive/negative wires over as they wrap around. Positive and negative should remain parallel all the way around the toroid. I was able to accomplish that with 14 AWG wire. I used zip ties to hold the turns in place.
It is essential that BOTH the positive and negative supply wires are wrapped in parallel around the toroid because, otherwise, the supply current to the autopilot will saturate the core and it will provide no RFI benefit. With the positive and negative wires running parallel, the DC current in the coil's turns has an algebraic sum of zero. This is called a "common-mode configuration" which prevents core saturation from the high DC currents drawn by the autopilot. Once a magnetic material saturates, it is no more effective than air.
For the data leads (NMEA and Seatalk) I used snap-on ferrite suppressors from DX Engineering: http://www.dxengineering.com/search...on-ferrite-beads?autoview=SKU&keyword=ferrite. Those too should be installed as close as possible to the autopilot connector. Be sure to use ferrites that have an inner core diameter that is just large enough to fit over the wires, i.e., a tight fit. This is to get the best coupling between the wires and the core material ("inductive coupling is reduced by the square of the distance"). Now I know Raymarine says not to install any ferrites they don't supply. But I believe my results are better than the outcome of asking Raymarine to: "please redesign your product to work on my boat."
Good luck!
If you're interested in the physics of what's happening inside a toroid, I suggest reading this article: http://hyperphysics.phy-astr.gsu.edu/hbase/magnetic/toroid.html
Last edited:
