Electrical issues with Autopilot 2 days into pacific crossing

[Sppro]

**I am relaying this question for my friend who is 2 days into a Pacific Crossing on a new yacht, he is communicating to me via a Inreach messaging device so can only receive/send SMS.. They do have a pretty technically savvy crew. this is going to be fun**

We experienced a massive electrical storm early into our pacific crossing from Panama to French Polynesia. We managed to get most of the electronics back online but are experiencing issues with out autopilot. It's reading inconsistent magnetic bearings which might be best described as having multiple Norths.

Our autopilot is a auto-helm type 300 with flux gate compass. Should the compass be the cause of the issue, can we bypass it by connecting to GPS or wind directly?
Or can we engineer something with the project electronic on-board. We have a mpu9250 the code, libraries and experience programming it for other applications with arduino but I need to know what protocol the flux compass uses to communicate with the autohelm unit. Can anybody assist please?

[Adelie]

What kind of boat?

[Paul L]

You can test the fluxgate compass with a multimeter. See the table here for expected resistance readings
http://www.raymarine.com/knowledgeba....cfm?view=3798

[cyan]

Originally Posted by Sppro

Our autopilot is a auto-helm type 300 with flux gate compass. Should the compass be the cause of the issue, can we bypass it by connecting to GPS or wind directly?
Or can we engineer something with the project electronic on-board. We have a mpu9250 the code, libraries and experience programming it for other applications with arduino but I need to know what protocol the flux compass uses to communicate with the autohelm unit. Can anybody assist please?

Unfortunately, the Raymarine Type 300 is an older course computer which can ONLY get heading information from an analog fluxgate compass, which has several coil connections. This unit does also use NMEA 0183 digital serial data protocol, BUT it can only SEND the heading info (as decoded from the fluxgate), not RECEIVE it over that protocol. (or the Seatalk)

My guess is that the lightning fried at least one of the delicate coils wrapped around the iron fluxgate core (or driver/receiver circuitry inside the course computer itself) This circular core has super-thin gauge magnet wire wrapped in several potted coils, all this suspended on a gimbal.

The fluxgate compass coils can easily be checked with a multimeter:
(I got this a while back from a Raymarine help forum- probably the same one as Paul_L above)
Red to Green - 4 ohms
Red to Yellow - 4 ohms
Green to Yellow - 8 ohms
Screen to Blue - 8 ohms
All other combinations - OPEN

If it was life-or-death, I would give it the old college try...
1. Verify that there is a bad coil with the ohmmeter test above
2. Pray that a ONLY a wire connecting a coil was zapped and NOT the whole coil.
3. Open the compass housing and see if my prayer was answered
4. If my prayer WAS answered, then re-solder a connection to the coil whose wire was zapped, test with meter, reassemble, sail away with working autopilot, crack a beer.
5. If my prayer was NOT answered (only a blob of copper left inside), chuck the whole works into the sea, sketch a makeshift windvane made from a broom handle + kayak chunks + duct tape, crack a beer.

[deblen]

Other possibilities:


1. Lightening current magnetized something in the vicinity of the compass.
Try relocating the compass.


2. Do the initial setup "circle" procedure as detailed in Paul L's link.


Keep in mind that the autopilot will work fine even if the compass displays wrong course-as long as that course is stable & changes with the boat's course.


Len

[ka4wja]

sppro,
In addition to the possibilities of damaged fluxgate compass, and/or something ferrous being magnetized near the compass, it is also possible that the corpak / course computer needs recommissioning??

Although not the easiest thing to do at sea, I would do a total reset and recommission the unit as if it was a new install...

And, in this instance the important part here is re-deviating the fluxgate compass (know as "swinging" the compass)...
(I did this on my old unit many years ago, when at sea...but in benign conditions...so, it can be done, just not as easy as when in harbor)

BTW, if you can give their approx. location (within couple hundred miles), I assume a few hundred miles SW of Panama...near Galapagos??
I can call them on HF radio ("SSB") and assist live...(12mhz and 16mhz in late afternoon, and then 12mhz and 8mhz at / after sunset??) If we cannot make contact, then WLO should have excellent coverage, and we can work thru them.

fair winds.

John

[ilenart]

If they are two days out, cannot fix it and have no backup they should turn around / head to Galapagos and get it fixed

[Dsanduril]

Uncalibrated fluxgate compasses don't know N from S - which can provide the exact symptom of 'multiple norths'. This is one thing that gets sorted when you do the initial calibration routine/compass swing (along with the deviation table). And a nearby lightning strike is a known cause of loss of calibration.

Recalibrate your fluxgate compass from time to time, since several factors could cause it to lose calibration. These could be as benign as a crew member inadvertently stowing metal objects nearby, or as serious as shore power surges or nearby lightning strikes.

While doing a couple of 360s in the middle of the ocean can be a little bit of a pain it's an overall easy thing to see if you can get the AP back.

[AndyEss]

From my Autohelm 100/300 manual, the compass heading alignment procedure Dsanderil refers to is as follows:
1 push and hold DISPLAY for 2 seconds until the display shows HDG, DEV. Slightly different for the st6000 or st7000 display/control modules
2 Press DISPLAY
3 Keeping boat speed below 2 knots, turn the boat slowly so it takes at least 3 minutes to complete a 360 turn. Keep turning (maybe as much as two full turns) until the DEV is updated. If DEV is more than 15 degrees, move the fluxgate compass because there is something magnetic around it
4 Steady boat on one heading, and use course change keys to adjust the displayed heading until it agrees with the steering compass (assuming that wasn't effected by the lightning strike also)
5 To exit fluxgate alignment /correction/calibration and save new settings, push and hold STANDBY for 2 seconds until display reads Standby
6 To exit without saving new settings, push STANDBY momentarily

[AndyEss]

PS - i keep an extra fluxgate compass as a spare. They're pretty cheap on Ebay. My non-electronic Monitor also does a great job steering when i single hand.
Redundancy of critical components / systems is key for existing in remote situations. Sailing the Pacific is pretty remote

[sheldon957]

As an old mechanic, the #1 rule in the shop is DO THE EASIEST things first! Try the recalibration with the 360's as advised above before taking apart things.

[hamburking]

The coils of the electronic compass may have become gaussed. You can do a deggausing procedure if you have a strong magnet or electomagnet. I did this once, long ago, when my TV got guassed (long story, not relevant here). I borrowed a degaussing coil and did the little degaussing dance where you pull the flux away from the coil...and it really worked.

Just a thought.

[cyan]

A ton of interesting suggestions, ranging from harmless to violent, none of them involving writing code as hinted at by the OP.
Hey Sppro, any word from our hand-steering brothers at sea?
Your one-post question has a few of us curious.

[simonthepom]

I would try just putting it on windvane mode (auto and standby buttons together), don’t think it needs the compass to maintain wind angle.

[billknny]

Originally Posted by simonthepom

I would try just putting it on windvane mode (auto and standby buttons together), don’t think it needs the compass to maintain wind angle.

Not true. "Wind mode" just uses wind data to adjust the set point of the magnetic compass heading. It still requires a working compass.