Alternator interferes with electronic compass

[colemj]

I have some experimenting to do.

As far as moving the compass, it is already 20' away from the offending alternator. The only high current wires near it come from the alternator 5' away that does not cause any issue.

I thought of another experiment to help pinpoint things - if I recalibrate the compass while it is being effected by the offending alternator, the calibration settings will tell me the magnitude of the magnetic field compensated for and which direction it is from the compass mount. However, it will be a week before I can do that one. In the meantime, I will proceed with some of the other experiments.

Mark

[jeremiason]

Originally Posted by colemj

I have some experimenting to do.

As far as moving the compass, it is already 20' away from the offending alternator. The only high current wires near it come from the alternator 5' away that does not cause any issue.

I thought of another experiment to help pinpoint things - if I recalibrate the compass while it is being effected by the offending alternator, the calibration settings will tell me the magnitude of the magnetic field compensated for and which direction it is from the compass mount. However, it will be a week before I can do that one. In the meantime, I will proceed with some of the other experiments.

Mark

Mark:

Two things... One did the fluxgate ever work correctly and if it did, what did you change since it started being influenced???

An easy troubleshooting method would be to put a temporary long wire on the compass and move it away from all wires and anything electric... Including the secondary alternator.

Then see if it works... If it doesn't your Fluxgate compass is probably bad, if it does you know you have to move the compass because it is being influenced by something.


Good luck!

[Dsanduril]

Where is the engine's ignition/instrument panel in relation to the compass? The field current for the alternator is usually controlled by the key switch, and usually passes through the system mono-polar (no associated ground return). On our boat when this happened the battery cables and main alternator cable were just fine, it was the alternator field current, flowing in a wiring harness near the compass, that was causing the problems.

[Edit] an easy test would be to remove the field wire from the input to the regulator, and then source it from some other location. That way you could run the alternator at power without field current in the normal harness[/edit]

[St. Elsewhere]

Originally Posted by jeremiason

I had a similar problem when I had my C-36...


The fluxgate compass was not located near the alternator, but then I realized that the Battery cables were.


When the batteries were discharged the alternator was kicking out about 75 or 80 amps of charging power it would crate a electronic field around the cables and throw the fluxgate compass off. This only occurred when the batteries were discharged and the alternator was in high output.


I moved the fluxgate compass and had no further problems.

Was the compass near the battery cables? How did moving the compass resolve the issue? Were all DC wires creating sufficient magnetic interference to cause trouble, or only high current conductors?

[colemj]

I just did a test with a HH electronic compass next to the main compass. It also changes its bearing when the alternator is running. So that suggests a true magnetic field is present.

However, I don't know what would be causing this. The only wires that run close by are from the alternator that does not cause any problem.

I have a difficult time believing that the alternator is projecting a strong magnetic field 20' out. It is not possible to move the compass any further away.

I will dig through the conduits and related areas to determine if there is in fact a current-carrying cable related to that alternator. Even if there is, I don't understand why the much closer alternator and its cables are not causing any problem. It can't be just that the other alternator is larger because the problem occurs even when the output is only 10A or so.

The field current is supplied by the voltage regulator mounted near the alternator. The engine panels are at the helm.

Mark

[David M]

Originally Posted by Steve Bean

Re twisted wires: I'm referring to a recommendation I heard years ago to twist the positive and negative wires of a DC circuit around each other to avoid interference with a compass. Does anybody here know if this is valid, or an old wives' tale? Might be easy to construct a test.

It's true, twisting wires in a DC circuit eliminates any magnetic field.

It is more likely that a DC wire which goes from the alternator, near the compass and to a battery is causing the magnetic interference.

If you cannot find a solution you can always go back to the old days where you had a Standard compass and a Steering compass. The Standard compass was put in a place away from stray DC caused magnetic fields affecting deviation. The Standard compass was put in a place where much of the deviation could be zeroed out with magnets. For what could not be zeroed out, a deviation card was created. The steering compass was put at the helm. When you were on the heading that you wanted, you "marked" where that heading was on the steering compass. When you had a large course change (greater than 15 degrees or so) you had to do this procedure again. It did not matter the accuracy of the steering compass this way, as long as the error remained constant, which it did as long as you had no large course changes.

[Xthewater]

Twisting cable does help nullify induced fields/signals that's why Ethernet cables are rated at twists per foot.
It may be you've lost shielding ground connection on the cable, which bleeds off inductive field. Likewise if cable shield is connected on both ends then the current loop is actually causing a problem rather than preventing it.

[Rockyscience]

I think it is magnetic, high current loads can generate large mag. fields.
The easy test is a faraday cage around the fluxgate compass or find a metal box to completely surround it with a hole just big enough to see the readout, if that stops the problem you can have a nice one fitted to it later.
Rocky

[colemj]

The test with the HH compass does suggest there is a magnetic field interference. While putting the compass in a faraday cage may solve the interference problem, it would defeat the purpose of having a compass at all.

Mark

[Terra Nova]

Originally Posted by Terra Nova

You might try isolating from that alternator output the 12-volt source which powers your autopilot compass.

This.

[ex-12m-guy]

Another experiment: Swap the alternators and regulators between engines and see if the problem moves with the swap.

If so, then replace or rebuild the alternator / regulator pair that is causing the
problem.

If swapping alternators does not move the problem, then investigate the wiring closely. Look for a bad ground somewhere in the 12 v. charging and distribution system in the side causing the problem.

[colemj]

I just powered up the compass (the entire N2K bus actually) from a stand alone battery that was connected to nothing else. The problem still exists.

So it has nothing to do with the power source, and appears to be a magnetic interference. From 20' away.

I will try to move the compass away from it, as well as see what cables could be running near it that I am unaware of.

I'm approaching these experiments by how easy and how much time they take instead of being more methodical because we have several things going on right now.

I will save swapping the alternators to later down on the list because that is much more work. I will investigate the wiring on the problem alternator.

I doubt the compass is bad because it works just fine as long as the alternator on that one engine is not running.

There isn't a need for a "standard/steering compass" because this only happens when we are motoring on that engine, and even when the compass is off from the actual heading, the AP either uses a relative heading or nav data, so the course steered is fine, and we can always choose to use the other engine if it is a problem for some reason and we need to motor.

I appreciate the suggestions and troubleshooting help.

Mark

[Wind River]

Originally Posted by jeremiason

I had a similar problem when I had my C-36...


I changed the stock alternator to a high output one.


The fluxgate compass was fine when we left the dock, but if we discharged the batteries, say at anchor, the electronic chart heading would be off as much as 90 degrees.


The fluxgate compass was not located near the alternator, but then I realized that the Battery cables were.


When the batteries were discharged the alternator was kicking out about 75 or 80 amps of charging power it would crate a electronic field around the cables and throw the fluxgate compass off. This only occurred when the batteries were discharged and the alternator was in high output.


I moved the fluxgate compass and had no further problems.

I just installed the Airmar H2183 heading sensor . I noticed a similar problem while I had my boat plugged in to dock power last weekend. While the battery charger was putting out 80 amps or so I noticed my heading, compared to the dock shown on my chart plotter, was off by about 45 degrees. Once the batteries were reaching full charge and the current they were accepting was less than about 10 amps, the heading seemed to be more accurate.

I did try moving the sensor as far away from the charger as the current cable would allow (while there was a heavy charge current) and the discrepancy was far less.

Just to be clear though, I have not done any type of calibration on this sensor yet.

I'm sure this does not help you at all, but it is just another case of interference from a high amperage charging source messing with a heading sensor.

[colemj]

Well, that was unexpected.

Problem solved. Short story is a ground loop caused by a vestigial second ground wire running near the compass.

Long story: Outside the engine rooms are power terminal posts that connect to the alternator and run to the house batteries. These exist because the wiring runs in a catamaran are long and complicated, and the posts simplify things a bit.

Also connected to these posts (I will focus on the negative post here) in each engine room are a few other wires, such as the ground to the VSR combiners and a cross-over ground to allow starting the engines from the house battery if a start battery dies.

So, when I connected the new alternator, I connected the negative wire to the negative terminal post with the intent that it would then go directly to the house batteries (which it does).

HOWEVER, although I knew there were also negative cables attached to this post and crossing over to the other hull, what I didn't realize was that one of them was a vestigial wire from the factory that not only crossed to the other hull, but CONTINUED STRAIGHT TO THE HOUSE BANK!

Going through a conduit right by the compass, of course.

I don't know why this wire was originally run this way - it is possible the PO ran it for some reason, or reconfigured it - but the result was that there were now two paths from the alternator negative back to the house, and one of those paths ran by the compass without a corresponding positive wire.

So there was a ground loop next to the compass.

Disconnecting that wire solved the problem.

Thanks for all the help and advice.

Mark

[Snowpetrel]

Thanks for the update. Interesting Thread, and a great bit of troubleshooting. A good reminder to keep an eye on your compass error!

Sent from my HTC_0PCV2 using Cruisers Sailing Forum mobile app