"Crazy Ivan" the Autopilot

[reedbr]

I wasn't able to tell if my original "A" heading number on the control head was maintained as I didn't write it down every time I set it.  However, I can tell  that over the two minutes that it took to return to course, the "A" heading never changed.  I think that puts it in case 2 from a previous post.

On fluxgate location, there seem to be wires everywhere.  There is bundle near the rail and the fluxgate now being at counter height in the port aft corner of the head is near it.  One easy mounting location option would be the flat bulkhead that I think is in front of the fuel tank.  I had assumed the metal tank might cause some issues.  I could also attach to the outer hull almost anywhere in the head using epoxy and a monting block.  I'm wondering if the head location in general is my problem though.  The main panel for the whole boat is only 16" in front of of the head.

Who has a MkII with a happy ST4000 and where is your fluxgate mounted?

[Wayne]

Mine is mounted in the bilge area.

[Ken Heyman]

mine is mounted on the bulkhead- in the vbirth- port side. It has always helped my wife and I keep "our bearings" while spending the night.

Ken

[sail4dale]

This problem sounds like my life some time ago.  I finally concluded that the fluxgate compass was my problem and relocated it from LOW in the wet locker aft of the nav station to HIGH in the same locker on the fwd wall.  Voila!  It worked and Ivan quit taking a notion of its own ever since.

I have concluded having it low put it too near the cables under the floor boards from the battery to the instrument panel.  I reasoned that every time something electrical happened the surge in the cable gave the fluxgate something to worry about.

After relocation, I re  oriented the compass by making the prescribed turns until the resulting number was reasonable.We are now at peace with each other.   

[Ken Juul]

Brian,
The fuel tank should be aluminium, should not cause any problems.  The installation instructions say the fluxgate should be at least 3' from electrical/ferrous metal sources. Not sure you can get that much under the sink.  I don't know how the space under the aft cabin in the Mk II is set up.  Can you move the fluxgate aft away from the engine/trans and wiring?

[Tom Clay]

My fluxgate for the auto is also located in the bilge area, and has worked fine for the last 3 years

[Mike and Joanne Stimmler]

Mine is mounted under the setee in the rear berth facing the oven. It's been working good there so far.

Mike

[Roc]

From the factory, mine is located under the port settee, forward of the holding tank.

[Bobg]

I don't know where mine is, still looking, the auto pilot has lately been working, although the heading display is off 60 degrees from the compass reading.  the + and - both work, everything is ok, keeps my course, if I didn't look at the display heading I would think everything is ok,  if I go 10 degrees either way, the display heading acknowledges albeit, wrong.  I do see in the manual I may be able to correct this.  Just not sure how it all happened in the first place.

[Momentum M]

Brian, mine is mounted in the aft cabin, on the starboard side in the storage under the seat.  It's pretty far from any "field" still a reasonable distance (almost under when you think of it) from the pedestal.
It work perfectly and it's always bang on with the compass...if you give time to the compass to ajust to your new bearing...since the digital is instant.
The only thing that I didn't notice is route for the cable..I'll look for it during the w-e....but I suspect that it's under the floor, bed all the way to the stern and then climb up from there.
Good luck

[jmnpe]

Hi guys,

A couple of comments on the last several posts.

Once you have run the auto-calibrate on the fluxgate sensor per the manual instructions, if you have a reasonable deviation value per the instructions, the next thing to do is to adjust the heading correction. During this part of the calibration procedure, you align the autopilot heading readout with your ship's compass readout. If your deviation number is relatively small ( <5 degrees ), you can pick almost any compass heading on which to make the AP heading agree with the ship's compass and the AP heading should agree with the ship's compass heading within about a twice the deviation number you get after the auto-calibrate sequence. If you want to be a little more accurate, you can record the difference you observe between the 2 heading source on 8 equally spaced headings and average the net differences you see, then offset the AP heading readout by the negative of that number at the heading with the largest absolute difference recorded. If you are like me, you only use the AP heading as a relative number, and the ship's compass is the primary heading reference, so if you are reasonably close on the AP heading display, that's good enough. If you read your manual, the heading correction process for the AP is pretty easy after you get a reasonable deviation number using the auto-calibrate procedure.

On the second point, the routing of the fluxgate cable in general shouldn't matter too much relative to heading accuracy or intermittent heading disturbances unless the cable is very close to a very noisy power lead: a high power DC motor drive, a VHF or HF antenna feed coax cable, or the input or output of a square wave ( i.e. - "quasi-sinewave ) inverter, as examples. Other than that, it is pretty hard to disturb the heading signal from the fluxgate compass through coupled interference into the output cabling of the fluxgate sensor.

I am not thrilled about mounting the fluxgate compass sensor in the bilge area having had bilge pump and/or switches fail on several boats leaving water levels in the bilge almost up to the floorboards on unattended boats, both mine and customer boats. While the fluxgate sensor is nominally sealed, I wouldn't want to place any bets on its survival for any lengthy submersion in salt water...... Your boat, your decision.

I think the common thread here on "Crazy Ivan" so far is that magnetic disturbances are probably the primary root cause of the problem, with Brian's problem being the possible outstanding exception since his problem involves a change in displayed heading on the AP control unit during the "Crazy Ivan" thing. I would encourage others who have experienced this problem in its various forms to tell us about what you have experienced in the context of the previous postings describing the 2 possible types of problems. The more information I have the better, the better I may be able to sort this all out.

Thanks to all who have made postings on this topic. Keep them coming!

Regards,

John

[Matley]

Just to add to the data base on this topic, I have also been visited by "Crazy Ivan".

My fluxgate compass is mounted in the storage compartment at the base of the mast.  (I have a Catalina 36 so my interior may be different)

I had replaced the fluxgate compass a few seasons ago so it is relatively new. I have an ST 4000 autohelm about 1993 vintage.

I get a Crazy Ivan (always to port) whenever I am in auto mode on a magnetic heading.  About 15 to 30 minutes after engaging the autohelm seems about the right time.  It happens on any point of travel but I do much more north and south as apposed to east and west.

It does not happen when I am in a "go to" mode of autohelm operation.  My solution is to place the cursor on my Raymarine C80 Chart Plotter, to a point I want to go to.  I engage the autohelm and immediately the chart plotter makes its initial course correction while I manually maintain a course that is very close to that shown by the plotter.  After a few minutes the autohelm makes a second course correction and then I engage the clutch.  In this mode I have gone 30 miles or more with out touching the wheel or the autohelm.

As soon as a crazy Ivan begins I disengage the autohelm clutch, manually correct the course and in a minute or two, the autohelm returns to the previous course.  The interesting thing is that when the autohelm starts a crazy Ivan, it try's to turn the wheel (and therefore the boat) much more to port than I let it, but seems to quickly accept the manual heading I hold, I does not "give back" what it takes from the original course.  I know this may sound a little confusing but if the crazy Ivan try's to turn the boat 90 degrees to port, if I hold the proper course manually by quickly disengaging the clutch, when the autohelm comes back to starboard it seems to know I am not that far off course and picks up the manual course (I reengage the autohelm) and it makes a few small course adjustments before all is forgiven and the original course is followed again.   

This small give back after a crazy Ivan doesn't smell like a magnetic interference or magnetic field fluctuation to me. It seems to be more connected to a GPS vs magnetic disagreement.  Since I get no crazy Ivan when in a "go to cursor" mode of navigation, I think that the Autohelm 4000 unit gets confused when both magnetic information and GPS heading information is available to it.

My question, does anyone know, will the Autohelm 4000 work without the fluxgate compass attached?  Won't the Autohelm just use the GPS heading information? 

Also, I am curious, is anyone having Crazy Ivan's with no chart plotter installed? 

[Mike and Joanne Stimmler]

I have one of the newer X-5 auto pilots and I was having some strange issues when I would hit the track button on the auto pilot to go to a waypoint. It would veer off about 40 degrees or more to one side or another.
There is a function on mine called "reset XTE" and I don't know if that function on the older models but that fixed my problems.
I think that if you have the AP set for a compass course for awhile and then switch to "track" it will mis calculate XTE and cause the problem of the course differential.
This may be altogether different than the trouble the rest of you are having but I just thought I would put it on the table.

[Matley]

Mike, the XTE is the cross track.  This is the distance your boat is "off track" from the course you originally put in to the autohelm.

Hitting the Reset XTE button tells the system to recalculate your course to the destination from this new point.

Perhaps a simple way to think of XTE it is to imagine you are in the 5th lane of a 10 lane highway (counting left to right).  If you drift to lane 8 you may still be heading toward your destination but you are 3 lanes off your original track.  This distance from lane 8 to the original lane 5 is the XTE (Cross Track Error).  The Autohelm will overstear to the left until your boat is back in lane 5. Resetting the XTE tells your Autohelm that your present position in Lane 8 is fine and it should navigate to your destination from Lane 8.

Veering off the 40 degrees you mention sounds like the "Crazy Ivan" maneuver we are trying to solve.

Bill Matley
Catalina 36
1252

[Mike and Joanne Stimmler]

Thanks Bill,
You're absolutely right! And I believe cross track error is also the difference in your compas heading and the course the boat is actually moving. So I'm thinking it (the gps/chart plotter together with the auto pilot )actually compensates for set and drift and that could explane the difference in heading, although 40 degrees is a bit much.