Faulty Flux gate Compass unit

Slipstream 34

Slipstream 34

Well-Known Member
Joined
15 Feb 2006
Messages
682
Location
Clyde
Visit site
I have a Cetrek 7000 autopilot fed by a flux gate compass unit. Unfortunately the heading information during a 360 deg turn will read 050 for first half of the turn and 230 for the other half. Obviously these are reciprocals, so my first thought was a wiring fault. I dismantled the unit and tried to ensure the miriad of small copper wiring were not touching....tricky job see photo... I refitted it today and the problem is still apparent.

Has anyone had a similar problem ?

Any advice gratefully received.
 
Last edited:
BJ4U said:
Anymore pictures . Name on unit etc . Any other electronics inside
Click to expand...
The Left hand photo is of a similar unit. The right hand photo of the PCB above the coils.

There is no makers name on the unit and no instructions in the Autopilot manual other than to remove the plastic transit screw from the base after installation...it was still fitted !

The output plug has only 5 pins, others units I have searched generally have 7 pins.
 
Last edited:
Been having similar problems for the last year, all of which were supposed to be fixed by the firm we bought the boat off, but we have now given up with them and are trying to get the problem sorted ourselves.

Different make but the problem with ours was that a slow turn of the boat just caused the compasss to move within a 60ish degree range and back again. I took away everything that was removable and possibly would affect the field. I took the compass off the bulkhead and rotated it by hand. I recalibrated the whole unit from factory settings, and I checked the wiring connections for breaks on the cable, for correct fitting, and for any loose connections.

Doing all the above was done on a whole series of trips to the boat and complicated by my strong belief that the boats supplier was coming back to the boat and making changes to try to fix it, without saying what those changes were, or that they had been on the boat.

The (unsuccessful) techie was convinced it was a computer, not a compass problem, but changing a compass is far cheaper so on the last trip I brought down a new compass. The unit then all appeared to work perfectly which was a little frustrating as that happened before I had even taken the new compass out of the box, let alone fitted it.

Hope some of the suggestions above help a little, and that unlike me you end up understanding more , not less, about the autopilot.
 
RupertW said:
Been having similar problems for the last year, all of which were supposed to be fixed by the firm we bought the boat off, but we have now given up with them and are trying to get the problem sorted ourselves.

Different make but the problem with ours was that a slow turn of the boat just caused the compasss to move within a 60ish degree range and back again. I took away everything that was removable and possibly would affect the field. I took the compass off the bulkhead and rotated it by hand. I recalibrated the whole unit from factory settings, and I checked the wiring connections for breaks on the cable, for correct fitting, and for any loose connections.

Doing all the above was done on a whole series of trips to the boat and complicated by my strong belief that the boats supplier was coming back to the boat and making changes to try to fix it, without saying what those changes were, or that they had been on the boat.

The (unsuccessful) techie was convinced it was a computer, not a compass problem, but changing a compass is far cheaper so on the last trip I brought down a new compass. The unit then all appeared to work perfectly which was a little frustrating as that happened before I had even taken the new compass out of the box, let alone fitted it.

Hope some of the suggestions above help a little, and that unlike me you end up understanding more , not less, about the autopilot.
Click to expand...

Thanks for your reply. I have not been able to source a replacement as it is circa 1985. I could only find similar not identical units on an American website, they have 7 pin connections not 5. Hopefully I can get it working again.
 
akela said:
Thanks for your reply. I have not been able to source a replacement as it is circa 1985. I could only find similar not identical units on an American website, they have 7 pin connections not 5. Hopefully I can get it working again.
Click to expand...

Might be worth checking what wiring/signal protocol the unit accepts from the compass. There may be modern compass units which are compatible. Also I bought my replacement from ebay so might be worth googling for old secondhand sensors, though you've probably already done that.
 
From what I have found it is a UK company that has ceased trading. As the compass unit has no makers name unless you have an installation guide for the main unit you are going to have trouble sorting out a replacement. By the look of the board there does not seem to be any intelligence there so you could assume that the main unit does some of the signal processing. It difficult to read the numbers on the ic's but you could try looking them up and seeing what they are. If one is a line driver of some sort you may be able to work out what the 5 wires are.

At the end of the day its probably a basket case and you may as well fit something a little more modern.
 
Last edited:
akela said:
I have a Cetrek 7000 autopilot fed by a flux gate compass unit. Unfortunately the heading information during a 360 deg turn will read 050 for first half of the turn and 230 for the other half. Obviously these are reciprocals, so my first thought was a wiring fault. I dismantled the unit and tried to ensure the miriad of small copper wiring were not touching....tricky job see photo... I refitted it today and the problem is still apparent.

Has anyone had a similar problem ?

Any advice gratefully received.
Click to expand...

If you have a ohmmeter (resistance tester) try the continuity of the coils. To be more specific: two of the input wires should be interconnected through a very small resistance (a few Ohm) and the other three should be interconnected as well. No connections is expected between the two groups. If you are able to read the resistance the better. From the problem you describe I expect that you will find that one of the group of three is disconnected enabling only one axis. Your compass seems to have two axes displaced 90 degrees.

Daniel
 
akela said:
I have a Cetrek 7000 autopilot fed by a flux gate compass unit. . . . . . . . . . . Any advice gratefully received.
Click to expand...

I spend 25 years in radio communications and electronics and a load of time I was dealing with the forensic side of problems and often I found that 'someone' had 'got at' the circuitry before it came to me in the government research laboratories! :mad:

I do not think the problem is inside the 'head' but if you take a very, very good look at the wiring at the plug, where it fits on the PCB, you might find that someone has possibly re-worked' the wires and they are now incorrectly arranged on the 5-pin female that fits the male 5-pin on the PCB.

All is not lost as two of the five wires will surely be the 'pos' and 'neg' supply for the board leaving you only to worry about the three data wires, one of which is probably 'a common' to the 'head unit'. The 'pos' or 'neg' supply to the PCB can be swiftly established by seeing which connects via the trackwork on the PCB to the -ve side of the large electrolytic capacitors. ;)

The Hex Inverter Buffers MC14049UB) and Quad Op-Amps (LM324), as seen in the photograph, will require a 'positive' supply as will the two transistors. :)

The PCB is not too complex and can be drawn out on paper within half a day and then with that info, it is only a question of probing with a 'scope?

Another strong possibility will be a dry-joint on the PCB as the unit has been in a vibrating, hostile environment for the past 20-30 years?

You are not near Gosport so I cannot say bring it in for appraisal? :(

Happy New Year to you. Get back if you require any further help?
 
Last edited:
lenseman said:
I spend 25 years in radio communications and electronics and a load of time I was dealing with the forensic side of problems and often I found that 'someone' had 'got at' the circuitry before it came to me in the government research laboratories! :mad:

I do not think the problem is inside the 'head' but if you take a very, very good look at the wiring at the plug, where it fits on the PCB, you might find that someone has possibly re-worked' the wires and they are now incorrectly arranged on the 5-pin female that fits the male 5-pin on the PCB.

All is not lost as two of the five wires will surely be the 'pos' and 'neg' supply for the board leaving you only to worry about the three data wires, one of which is probably 'a common' to the 'head unit'. The 'pos' or 'neg' supply to the PCB can be swiftly established by seeing which connects via the trackwork on the PCB to the -ve side of the large electrolytic capacitors. ;)

The Hex Inverter Buffers MC14049UB) and Quad Op-Amps (LM324), as seen in the photograph, will require a 'positive' supply as will the two transistors. :)

The PCB is not too complex and can be drawn out on paper within half a day and then with that info, it is only a question of probing with a 'scope?

Another strong possibility will be a dry-joint on the PCB as the unit has been in a vibrating, hostile environment for the past 20-30 years?

You are not near Gosport so I cannot say bring it in for appraisal? :(

Happy New Year to you. Get back if you require any further help?
Click to expand...

I am afraid but what you describe does not seem to fit the actual situation. The author of the post writes that the fluxgate compass, depending on its orientation, feeds a signal that the PCB interprets only as two fixed positions 180 degrees apart. Most of the times this can be attributed to a failure in one of the "sense" coils inside the compass or its connection. If he is able to find the defective coil/connection he might be able to pinpoint the problem and maybe fix it. Of the five wires, two carry an alternating current (possibly 400 Hz) from the PCB to the "exciting" coil and three carry x-y signals from the two "sense" coils to the PCB. My interpretation is that the "exciting" part of the circuitry works fine but the PCB receives the signal from one axis (sense coil) only. Among the five wires there is no whatsoever positive or negative supply to the PCB because the power supply is taken care elsewhere.

Daniel
 
nigelmercier said:
In that case, first suspect an install problem.
Click to expand...

Thanks Nigel. After many hours of fault diagnosis I am fairly confident the compass head is faulty. I have managed to source a replacement compass unit from the USA, marine-plus.com. They have lots of antique electronics and are extremely helpful. Item should arrive next week.....fingers crossed that will cure the fault.
 
You must log in or register to reply here.

Share this page

Top