Autopilot wiring question

[Snowpetrel]

Trying to troubleshoot an autopilot on my folks 45 foot ketch. The wiring was recently completely redone for the entire vessel, and a very nice professional job was done. The marine electrician rewired the Navman Autopilot into the main panel, and since then it has been playing up and shutting itself down completely at awkward moments. Initially the AP was shutting down while the engine was running, but not under sail. A voltage conditioner fixed that but now it struggles, and still occasionally dies.

It also seems the high amp alarm is going off in only moderate conditions with 20 amps of load. I improved things by reducting gain and counter rudder to reduce the amount of work the AP did, and it got us across the Tasman, but a proper fix is needed.

I suspect a combination of voltage drop from the long wiring run now that it starts from the electrical panel, given the very high startup currents that might be needed to overcome the torque of the big barndoor rudder. And also the potential spike from the new high amperage alternator.

I suggested they revert to wiring (via a 30? amp or so circut breaker) directly to the batteries that are very close to the core pack.

Any thoughts? How do most of you wire the system, though the panel, or direct to the batteries?

Cheers

[RaymondR]

You could run short relatively heavy wiring direct to the battery with a fairly high ampage mechanical or SSD relay in the circuit switched from the panel.

[Blue Stocking]

Core pack amp outputs are adequate for the steering demand ? Just went thru this with a Mamba/Whitlock system where the specified core pack was not of adequate output rating.

[Snowpetrel]

Thats a good idea Raymond on the remote isolater. Fortunately in this case the batteries are pretty accessible so there is no real reason not to just fit a breaker near the batteries under the saloon table for the AP.

Blue stocking, the core pack is adequate, or at least it was originally, so unless the system has degraded due to faulty relays or a faulty clutch soliniod or some other reason this shouldn't be the issue, it always used to overload in some cases if the boat was sailed hard, or was unbalanced. But now in anything real wind it can't cope, even with moderate sail and a helm that is very light by hand.

Blue stocking, how long and what guage is the wiring for your system, and what sort of amps is yours rated for?

Sent from my SM-G930F using Cruisers Sailing Forum mobile app

[sailorboy1]

What size wiring are you using to the autopilot?

Of course it could just be that the unit is starting to go bad. If it wasn't for the recent rewiring this is what you would think of before your thought of a wiring related problem. Mine worked for over a year after it started going bad.

[Snowpetrel]

Thanks Sailorboy, I believe its about 8mm^2 or 8 guage wire. But I haven't verified this. It could be faulty unit as it spent a while during the refit without any use.

There would have to be about a 20 foot distance one way from the panel to the core, and another 20 feet back to the hydralic motor. So we are looking at at least 40 foot one way and 80 feet round trip if we include the motor.

This could be cut down to about 50 foot round trip if it was wired directly to the batteries skipping the main panel. This distance seems to be ok from a voltage drop aspect if you use the normal 20 amp max amperage, but it may be that the startup currents are causing a big momentary voltage drop that our guages arent picking up. And its enough to trip the unit off.

http://www.cruzpro.com/goodbad.html

This is an interesting article that suggests wiring directly to the battery for heavy load motors is the best way.

[sailorboy1]

Originally Posted by Snowpetrel

Thanks Sailorboy, I believe its about 8mm^2 or 8 guage wire. But I haven't verified this. It could be faulty unit as it spent a while during the refit without any use.

There would have to be about a 20 foot distance one way from the panel to the core, and another 20 feet back to the hydralic motor. So we are looking at at least 40 foot one way and 80 feet round trip if we include the motor.

I feel that 8 gauge would be very marginal for a 45' ketch that I assume is fairly heavy displacement and has a barn door rudder. For my lighter boat with a spade rudder the AP calls for 6 gauge. Even if the wiring is the same as it had before it has maybe run at low voltage due to voltage drop a long time that would had taken a toll on the unit.

Maybe the reason it has had more problems while motoring is that you motor in worst conditions than you sail in. So even though the battery voltage is higher that AP is still trying to work harder.

[Cap Erict3]

What are the specs on the hydraulic pump?

[Stu Jackson]

Either a bigger wire and new connections using the "new" installed routing or a separate line direct from the battery bank with its own switch and adequately sized wire.

I don't see any other options.

Good luck.

[DeepFrz]

I vote for the direct to the battery with breaker method and check the wiring size. That is quite a long run.

[Snowpetrel]

Originally Posted by sailorboy1

I feel that 8 gauge would be very marginal for a 45' ketch that I assume is fairly heavy displacement and has a barn door rudder. For my lighter boat with a spade rudder the AP calls for 6 gauge. Even if the wiring is the same as it had before it has maybe run at low voltage due to voltage drop a long time that would had taken a toll on the unit.

Thanks sailerboy, thats really good info on the wire size you use. To me 8 ga seems a little light as well given the very long run. Looking again at the entire run at 20 amps it gives a 8% drop. Thats too high, but it still shouldn't kill the computer which according to the specs is rated down to about 8 volts. So maybe it is a transient voltage drop of way more than this? Probably due to the autopilot motor, but since its running through the switch panel it could also be due to other things like the fridge motor triggering, or the smart regulator doing something funny.

I guess the real mystery was why initially it was shutting down due to overvoltage when the engine was running, but the link I posted suggests transient voltage spikes can also happen when the motor stops. So maybe that was the problem. Its also weird that the it now doesn't seem to matter if the engine is running or not. But maybe this is due to the voltage conditioner that was installed in a desperate attenpt to correct the overvoltage problem.

Stu, I am more and more convinced the problem is exactly this. Bigger wires and a shorter run directly to the battery. Hopefully this will also cure the HF interference issue as well.

Still curious if most of you run dedicated heavy duty wire direct from the batteries on its own seperate breaker, as you might with an anchor winch, or do you run it off the main switch panel wiring?

The whole unit was at the time (10 years ago) about the biggest system easily available without going to very expensive semi custom commercial gear. Nowdays with more bigger boats around its probably easier to pick up heavier core packs.

It worked well for a few thousand miles of offshore use, if reasonable care was taken with balancing the boat. So it is basically sized about right, though a little bigger would be useful at times.

[smac999]

Put A volt meter on the pump connections and have someone steer the pilot back and forth. Read volts at pump input. Without this info you are guessing. Do the same at input and output of core pack. and at DC panel. And at batteries. You'll quickly see where losses are if any

Couple days ago I had a boat with 9v at pump, 9v at core pack. 9 volt at pilot breaker, 12v at input wire to DC panel, 12v at battery. Bad jumper on panel. Easy to fix when you have data. Right now you're just guessing

[Snowpetrel]

Yes thats the best way to do it Smac, unfortunatly the boats about 1300 miles away in another country.. I will see if they can find the time to get some numbers. But I dont think the voltages are going that low (9v), at least not for long enough to register on a meter, but until they test it properly I don't know for sure.

Both the panel and the main autopilot display are showing voltages not dropping below about 11.8 volts. But they are probably averaged and aren't showing any transient spikes. And I dont know exactly were they sense the voltage from.

I am not sure a normal voltmeter will show a very short spike or crevase in the supply either?

[Neptune's Gear]

Hey, SP, are your folks coming to Auckland? I'd be happy to sort it out for them if you like.
On a system such as this, if the helm can be used and is not isolated by a solenoid, load can be applied to the helm to resist the AP. Then actual current measurements, and voltage measurements can be taken to identify any issues. If the AP would give up in heavy conditions, then the AP computer is likely undersized. This can be remedied with the use of a mosfet H bridge, so that the AP computer only instructs the pump to turn to port or stb, but the current to do that is supplied directly from the battery bank.
However, there could be a connector, wiring, computer or pump issue of course. Fault finding from other there is, um, rather difficult!
Cheers
Matt.
PS Josh knows how to contact me if you don’t want to do it on here

[Snowpetrel]

Hi Matt, yes they are headed up to Auckland. As far as I know they are planning to spend the summer up there. So if they haven't got it all sorted then for sure you sound like the person to talk to. Thats the problem, trying to find someone who really knows what they are talking about, rather than the run of the mill marine electrician who might not appreciate the importance from a safety point of view of the AP working properly all the time.

That mosfet H bridge idea sounds like it could be the way to go in the future to make the unit much more reliable and bypass the 20 amp software limit. Ideally some sort of soft start or hydraulic soft relief valve would be nice as I noticed it seems pretty jerky, and that has to really increase the loads on the whole system.

One question with the mosfet idea, how do you protect the system from an overload, say from a rope tangled in the wheel?

That test using the wheel to provide a steady load sounds like a good one. At the moment I am sure it would easily fail with two fingers...

Thanks for the tips. What size wiring circuit breaker and is it wired direct to the batteries?

Cheers

Ben