+- Raymarine forum (http://forum.raymarine.com)
+-- Forum: Raymarine Forums (/forumdisplay.php?fid=1)
+--- Forum: Autopilots (/forumdisplay.php?fid=23)
+--- Thread: Gathering data to help diagnose autopilot performance problems (/showthread.php?tid=4576)
Gathering data to help diagnose autopilot performance problems - Tom - Raymarine - Moderator - 01-19-18 01:48 AM
Autopilots are complex systems with behaviour that depends on input from multiple streams of sensor data, vessel speed, steering characteristics and much more as well as weather and sea-state. If a pilot's not steering a good course, therefore, there are a lot of potential causes that should be checked before assuming a fault. In fact, (I can say this because I didn't design them) the current Evolution pilots are brilliant and massively reliable, so if yours isn't performing well there is almost certainly a problem with its setup or incoming data, which we can help find and solve.
Some other FAQs (e.g. http://forum.raymarine.com/showthread.php?tid=1772) deal with common things that can cause steering problems, but if it's not something plain and simple like this, what can be done to understand the problem and the cause?
The answer is to gather a recording of NMEA2000/SeatalkNG data whilst the pilot is in use, and the problem is apparent.
WHY IS RECORDING SEATALKNG DATA USEFUL?
A verbal description of how a pilot performs is very hard to convey clearly, but contained within a data recording will be time-stamped data which shows:
- pilot heading versus GPS course-over-ground
- speed-through-water versus speed-over-ground
- pilot mode (Standby, Auto, Vane, Track, Calibration)
- requested and actual rudder angles and rudder slew rate
- vessel yaw rate
- vessel pitch and roll, wind data
- any pilot alarms and error messages
- bearing and distance to waypoint, cross-track error
- product part numbers, serial numbers and software versions
and much more. In other words, if you can log that data and send it to us, we can 'replay' your trip, seeing how the pilot and vessel performed almost as if we were aboard ourselves, and graph any combination of data over time.
WHAT SHOULD YOU DO?
1. Turn on pilot 'debugging', in Setup > Calibration > Commissioning. Set to debug level 7. To see this option, Calibration Lock (in Vessel Settings) needs to be Off.
2. Insert a blank MicroSD card into the chart slot of a Raymarine Lighthouse2 MFD (if you have more than one MFD, this needs to be the Master unit which is connected to STNG)
3. On the MFD, go to Home > Setup > Maintenance > Diagnostics > Interfaces > Record File and make sure that the correct slot is selected (1 is top), then go to N2K/STng > Buffer > Start Buffer > Record Buffer
4. Go to Home > Setup > Maintenance > Diagnostics > Select Device. This causes the system to poll all NMEA2000 devices, which will then appear in the recording.
5. Use the pilot, for as long as you need to in order to demonstrate the problem. You must be underway, and you must log a minimum of several minutes of data. If you log for a long time, please make a note of when we can expect to see the problem (e.g. 4.30pm or 25 minutes into the log.)
6. Go back into MFD diagnostics and stop the log recording.
7. Turn Debug Mode back off on the pilot.
8. Create a post in the Autopilot forum, describing the problem and attaching all of the NMEAxxxxxx.log files that you will find on your MicroSD card.
IF YOU DON'T HAVE AN LH2 MFD
Many Raymarine technical service dealers have hardware which can make recordings, or we may be able to supply a 'data logger' which can do the same thing. If you think you may need to go this way, contact Technical Support.
Raymarine LH3 MFDs do not yet have the capability to make recordings, but this is coming to the software in due course.
WHAT PILOTS IS THIS GOOD FOR?
Any SeatalkNG-connected pilot: SPX pilots with colour pilot displays and the Evolution range.
WHAT NEXT?
We'll take a look at your log-file, replay it through our test systems, and come back to you in due course.
RAYFAQ#