Cruisers Forum
 


Join CruisersForum Today

Reply
 
Thread Tools Rate Thread Display Modes
Old 20-01-2013, 14:37   #16
Registered User
 
rgleason's Avatar

Join Date: Mar 2012
Location: Boston, MA
Boat: 1981 Bristol 32 Sloop
Posts: 9,320
Re: Test Polauto by using OpenCpn w/o instruments

Goal

The goal is to play previously recorded Nmea 0183 text files and make that data stream available to other programs which utilize Serial Com ports to receive Nmea 0183 data. In this example we will be using three programs, NavMonPC, OpenCpn and Polauto and will connect using Virtual ports provided in NavMonPC.
  • NavMonPC - Play Nmea 0183 file. Create two Virtual Ports A & B to send the Nmea data out to two other programs.
  • Polauto - Polar creation program which will create polars from NavMonPC's nmea data stream.
  • OpenCpn - Charting program which will show the Boat location and AIS on a chart and display instrument data from NavMonPC's nmea data stream.
Install Programs

First install and setup the programs, and download a sample nmea0183 file:
Data Flow conceptual framework
  • NavMonPC (Options - Play - Select a Nmea0183 file to play) Check that the file is being streamed.
  • NavMonPC (example: Virtual Port A - Com7) ----> OpenCpn (Com7, 38400 baud)
  • NavMonPC (example: Virtual Port B - Com8) -----> Polauto (Com8, 38400 baud)
Get everything connected.

Start NavMonPC,
  1. Open and play a Nmea0183 file - "File > Play File > Nmea File >/ [Select the file]" The hit "Play" Hit "10x" to speed up flow. Instruments should operate.
  2. Setup Virtual Ports - "File > Connections > Virtual Ports ABC"
  3. On "Virtual Port A" where it says [Closed] select [Com7]. Check "Autostart" then pick "Connect". Pick "Moniter" see "Vport A >> ..." streaming.
  4. On "Virtual Port B" where it says [Closed] select [Com8]. Check "Autostart" then pick "Connect". Pick "Moniter" see "Vport B >> ..." streaming.
  5. For yucks in the buttons above the command bar, select all Vports and Client and Server to see data stream. Then hit "Done"
  6. Leave program open and runnning with instruments showing.
Start OpenCpn
  1. Click on "Settings > Connections" Then pick "Add a connection"
  2. Then * Serial and at Dataport enter "Com7", Baudrate "38400", Priority "0", Check "Control Checksum", * "Accept only sentences"
  3. Make sure Enable is checked in the Data Connections window. You can leave "Use Garmin Host Mode for uploads" checked.
  4. Then Hit "Apply" Then "Ok" The Options window closes.
  5. Click on the Red Boat (AutoFollow) or F2
  6. Turn on "Show AIS Targets" - 9 buttons over.
  7. Now you should see the boat moving, with AIS targets and the instruments show wind and boat speed (hit Dashboard icon).
Start Polauto
  1. In upper right corner, pick Port Com to set the port, select Com8. 38400 baud. (-For some reason Com3 seems to work.)
  2. See nmea data streaming in the little next to "+"
  3. Now the Data Used box should show Wind speed and Boat speed.
  4. If not, go to "Conf" and try changing Real Wind to Apparent Wind.
  5. Select "Full auto recording"
  6. Now Polauto should be recording a Polar from the data.
Notes

Virtual Port
-A "Virtual Port" is used to create a "Com Port" on a computer that does not have to be an actual physical plug (virtual), such that another program running on the same computer, can connect to the assigned virtual "Com port" using a standard "Com Port" connection such as Com7 or Com8. for example.
Attached Thumbnails
Click image for larger version

Name:	NavMon-OpenCpn-Polauto-Kirke-1.jpg
Views:	225
Size:	410.0 KB
ID:	53225   Click image for larger version

Name:	NavMon-OpenCpp-Polauto-Kirke-2.jpg
Views:	160
Size:	408.0 KB
ID:	53226  

Click image for larger version

Name:	NavMon-OpenCpn-Polauto-Kirke-3.jpg
Views:	198
Size:	419.8 KB
ID:	53227  
__________________

__________________
rgleason is offline   Reply With Quote
Old 20-01-2013, 14:46   #17
Registered User
 
rgleason's Avatar

Join Date: Mar 2012
Location: Boston, MA
Boat: 1981 Bristol 32 Sloop
Posts: 9,320
Re: Test Polauto by using OpenCpn w/o instruments

Many thanks to Hartmut [hwecken] and Patrick [tiersp] for their help getting this to work. Hartmut helped with a large recorded file and advice, and Patrick with an updated Polauto version. What follows is an email dialogue about some details to consider for recording polars with Hartmut's comments added -which he allowed to have posted.
-----

Hartmut,

The first thing I did this morning was download and run your nmea file with NavMonPC, OpenCpn and Polauto running in tandem each using a separate Virtual com port set up in NavMonPC, as you suggested. Watching Polauto work graphically with Opencpn actively charting is quite a treat, how exciting. I did not put any filters on (10-10-10-1s) and the Polauto drew some reasonably good polars with your file without my intervention. They were not nearly as complete as the one you posted, but with some additional work, I can see how more accuracy would be developed. Thank you for the file and your description of sail conditions and route, that kind of notation appears to be essential for interpreting and obtaining good polar recordings.

I've been thinking about your response to my questions and observations about filtering:
"The more you filter all the values in polauto the more realistic the polar will be. In upwind conditions you must filter more because when you are coming to the edge you are overshooting in boat speed. I think for broad reach or run courses filtering is not so essential.
For me, it makes no sense to filter the raw data, because you will loose information and you will need a lot more data."

Without having any experience in this, I think I would agree with you about the need for filtering upwind more than downwind. Do I understand that correctly that normally you do not use filters upwind either?
I do not use filters in NavMonPC.
In Polauto you must use filters otherwise you will get too high values due to overshooting when changing course.

Polauto - under Conf - Data Filters -
Maximum Difference in Boat Speed 1, 3, 5 & 10 nds
Maximum Difference in Wind Speed 1, 2, 4, & 10 nds
Maximum Difference in Wind Direction 1, 2, 4 & 10 nds
Average on Time 1, 5, 10 & 15 seconds
Patrick's note: "No filter use 10-10-10-1s"
The more you filter the more data you need, the longer time you need to get an polar but this polar will be more accurate than with lower filters. I use filters in all directions (upwind and downwind).

NavMonPC - under Options Filters -
SOG Filter - 2, 5, 10, 15 & 30 seconds
COG Filter - 2, 5, 10, 15 & 30 seconds
HDG Filter - 2, 5, 10, 15 & 30 seconds
AWA Filter - 2, 5, 10, 15 & 30 seconds
TWA Filter - 2, 5, 10, 15 & 30 seconds
TWS Filter - 2, 5, 10, 15 & 30 seconds
TWD Filter - 2, 5, 10, 15 & 30 seconds
my configuration is no filter

These filters average the value over 2,5,10 or 30 seconds.
NavMonPC's Windchart and Speedchart graphics are pretty good.
It appears the wind direction oscillates much more than we may realize just looking at tell tails.
NavMon allows you to Play a nmea file, Options-Filter it and Log the output.
That Log file could then be sent to Polauto.
NavMon appears to be very feature rich.
Yes, but I use it only to bring the nmea streams to the virtual com ports, and somtimes to log windchart as an overview what had happen on a trip.

I suppose only experience will inform us what filtering is best, but Polauto seems to work pretty well. I had thought that to get good polar data one should record sea state, sails, sail trim, current (preferably slack) and any small changes. Then I had thought it was best to get a segment of data with all these variables held constant for awhile and then pick that data from the recording.
I think it is good to have som variables like seastate or sail usage. Seastate is realized now in polauto 4.
I will separate in future only between offshore and inshore because seastate corresponds in my opinion very close to windforce, but ther is a relevant factor how shallow/deep is the water. In my case on Ijsselmmer you get up from 5Bft a short steep breaking wave, on the other hand with same conditions in the northsea the waves are longer, may be higher but you can sail faster.

Have you had Polauto running long enough so that you know whether judicious use of the Polauto Record on/off button is effective?
Yes this is a good feature to exclude maneuvers like tacking or setting sail.

Also, perhaps you can advise regarding connecting my instruments. I have Raymarine Seatalk instruments and a Garmin handheld GPS, with a Globalstar 353 puck connected to VHF/DSC. The GPS is not a problem, but I need to convert seatalk to Nmea0183 and I will want to get AIS sometime. Should I be considering a Multiplexer rather than a Raymarine seatalk - Nmea converter? If so, which one?
Have a look at this link http://www.busse-yachtshop.de/pdf/seatalk-nmea-interface.pdf
Not realy cheap, but I think it fits your problem

Also do you mind if I post the above, with your comments, along with some shots of the 3 programs running your file?
It is ok for me

Thanks, so much.
Rick
__________________

__________________
rgleason is offline   Reply With Quote
Old 21-01-2013, 07:00   #18
Registered User
 
rgleason's Avatar

Join Date: Mar 2012
Location: Boston, MA
Boat: 1981 Bristol 32 Sloop
Posts: 9,320
Re: Test Polauto by using OpenCpn w/o instruments

This morning I realized that the difficulty I had yesterday with Polauto stuck on Com3 and being unable to select Com7 was due to my OpenCpn connections setup. OpenCpn was streaming nmea out on Com3 I believe (couldn''t reconstruct this).
__________________
rgleason is offline   Reply With Quote
Old 22-01-2013, 14:38   #19
Registered User

Join Date: Nov 2010
Posts: 74
Re: Test Polauto by using OpenCpn w/o instruments

Hello, Rick
here is the last Polauto.exe ( it's not an install !)
I'll send the new install if it's OK
__________________
tiersp is offline   Reply With Quote
Old 22-01-2013, 14:42   #20
Registered User

Join Date: Nov 2010
Posts: 74
Re: Test Polauto by using OpenCpn w/o instruments

The Polauto.exe file is not here,
I put it on Hisse et ho.
__________________
tiersp is offline   Reply With Quote
Old 10-02-2013, 06:02   #21
Registered User
 
rgleason's Avatar

Join Date: Mar 2012
Location: Boston, MA
Boat: 1981 Bristol 32 Sloop
Posts: 9,320
Re: Test Polauto by using OpenCpn w/o instruments

Patrick has uploaded English Version 4.2 with some significant improvements.
Thank you Patrick.

Polaires auto
__________________
rgleason is offline   Reply With Quote
Old 10-02-2013, 09:24   #22
Registered User
 
hwecken's Avatar

Join Date: Feb 2012
Location: Germany
Boat: HR382
Posts: 110
Re: Test Polauto by using OpenCpn w/o instruments

First of all thanks to Patrick, big improovment and also for Rick with the "help file"

I saw that there is an export function to OCPN. I do not know any function in OCPN which uses polar data. Have i overseen something ?

Hartmut
__________________
hwecken is offline   Reply With Quote
Old 10-02-2013, 10:18   #23
Registered User
 
rgleason's Avatar

Join Date: Mar 2012
Location: Boston, MA
Boat: 1981 Bristol 32 Sloop
Posts: 9,320
Re: Test Polauto by using OpenCpn w/o instruments

Maybe this OpenCPN format is hopeful of inclusion? I would have to go back to see where it came from. It was the same as Bluewater format, I believe, but I can't recall how it started, and I did not check OpenCPN features.
__________________
rgleason is offline   Reply With Quote
Old 19-02-2013, 15:30   #24
Registered User
 
rgleason's Avatar

Join Date: Mar 2012
Location: Boston, MA
Boat: 1981 Bristol 32 Sloop
Posts: 9,320
Re: Test Polauto by using OpenCpn w/o instruments

Patrick has a new version, very nice, with a Current feature showing current vectors and history!

Give it a spin and let him know how it works...and what you think.

Here is a photo. The big brown area in the current diagram is the power from the engine, I finally figured it out, Engine ---> current.

Don't know if this is actually Current + Leeway Vector, but this promises to get better.

Thank you Patrick. Let us know where you are posting the new version please.
Attached Thumbnails
Click image for larger version

Name:	Polauto-NewVersion3-TheCurrent.jpg
Views:	175
Size:	428.5 KB
ID:	55237  
__________________
rgleason is offline   Reply With Quote
Old 06-03-2013, 15:49   #25
Registered User
 
rgleason's Avatar

Join Date: Mar 2012
Location: Boston, MA
Boat: 1981 Bristol 32 Sloop
Posts: 9,320
Re: Test Polauto by using OpenCpn w/o instruments

Since Feb 19 Patrick [Tiersp] seems to have been busy with more improvements. He has added a nema recording feature because someone suggested a different way of establishing Polars; collect polar data in a point cloud and then post process and filter all of the points.

There are now several additional threads, an instructional website, and another download website for his current version listed below:

Polauto - Measure the Polar 4.3.1 (Eng Translation) Software Description

Polauto - Thread (Eng Translation) - Thread with downloads for French and English

Polauto - Development Thread (Eng Translation) - Development thread

Polauto - Download and illustration in French

Current English version of Polauto is 4.3.2 at the top of the page.
__________________
rgleason is offline   Reply With Quote
Old 17-03-2013, 19:29   #26
Registered User
 
rgleason's Avatar

Join Date: Mar 2012
Location: Boston, MA
Boat: 1981 Bristol 32 Sloop
Posts: 9,320
Re: Test Polauto by using OpenCpn w/o instruments

Here is the other longer development thread on hisse-et-oh.com
__________________
rgleason is offline   Reply With Quote
Old 17-03-2013, 20:56   #27
Moderator
 
Paul Elliott's Avatar

Cruisers Forum Supporter

Join Date: Sep 2006
Posts: 3,864
Images: 4
Re: Test Polauto by using OpenCpn w/o instruments

Just in case there's any question about NavMonPc filtering, in general, when you play back a recorded NMEA file there is no filtering on the output stream. The filters are in the display and calculation path, but the output NMEA is exactly the same as the received and captured NMEA.

There is one case where the NavMonPc filters will affect the playback data, and that is when you have NavMonPc generating NMEA messages MWD and MWV. The generators use the filtered values for true and apparent wind to build these NMEA sentences.
__________________
Paul Elliott, S/V VALIS - Pacific Seacraft 44 #16 - Friday Harbor, WA
www.sailvalis.com
Paul Elliott is offline   Reply With Quote
Old 18-03-2013, 07:07   #28
Registered User
 
rgleason's Avatar

Join Date: Mar 2012
Location: Boston, MA
Boat: 1981 Bristol 32 Sloop
Posts: 9,320
Re: Test Polauto by using OpenCpn w/o instruments

Yes, Paul. Thank you for clarifying. I've been using NavmonPC to stream nmea data to Polauto for testing purposes. The stream has been unfiltered, with Polauto providing the filters. I believe Opencpn plugins will do a similar task as well.

I should try out the other robust features of your robust and useful software.
__________________
rgleason is offline   Reply With Quote
Old 01-04-2013, 17:27   #29
Registered User
 
rgleason's Avatar

Join Date: Mar 2012
Location: Boston, MA
Boat: 1981 Bristol 32 Sloop
Posts: 9,320
Re: Test Polauto by using OpenCpn w/o instruments

Polauto Version 5 - Patrick has announced a new version of his Nmea recording Polar Diagram Program. It looks more flexible, reading and writing Nmea files, and will result in better and more detailed diagrams.

I see the wind speed increment is much less, and thus the polar data points collected are greater and will result in better polars. Also you've got the polar graph as a popup, makes sense.

Right now I am puzzling about the black background. Is that normal ... Now I recall that you can click on most any of the background areas and then right click and pick a different color. I've reconfigured the colors and that is better.

I also see that you can record and read Nmea Files. -That will be great help because you won't need run another program like NavMonPC and Virtual Ports to read the Nmea data from Harmut.

I also see you've added a handy Wind and Optimum Boat Speed window for sailing optimally.

I am very interested in the Current + Leeway window and need to think more about how we can use this in other ways, in addition to being sure that current is less than 1 nds when recording data!

This version appears to be quite stable. I hope to use it this summer to create some good polars for the boat. Very useful! Version 5 seems to be quite stable.

Thank you very much, Patrick.

Here is the link to download PolAuto Version 4.3 Patrick may update it.
Attached Thumbnails
Click image for larger version

Name:	Polauto-5-screenshots.jpg
Views:	160
Size:	424.2 KB
ID:	58391   Click image for larger version

Name:	Polauto-5-Screenshots-2.jpg
Views:	169
Size:	437.4 KB
ID:	58392  

__________________
rgleason is offline   Reply With Quote
Old 04-04-2013, 05:35   #30
Registered User

Join Date: Nov 2010
Posts: 74
Re: Test Polauto by using OpenCpn w/o instruments

Polauto version 5
the update is effective now.
Please not that if you have install an old version, it's better to save your
polars and then uninstall the old program and then install the version 5.
After you should import ( not copy in the polaires directory ) your old polars and they where transform in the new format.

Patrick.
__________________

__________________
tiersp is offline   Reply With Quote
Reply

Tags
opencpn

Thread Tools
Display Modes Rate This Thread
Rate This Thread:

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off




Copyright 2002- Social Knowledge, LLC All Rights Reserved.

All times are GMT -7. The time now is 05:49.


Google+
Powered by vBulletin® Version 3.8.8 Beta 1
Copyright ©2000 - 2017, vBulletin Solutions, Inc.
Social Knowledge Networks
Powered by vBulletin® Version 3.8.8 Beta 1
Copyright ©2000 - 2017, vBulletin Solutions, Inc.

ShowCase vBulletin Plugins by Drive Thru Online, Inc.