• Announcements

    • Zapata

      Abbreviated rules   07/28/2017

      Underdawg did an excellent job of explaining the rules.  Here's the simplified version: Don't insinuate Pedo.  Warning and or timeout for a first offense.  PermaFlick for any subsequent offenses Don't out members.  See above for penalties.  Caveat:  if you have ever used your own real name or personal information here on the forums since, like, ever - it doesn't count and you are fair game. If you see spam posts, report it to the mods.  We do not hang out in every thread 24/7 If you see any of the above, report it to the mods by hitting the Report button in the offending post.   We do not take action for foul language, off-subject content, or abusive behavior unless it escalates to persistent stalking.  There may be times that we might warn someone or flick someone for something particularly egregious.  There is no standard, we will know it when we see it.  If you continually report things that do not fall into rules #1 or 2 above, you may very well get a timeout yourself for annoying the Mods with repeated whining.  Use your best judgement. Warnings, timeouts, suspensions and flicks are arbitrary and capricious.  Deal with it.  Welcome to anarchy.   If you are a newbie, there are unwritten rules to adhere to.  They will be explained to you soon enough.  
WHK

Any Data for Triducer versus single Transducer Speed Accuracy?

Recommended Posts

I'm helping a friend upgrade old sensors & instruments on his J/109.  He is replacing 15 year old NKE instruments.  The boat speed has always suffered from tack-to-tack speed accuracy and the corrections made aren't that accurate so we are moving the paddlewheel to the centerline.  I did this on my own boat and it really made a difference.  I plan on mounting an Airmar P217 housing (completely flush on the bottom with no mushroom) and fairing it smooth to the hull.  Since we can start from scratch on the transducer, I want to be sure that the most accurate and most responsive speed insert is used.  Depth accuracy is good enough for either the triducer ior stand alone, it is most accurate speed that we want.

Is there any real data available that compares accuracy and response of a Triducer with speed, depth and temperature to a stand alone speed and depth transducer?  It would not be a problem to keep the original depth transducer since it works fine, although we do plan on replacing the housing with the P217 similar to the depth in order to get a smooth faired surface.  Alternately, if there is no speed performance difference, the Triducer could be used.  I plan on using a NMEA 2000 smart sensor so the flexibility exists for future instrument upgrades.  How good is the data on the new Ultrasonic speed transducers?

P217 Housing

33217.gif

Airmar ST850 Smartsensor

airmar-st850-speed-transducer.jpg

Airmar DST800 Smartsensor

airmar-dst800-triducer.jpg

 

Share this post


Link to post
Share on other sites

Not seen any real data from the companies, but did have the misfortune of sailing a 100ft Baltic with a triducer, we switched to SOG immediately.   The unit was crap and very noisy.

Share this post


Link to post
Share on other sites

I agree. That airmar triducer is, in my experience, pretty useless and I pretty poor experience of flushing the Airmar DST800 also. It became very very unlinear which I attribute to the small dimension of the paddlewheel compared to the boundary layer thickness

Share this post


Link to post
Share on other sites

Doing some more online research I discovered that the data update rate for all the Airmar "Smart" transducers is 1Hz which is to slow.  I found the Garmin GST43 transducer (same as the old Nexus TH43) has a bigger paddlewheel.  It can be coupled with the Garmin GST10 analog to NMEA 2000 converter that outputs at a 5hz rate.  This can be bought for under $200.  Now all I need to do is find the 43mm flush mount housing and I'm good to go.

cf-lg.jpg

Share this post


Link to post
Share on other sites

Just a thought...but Airmar is now selling paddlewheel-less speed sensors (UDST800).   One actually can determine leeway as well (DX900).  No experience with them but it would  be nice to get rid of the forever fouled paddlewheel

No I don't work for Airmar.

 

Share this post


Link to post
Share on other sites
7 minutes ago, sail69 said:

Just a thought...but Airmar is now selling paddlewheel-less speed sensors (UDST800).   One actually can determine leeway as well (DX900).  No experience with them but it would  be nice to get rid of the forever fouled paddlewheel

No I don't work for Airmar.

 

Thanks - I did see that but at an order of magnitude more expensive than a good paddlewheel, I don't know if I want to be a beta tester for this product.  If they want to offer it at the same price I can get the GST bundle combo, I would be happy to try it and report test results.

Share this post


Link to post
Share on other sites

Slight thread drift, but I'm considering buying a B&G Triton 2 system, which comes with the DST800 sensor. It doesn't seem to be getting great reviews, any further info/opinions?

(Nice to have only one hole in the boat but happy to have 2 holes and separate sensors if the DST800 is NBG!)

Share this post


Link to post
Share on other sites

I have the T41 Triton on my boat as a single display added to the existing Nexus NX2 displays.  I added the T41 2 years ago so I had the wind strip charts visible while racing - it is very good for that.  I can put any data I want on the T41, but use the Nexus displays for all other data.  I have installed the B&G Tritons and Zeus 2 / Zeus 3 on other boats - all were done using the existing analog transducers through an interface, so I can't comment on the DST800 sensor.  Bottom line is when the sensor inputs are good, the displays are great.  Hence - the thread drift here is kind of the reason I started with the original post.  If you are serious about good speed inputs, I don't know if the Triducer is the way to go.

Share this post


Link to post
Share on other sites

If you are serious. Get Signet paddlewheels and NOT the "SMARTSENSORS" with  NMEA outputs which are not high update rates, and have difficulty in adjusting the damping. The bigger the paddlewheel and the better the install, the more likely your bottom speed will be accurate. 

 

Check on compatibility with your instruments whether you need Hall Effect or simple magnet/coil. 

Paddlewheel Sensors
paddlewheel-icon.jpg

Self-Powered Paddlewheel Sensor
 

   RCA Plug with 35 ft. cable (for new traditional instruments)
   AMP Plug with 35 ft. cable (for older knotmeters)
   For SmartPak with 35 ft. cable (Conxall Plug)
 



1-2200.100-1 
1-2200.100-2
1-2200.100
 


$189
$192
$189
 
 
paddlewheel-open-collector-icon.jpg Open Collector (Hall Effect) Paddlewheel Sensor

   Square wave output for non-Signet Marine Instruments. Call

 
1-2200.200 $189

 

or get the equivalent that are available from B&G

Get the blanking plugs and remember to remove the paddles between sailings

put a DT800 in the old hole and get depth/temperature... 

 

Share this post


Link to post
Share on other sites

Two years ago I put in a B&G system with triducer, wind sensor, and 3 T41s.  I removed a Signet system with the depth and speed sensor and three instruments.  

Unfortunately, I was ignorant of the fact the the existing Signet speed sensor (from this conversation) is probably better than the B&G/Airmar triducer that replaced it.  

This winter I have removed the old Signet thru hulls and plan to fill them in.  Now I'm wondering if I just put the speed sensor back in?  I did like the fact that I would have only one hole vs two but I have been very disappointed with the triducer thus far.  I never removed the Signet sensor over a season (on Lake Michigan) but the B&G fails/fouls several times a season and can be intermittent even after cleaning.

Thoughts on this?  How do I connect the Signet sensor to the N2K network?  Another Actisense?

Share this post


Link to post
Share on other sites
34 minutes ago, Passport111 said:

Two years ago I put in a B&G system with triducer, wind sensor, and 3 T41s.  I removed a Signet system with the depth and speed sensor and three instruments.  

Unfortunately, I was ignorant of the fact the the existing Signet speed sensor (from this conversation) is probably better than the B&G/Airmar triducer that replaced it.  

This winter I have removed the old Signet thru hulls and plan to fill them in.  Now I'm wondering if I just put the speed sensor back in?  I did like the fact that I would have only one hole vs two but I have been very disappointed with the triducer thus far.  I never removed the Signet sensor over a season (on Lake Michigan) but the B&G fails/fouls several times a season and can be intermittent even after cleaning.

Thoughts on this?  How do I connect the Signet sensor to the N2K network?  Another Actisense?

I'm thinking of using the Signet with the Garmin  Garmin GST10 analog to NMEA 2000 converter that outputs at a 5hz rate.  That will tie to the NMEA 2000 bus with all your B&G stuff.  You could keep the triducer and just use the depth off it.  The B&G source menu should allow you to select the GST10 or the Tridcuer for speed input.  Just select the GST10 for speed and Triducer for depth.

Share this post


Link to post
Share on other sites
2 hours ago, WHK said:

I'm thinking of using the Signet with the Garmin  Garmin GST10 analog to NMEA 2000 converter that outputs at a 5hz rate.  That will tie to the NMEA 2000 bus with all your B&G stuff.  You could keep the triducer and just use the depth off it.  The B&G source menu should allow you to select the GST10 or the Tridcuer for speed input.  Just select the GST10 for speed and Triducer for depth.

Will that require a Garmin MFD to calibrate the Signet paddel wheel via the GST10?

Share this post


Link to post
Share on other sites

I don't think a Garmin MFD is needed to calibrate it over the N2K network.  On another installation I was able to calibrate analog speed using a B&G Triton connected to a Raymarine ITC-5.  The Triton brought up the calibration menu option and sent the calibrated speed to the other Triton on the network.  The B&G Vulcan did not get the calibrated speed - B&G acknowledged the S/W limitation in the Vulcan when I called about it.  Worst case, I have a Garmin GND10 on my boat that I know will talk to it for calibration.

Share this post


Link to post
Share on other sites
5 hours ago, WHK said:

I don't think a Garmin MFD is needed to calibrate it over the N2K network.  On another installation I was able to calibrate analog speed using a B&G Triton connected to a Raymarine ITC-5.  The Triton brought up the calibration menu option and sent the calibrated speed to the other Triton on the network.  The B&G Vulcan did not get the calibrated speed - B&G acknowledged the S/W limitation in the Vulcan when I called about it.  Worst case, I have a Garmin GND10 on my boat that I know will talk to it for calibration.

Please let us know if you are successful. 

Share this post


Link to post
Share on other sites

The only B&G speed sensors shown on its Oz site are the DST800 and the ST800, which looks just about the same! Paddlewheel looks identical. Looks as though I'll have to chat to the local B&G dealer, who fortunately knows his stuff.

Signet doesn't appear to have a local presence, importing instruments can be a pain (tho' I've done it before.)

Share this post


Link to post
Share on other sites

If you're concerned about accuracy at low speed, you need not bother much with the Airmar sensors. Signet is easy to deal with and will ship from stock. Getting through Australian customs isn't much drama if you call them before you order and follow their process so that they can find you when the parcel arrives in order to charge the (obnoxious) duty.

 

  • Like 1

Share this post


Link to post
Share on other sites

I've brought heaps of stuff in from the US as most marine gear is 30+% cheaper there even accounting for exchange rates and shipping, than in Oz.  Most times customs just let it through - I think they're changing the GST exemption so that may change.

The pain is more customer service: timezones to call (not so bad with CA) and shipping delays,. All manageable tho'.

I hadn't considered Signet before, maybe I should look closer.

Share this post


Link to post
Share on other sites
5 hours ago, Itsabimmerthing said:

I was thinking about going from 2 transducers down to one triducer, asked B&G and they told me not to do it. Accuracy is not as good as with pure speed log only.

Is that the case on a monohull too if the xducer is mounted on the centerline?

 

Edit: nevermind - I reread and you are talking speed and depth separate.  Also the tri-ducer which based on comments here is not as accurate.

Share this post


Link to post
Share on other sites

Now a related question - I received a comment about flush mount versus the "mushroom" mount.  The comment indicated that the mushroom was actually better because it projected the paddlewheel better into the water flow.  I was thinking that the flush mount wouldn't disturb the flow and provide more accurate readings in addition to the advantages of an overall faired surface.  Since I've taken the time to longboard the hull for a faired surface, it made sense that the flush mount would seem better.

Any comments on flush versus the standard "mushroom" housing?

Based on what I've gained from comments on the forum: 

  • Triducer performance is not as good as a dedicated speed sensor
  • NMEA 2000 Smart Transducer data rates are 1 sample per second on N2K bus - not enough for meaningful performance processing
  • The Garmin GST10 provides analog to N2K conversion at 5 samples per second on N2K bus and seems to be a reasonably priced solution for performance processing inputs
  • The best (most responsive) paddlewheel sensor is the Signet - blue top is the Hall effect sensor that should work with standard analog speed inputs on various manufacturers processors.  The Garmin GST43 is a close second with its large paddlewheel. Both the Signet and GST43 are analog Hall effect sensors.

I am going to get a Garmin GST10 analog xducer to N2K adapter and check it with the boat on the hard to confirm that the B&G Triton brings up the speed calibration menu.  I have previously done this with a Raymarine ITC-5 so I am confident this will work.  I will document the GST10 wiring color codes so when the analog xducer connector is cut off the sensor connections are known.

Share this post


Link to post
Share on other sites

WHK,  thanks I'm really interested in what you find.  I have the red top Signet speed sensor.  Is this self powered?

If the Signet works with the Tritons I may very well put the speed sensor back in and just fill the one hole for the old depth and as you say just leave the triducer for depth and water temp.

Share this post


Link to post
Share on other sites
5 hours ago, Fleetwood said:

WHK, will keenly await your results.......!

GST10 is on order so should arrive for testing late next week.  Boat is on the hard but we'll be able to do communications tests and determine if the calibration menu is visible on the B&G displays. 

I also sent a message to Garmin tech support asking for the pinout / color code on the analog side of the GST10, since it isn't in the GST10 manual.  I did some research on the Airmar cross reference page and found the pinout for the Airmar T80 temp sensor that Airmar makes for Garmin and can plug in to the GST10.  Here is the analog connector pinout.  Pins 1-4 on the 8 pin connector are used as follows: 1 - Shield/Ground, 2 - White/Temperature, 3 - Green/Speed Sensor, 4 - Red/+ Voltage Speed

Share this post


Link to post
Share on other sites

If you want accuracy run a port and starboard speed sensor and have a switch to always use the leeward one. 

Share this post


Link to post
Share on other sites
12 hours ago, WHK said:

GST10 is on order so should arrive for testing late next week.  Boat is on the hard but we'll be able to do communications tests and determine if the calibration menu is visible on the B&G displays. 

I also sent a message to Garmin tech support asking for the pinout / color code on the analog side of the GST10, since it isn't in the GST10 manual.  I did some research on the Airmar cross reference page and found the pinout for the Airmar T80 temp sensor that Airmar makes for Garmin and can plug in to the GST10.  Here is the analog connector pinout.  Pins 1-4 on the 8 pin connector are used as follows: 1 - Shield/Ground, 2 - White/Temperature, 3 - Green/Speed Sensor, 4 - Red/+ Voltage Speed

Thanks, I'm about to start a file on this stuff, expect it to grow a bit.....

Share this post


Link to post
Share on other sites

I received the following reply from the Garmin support technician regarding the analog connector and wire colors for the GST10.

Calibration is stored in the GST 10, not the display.  I was able to verify that by plugging into several isolated networks. Also, note that it’s not possible to calibrate it unless the paddle wheel is spinning.  As for the transducer cable, here are the wire colors and functions. 

image001.png

Share this post


Link to post
Share on other sites

Quick update here.  The GST43/GST10 combo pack arrived.  I connected it to the N2K bus on a friends boat with a B&G Zeus3.  The good news is we can see the GST10 on the device list.  The bad news is we only see temperature and not speed.  I took the combo over to my boat and plugged it in where I have a Nexus Server interface to N2K via Garmin GND10, B&G Triton and Raymarine e125.  Same thing.  I called Garmin and they said "you need to have it connected to a Garmin display to calibrate it first".  I said that doesn't make sense.  I have an Actisense NGT-1 with the reader software that shows PGN 128259 is not being sent  on the N2K bus.  How can you calibrate something you can't see?

Doing some more troubleshooting I bench tested the GST43 using the pinout in the post above.  I get expected resistance readings on the temperature sensor.  When I connect 12V to the hall sensor, I should get a square wave of pulses when the paddlewheel is turned.  Nothing.... I had an Airmar ST800 analog transducer in my parts bin and ran the same tests.  As expected, there were pulses out of the Airmar xcdr when the paddlewheel was rotated.  My guess is the GST10 won't send the water speed PGN if it does not see the analog sensor connected.

Called Garmin Tech support again - they are shipping a replacement GST43/GST10 kit and we'll continue this journey once the it arrives.

Share this post


Link to post
Share on other sites
2 minutes ago, argh said:

@WHK did you find a flush housing for the GST43? 

No - Blur posted in another thread that he mounted the GST43 on both his J/109 and J/111 using the standard housing and ground/faired it flush.  That is what I plan to do.

Share this post


Link to post
Share on other sites

So we're not really going 11 knots upwind? Damn you B&G!

Curious to see what options work. Appreciate the info.

Share this post


Link to post
Share on other sites

I had the exact same problem when I tried to connect a GST43+GST10 to a NMEA 2000 network with B&G Triton display (appear in the device list, but only temp data is appearing on the screen). Have not solved it yet, but here is a thread from another forum that might contribute to some insight: https://www.panbo.com/forum/2016/12/garmin-gst43-to-bg-triton-2.html

Share this post


Link to post
Share on other sites
5 hours ago, bfu01 said:

I had the exact same problem when I tried to connect a GST43+GST10 to a NMEA 2000 network with B&G Triton display (appear in the device list, but only temp data is appearing on the screen). Have not solved it yet, but here is a thread from another forum that might contribute to some insight: https://www.panbo.com/forum/2016/12/garmin-gst43-to-bg-triton-2.html

Great find bfu01.  I will know more when I get the replacement GST43/GST10.  If it does turn out that I need a Gramin display to initialize the GST10 I will borrow one and then disconnect it to see how it works.

Share this post


Link to post
Share on other sites

I may be in left field here but @WHK, is the problem with the GST10 adapter, the GST43, or both.  My use case is with the Signet sensor so I just want to make sure I'm understanding all this correctly.

Share this post


Link to post
Share on other sites

I believe the GST43 sensor was bad.  New pair is scheduled to arrive tomorrow.  Weather looks iffy so may not be able to test until later this week.  I also now have my hands on  a Garmin GPS Map 5208 chartplotter that Garmin has confirmed will allow the GST10 to be "seen" and calibrated.  First I will try without the Garmin 5208 on the N2K bus and if the GST43/10 combo does not display speed on the Zeus3 and Triton 2 displays, will then connect the Garmin 5208 to run the initialization of the GST10.  I'll report back what I find. 

Once I get this all working, I'll rig up a test with an old Airmar ST800 analog paddlewheel connected to the input of the GST10.  That should prove that any old Hall sensor like the Signet should work.

  • Like 1

Share this post


Link to post
Share on other sites

Great stuff WHK, you 're answering all my questions before I even think to ask them! Keep it up.

(Seems as tho' optimum solution for me is to buy a Garmin or Signet speed transducer and plug it into a Triton system, if it can be made to work.)

Share this post


Link to post
Share on other sites

Here is a status update.  Garmin sent a replacement GST 43 transducer.  This does appear to provide the proper analog output.  I can confirm that the GST 43/GST 10 combo when plugged into the N2K bus will NOT display any speed information or the speed calibration menu on non-Garmin displays (Zeus3 or Triton2).  So - I was prepared and had a Garmin GPS Map 5208 that was reported by Garmin as being able to calibrate the GST 10.  I added a N2K T-connector on the bus and supplied 12V to the 5208.  The unit powers up, but there is no sign of the 5208 on the bus, and the 5208 doesn't see anything on the bus.  

I called Garmin with the GPS Map 5208 problem and was instructed to go through a bunch of screens I had already done.  While I was on the line I asked if he wanted the S/W version - yep that would help.  This 5208 was version 2.2 with a 2007 date.  The tech said that was really old and that when I updated to the latest software available on the Garmin site, it should communicate with the N2K bus.  He said I can download the S/W and put it on an SD Card to do the update.  The problem is for the initial update it required an SD card of 2GB or less.  The smallest SD card (or micro SD with adapter) I have is 32GB and I confirmed it will not read that.  So, I'm ordering a 2GB card on Amazon just to flash the Garmin upgraded S/W.  Stand by for future updates once I get the memory card.

  • Like 1

Share this post


Link to post
Share on other sites

Had a chat to our local (knowledgeable) dealer the other day about the DST800 and he said the problems were all traceable to a bad batch of sensors about 18 months ago, which refused to talk reliably to the N2K network. The problem is now resolved and he's had no problems since. He's installing one on a friend's boat next week, so I'll get a chance to see how well they perform and can be calibrated.

Share this post


Link to post
Share on other sites

It Works!  That's the bottom line.  The GST10 DOES require a Garmin device to initialize & calibrate, but after that, you no longer need to have the Garmin device on the network.  I had access to an old Garmin GPS MAP 5208 that was used for this testing.  Virtually any Garmin chartplotter or display with an N2K port should support the calibration. 

 I've posted photos of the sequence followed below.  "Calibration" was done by blowing air across the paddlewheel since the boat is on the hard.  Real calibration will be done on the water, and could be done via the Zeus or Triton.  The difference is if you do it with the Garmin, the value is stored in the GST10 so theoretically, everything gets accurate speed from the sensor output.  Calibration on the Zeus or Triton is just for the displays.  I plan on doing a semi accurate calibration using the Garmin 5208 when the boat is launched, then will remove it from the system.  After that, any calibration needed will be done vie the B&G displays over the network.

1.  Garmin 5208 Chartplotter after upgrading software finally sees stuff on the N2K bus.  The Initial power up screen showed the message in the bottom of the display.  None of the B&G devices saw the GST10 at this point.

1-Garmin5208-Initial-Display.jpg

 

2.  This is the NMEA 2000 device list displayed on the Garmin.  Note the GST10 is shown as "Uncalibrated"

2-Gramin5208-Device-Uncalibrated.jpg

 

3. This is the GST10 status by selecting "Review" from the previous screen.  It brings up the menu with an option to "Calibrate Water Speed"

3-GST10-Status.jpg

 

4.  This is the calibration menu.  I just blew on the paddlewheel to generate a speed input, then clicked on "Finish Calibration"

4-GST10-Cal-Screen.jpg

 

5.  The screen after "Finish Calibration" allows you to set a speed that would match the GPS.  I just used 5 kts so it would store a value.

5-GST10-set-cal-value.jpg

 

6.  After clicking done on the above screen the devise list is displayed showing the GST10.  Note the "Uncalibrated" label is now gone.

6-GST43-after-cal.jpg

 

7.  This is the Triton2 - It always showed the GST10 on this screen.  The next screen is the key.

7-Triton2-GST10-Device.jpg

 

8.  The data list now shows both temperature and water speed.  Here the paddlewheel is stopped.

8-Triton2-GST10-Data-List.jpg

 

9.  The Triton2 shows water speed with the paddlewheel spinning

9-Triton2-GST10-Spinning.jpg

10.  The Triton2 can now select the GST10 for Boat speed.  Previously this came via NMEA 0183 from the NKE instruments.

91-Triton2-Speed-Source.jpg

  • Like 1

Share this post


Link to post
Share on other sites
2 hours ago, Fleetwood said:

Well done! (If the DST800 option doesn't work I'll have to find a Garmin display....)

Fleetwood - I see no reason why a NMEA DST800 won't work.  The Airmar stuff should connect with anything on the N2k bus.  It's only the funky initialization requirements that Garmin puts in the GST10 that requires a Garmin display to do the initial setup.  I need to do some more work on the boat including drilling on the centerline to mount and fair the GST43.  In a few weeks when I get stuff cleaned up, I will connect an analog ST800 to the same GST10 that has been initialized to show that speed works with any Hall effect sensor and I'll report back here.

Share this post


Link to post
Share on other sites

The issue for me is whether the DST800 gives accurate, clean data. I should get a good data point from my mate's boat; the installation is being done by a very good guy.

Share this post


Link to post
Share on other sites
11 hours ago, WHK said:

Real calibration will be done on the water, and could be done via the Zeus or Triton.  The difference is if you do it with the Garmin, the value is stored in the GST10 so theoretically, everything gets accurate speed from the sensor output.  Calibration on the Zeus or Triton is just for the displays.  I plan on doing a semi accurate calibration using the Garmin 5208 when the boat is launched, then will remove it from the system.  After that, any calibration needed will be done vie the B&G displays over the network.

10.  The Triton2 can now select the GST10 for Boat speed.  Previously this came via NMEA 0183 from the NKE instruments.

I don't understand how the B&G displays will successfully calibrate the GST10 once you pull the 5208 off the system.  Sounds like you say the displays will calibrate what they show, but not calibrate the actual NMEA 2000 data stream from the GST10, so if any other instruments rely on boat-speed, they will be getting the old calibrated values from the 5208 calibration, which will be incorrect.  Also, if your B&G (or other instruments) do TWD/TWS calculation, I'm not sure which value they'll use...  So it seems a bit risky to use this approach in a system without a Garmin display...  Though it might well work for your specific system (sounds like you've thought it through!)

I've been thinking of replacing an old Airmar DT800 (analog version) with a NMEA 2000 version from Airmar directly (or perhaps B&G branded DT800) to avoid calibration issues across our mixed system.  We have an old Nexus system, Garmin 721xs plotters, but also B&G Triton displays.   So my own bias would be to avoid transducers that can only be calibrated by one brand...

Share this post


Link to post
Share on other sites
6 hours ago, galacticair said:

I don't understand how the B&G displays will successfully calibrate the GST10 once you pull the 5208 off the system....

What I found is the B&G Triton will calibrate the speed value it receives for all the other B&G products on the bus, and depending on the vendor, may or may not for other vendors.  For example - I setup an upgrade on a friends J/109 that has the following.  Speed source is analog Airmar ST850 -> converted to NMEA 2000 (Seatalk NG) via Raymarine ITC-5.  On the N2K bus are two B&G T41 Tritons and Vulcan 5.  Raymarine E22158 Seatalk NG to Seatalk converter to drive S1G Autohelm and old Raymarine RL70 display and old Raymarine Seatalk to NMEA 0183 converter to drive TackTick T122 NMEA 0183 to TackTick converter.  Multiple vendors, multiple bus types all talking to each other.

  • Either T41 Triton allows the calibration of the speed data out of the ITC5 on the N2K bus so that all the B&G stuff gets the same calibrated speed
  • RL70 display calibrates speed for anything on the Seatalk bus
  • TackTick has its own speed calibration for all the TackTick displays

Thus when I say the B&G displays will calibrate the speed without the Garmin display once it is setup with the GST10, I mean is it sees the GST10 speed output and will adjust whatever is out of the GST10 for the other stuff on the N2k bus.  I don't mean it will change the stored value in the GST10.  Unfortunately out of the box, no product except a Garmin display will see the GST10 except for a Garmin display.  Once the calibration value has been stored in the GST10, it may be put anywhere without a Garmin display.

Share this post


Link to post
Share on other sites

Adding a couple of Triton 2 calibration screens here, as I neglected to put these in the post above with the other pictures.  These were taken without the Garmin 5208 connected to the N2K bus.  The Triton 2 has a calibration option menu as does most every other NMEA 2000 display.  The two photos below show the temperature and speed calibration options for the GST10.  I was blowing on the paddlewheel when I took the picture so you can see it is getting speed from the paddlewheel.  This calibration stores values in the display, not the GST10 as the Garmin displays would.

Water Temperature Calibration:

92-triton2-temp-cal.jpg

 

Speed Calibration:

93-triton2-speed-cal.jpg

Share this post


Link to post
Share on other sites

Cool stuff - thanks for clarifying WHK.  Thinking of changing to a new Airmar DT800 and a separate speed log soon (both N2K, currently analog), so I'm looking forward to using the Triton display to calibrate -- hopefully everything on the N2K network will align to that data.

Share this post


Link to post
Share on other sites

So if one was going to add a speed log sensor, in addition to the BT-800 triducer, and that worked on the N2K with B&G, what would it be?  

Want to keep cost down as much as possible.  I have the old Signet, should I just try that with the associated adapters?

I'm really disappointed in the DT-800 speed log reliability.

Share this post


Link to post
Share on other sites

It would help if you described the problem you're trying to solve rather than the symptoms you're experiencing. If you're concerned about accuracy, the problems with NMEA2000 for communicating speed (water or wind) are significant and insurmountable without the use of proprietary sentences. This is why high end instrument system have direct connect for wind and speedo sensors.

Adding a second speedo more likely to reduce reliability rather than improve it. What do you mean when you say "DT-800 speed log reliability"? What are the failures that you've seen? 

Share this post


Link to post
Share on other sites

I asked Blur in another post what he did on his previous J/109 and J/111 for mounting the sensors since I was aware he had them on the centerline.  There are pictures on his blog for the J/109 installation using the Nexus TH43 (now Garmin GST43) speed sensor.  I asked him if he found a flush housing and he indicated that he ground the mushroom housing to match the hull contour.  Pictures below credited to the Blur blog.

j109details_091.jpg

j109details_08.jpg

Below is a Google Swedish to English translation from the post on the blog:

Drill the hole for the run (as usual).

Insert the lead, mark with the stylus on the outside.

With angle grinder (!) Phase out into the hull in the front and aft sides so that the hole in the lead comes edge to edge with the hull. The flange on the sides will then protrude. You can even erase the hole in the hull a little bit to get better strength (as if needed).

Stick all the way with some epoxy spackle (I ran Hempel's Epoxy Filler, strong but hard-cut). Make sure it fills the hole and comes on the inside.

Spackle especially in front and aft. Plaster the dum plug with a gown and hold it to protect the hole. Here I chose a little more slim filler.

Slip down the flanges on the sides (remain sufficient).

Finspackla & grind down.

Stop the transducer without wheels and grind the edges on it as well.

NOTE There are those who refuse to recalculate "as it affects the function on the log". However, we got very good accuracy on Novan (calibration factor 1.51 to compare to original 1.3). So you do it at your own risk :-)

Edited by WHK
Added text from Blur's blog

Share this post


Link to post
Share on other sites
8 hours ago, Moonduster said:

It would help if you described the problem you're trying to solve rather than the symptoms you're experiencing. If you're concerned about accuracy, the problems with NMEA2000 for communicating speed (water or wind) are significant and insurmountable without the use of proprietary sentences. This is why high end instrument system have direct connect for wind and speedo sensors.

Adding a second speedo more likely to reduce reliability rather than improve it. What do you mean when you say "DT-800 speed log reliability"? What are the failures that you've seen? 

It stops working a lot.  I said Lake Michigan and never had to touch the old Signet sensor for the entire season.  

For the last 2 years, the Airmar sensor is intermittent at best.  Usually ok for the first month or so of the season.  Then it will stop so I pull it and find a zebra muscle or two (my fault, should keep it clean but somehow they were never in the Signet sensor) clean it and then it will be intermittent for the reset of the season.  By that I mean I will loose speed data 25% of the time the last 25% of the season.

Maybe the reliability of the Signet was unusual and most transducers need more frequent servicing or I should put the dumby plug in more often.  That is not a good explanation from my point of view which I went from something that work to something I'v got to fuck with all the time.  The have been a few short port to port races where I've literally been in the bilge yanking this thing and trying to make it work again 10 minutes after the start.

Share this post


Link to post
Share on other sites

@WHK when you use the Garmin plotter to calibrate the GST43/GST10 combo, how does the plotter take current into account? Also, is there any ability to enter different calibrations for port and starboard tack? Since the calibration is stored in the GST10 I'm thinking the answer is no. 

With the Nexus server and the log wired directly to it, the server would allow calibration on each tack. The software that did the calibration would have you drive the boat for a distance in one direction, then do a 180 turn and go back the same distance, so current was accounted for. 

 

Share this post


Link to post
Share on other sites

On this topic. 

My new boat seems to have a set of Raymarine ST60 instruments which I expect that I will not be happy with. Planning to replace with Tritons but can I keep the transducers if they work and hook them up via a GST10 or something similar? Especially as the dst800 which comes with the Tritons does not seem too impressive based on the replies above.

 

If I cannot / should not keep the ST60 transducers, what would be the best option? Anything that fits in the same housing? Or if not, which transducer would be the best fit considering we race some and I expect to have accurate TWA, TWD, BSP and so on...

Share this post


Link to post
Share on other sites
6 hours ago, jarcher said:

@WHK when you use the Garmin plotter to calibrate the GST43/GST10 combo, how does the plotter take current into account? Also, is there any ability to enter different calibrations for port and starboard tack? Since the calibration is stored in the GST10 I'm thinking the answer is no. 

With the Nexus server and the log wired directly to it, the server would allow calibration on each tack. The software that did the calibration would have you drive the boat for a distance in one direction, then do a 180 turn and go back the same distance, so current was accounted for. 

 

Thanks - I am aware of that.  This install is not on my J/109 but a friend who has NKE equipment he is upgrading.  The bottom line is the calibration should be done for the sensor, and you need to compensate for current during calibration - either test with no current or go perpendicular to current to minimize.  You can also see the difference between reciprocal courses and split the difference to correct for current.

Share this post


Link to post
Share on other sites
3 hours ago, Nubben said:

On this topic. 

My new boat seems to have a set of Raymarine ST60 instruments which I expect that I will not be happy with. Planning to replace with Tritons but can I keep the transducers if they work and hook them up via a GST10 or something similar? Especially as the dst800 which comes with the Tritons does not seem too impressive based on the replies above.

 

If I cannot / should not keep the ST60 transducers, what would be the best option? Anything that fits in the same housing? Or if not, which transducer would be the best fit considering we race some and I expect to have accurate TWA, TWD, BSP and so on...

If you are interested in easy interfacing of the existing sensors. the Raymarine ITC-5 will allow you to use the existing sensors and put the data on the N2k bus.  The Raymarine ITC-5 has the proprietary Seatalk NG connectors.  You would need to take a Seatlak NB bus cable, cut it and wire in a NMEA 2000 field installable connector.  I did this on a friends boat to eliminate the ST60 displays that were replaced with B&G T41 Triton displays and a B&G chart plotter.

Share this post


Link to post
Share on other sites

I have some observations that might be of interest to someone reading this thread although sounds like the OP has his solution in hand.  I have both an Airmar Speed/Temp sender and a Signet speed.  One on each side of my rather long keel.  I am not cursed with interfacing to Raymarine or B&G equipment as I built all my own electronics.  So I cannot comment on that part of the problem.  What I will say is that on my boat, one sensor reads high on port tack and the other on starboard tack.  I average the two and I think that should solve that problem.  The Airmar has about twice the noise as the Signet.  I linearize with an equation of the form y = m*x + b. Ideally b is zero as it is an offset.  b is much larger on the Airmar, I think about three times.  On my TackTick system, which uses just the Airmar, it just quits reading below 1.3 knots because of this.  I can read on my custom system down to about 1/4 knot.  m is about twice with the Signet. This is partially responsible for less noise and due to them having 4 magnets instead of just two.  In terms of readings per second, I believe that is a fool's errand.  To get low noise, you have to average on the order of 4 seconds.  Who cares if you readout 4 readings or 40 readings if the response is the same.  I do not count pulses, I measure the time between pulses. I don't know how B&G or anyone else does it but if they count pulses in a given time, get another system.  That is just not going to give good performance.  On my boat I average the two senders and that has the effect of giving my about the same total S/N as the Signet as the signal is correlated so I get twice the reading but the noise is not so I get root sum square noise and the net is more signal with a little more noise and about the same S/N as just the Signet (about 10% worse).  I hope someone finds this useful or at least interesting.  Cheers.

Share this post


Link to post
Share on other sites
42 minutes ago, allene222 said:

I hope someone finds this useful or at least interesting.  Cheers.

I found it both, thanks!

 

Share this post


Link to post
Share on other sites
5 hours ago, WHK said:

  The Raymarine ITC-5 has the proprietary Seatalk NG connectors.  You would need to take a Seatlak NB bus cable, cut it and wire in a NMEA 2000 field installable connector.  I did this on a friends boat to eliminate the ST60 displays that were replaced with B&G T41 Triton displays and a B&G chart plotter.

Raymarine actually sells a very affordable N2K adapter cable that does the same thing without the hassle of a custom field install: see their SeaTalkNG to NMEA 2000 (male) adapter cable (model A0607): https://www.defender.com/product3.jsp?path=-1|344|2028705|2028879&id=2640087

They also have female versions and longer versions -- you need the male version to plug into a regular NMEA 2000 backbones.  The field install may still be better if you want a custom length.

Side-note: I'm not sure why Raymarine keep's calling NMEA 2000 "Devicenet" -- it sounds like that's the technical term for the N2K cabling standard, but my hunch is it's just another of their many attempts to prop up SeaTalk in its various versions (though recently they've finally folded and their latest Axiom MFD is pure-NMEA 2000)...  It certainly makes it harder than it needs to be to understand how to interface Raymarine with NMEA 2000 devices.  It's fairly easy actually! (if you have SeaTalkNG)

Share this post


Link to post
Share on other sites
3 hours ago, allene222 said:

The Airmar has about twice the noise as the Signet.  I linearize with an equation of the form y = m*x + b. Ideally b is zero as it is an offset.  b is much larger on the Airmar, I think about three times.  On my TackTick system, which uses just the Airmar, it just quits reading below 1.3 knots because of this.  I can read on my custom system down to about 1/4 knot. 

I read SeaTalk data from my ST60 system and then show corrected data on a NMEA display. For that b is 0.45 knots and speed is accurate down to 0.5 knots (compared to GPS, currents are very small around here). I have separate depth and speed sensors. I don't there is any offset in the SeaTalk data so 0.45 knots should be comparable with your 1.3 knots, which sounds very big. Have you cleaned the sensor well? Is the paddle wheel turning with low friction?

I use exponential smoothing with 5 seconds time constant to get stable 0.01 knots reading. Probably could use smaller time constant, but I'm quite happy with the response rate.

Share this post


Link to post
Share on other sites
6 hours ago, allene222 said:

...  I hope someone finds this useful or at least interesting.  Cheers.

I have to give a shout out here to Allen.  He has done some amazing things with the systems on his boat.  He also has a website L-36.com with valuable information for all sailors.

Allen - thanks for the great contribution to this post and your website!

Share this post


Link to post
Share on other sites
4 hours ago, WHK said:

I have to give a shout out here to Allen.  He has done some amazing things with the systems on his boat.  He also has a website L-36.com with valuable information for all sailors.

Allen - thanks for the great contribution to this post and your website!

Thank you.

6 hours ago, Joakim said:

I read SeaTalk data from my ST60 system and then show corrected data on a NMEA display. For that b is 0.45 knots and speed is accurate down to 0.5 knots (compared to GPS, currents are very small around here). I have separate depth and speed sensors. I don't there is any offset in the SeaTalk data so 0.45 knots should be comparable with your 1.3 knots, which sounds very big. Have you cleaned the sensor well? Is the paddle wheel turning with low friction?

I use exponential smoothing with 5 seconds time constant to get stable 0.01 knots reading. Probably could use smaller time constant, but I'm quite happy with the response rate.

What I do is add .6 knots to the Airmar unit output and adjust the gain.  That matches it to the Signet all the way from 6 knots to 1.3 knot where I quit using the Airmar in my averaging.  Below 1.3 knots I just use the Signet as I figure the boat is not healing at that speed.  But they match very closely down to that point.  Then I add .3 knots to the average to get my final reading.  That number was picked to get my readings of current at 6 knots and at 3 knots to agree. 3 knots is as slow as my boat goes under power at idle so the lowest constant speed I can go for testing current.   I am unsure about the .3 number but confident that the Airmar needs .6 knots more offset than the Signet. Current is not small here so it is difficult to know for sure.  I would not assume there is no offset in the SeaTalk data but I don't know.

Share this post


Link to post
Share on other sites

What a bunch of crap.  I took the green curve and adjusted the gain and offset and it shows that the paddlewheel is significantly better than theUST800, which is noisy as hell. The new curve is shown in red and was derived in Photoshop by just changing the gain and offset of the green curve.  I did not draw anything.

image.thumb.png.8972905320db65d3b7a5ed62eaa4c83c.png

Share this post


Link to post
Share on other sites

Nice work.

Hmm, I wonder why Airmar didn't do that; maybe they want to promote the (much) more expensive sensor?

Share this post


Link to post
Share on other sites
2 hours ago, carcrash said:

Sure looks like GPS SOG is the way to go.

Not around here where there is current.  I am actually not sure which has less noise, SOG or paddle wheel.  It is not a straight straightforward thing to determine as you have to compare systems with similar response times.

Share this post


Link to post
Share on other sites
On 3/2/2018 at 10:20 PM, carcrash said:

Sure looks like GPS SOG is the way to go.

Nah, a good paddlewheel like the Garmin with the red, vented paddle is the way to go. Even if the ultrasonic gadgets worked, they are about $900+. I'm a fan of accurate input data but wow, that's spendy even by instrument standards. 

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now