• 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.  
roblpm

New Instruments - Calibration

Recommended Posts

So I need some new nmea 2000 instruments for my 28 foot sailing boat.

The system will be based around a shipmodul multiplexor so I can send data over wifi to phone, pc, ipad and allene's racebox.

So the cockpit display is important but not the only thing.

I started looking at the different systems and it seems calibration is an ignored and little understood issue (including by me!).

It seems that some of the systems store any calibration information in the display, and then maybe send the corrected items to any other displays from the same manufacturer. If this is the case then the output on my cockpit display could be different from say the output on my ipad. I don't really want to enter calibration information on 5 devices?!

The DST doesn't seem like so much of an issue as a DST800 can store the calibration information and output it directly onto the nmea2000 bus. So just need something that can directly calibrate one.

I am not so sure about Raymarine I70s and B&G Triton. They both seem to calibrate in the displays. The guy at raymarine told me the corrected data was then put on the bus but is there a PGN for wind offset for example? I just looked at a 130306 sentence and it doesn't seem like there is an offset setting? Wind Data:  SID = 177; Wind Speed = 0.92 m/s; Wind Angle = 353.4 deg; Reference = Apparent. So I am not sure he knew what I was talking about? To me the NMEA2000 is coming from either the ITC5 in Ray or the wind instrument in B&G. Aha but does the instrument send out corrected data as the same PGN but from a different instrument? So can you select that on the shipmodul I wonder? Yes you can select a datasource. So the question here is do these displays reoutput corrected NMEA2000 data?

Garmin support knew what I was talking about and though you need a Garmin display to calibrate he assured me that the offset etc was stored at the sensor.

So I sort of like the idea of the calibrations being at the sensor. But maybe it will work with Raymarine or B&G by selecting the display as the datasource? Or am I missing something? 

 

 

 

Share this post


Link to post
Share on other sites

Raymarine calibration UI is on the i70s display, but the calibration is stored in the itc5 (what the transducers plug into) and sent out over the network after calibration math is done. 

A few years ago I was building my own nmea2000 data logger and spent a lot of time watching my boats network traffic. 

One wrinkle in this is that you need a mfd in at least Raymarine and B&G networks to update firmware on all devices. I assume Garmin is the same. 

Share this post


Link to post
Share on other sites
Quote

So I sort of like the idea of the calibrations being at the sensor. But maybe it will work with Raymarine or B&G by selecting the display as the datasource? Or am I missing something? 

Your sensors output either a raw signal such as a pulse train/voltage or a NMEA 0183/2000 string. 

When you use the term "calibration", it can be as simple as a one point offset for depth transducer (vs keel? vs waterline?) or as complex as the corrections in a B&G H5000 for True Wind Direction as a function of:

  • Heel corrected heading & Boat Speed,
  • Measured Wind angle corrected for static sensor orientation and mast rotation if any
  • Sensor reading then corrected for dynamic motion of the boat at wind sensor's height and offset from center of mass/rotation
  • input for the leeway coefficient. 

Since that's a fusion of many inputs; to give a "better" output you probably can't do that in one sensor, though Airmar tries in their masthead wind instruments with associated accelerometers and fluxgate. 

That's why a processor, such as the H5000 with Hercules, +/or Expedition or similar running on a PC are a means to improve your data's quality, by including damping, processing as above and then distributing out to multiple readouts in appropriate formats, and it doesn't come cheaply in terms of either dollars or time. 

Your helmsman may want undamped speed to watch, while your tactician/navigator want averages to detect trends. 

 

 

Share this post


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

Raymarine calibration UI is on the i70s display, but the calibration is stored in the itc5 (what the transducers plug into) and sent out over the network after calibration math is done. 

A few years ago I was building my own nmea2000 data logger and spent a lot of time watching my boats network traffic. 

One wrinkle in this is that you need a mfd in at least Raymarine and B&G networks to update firmware on all devices. I assume Garmin is the same. 

That is what I hoped until I read this.....

But i hope you are right and raymarine are wrong??!!!! The guy I spoke to at raymarine support in the uk this morning didn't understand what i was talking about........ 

http://raymarine.ning.com/m/discussion?id=6492755%3ATopic%3A110405

Underway - Raymarine, US

The iTC-5 is a analog transducer converter from analog to Seatalk NG, it does not hold calibration data, that is held in the i70 instrument display.  So the i70 will need to be present on the Seatalk NG backbone to keep calibration. The only other option would be is using an i50/i60 instrument display and connecting the analog transducer directly to the back of the display, and calibrating from that particular display.

 

Share this post


Link to post
Share on other sites
1 hour ago, LionessRacing said:

Your sensors output either a raw signal such as a pulse train/voltage or a NMEA 0183/2000 string. 

When you use the term "calibration", it can be as simple as a one point offset for depth transducer (vs keel? vs waterline?) or as complex as the corrections in a B&G H5000 for True Wind Direction as a function of:

  • Heel corrected heading & Boat Speed,
  • Measured Wind angle corrected for static sensor orientation and mast rotation if any
  • Sensor reading then corrected for dynamic motion of the boat at wind sensor's height and offset from center of mass/rotation
  • input for the leeway coefficient. 

Since that's a fusion of many inputs; to give a "better" output you probably can't do that in one sensor, though Airmar tries in their masthead wind instruments with associated accelerometers and fluxgate. 

That's why a processor, such as the H5000 with Hercules, +/or Expedition or similar running on a PC are a means to improve your data's quality, by including damping, processing as above and then distributing out to multiple readouts in appropriate formats, and it doesn't come cheaply in terms of either dollars or time. 

Your helmsman may want undamped speed to watch, while your tactician/navigator want averages to detect trends. 

 

 

Yes I get your point. I have Allan Edwards race box to do further processing. I really just mean the calibration for the log, depth and wimd direction if I manage to mount it wonky! It would be a bit irritating if the basic outputs were different on the displayed than on the nmea 2000 network. Maybe i am overthinking this though! 

Share this post


Link to post
Share on other sites
22 hours ago, roblpm said:

I really just mean the calibration for the log, depth and wimd direction if I manage to mount it wonky! It would be a bit irritating if the basic outputs were different on the displayed than on the nmea 2000 network. Maybe i am overthinking this though

Other than unit conversion and possibly local offsets, unlikely to have differences. More reason to aggregate in single system and distribute outwards. 

Share this post


Link to post
Share on other sites

It's possible that the i70s broadcasts the calibration into it to the itc5 at boot time. The data on my network in the normal boat speed and depth pgns was calibrated. 

Share this post


Link to post
Share on other sites

So here is a reply from b&g. Exactly the opposite of what i expected.... 

Hi Robert, 

 
Assuming you are planning to use a 608 mast head unit the depth and wind calibration is stored in the sensors so this data would appear correct. 
 
Boat speed calibration is slightly different in that the calibration value is transmitted system wide and the calculation to correct is locally carried out in each display. This means boat speed calibrated via Triton 2 o Zeus 2/3 will display correctly on all B&G displays but likely not come through correct on your N2k to WiFi device. 

Share this post


Link to post
Share on other sites
14 minutes ago, Slim said:

You might also look at Allen's RaceBox - http://l-36.com/RaceBoxVideo.php 

He has one.  The issue is he wants the readings on his displays pre RaceBox to match the readings on his display post RaceBox.  In my case, I am using TackTick and, however bad they are, there is an open protocol and I can send true wind correction numbers back into the TackTick system so the readings pre RaceBox and post RaceBox are the same, at least in the case of true wind.  Speed and depth calibration can be stored in the TackTick system and the corrections incorporated in the data stream into the RaceBox.  The issue with wind is that the wind vane is extremely inaccurate with errors of 10 degrees not uncommon but it is correctable in the RaceBox so all is good.  One of the reasons I did not get the Nexus (now Garmin) system is that the protocol to send calibration data to the system was closed.  If these protocols were published for the B&G or other systems I am perfectly willing to modify the RaceBox code to work with whatever system he has.

Personally, I mostly use my own displays.  The Stowe terminal (developed for the SF AC) and some LCD displays of my own making are battery powered and communicate with the RaceBox over WiFi.  I do get different speed readings on the TackTick display that what I get on my own display but at this point, I find that interesting.  They are different because I have both port and starboard paddle wheels and I average them and I also use different linearization algorithms.

I guess the question is if the calibration can't be stored in the sensors, can correction factors be sent back to the displays and is that protocol published.  The second question is are is the calibration in the sensors enough to correct the errors.  For the TackTick wind, there is a system calibration that is basically an offset so you can set 0 degrees as into the wind.  But there is no nonlinear calibration.  The RaceBox has 4 non linear terms (sinx, cosx, sin2x, and cos2x plus the dc term).  That was necessary to take out the nonlinearity of the display.  Even in the case of speed, the TackTick is not as linear at low speeds as I want.  It won't read anything below 1.3 knots to start with and isn't particurally accurate down there.  So even there, the RaceBox calibrated data and the TackTick calibrated data will differ at low speed even if I did not average two sensors.

It is a complicated problem.

Share this post


Link to post
Share on other sites

To be honest its really just irritating now. 

All I really want is to understand what will happen before i shell out for some new instruments but it appears difficult for the manufacturers just to answer a straightforward question. 

Ie is the nmea 2000 information on the network corrected or uncorrected. (ie is the information corrected in their display). 

Share this post


Link to post
Share on other sites

Ok garmin say their wired gwind and dst800 both store the calibrating in the sensors. So looking like the favourite. 

If only I could find if Raymarine ITC5 definitely stores the calibrations they would be a contendor. 

Though garmin uses standard nmea2000 cabling i think. So maybe that's the way to go anyway. The displays aren't as flash but i need to remind that's not why i am doing this anyway! 

Share this post


Link to post
Share on other sites

If you are specific about what you want me to test on a Raymarine setup I'll be happy to do so. It is easy to get a nmea 2000 trace and give you video of the instruments (especially if I can do it with dockside compatible stuff like wind).

I'm not sure that it matters if the i70 or itc5 remembers the calibration, what matters is that the network traffic is calibrated. I'm 95% sure that is the case based on the network sniffs that I was doing two years ago.

SeatalkNG cabling is nicer than NMEA2000 standard too.  The ends are smaller, it's about the same price, and it is nearly impossible to build illegal topologies with it.  It is trivial to mix both wiring standards in the same network.

 

Share this post


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

If you are specific about what you want me to test on a Raymarine setup I'll be happy to do so. It is easy to get a nmea 2000 trace and give you video of the instruments (especially if I can do it with dockside compatible stuff like wind).

I'm not sure that it matters if the i70 or itc5 remembers the calibration, what matters is that the network traffic is calibrated. I'm 95% sure that is the case based on the network sniffs that I was doing two years ago.

SeatalkNG cabling is nicer than NMEA2000 standard too.  The ends are smaller, it's about the same price, and it is nearly impossible to build illegal topologies with it.  It is trivial to mix both wiring standards in the same network.

 

Great offer. 

I am not sure that the I70s and other nmea2000 displays have the ability to output nmea2000 pgns themselves? (is this right?!). 

So my research so far implies that the sensors output nmea2000 or in the case of raymarine the itc5 does it. 

Therefore when you run a calibration on the display it is doing one of two things. Sending the sensor a calibration factor which it will store (or the iTC-5) or just keeping it local. 

I am sure you know this already but I am trying to get it straight in my mind. 

So really the two easiest  things to test would be the wind direction offset and the log speed calibration. 

If you can set a calibration factor in each, then sniff the network and see if the calibration factor is included in the pgn on the network. Aha as you say log speed will be difficult at the dock! 

And i wonder what happens if you power up the system without the i70s however it is going to be difficult to tell whether the calibration is set or not. Maybe you could set the wind direction calibration to 180% repower the system with the i70s switched off and see of it is still there. If so the iTC-5 must be saving it. 

If this all works and i buy raymarine you should call them for some commission!! 

 

 

 

Share this post


Link to post
Share on other sites

The i70 can output PGNs for sure, this is done when they broadcast display brightness to the network so that other displays in the same group get the same brightness. If calibration math is done per i70 then also must write a broadcast PGN to make them all show the same number (since you can't put different calibrations in different displays).

Share this post


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

The i70 can output PGNs for sure, this is done when they broadcast display brightness to the network so that other displays in the same group get the same brightness. If calibration math is done per i70 then also must write a broadcast PGN to make them all show the same number (since you can't put different calibrations in different displays).

Ok sure. So here is the list of itc5 nmea sentences:

5.7 NMEA2000 sentences
The converter supports the following NMEA2000 sentences.
Protocol PGN PGN name Receive Transmit
NMEA2000 59904 ISO request •
NMEA2000 59932 ISO
acknowledge

NMEA2000 60928 Address
claim
• •
NMEA2000 65240 ISO
commanded
address

NMEA2000 126464 Transmission
PGN list

NMEA2000 126464 Received
PGN list

NMEA2000 126996 Product
information

NMEA2000 126208 Acknowledge
group
function

NMEA2000 126208 Command
group
function

NMEA2000 126208 Request
group
function

NMEA2000 127245 Rudder •
Protocol PGN PGN name Receive Transmit
NMEA2000 128259 Speed, water
referenced

NMEA2000 128267 Depth •
NMEA2000 128275 Distance log •
NMEA2000 130306 Wind data •
NMEA2000 127250 Vessel
heading

NMEA2000 130310 Environmen-
tal parame-
ters

NMEA2000 130312 Temperature •

Share this post


Link to post
Share on other sites

And here is the list of i70s. Where there are 2 dots it can transmit as well. 

Appendix A Supported NMEA
2000 PGN list
PGN Description Received
Transmit-
ted
59392 ISO
Acknowledgment
● ●
59904 ISO Request ●
60928 ISO Address Claim ● ●
126208
NMEA - Request
group function
● ●
126464 PGN List – Receive
/ Transmit PGN’s
Group function
● ●
126992 System Time ● ●
126996 Product Information ● ●
127237 Heading/Track
Control

127245 Rudder ● ●
127250 Vessel Heading ● ●
127251 Rate of Turn ● ●
127257 Attitude ●
127258 Magnetic Variation ● ●
127488 Engine Parameters,
Rapid Update

127489 Engine Parameters,
Dynamic

127493 Transmission
Parameters,
Dynamic

127496 Trip Parameters,
Vessel

127497 Trip Parameters,
Engine

127498 Engine Parameters,
Static

127505 Fluid Level ●
127508 Battery Status ●
128259 Speed ● ●
128267 Water Depth (below
transducer)
● ●
128275 Distance Log ● ●
129025 Position, Rapid
Update
● ●
129026 COG & SOG, Rapid
Update
● ●
129029 GNSS Position Data ● ●
129033 Time & Date ● ●
129038 AIS Class A Position
Report

PGN Description Received
Transmit-
ted
129039 AIS Class B Position
Report

129040 AIS Class B
Extended Position
Report

129041 AIS Aids to
Navigation

129044 Datum ● ●
129283 Cross Track Error ● ●
129284 Navigation Data ● ●
129291 Set & Drift, Rapid
Update

129801 AIS Addressed
Safety Related
Message

129802 AIS Safety Related
Broadcast Message

129809 AIS Class B CS
Static Data Report
Pt A

129810 AIS Class B CS
Static Data Report
Pt B

130306 Wind Data ● ●
130310 Environmental
Parameters
● ●
130311 Environmental
Parameters
● ●
130576 Small Craft Status ●
130577 Direction Data ●

Share this post


Link to post
Share on other sites

So this doesn't clear things up at all. 

Does your sniffer see two sources? The itc5 and the i70s? 

Share this post


Link to post
Share on other sites

May I suggest another way to look at the problem.  Get the best instruments and don't worry if the calibrations are in the transducers or in the displays.  They have to be calibrated somewhere for the systems to work.  After you install it, you can see if the nmea sentences are calibrated or not.  The RaceBox can duplicate the calibrations if the nmea data is uncalibrated or use it as it is if it is already calibrated.  You may still have differences between the standard displayed values and the displayed values after the RaceBox but that would be because the RaceBox can calibrate things more accurately.  For example, in my case the wind display is just not accurate even after it gets the TackTick calibration, which is just an offset.  But the TackTick system can receive calibration data from the RaceBox so it ends up being corrected in the displays if the RaceBox is connected.  The point is, it is not enough to "calibrate" a sensor, you need to be able to take out its errors either because you are starting with an inherently good sender, like the Signet log sender, or you can calibrate errors out sufficiently unlike the TackTick-Airmar log sender which just quits reading below 1.3 knots.

Knowing what I know now, I would figure out if it would be possible to get a B&G wind, a Singnet log, and a smart Airmar depth.  Get all the data into the RaceBox, and run repeater displays on some output of the RaceBox.  I will give you a PC Board that interfaces between the Signet log and the RaceBox if you want.

Just a different way to think about it.

Allen

Share this post


Link to post
Share on other sites
2 hours ago, olsen370 said:

allene222

Sailmons E4 processor will do the calibration for most of the sensor out in the market. The E4 is also capable to distribute the calibrated measurements and the calculated data to your favorite displays.

Sailmon E4

Instrument calibration in detail explained by sailmon 

Out of my league pricewise. 1995 euros. 

https://www.segelwelt.at/en/shop-en/1228/118/navigation-instruments-antenna-internet/sailmon-model-e4-processor-detail.html

This is really a fun experiment to understand how it all works rather than a high budget sailing programme!!! 

Share this post


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

May I suggest another way to look at the problem.  Get the best instruments and don't worry if the calibrations are in the transducers or in the displays.  They have to be calibrated somewhere for the systems to work.  After you install it, you can see if the nmea sentences are calibrated or not.  The RaceBox can duplicate the calibrations if the nmea data is uncalibrated or use it as it is if it is already calibrated.  You may still have differences between the standard displayed values and the displayed values after the RaceBox but that would be because the RaceBox can calibrate things more accurately.  For example, in my case the wind display is just not accurate even after it gets the TackTick calibration, which is just an offset.  But the TackTick system can receive calibration data from the RaceBox so it ends up being corrected in the displays if the RaceBox is connected.  The point is, it is not enough to "calibrate" a sensor, you need to be able to take out its errors either because you are starting with an inherently good sender, like the Signet log sender, or you can calibrate errors out sufficiently unlike the TackTick-Airmar log sender which just quits reading below 1.3 knots.

Knowing what I know now, I would figure out if it would be possible to get a B&G wind, a Singnet log, and a smart Airmar depth.  Get all the data into the RaceBox, and run repeater displays on some output of the RaceBox.  I will give you a PC Board that interfaces between the Signet log and the RaceBox if you want.

Just a different way to think about it.

Allen

Yes i am beginning to think that due to the proprietary nature of all the systems it is difficult to find out how they actually work! I will see what alex comes back with. Not quite ready to purchase anyway so no hurry!! 

Share this post


Link to post
Share on other sites

I'll be at the boat today and do a trace for you if I get a chance. Otherwise there will be lots of time for it in a couple of weeks when I'm cruising. 

Share this post


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

I'll be at the boat today and do a trace for you if I get a chance. Otherwise there will be lots of time for it in a couple of weeks when I'm cruising. 

Ok great! 

Share this post


Link to post
Share on other sites

once you are down the road it will either be B&G because you need the data and accurate calibration or back to Raymarine coz you dont

Post in 2 years and lets us all know

Share this post


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

once you are down the road it will either be B&G because you need the data and accurate calibration or back to Raymarine coz you dont

Post in 2 years and lets us all know

Ok but b&g emailed me and told me their triton display holds the calibrations from the dst800. Ie the data on the network is uncorrected. So great if you have a full b&g setup, h5000 etc. But if you want cheap and cheerful calibrated data across multiple devices i am not so sure. 

All 3 of b&g, raymarine and garmin use the dst800 anyway so i can't really see that the actual data is going to be much different. Different wind sensors though. 

Tbh when I am speaking about calibration on here I am just meaning basic calibrations like wind offset, depth offset and log speed. All I wanted was those basics to be the Samsung across all devices. Then more calibration could be done for racing in whatever setup I felt like. Even that is not so easy though! 

Share this post


Link to post
Share on other sites
On 08/07/2017 at 1:30 PM, Alex W said:

The i70 can output PGNs for sure, this is done when they broadcast display brightness to the network so that other displays in the same group get the same brightness. If calibration math is done per i70 then also must write a broadcast PGN to make them all show the same number (since you can't put different calibrations in different displays).

Alex. If you are on your boat yet I am still keen to know what shows on the network. Maybe there are two instance numbers for the iTC-5 and the I70. 

I am actually considered keeping my old instrument sensors. Getting an i70s and then using the corrected data from the network. 

In the manual it says for example on the i70s that it can input and output log speed. So this must mean it can be two different values with two different instances? 

Share this post


Link to post
Share on other sites

I'm on the boat now and will do it soon.  I'm cruising with a 3 month old baby and we're still finding our rythym.

Share this post


Link to post
Share on other sites
20 minutes ago, Alex W said:

I'm on the boat now and will do it soon.  I'm cruising with a 3 month old baby and we're still finding our rythym.

I think that rhythm is a lot like a three hour watch schedule on passage.

Share this post


Link to post
Share on other sites

I got a log today. My good python scripts for parsing it aren't on the laptop that is with me (sadly), but I found a really old version online. It looks like the nmea traffic is uncorrected.

The corrections are being broadcast to all instruments, and I can probably figure out how when I'm back at home.  Its interesting to note that the adjustments are done on every button push, so all instruments were showing the calibration as I made them.

Notes for the log which I'll share later. 1:00 in is the first adjust, going from -6 wind adjustment to -96.  2:00 I go back. 

Share this post


Link to post
Share on other sites
16 hours ago, Alex W said:

I got a log today. My good python scripts for parsing it aren't on the laptop that is with me (sadly), but I found a really old version online. It looks like the nmea traffic is uncorrected.

The corrections are being broadcast to all instruments, and I can probably figure out how when I'm back at home.  Its interesting to note that the adjustments are done on every button push, so all instruments were showing the calibration as I made them.

Notes for the log which I'll share later. 1:00 in is the first adjust, going from -6 wind adjustment to -96.  2:00 I go back. 

Great stuff but not what I want to hear! 

I am not surprised though. Raymarine haven't specifically told me that you could turn off the i70s for example after running the calibrations and use wifi off the nmea2000 network with the corrected data. 

I suspect Raymarine and B&G both have fancy calibration routines but that all the adjustments aren't  written to the sensors for use on the network. 

I think Garmin are the one of the three who understood what my question was and that the corrections are written to the sensors. (but whadda I know!) 

Share this post


Link to post
Share on other sites

There are lots of good reasons why you wouldn't want corrections done at the transducer. It's a feature, not a bug.

Share this post


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

There are lots of good reasons why you wouldn't want corrections done at the transducer. It's a feature, not a bug.

Respectively, that is one of the craziest things I have heard.  You always want corrections done at the earliest place.  That said, and why I say it respectfully, there are corrections that cannot be done at the source.  Let me elaborate with a simple example.  A wind vane should be calibrated so that if you turn it by hand 90 degrees, the output changes by 90 degrees.  On the other hand, if you want to know what apparent wind angle, you might need to know the boat heel or the position of the mainsail to get the utmost accuracy.  That cannot be done in the transducer unless every transducer is getting every other transducer reading that is just asking for trouble.  In my software, I correct each transducer first, then I correct again for whatever reading is desired taking account whatever variables are important.  Of course, if you are just correcting with a multidimensional lookup table, it doesn't matter but I doubt displays are doing that.

Share this post


Link to post
Share on other sites

It seems to me that there are some basic calibrations that should be done at the transducer. 

Log speed - the dst800 has a multiple point calibration table built in. 

Apparent wind direction offset - to correct for installation error. 

Heading - again to correct for installation (I already have an airmar heading sensor). 

I can't see the advantage of having these set in a display. 

Everything else I will be happy to do in software. 

However I have yet to confirm a system that will even do the 3 above! I am not sure there is one display that will calibrate the dst800 and airmar heading sensor and has a wind sensor that will store an offset. Amazing. 

I think I should get a new nmea2000 wind sensor designed. Rebadge the dst800. Sell a package with no displays but some pc software to calibrate them. Then people could add whatever displays they like! 

Share this post


Link to post
Share on other sites

Actually I think Garmin might work.... 

DST800 calibrated at sensor 

Gwind wired calibrated at sensor

H2183 heading sensor calibrated at sensor 

Maybe...... 

I have emailed them to double check. 

B&G (log speed) and Raymarine (wind offset) don't do this I think. 

Share this post


Link to post
Share on other sites

I didn't say that no corrections should ever be done at the transducer. Things like basic output linearization of nonlinearities inherent in the transducer design or corrections for invariant external distortions (e.g. boxing a compass to compensate for a nearby engine block) are fine IMHO. However, if you're talking about using the output of multiple transducers at a time or correcting for external factors that change faster than once a year, I prefer the math to be done outside the transducer.  

Share this post


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

I didn't say that no corrections should ever be done at the transducer. Things like basic output linearization of nonlinearities inherent in the transducer design or corrections for invariant external distortions (e.g. boxing a compass to compensate for a nearby engine block) are fine IMHO. However, if you're talking about using the output of multiple transducers at a time or correcting for external factors that change faster than once a year, I prefer the math to be done outside the transducer.  

I think we are all agreed. 

Apart from B&G and Raymarine who seem to not even do the basic ones at the sensors consistently. 

Share this post


Link to post
Share on other sites

why not apply a custom calibration profile prior to the display? So, sensor->some custom box->(boat box if not display only)->display & whatever it was you were trying to do. This way the data that hits the screens is the same data that hits whatever it is you're trying to do. 

Share this post


Link to post
Share on other sites
5 hours ago, mustang__1 said:

why not apply a custom calibration profile prior to the display? So, sensor->some custom box->(boat box if not display only)->display & whatever it was you were trying to do. This way the data that hits the screens is the same data that hits whatever it is you're trying to do. 

Yes. This is what I will end up doing if I ever get round to it. Sensors - Allen Edwards racebox - nmea multiplexor to display. 

But meanwhile i have given up! Got b&g system and will use b&g display for depth etc while racing and using second display off racebox with calibrated data. 

I have mentally designed a base system with native nmea2000 sensors that can be calibrating on a pc and the basic calibrations stored in their sensors regardless of which displays you chosen to use. 

 

Share this post


Link to post
Share on other sites

If I were designing the software for a boat instrument system, I doubt I'd use calibration in the sensors unless it's something that can/should be done to the sensor before it's shipped to the customer.

The transducers are made by 3rd parties, Airmar, etc.  They have different calibration abilities.  One speed transducer might have nothing, another a single programmable multiplier, another a multi-point look-up table, and easily a dozen variations on how that table works exactly.  The goal is to have one interface for the customer to calibrate their sensors.  The complexity of that interface is something determined by product development.  It's a trade of capability vs complexity.  What prod dev thinks the market segment they are trying to sell to wants.

Maybe this product should have a very powerful calibration with a multi-dimensional table of offsets and multipliers, indexed by speed and heel angle (tables for both port and starboard of course).  

Maybe this product should say "motor at cruising speed in still water and push calibrate" and then it sets log speed to gps speed with a simple multiplier and that's that.

Am I going to try to support my desired calibration design merged with all the different sensor calibration methods?  Come up with a custom system for each sensor I need to support?  Deal with transducer abilities that change unpredictably from model to model according to Airmar's whim?  Or where we buy transducers from this quarter?  Sounds too hard to support.  I'm going to just get the data from the sensor and run it through my calibration system.  I want the data path for any sensor to be as similar as possible.

Share this post


Link to post
Share on other sites

The transducers are dumb inputs, but on my Raymarine system (using current hardware) all of the transducers interface into the same box, the itc5.  Going into this thread I was thinking that calibration was done there so that the network had calibrated data.  It doesn't look like it works this way, but I'm still not at home to do a full log dump and see. 

The advantage of calibrating at a central place is that all devices on the network would see identical data and you wouldn't have different calibrations per display (especially across brands of network hardware).

The downside is that while the itc5 has most of the raw data it doesn't have all of it. The heading, pitch, and roll come from the autopilot and the GPS data comes from the chart plotter. 

I'm not sure how this is supposed to work on a cross platform network, but those seem to be pretty rare in the real world.

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