Panbo

Airmar PB200 hands-on #1, & into the delivery box!

... written for Panbo by Ben Ellison and posted on Apr 22, 2009
Airmar_PB200_w_splitter_box_cPanbo.jpg

I'm once again way behind on testing borrowed products, and my apologies to all manufacturers involved.  The Airmar PB200 Weather Station above is a case in point.  I mentioned how well it feeds NMEA 2000 data to the lab network back in February, and Dan Corcoran shared his beta testing experience before that, but there's a lot more to say.  For instance, I didn't really understand how it delivered both NMEA 0183 and 2000 data -- did it use one of the various junction boxes available to translate from one protocol to the other? -- until I got my hands on the sample.  Now I know that the PB200 flat out does it all! ...

What you can see in the photo above, especially if you click it big, is that the PB200 unit itself (upside down) fits to a nine pin cable that simply breaks out in the junction box to the standard five NMEA 2000 wires and four 0183 TX/RX wires.  Airmar can, and does, make cables (probably thinner?) that only bring down 2000 or only 0183 (and probably with whatever custom 0183 plug a manufacturer might want). The 0183 cable shown here has the same Airmar plug as the original PB100, which means I can plug it into the Combiner box I already had in the lab. The results were impressive.
    Picture the N2K cable above feeding GPS, heading, wind, air temp, and barometric pressure to every N2K device that knows those PGNs, plus to CE 2009 via the Maretron USB100 gateway.  Meanwhile the Combiner box's USB output is going to another laptop running Airmar's Weather Station software, and it could be feeding most any other 0183 aware program. Plus it could have, say, 0183 depth and speed info combined in at the Combiner box, and/or could be using the box's 0183 output to feed a plotter.  In other words, the PB 200 could supply a lot of data to a big yacht full of NMEA devices, or it can be set up simply to plug into one flavor or the other. Damn!
   I look forward to seeing for myself how good that data is on the water, but I wasn't done with the lab experiments.  I thought I'd stress test CE 2009 by adding the Combiner's USB output...and the output of the Digital Yacht AIT250 (now being sold in the USA by Cactus Navigation)...and...what the hell...the IP AIS feed mentioned recently.  Check out the screen shot; CE handled all four inputs with style and grace.  By right clicking on my boat icon I can choose from any of the three GPS feeds (and CE will automatically roll over if the active feed fails).  And note how much network data, mostly N2K, is showing in the panels.  GPS signal strength would be there too, if...

CE2009_multi_inputs_2_cPanbo.JPG...if I was using the Maretron GPS feed. For some reason, signal strength doesn't get through from the Class B AIS (aka Keyspan USB) or from the PB200 via the Combiner, but does from the USB100 (though it's all 0183 when it gets to CE).  There were a few other fairly minor glitches like this, discussed a bit more in the slide show below (I got a little carried away!), but I'm not too concerned about them.  That's because as sweet as the above is in terms of PC charting, I'm hoping that we're about to go to the next level, which will be straight up NMEA 2000 data to programs like CE via an Intelligent Gateway (NMEA claims it's finally coming).  But (more than) enough for now.  Am I going to take CE and the PB200 along on the Gizmo delivery tentatively routed above? Hell yeah.

Comments

Hi Ben,

About CE displaying TWA even though it has no HDG and STW information. Since the PB200 has its own built in GPS and heading sensor it can calculate and output the shown wind information by itself.

The reason why CE is not displaying HDG and STW when you have disconnected the NMEA2000 gateway is probably that these sentences are disabled on the 0183 output in the PB200. By using the Weathercaster software you can configure the PB200 to which sentences it should output on the NMEA0183.

Marinplus AB distribute Rose Points CE 2009 and ECS products in Sweden. We also distribute Airmar's products and have done quite some tests on the setup with CE 2009 and PB200. It is definitely a great combination of products.

Thanks for the superb Panbo blog, I read it every day!

Posted by: Christer at April 23, 2009 4:23 AM | Reply

You don't even need a large boat to make use of both 0183 and 2000.

I have been thinking to use the NMEA-2000 output as the primary feed to my boat network, and use the NMEA-0183 output exclusively for the autopilot.

This would essentially be like providing a dedicated heading source to the autopilot (like my fluxgate is today), overcoming the need for every bit of my boat network to be powered on and working flawlessly to get heading data from the 2000 side of my network, thru the chartplotter (that converts to 0183), thru my other 0183 devices, and finally to the autopilot. In my particular case, it would also do away with the need (when I get AIS) to deal with an additional component to bridge between 38,400 baud AIS and a 4,800 baud Autopilot.

Posted by: Dan Corcoran (b393capt) at April 23, 2009 6:53 AM | Reply

Thanks, Christer. I just tried turning on every single NMEA 0183 message going out the USB port on the Combiner (had to slow some down of course), but CE still does not see heading or individual sat positions and signal strength.

I also confirmed that the PB200 was set up NOT to use COG as a substitute for Heading when computing True Wind. So how is TWA showing up in CE?

Posted by: Ben at April 23, 2009 9:19 AM | Reply

OMG, telephone company style crunch connectors for a junction box? How primitive. We have installed two PB200s for a local yacht clubs mark set RIBs. So far, it looks like a really good solution, especially for a small boat.

ems

Posted by: emsusa at April 23, 2009 9:20 AM | Reply

Have you tried the Actisense NMEA 2000 to USB adaptor?

We've been using it to interface directly to NMEA 2000 networks and it seems to be very good.

Posted by: nick at April 23, 2009 3:32 PM | Reply

This is a little off topic, Nick, but does the PB200 produce enough data to accumulate and build polars from a period of normal sailing, given the approprate software?

Can you envision a portable set up that could record actual performance data for a boat test article?

Posted by: Sandy Daugherty at April 25, 2009 9:26 AM | Reply

Hello,

I haven't used one yet and I suspect may never. I can't see it replacing a normal wind sensor on a sail boat because of windage, weight and the heel issue.

However, there is no reason why it couldn't be used to a certain extent. It may well be better than the cheaper instrument systems available and require less calibration effort.

I'm not a big fan of doing it automatically for obvious reasons.

Nick

Posted by: Anonymous at April 25, 2009 4:13 PM | Reply

I'd use it excep that I have a sailboat, and my anchor/tricolor light would impact the airflow for this thing. It needs an elevated wand mount to be out in clear air, but then the weight becomes an issue.

Posted by: Mark at April 27, 2009 11:40 AM | Reply

Sandy asked "does the PB200 produce enough data to accumulate and build polars from a period of normal sailing, given the approprate software?"

--> The answer becomes yes, if you add to this a speed over water sensor either paddlewheel or better. Some people might argue you need two speed sensors (one for sailing on port, and the other on starboard .. I would think one is close enough)

In fact, I don't beleive it's practical to build polars with a normal wind sensor and marine electronics gear unless it's tied into a top of the line system, like those from B&G, that compensate true wind angle and speed for motion, heal, and so forth like the PB200 has accomplished.


Nick wrote "I can't see it replacing a normal wind sensor on a sail boat because of windage, weight and the heel issue."

--> I 100% disagree Nick. Read ...
http://www.panbo.com/archives/2009/02/sailing_with_an_airmar_pb200_.html

Posted by: Dan (b393capt) at April 27, 2009 12:58 PM | Reply

Well, that depends on the sail boat and what you want to achieve ...

Posted by: Nick at April 27, 2009 8:44 PM | Reply

Nick: I'm thinking as a Boat Review writer. I'm looking for something I can tape to a stanchion for a couple hours during a boat test, that could provide a numerical comparison of performance. I envision a PB200 on a short stick, feeding a data logger or laptop, running off a small 12V battery.

Part of the question is whether crab angles and the effect of current can be derived from the data available, or if not, would the data still be good enough to compare different sailboats.

Good points Dan.

Posted by: Sandy Daugherty at April 28, 2009 7:52 AM | Reply

dan,

you may disagree with nick on this but he is right, your "review" of the sensor you cite as evidence was very subjective and the facts are that those sensors are not really very good in real world racing mission critical sailing...(if there is such a thing in this game)

If those sensors were any good they would be used in top level sailing. They have all been tested and found wanting and that is why they are absent at the high end.

We would like them to work because they would solve a lot of problems but for now the traditional wind vane and cup anemometer systems are superior.

Posted by: bobc at April 28, 2009 10:06 AM | Reply

To calculate True wind speed and direction when on a moving vessel you need to know your own vessels heading and speed. The PB200 uses its built-in GPS as SOG input to this calculation and its built-in solid state compass as the HDT input. You could actually disable all other NMEA output sentences from PB200 except TWA and the TWA would still show up nice and neat in CE 2009 and you wouldn't need any other sensor providing additional NMEA information either.
The PB200 is promoted as a "Weather Station" but I think that is a bit of an understatement. PB200 is a really smart multisensor that can provide lots of useful information to a boater and definitely more than "just weather" related info.

Why CE 2009 doesn't show HDG from PB200 is a bit mysterious though and I wonder if it can have something to do with the fact that you have had another sensor providing this information to CE 2009 before. Disabling the telnet port in CE 2009 might solve the problem but I haven't yet had time to test this myself. Another very good feature in CE 2009 is the port troubleshooter. You can use this tool to monitor NMEA sentences and even the raw data being received on a specific port. In fact we often use CE 2009 as a tool to troubleshoot connectivity problems in installations even if CE 2009 isn't included in the setup on the ship, it is really powerful and very easy to use.
Is HDT coming in on the port connected to the PB200 in your CE 2009 setup, Ben?

Posted by: Christer at April 28, 2009 6:56 PM | Reply

Christer, the subject of how to calculate True Wind deserves an entire entry, as there's so much confusion about it. But I'll just note here that sailors, racers particularly, want True Wind Relative to Boat -- which is what is being shown in the CE 2009 TWA gauge -- to be calculated by removing a boat's Speed Through the Water from the Apparent Wind being measured on the boat.

True Wind Relative to Boat ignores Set and Drift because the goal is a reference point for sail performance regardless of set and drift. Ground Wind, which most mortals would call True, truly corrects Apparent Wind for all the boat's motion vectors by figuring in the boat's actual heading along with COG and SOG. When and if I write an entry on all this, there will be vector diagrams.

Meanwhile, I added to the confusion in this thread with my first comment's "I also confirmed that the PB200 was set up NOT to use COG as a substitute for Heading when computing True Wind." What I should have written is "I also confirmed that the PB200 was set up NOT to use SOG as a substitute for STW when computing True Wind."

Posted by: Ben at April 28, 2009 11:09 PM | Reply

bobc, in regards to you comments, my first reaction is Ouch! ... what did I do to deserve that? Maybe I did something, but you would need to elaborate for me to understand your point of view.

You apparently thought poorly of the "review".
1. Did you mean that in it's entirety, or just with respect to racing mission critical sailing?
2. What facts back up the "windage, weight and the heel issue" ? How would you describe the issue ?
3. Did you catch the element of the review
a) It has an integrated heel correction feature ?
b) The review was done on the water ... side by side ... with a mechanical vane sensor ?
c) The PB200 is unlike any other wind sensor before it, in that it has integrated motion correction at the masthead ?
d) In a side by side comparison of wind direction measurement performance, the mechanical vane was absolutly crushed, not so much by the ultrasonics vs cup and vane, but by the motion correction feature ?

It would be fair to find fault that the review wasn't done in a wider range of conditions .. but I believe I am breaking new ground in evaluating this motion correction capability on a sailboat.

In regards to "If those sensors were any good they would be used in top level sailing." I would answer that's not a fair summation ...

#1 In top level sailing the wind and cup sensor dosn't stand on it's own ... it is backed up by an array of additional sensors and sailing optimized computer software that corrects the wind measurement with the data from the other sensor.

#2 Before the review on Panbo ... nothing has appeared in print about the PB200's motion correction capabilities application to sailing ... you read it here first !

Posted by: Dan Corcoran (b393capt) at April 28, 2009 11:44 PM | Reply

Dan,

I for one REALLY appreciate your review. I'd been waiting for some good information on the feasibility of the PB for sailing for some time and you seemed to prove it's worthiness and then some.

Good on ya! and please keep us posted of any further developments!

richard.

s/v hacienda II, Toronto

Posted by: Richard at May 1, 2009 11:48 AM | Reply

Thank you Richard.

In retrospect I overreacted. I need a thicker skin.

I am wondering now if the bobc that commented here is the same bobc from NKE (Regatta processor).

Posted by: Dan Corcoran (b393capt) at May 5, 2009 9:17 PM | Reply

Dan,

Keeping in mind the civility of Ben's site I'll be brief.

The motion correction mentioned in Airmar's literature is for the compass only and will have no affect on the wind data. It is a stabilized compass and because the heading is better it will improve any functions that rely on heading for calculation but that is all. The basic wind measurements will still be suspect.

I repeat, if these sensors were better they would be used at the highest levels, they are not.

Your reference wind vane was a Raymarine ST 60 system, right? This is sailing's equivalent of a random number generator and you could reasonably expect to beat it with a damp finger held up in the air and that alone would probably be more reliable over time.

You said, "#1 In top level sailing the wind and cup sensor dosn't stand on it's own ... it is backed up by an array of additional sensors and sailing optimized computer software that corrects the wind measurement with the data from the other sensor."


This is muddled at best but kinda sorta correct. However, the cup and vane devices used in grand prix sailing, generally from B&G, still provides the primary data and are superior to what is on the market today from any producer of sonic wind units.

Solid state wind is a holy grail for instrument manufacturers and once something is developed that meets the requirements of accuracy and reliability and is light enough not to be a problem on a masthead it will be adapted. Then all the other "array of additional sensors", which is actually only one, will come into play and further refine the data.

It may well be that the Airmar unit is a better sensor than your existing lowest common denominator Raymarine equipment but that does not make it ready for prime time.

Right now it's like a dog who plays piano, the amazing thing is that he can sit at the keyboard, not that he is good at it. I'm sure that the future of this technology is bright.

Posted by: bobc at May 5, 2009 11:53 PM | Reply


Dan,

Just saw your post after I hit the button after a break for dinner or I would have mentioned this, yes I am that bob and I have some experience in these matters. Keep up the good work.

cheers

Posted by: bobc at May 6, 2009 12:00 AM | Reply

Right now it's like a dog who plays piano, the amazing thing is that he can sit at the keyboard, not that he is good at it. I'm sure that the future of this technology is bright.

That pretty well sums up all technology.
As a geek, it's incredible to see progress happen so fast.
As a consumer, whatever you buy today is obsolete tomorrow. ;-)

Posted by: norse at May 6, 2009 12:29 PM | Reply

Bobc: Ouch! But, with all sincerity … Thank You for elaborating as I asked.

I have no regrets, I now completly understand the disconnect … which centers on the apparent (but not true) error you saw in my attributing the motion compensation feature of the PB200 to wind measurements as well as compass heading.

You wrote “The motion correction mentioned in Airmar's literature is for the compass only and will have no affect on the wind data”

Probably due to the literature being focused on power boats, it’s easy to miss, but buried in the literature it is written that the motion compensation applies to wind measurement.

Perhaps I should offer to write some sailboat oriented literature for Airmar, but in this case I find they placed the right priority on placing emphasis on the other features of the PB200 weather station when it comes to power boats.

That aside … in addition to the literature, Airmar confirmed to me the motion correction feature applies to reporting apparent and true wind and their wind tunnel testing. Also Ben reviewed, questioned, and asked me to validate the stunningly glowing review I provided before I had the opportunity to publish it here on Panbo. So in short, it is true, motion compensation applies to wind measurement.

Your comments about the cup and vane technology being comparable to a “random number generator’ is understood. Before I had the opportunity to evaluate the PB200 I might have thought that unreasonable … now I think it’s only just a little harsh. I am (no longer) a believer in the damping feature that is including in my basic sailboat electronics. The existing cup and vane on my boat was an available reference point, but I didn’t rely upon it solely. Under sail the autopilot that used the PB200’s wind information demonstrated remarkable abilities not available when connected to my existing sensor, and at the dock I could do other things.

In fact, at the dock, it was plain to see (as waves rocked my sailboat and the wind was not shifting), the PB200 consistently reporting the identical wind direction without a single degree of angular change thru the five or more oscillations of the mast it took before my sailboat leveled out.

I realize high end racing solutions have motion correction capabilities done in a different manner (better ??), provide additional correction factors for wind and for water speed measurements too (better !!), and such solutions are purpose built for sailing in ways I have not explored.

But, if the PB200 won’t so much as flinch when the mast shakes 15 feet in either direction while the ST60 is deflecting +/- 20 degrees and more, I wonder if the PB200 hasn’t gotten me 80, 85, or even 90% of the way to apparent and true wind measurement nirvana.

I am probably not the right person, but, your welcome to entice me into evaluating your product … but geez … until then you can’t blame me if all I can hear from the dog at this piano is a symphony like no other I have heard before.

Best Regards
Dan

Posted by: Dan Corcoran (b393capt) at May 6, 2009 4:33 PM | Reply

Dan,

I've been following this thread and the original, with great interest. While bobc's comments may or may not be technically correct with regards to the Airmar sensor, it sounds to me as though the bottom line is, that the Airmar works considerably better than a standard cup and vane unit.

I'm in the market for a wind sensor and your review has made me think twice about purchasing a cup and vane unit. While in bobc's opinion the ST60 may not be as good as a wet finger, I would bet that 70% of your average boats have them installed! I don’t know what a full set of B&G or other race class instruments costs, but I can guess. I’m not a grand prix racer, nor are 90% of the sailing public. If this device provides the general public with a greatly improved sensor, so much the better.

If I had a full out racing boat, gobs of money, and a crew, I might worry about which is technically superior, but I’m just looking for the “best bang for the buck” that I can get!

Thanks for your contributions; I look forward to seeing them in the future.

Posted by: Dave at May 6, 2009 8:02 PM | Reply

Hello,

I had thought the Airmar apparent winds did use the accelerometers to correct for motion to some extent. If not, that isn't very useful and somewhat defeats the purpose.

We tested sonic devices in a wind tunnel a couple of years ago for an AC application and the heeling behaviour wasn't ideal - but then most compasses don't like to be heeled either.

However, the Airmar sensor, IF it does correct for motion, should be one step better for low end applications. The gyro isn't in an ideal place at the top of the mast and you still have the problem with N2k that boat speed is only at one hertz.

The Raymarine wind sensor is surprisingly ok in testing and it is a little more waterproof than others. What lets it down is lower down.

Nick

Posted by: Nick at May 7, 2009 5:18 PM | Reply

Nick, I think 1Hz is just NMEA 2000's default update rate for wind. It can be changed and I doubt there's a meaningful limit to fast you could go (certainly the bandwidth is there). Some of the most detailed information I've ever seen on this and other N2K subtleties is contained in Airmar's technical manual for the PB200, which you can get by downloading their WeatherCaster software. I think all the info is in there such that you could program Expedition to calibrate and manage a PB200, once there's a straight thru N2K-to-PC gateway available. Which is very soon, I'm told (but I've told that for a long time).

Posted by: Ben at May 7, 2009 6:35 PM | Reply

Hello,

The default rate for 130306 Wind data is 10Hz.

The issue I was referring to above is that the speed sentence 128259 default is at 1Hz. So, if you could change the default rate for the paddle-wheel ...

On gateways, the Actisense N2k to USB interface works well and is cheap - it is certainly cheaper than the CAN to USB option we have been using.

Rgrds,

Nick

Posted by: Nick at May 7, 2009 8:47 PM | Reply

Dan...I am planning to upgrade my existing sail instruments from current B&G Network to SIMRAD. I plan to use their AP 28 Pilot. They want to sell that with their own rate compass, and they want to sell their IS20 Wind display with their own masthead wind sensor (anemometer cups and vane). Since they advertise their network to run NMEA 0183 and 2000, I am thinking that a person could use an Airmar PB200 and get course information as well as necessary wind information all in one unit. But before I made that purchase, I'd want to be certain that the PB 200 will be compatible with the SIMRAD network and instruments, namely Pilot and Wind. I'd appreciate very much your thoughts on that subject.

Posted by: Jon TC at June 29, 2009 12:59 PM | Reply

Dan, I know the IS20's can use PB200 data fine, and will even let you choose between multiple data sources. Illustrations of the latter here:

http://www.panbo.com/archives/2008/04/simrad_is20s_middle_school.html

I would guess that the same is true of the AP 28, but have not actually seen it done.

Keep in mind, though, that the Simrad gear can not calibrate the Airmar sensor. But that can be done with a laptop and a USB Gateway, as shown in today's Tacktick entry.

(And I think that Airmar has made all the code available so that other manufacturers can program PB200 calibration routines into their own devices. I have no idea if or when other manufacturers will do this, however.)

Posted by: Ben at June 30, 2009 9:59 AM | Reply

The Furuno FI50 Instruments can calibrate to the PB200 easily without any external laptops or gateways.

I just tested it with a new system - no problems.

Techhead

Posted by: Techhead at June 30, 2009 3:13 PM | Reply

Furuno does N2K calibration in an unusual way:

http://www.panbo.com/archives/2008/04/furuno_fi50s_another_calibration_road_taken.html

It's a useful feature, but the calibrations don't reside in the sensor, so non Furuno gear won't see the calibrated values. But maybe Furuno is doing sensor level calibration with Airmar sensors as it sells them under its own brand.

Posted by: Ben at June 30, 2009 3:50 PM | Reply

Leave a comment