Satellite Compasses?

[Paul L]

It seems that you could mount two cheap GPS antennas on either side of an arch or a pushpit and get superb roll/pitch/yaw info from sat compass.

[Snowpetrel]

Originally Posted by Dockhead

Actually I would like roll and pitch. When sailing hard I like to set one cockpit display to show heel. Any way to get that onto the system somehow?

http://thumbnail.image.rakuten.co.jp...g/51494a_w.jpg this and some bluetack...

A few of the fishermen I work with are using satcompasses. The main advantage to them is to be able to stabilise there depth sounder and remove the effects of the southern ocean swell to input into their own charting software, such as piscadis. Very impressive technology, it will change the scene for sure in a few years. Interesting thread thanks dockhead.

[smac999]

I'm guessing they have improved since then. but I tested a jrc sat compass about 7 years ago. and a few weeks later pulled it off and hooked the 1970's sperry step gyro back up. with step to nmea 183 converter. because it was so still much better then the "new" garbage.

[bcn]

Originally Posted by Dockhead

Cool! Then why not?! I'll keep my H2183 hooked up, so I can afford to experiment. Just $1000 for centimeter precision??

I would not expect real time cm precision at this price point.
For that quality you need
- surveyor grade gps cards
- differential corrections in real time and from a point not more than 30km away (or virtual base stations).

With standard GPS equipment expect 2m precision, bundled with a 6 degree IMU(3 gyro, 3 acc.) and WAAS/EGNOS corrections you can achieve 50cm without differential corrections.
Heading 0,15° and pitch/roll 0,05°.
And prepare 15.000€.
An IMU is as well required to achieve update frequencies as high as 100Hz or even more. Bare GPS is able to deliver up to 15Hz. Well, there are GPS cards with higher output rates, but there is not more information than at 15Hz.

Goboatingnow has already explained that heading from two GPS sources with a known vector between them is derived from carrier phase analysis. This implies that multi path reception and other noise is very critical for this type of equipment.

But in reality - if you don't want to dock automatically 2m precision is not that bad neither...

The other point to take into account perhaps is that on small boats the heading variation due to sea state and wind is much higher (angle and frequency) than compared with a vessel. The radar overlay has to be able to deal with this - being sufficiently fast.

Hubert

[goboatingnow]

Goboatingnow has already explained that heading from two GPS sources with a known vector between them is derived from carrier phase analysis. This implies that multi path reception and other noise is very critical for this type of equipment.

yes which is why they work well at sea.

Its important to remember that survey grade GPS is interested in absolute accuracy as well as relative accuracy. Compasses have no need for high absolute accuracy

[Dockhead]

Originally Posted by bcn

I would not expect real time cm precision at this price point.
For that quality you need
- surveyor grade gps cards
- differential corrections in real time and from a point not more than 30km away (or virtual base stations).

With standard GPS equipment expect 2m precision, bundled with a 6 degree IMU(3 gyro, 3 acc.) and WAAS/EGNOS corrections you can achieve 50cm without differential corrections.
Heading 0,15° and pitch/roll 0,05°.
And prepare 15.000€.
An IMU is as well required to achieve update frequencies as high as 100Hz or even more. Bare GPS is able to deliver up to 15Hz. Well, there are GPS cards with higher output rates, but there is not more information than at 15Hz.

Goboatingnow has already explained that heading from two GPS sources with a known vector between them is derived from carrier phase analysis. This implies that multi path reception and other noise is very critical for this type of equipment.

But in reality - if you don't want to dock automatically 2m precision is not that bad neither...

The other point to take into account perhaps is that on small boats the heading variation due to sea state and wind is much higher (angle and frequency) than compared with a vessel. The radar overlay has to be able to deal with this - being sufficiently fast.

Hubert

As Dave said, we don't care about absolute position precision at that level. We care about high precision of heading data with high rate of updates. Here the $1000 Hemisphere card is supreme. Modern plotters should have plenty of processing power to deal with a 20Hz heading data stream.

[colemj]

Originally Posted by Dockhead

I have never been on a recreational boat, even some very large expensive ones, where MARPA worked very well, or any better than mine does.

Disclosure - we have the same radar as PaulL. Our ARPA works extremely well up to 30 simultaneous targets. Even if I pick an AIS target, I get identical information from ARPA as AIS - with the exception of a quick change in speed or direction, where it takes ARPA a few seconds longer to catch up.

Our compass is a relatively low-tech Simrad RC42 - yours should perform even better.

Mark

[Dockhead]

Originally Posted by colemj

Disclosure - we have the same radar as PaulL. Our ARPA works extremely well up to 30 simultaneous targets. Even if I pick an AIS target, I get identical information from ARPA as AIS - with the exception of a quick change in speed or direction, where it takes ARPA a few seconds longer to catch up.

Our compass is a relatively low-tech Simrad RC42 - yours should perform even better.

Mark

That sounds clearly better than how my B&G system works. Two points for Furuno. Heading data clearly is not the only factor in MARPA performance. My guess would be that Furuno has some kind of unique algorithm for picking out real bearing out of a mass of inconsistent data. My system clearly doesn't do it that well -- data deviates often from AIS data, unless conditions are calm.

[Dockhead]

Here's a ready made solution:

http://www.navtechgps.com/assets/1/7/V102_DS.pdf

It outputs N2K data, too, although I don't know how to get that onto my network.

I've been quoted $2500; bet I could negotiate a discount.

[colemj]

Originally Posted by Dockhead

That sounds clearly better than how my B&G system works. Two points for Furuno. Heading data clearly is not the only factor in MARPA performance. My guess would be that Furuno has some kind of unique algorithm for picking out real bearing out of a mass of inconsistent data. My system clearly doesn't do it that well -- data deviates often from AIS data, unless conditions are calm.

If you are also having heading problems…

I am starting to suspect that your compass is not operating correctly. Others don't seem to have as much of an issue as you are having, and it is difficult to believe your boat is somehow a unique environment for a compass.

Our compass is mounted 5' from an engine and 2-3' below a locker containing lots of pots and pans. We don't have any problem compensating the compass for this.

Mark

[Dockhead]

Originally Posted by colemj

If you are also having heading problems…

I am starting to suspect that your compass is not operating correctly. Others don't seem to have as much of an issue as you are having, and it is difficult to believe your boat is somehow a unique environment for a compass.

Our compass is mounted 5' from an engine and 2-3' below a locker containing lots of pots and pans. We don't have any problem compensating the compass for this.

Mark

This compass seems to be very sensitive to interference. It was off by 15 degrees or something for a couple days because someone left an electric drill in one of the pullman cabin bunks, 3 meters away or so. I don't know why it would be more sensitive than the old one.

N.B. this is not a question of compensation.

[colemj]

3 meters away? It is hard to believe that something this sensitive, giving such a large error with the most minor interfering field, could be mounted any place on any boat and operate correctly.

Maybe all you need is another $500 less sensitive compass and not a $3000 satellite compass?

Mark

[goboatingnow]

Originally Posted by bcn

I would not expect real time cm precision at this price point.
For that quality you need
- surveyor grade gps cards
- differential corrections in real time and from a point not more than 30km away (or virtual base stations).

With standard GPS equipment expect 2m precision, bundled with a 6 degree IMU(3 gyro, 3 acc.) and WAAS/EGNOS corrections you can achieve 50cm without differential corrections.
Heading 0,15° and pitch/roll 0,05°.
And prepare 15.000€.
An IMU is as well required to achieve update frequencies as high as 100Hz or even more. Bare GPS is able to deliver up to 15Hz. Well, there are GPS cards with higher output rates, but there is not more information than at 15Hz.
.......

Just be aware that a satellite compass is not the same as a survey grade GPS, survey grade GPS uses RTK , but also has local receivers to receive corrected position signals from ground stations. Many countries have a network of such stations that survey gear can access to get corrections.

This is because survey grade GPS is not interested in relative accuracy, they are actually also interested in absolute accuracy.

Sat compasses, over come the issues of DIP, damping, interference , etc that flux gate compasses suffer from, even those that are assisted by solid state gyros. Their primary advantage is the effect of RTK to provide very accurate sub degree relative heading references. Their secondary advantages is a 20Hz typical processing rate and low settling time.

Dave

[tanglewood]

I'm not sure where to begin. Based on a wide range of supporting info including Dockhead's early experience, my own query here under my previous username of Twistedtree, and my own personal testing, I installed a complete suite of Simrad gear ("kit", for my UK friends) on my Nordhavn 60 (I know, it's a power boat, but it really doesn't matter to this discussion). After 6 months of use, I just finished removing all of it and am replacing with other equipment.

I'm slowly blogging about this at MVTanglewood.com, but will try to focus on one particular part that I think is relevant to this thread.

My take is that all the discussion is this thread is that the OP is looking for better ARPA/MARPA performance, therefore the interest in a Sat Compass.

The real issue, I believe, it that Simrad/Navico/B&G radar MARPA is fundamentally broken. A sat compass won't fix it. I have one, and can tell you for certain.

The Navico radars have erratic MARPA course vectors. These are the arrows emanating from the target showing calculated course and speed when you have acquired a target. These vectors tend to swing around in course, and shrink/grow in speed. I've been chasing this problem for 6 months, back and forth with Navico engineering, and they finally agreed they had work to do. I don't doubt they will fix this, but I don't know when. I can't wait, so have returned the equipment and am replacing with other.

Just in case you think I might be acting rashly, these MARPA issues represent one or two in a list of 25 issues reported to Navico with my radar, HS70 sat compass, NAIS400 AIS receiver, auto pilot, and NSO black box. To date only one of those issues has been resolved, or even has a plan to be resolved.

[Dockhead]

Originally Posted by tanglewood

I'm not sure where to begin. Based on a wide range of supporting info including Dockhead's early experience, my own query here under my previous username of Twistedtree, and my own personal testing, I installed a complete suite of Simrad gear ("kit", for my UK friends) on my Nordhavn 60 (I know, it's a power boat, but it really doesn't matter to this discussion). After 6 months of use, I just finished removing all of it and am replacing with other equipment.

I'm slowly blogging about this at MVTanglewood.com, but will try to focus on one particular part that I think is relevant to this thread.

My take is that all the discussion is this thread is that the OP is looking for better ARPA/MARPA performance, therefore the interest in a Sat Compass.

The real issue, I believe, it that Simrad/Navico/B&G radar MARPA is fundamentally broken. A sat compass won't fix it. I have one, and can tell you for certain.

The Navico radars have erratic MARPA course vectors. These are the arrows emanating from the target showing calculated course and speed when you have acquired a target. These vectors tend to swing around in course, and shrink/grow in speed. I've been chasing this problem for 6 months, back and forth with Navico engineering, and they finally agreed they had work to do. I don't doubt they will fix this, but I don't know when. I can't wait, so have returned the equipment and am replacing with other.

Just in case you think I might be acting rashly, these MARPA issues represent one or two in a list of 25 issues reported to Navico with my radar, HS70 sat compass, NAIS400 AIS receiver, auto pilot, and NSO black box. To date only one of those issues has been resolved, or even has a plan to be resolved.

Tanglewood has sent me this information in private correspondence, and I have found his blog to be also very informative.

I'm not sure whether I've had exactly the same problem as his, but I have certainly experienced what are undoubtedly very serious bugs in all of my Navico equipment, which is really disturbing. Some of those definitely identified bugs:

1. Radar grabs the wrong heading source, which ruins MARPA, overlay, and radar bearings.

2. Radar totally incompatible with GoFree and the Wifi1 module. If the Wifi1 module is online, the radar is offline.

3. Conflict between the Triton pilot keypad and the pilot controllers in the Zeus MFDs, with total crash of the keypad or ineffectiveness of the keypad when the MFD pilot controller is active.

4. Erratic pilot behavior, including crashes and sudden swerves.

5. Regular crashes of the MFDs.


My first attempt to deal with these things through Navico tech support put me in touch with a complete idiot and zero results. I gave up for while.

More recently, I was hooked up with an extremely knowledgeable and helpful person who really sorted a lot of things out for me. The solutions were generally software updates -- so various bugs have been eventually corrected.

Like that I have for sure solved problems 1 and 2, and maybe problem 3. I will know whether 4 and 5 are solved only when I get a chance to take the vessel to sea. We shall see.

I hope so because I love this equipment, and am not ready to give up yet.


Besides the bugs above, I have had less than satisfactory MARPA performance, the original subject of this thread. Whether that's a result of a bug or poor heading data I don't know yet. I'm not sure whether I've had the same problem as Twisted Tree or not.



By the way and off topic -- it is thanks to Twisted Tree's blog that I figured out how to use the relative motion extension lines in the Zeus plotters -- genius!!! I sailed 3000 miles last summer through some of the busiest seaways in the world (English Channel and North Sea), and was constantly frustrated by not being able to tell whether AIS targets were passing ahead or behind. I even posted on CF about this problem. Wow, I wish I had known how to set up those extension lines -- what a fantastic tool. I will certainly do it this summer.