Jump to content

Upgrade from Nexus Server to B&G H5000 Hercules - Please Comment!


Recommended Posts

I've posted a number of times with stuff I've done to integrate the Nexus FDX server the previous owner installed in 2005 with the J/109 I purchased in 2013.  It has 3 XL mast displays, on the spray hood 2 Mutli controls and the Nexus Wind display.  When the original Nexus nWind Race xducer died I installed the Garmin gWind race replacement with GND10, which gave some NMEA 2000 connectivity between the Nexus Server and the N2K side.    This summer my Nexus HPC compass died and I replaced it with a B&G Precision 9 on the N2K bus.  Unfortunately, the GND10 does not translate the roll and pitch from the N2K side back to the Nexus bus side, so all the advanced calculations done by the Nexus server no longer work.  I posted a while back on the J/109 forum with some interconnection diagrams.  The diagram is slightly behind what I have now, but gives a pretty good idea on the starting point.

Here is the grand plan I am considering this winter:

  1. Nexus XL Mast Displays (3) - Leave in place and drive from GND10 for Magnetic Heading, COG (Bottom), TWA & TWS (center), Boat Speed, SOG (Top).  Long term replace with N2k compatible displays.  I can control the displays for any infrequent settings changes via a laptop connected to the GND10 rather than using a Nexus Multi Control.
  2. Spray Hood Nexus Multi Control (2), Nexus Wind, B&G T41 Triton - Replace the 3 Nexus displays with B&G Triton 2 and keep the B&G T41 Triton.  One Nexus Multi Control has buttons that are intermittent when pressed so I need to do something to replace it.  Since I extended N2K bus to spray hood it is a no brainer to plug in additional N2k based displays.  A piece of White Starboard cut with the new holes for the display mounting is what I did in the past and will be done again.
    Possibly add B&G H5000 Graphics display and B&G pilot control (see 6 below)
    Question... What does the H5000 Graphics display have that the Triton 2 does not? If no additional features and I can take advantage of unique H5000 Hercules data on the display, I may just use the Triton 2s as I run out of real estate using the larger H5000 Graphic display.
  3. Nexus Server - Replace with B&G H5000 Hercules.  Connect existing analog paddlewheel to the H5000.  Replace analog depth with NMEA 2000 compatible depth transducer unless there is another way to easily get the Nexus (Airmar) depth xducer to play with the rest of the system.  Some of the chart plotters will drive the analog xdcer so I need to do more research.  The Garmin gWind Race and GND10 will feed wind data via N2k to the H5000.  I already have a B&G Precision 9 compass on the N2K bus for heading.
  4. Helm Pedestal Mounted Raymarine e125 - Could keep this if I keep the existing Raymarine S1G autohelm for integrated control.  Otherwise would get a B&G Zeus 3 12".
  5. Nav Station Raymarine e97, Surface Pro 4 running OpenCPN connected to N2kBus via Actisense Converter - All this would be kept.
  6. Raymarine S1G Autopilot with Remote - Potentially replace with B&G H5000 Pilot Computer, H5000 Pilot Controller on Spray Hood, RF25N rudder sensor (replaces Raymarine analog sensor).  Would use existing Raymarine linear drive - I know it works well with B&G NAC-3 so should work with H5000 pilot computer.  If I updated to the B&G Autopilot, I would definitely put a Zeus 3 at the helm for easy control there.
  7. I have a router and ethernet network that any stuff on the install can connect to (e.g. the H5000 Hercules, Zeus 3, etc.).  It provides local IP connectivity and when I activate the hot spot, provides Internet connectivity.

The intended use is as follows with estimated percentage of use:

  • J/109 One Design racing - 15%
  • Beer can racing - 40%
  • Double handed racing - 10%
  • Distance racing - 15%
  • Deliveries & Cruising - 20%

I'm interested in comments on the setup and what might not work well or might work better.  

Question... What calculations/features are available without the H5000 Hercules on the Triton 2 and Zeus 3? (e.g. what do I loose if I don't jump to the H5000 Hercules processor)

Below are pictures I took of an install I did on the spray hood of a friends J/109.  This is all B&G Triton 2 stuff without an H5000 Processor.  It also has the Triton 2 Pilot Controller and NAC-3 pilot computer.  There is a Zeus 3 at the helm.  Based on sea trial results, I really like the setup!

LycPP4pnwWFCAKhJVwH5G59bVKjQDm-3VLZs_dt9

 

Link to post
Share on other sites

The primary value in high-end sailing instrumentation is wind triangle calculations that result in reliable TWD and TWS. From these, you get VMG, Time-to-kill and lay line feedback.

When you use NMEA2000-connected mast head sensors, your wind data is close to worthless. You're installing a great processor that is specifically designed to provide a pretty good wind triangle solution, but you're feeding it garbage and rationalizing displaying this garbage using a hodge podge of displays. I don't get it.

Money is likely the underlying culprit; fair enough. Your 15% use case, J109 One Design racing, is the only pursuit you mention where nailing starts and lay lines will matter. Is your sail inventory and boat prep sufficiently good that instrumentation is the next place to invest? Do you sail well enough to believe instrumentation will help your cause? Do you have a navigator who can leverage instruments to get you a place or two around the course? Can you realistically afford to toss $10K on the fire to bolster 15% of your boat's usage?

If so, then you should do yourself a favor and design an instrumentation system that reflects these goals and work in that direction as your budget allows. It may take you 2-3 years to transition and the approach you describe above might be a good first step, it's difficult to say.

If not, then don't squander money on a great processor to which you're feeding garbage and that can only provide you with garbage out.

 

  • Like 2
Link to post
Share on other sites

Moonduster - great comments.

I had a pretty reliable setup with good sensor data working with the Nexus server.  The wind instrument was originally an analog input to the Nexus server.  When the sensor died after 12 years, Garmin had taken over Nexus and replaced the analog output wind sensor with a sensor that connected to the Nexus RS-485 bus.  Garmin also introduced the GND10 that translates between the Nexus RS485 bus and NMEA2000.  The wind data previously had correction for pitch and upwash using the Nexus server, so was pretty good and I took the time to calibrate it.  Bottom line - I had good data and feel as if I've stepped backwards since I had to replace a good compass that is now no longer available.  Rather than continuing to throw $ at trying to keep the old system working, I figured it was time to start making the switch. 

The problem was when the Nexus HPC Compass died, I couldn't get a replacement and decided to use the B&G Precision 9 compass.  I thought it would be a good replacement, then after it was installed I found out that the GND10 did not convert roll and pitch from the N2K side to the RS-485 side.  Thus all my true wind data is now less accurate than before.  I have an Actisense NGT-1 and NGW-1 that I can use to check the N2K bus load and data.  I plan to do some tests to find out what the data rate is for wind coming through the Garmin GND10.  If it is marginal, I was thinking of getting a Raspberry Pi and programming it so it would be dedicated to converting the RS485 wind data to N2K sentences that contained apparent wind speed and angle at a minimum 5 hz update rate.  I think this could work and if not, I can always add the B&G MHU sensor which is very pricey!  The H5000 should be able to do the advanced wind corrections assuming the N2K wind data input is accurate and at a fast enough rate.

My analog speed input is pretty accurate.  I moved the sensor to the centerline and have calibrated it.  I put in a blanking plug everytime I leave the boat, so have zero issues with a fouled paddlewheel.  The speed through the water data is very good.  I planned on using the analog output to directly feed one of the H5000 analog inputs.

For the autopilot, I wanted something that will help performance on the short handed races, and making delivery and cruising easier.  My Raymarine S1G is "ok" but could be better.  I also really like the user interface on the B&G Zeus to easily control the B&G autopilot.

On the other stuff - My sail inventory is pretty good.  I have new 3Di main and jib and a new spinnaker used for OD racing.  The PHRF sails are pretty good too.  I'm fortunate to have worked with the local North loft on testing new sail designs and have access to some pretty good stuff.  I'm retired so tend to spend my time doing boat projects on my own boat and for friends (e.g. the photo in my original post was work I did on a friends boat).  As a retired EE I can handle most of the electronics integration work and can write some software... if the vendors would just open it up so we could fix their mistakes!  I have the ability to design, install, integrate and test the various systems and enjoy doing it.  Thus the cost to me is the equipment as my labor is "free".

I'm fortunate to have built a good reliable crew over the past few years with some who should be able to help with the next level.

So - assuming I get good wind data....back to the original questions - any experience with the H5000 Graphic Display versus the Triton 2 displays?  

Link to post
Share on other sites

If you can, why not input heel and pitch from your new compass to NEXUS bus? That would take care of the heel compensation you have lost. I really liked the Nexus system I had on my previous boat and I did my own system to add targets and store all the instrument data without using a computer (I used NSLU2, a bit similar project to Raspberry Pi).

There may be a problem using your MHU with B&G, but I don't think leaving the old displays at the mast is a problem, if you do change the processor. I have NMEA0183 displays at the mast and even they work just fine for showing BSP etc, if you feed the data at a decent update rate.

Link to post
Share on other sites

Generally echo what Moonduster says with a few specific inputs. 

Analog wind & Speed are better than digitized, less lag and the derivatives are directly calculated by processor. 

Precision 9 is a good compass, I’d augment with a motion sensor near your center of rotation, which may not be best magnetic spot. 

12” at helm will be huge. I have one on bulkhead 8’ in front of helm, it’s perfectly readable unless obscured, you might go smaller at helm, put big elsewhere  

Triton and 5k GFD are both N2k, triton lacks start display etc. 

Advantage of single vendor is integration. With Zeus you don’t need pilot display.

Consider if PC based (Expedition) & Deck display  better than H5k or Zeus for starting line 

Some MFD support analog depth, otherwise a “Sonar Hub” will, and allows future expansion (forward scan for cruising), and acts as ether hub 

Radar would be nice for deliveries. Multiple MFD (small helm, big nav?) distributes. 

Link to post
Share on other sites
On 9/22/2018 at 11:01 AM, LionessRacing said:

...

Precision 9 is a good compass, I’d augment with a motion sensor near your center of rotation, which may not be best magnetic spot. 

12” at helm will be huge. I have one on bulkhead 8’ in front of helm, it’s perfectly readable unless obscured, you might go smaller at helm, put big elsewhere  

Triton and 5k GFD are both N2k, triton lacks start display etc. 

Advantage of single vendor is integration. With Zeus you don’t need pilot display.

Consider if PC based (Expedition) & Deck display  better than H5k or Zeus for starting line 

Some MFD support analog depth, otherwise a “Sonar Hub” will, and allows future expansion (forward scan for cruising), and acts as ether hub 

Radar would be nice for deliveries. Multiple MFD (small helm, big nav?) distributes. 

LionessRacing - thanks for the comments. 

Motion Sensor - When I spoke to the B&G rep he indicated that the mems sensors in the Precision 9 compass negated the need for a separate motion sensor.  He indicated that it was needed with the RC-42 rate compass.  Are you aware of why this recommendation would different than what the B&G guy told me?

Since I have a networked PC at the Nav Station, it can mirror the Zeus 3 display at the helm.  I may keep the Raymarine e97 at the Nav Station knowing that it won't control anything but will be a backup display and have a backup for the GPS that can be put on the N2K bus if there is ever a problem with the Zeus 3.  If I remove the Raymarine e97 I have a hole to fill in the nice cherry panel at the Nav station.

I did see what you reported on the H5000 Graphic versus the Triton 2 display by comparing the manuals page by page - Based on the start display being available on the H5000 Graphic, I will include one on the spray hood that will be used as another surface to display whatever is information is desired after the start, along with a couple of Triton 2 displays.

I'm still toying with putting Expedition on the Nav Station PC and know that I can add that anytime.  With the H5000 Hercules processor, what is the preferred connection to the PC for Expedition?  I have available Ethernet, an Actisense NGT-1 to USB and Actisense NGW-1 via USB.  The USB and Ethernet all run through the docking station I have for the Surface Pro 4 mounted at the Nav Station.

I'll play with the existing analog Airmar depth transducer and see if I can connect it to the sounder on the Raymarine e97 display.  If that doesn't work, I'll just get an Airmar N2k compatible depth transducer to fit in the existing housing and be done with it.

I currently have a Raymarine Radar on a removable mast by the transom that is normally off the boat and infrequently used.  With the current Raymarine e97 at the Nav Station and e125 at the helm, it can be controlled and displayed by both.  I know the Zeus display and Raymarine Radar will not play together.  I may in the future add a B&G radar in the system, to replace the Raymarine but not for a while.

On the H5000 pilot controller - I am mounting this on the spray hood so when I have deliveries in ugly weather, I can be dry under the dodger to control the pilot and see what the pilot is doing without being back at the wheel.

 

Link to post
Share on other sites

Regarding Motion detector vs Precision 9, I got one of the first compasses in 2016, so it was not clear how well integrated and accurate it was going to be, the 3D detector was well known, decided to do both  

Mine are mounted about 0.5 meter vertical separation on same bulkhead on starboard hanging locker, near max beam, offset perhaps a meter outboard from centerline, perhaps 1 meter above water line.   I was not sure if that was going to be good enough for magnetic field and it’s a bit more outboard than ideal for motion. Seems to work well, your location may vary. Note Lioness is a CCA centerboard Yawl, with high prismatic coefficient. 

Ask B&G about where to place, so that your rotational pitch, yaw, roll values are more distinct from common linear modes of heave, sway & surge.  Center of rotation may not be a point, as the three axes may have different loci. You might ask with respect to autopilot PID control  loop learning, and best correction for masthead motion back to wind measurements.  If it can calculate where the transducers are moving, it’s got a better chance to correct for it. One solution would be use one for pilot and one for wind and have backups  

For PC, I  use the “Websocket” connection on wired Ethernet. Also, think through your network bandwidths, If you start broadcasting sounder and Radar data, you may want a separate network  

WiFi also works, and depending on how dry nav station, or if you move about a lot, might be better as fewer make/Break connections to fail. My PC is  a fanless $1000 MacBook running boot camp, with SSD ram, only moving part is cover, only one USB-C Port  

You will want one of the Bluetooth pilot remote fobs, allows you to steer from anywhere. With a nav/Ais/radar overlay on a WiFi iPad to your Zeus, you can be briefly below decks, monitoring, dealing with issues, using head etc as needed during a solo passage.  My Zeus is under the Dodger when it’s rigged, accessible from companionway .  

Some type of masthead or spreader video look ahead, preferably IR (but the cost is dear) would round out.  The monitoring use and height of eye of an elevated camera for “lookout” is powerful. You can feed to Zeus video port. 

Link to post
Share on other sites
  • 3 weeks later...

Interesting debate.

I am still on Nexus, with an almost complete Nexus setup. Most parts still working, the log transducer keeps breaking down, and the instruments in the cockpit are often impossible to read due to damp. I still think Nexus was a good system. Often I find myself swaring at Mr Dean Barker and Richard MacAllister who sold Nexus Marine to Garmin. You can rightly criticise Garmin for killing an excellent system. My Nexus stuff was still under warranty when Gamin took over with their none existing service, support and customer care. Out of principle, I will never purchase any Garmin equipment.

Reading these comments is very helpful as I am planning my transition to B&G H5000 Hercules. The total price is more than what I can swallow right now. I need to make the transition is two or three stages.

Another question: How difficult is it to install B&G? When the Nexus system was installed, it was sold a package by a reseller including installation. Taking the system apart a few times, and getting it back in working order; I am more than happy to install the B&G stuff myself. How difficult is it? Any warranty issues doing it this way?

Link to post
Share on other sites
On 10/12/2018 at 10:20 AM, Mogle said:

Another question: How difficult is it to install B&G? When the Nexus system was installed, it was sold a package by a reseller including installation. Taking the system apart a few times, and getting it back in working order; I am more than happy to install the B&G stuff myself. How difficult is it? Any warranty issues doing it this way?

Moogle - I think the B&G system is easier to install and maintain then the Nexus system.  The Nexus system uses the RS485 bus and requires that you initialize the bus and assign addresses for the displays (e.g. the magnet needed for the XL displays, or push a button on the Multi-control displays).  The B&G H5000 Hercules based system has all the displays on a NMEA 2000 bus.  The stuff is much easier to setup and easier to select sources if you have multiple sources on the bus.  The H5000 Hercules is setup with a PC over Ethernet and a web interface.  I think this is much better than the Nexus Race S/W or using the setup option on the Nexus displays.

My Nexus system has a GND10 which has allowed me to put many N2K devices (displays, chartplotter, etc.) in communication with the Nexus equipment.  Since I already built a NMEA 2000 backbone, it is fairly easy to expand and plug in the new B&G stuff.

I can't answer the warranty question since I am buying the pieces from different sources and installing myself.  I have a local dealer / installer who sells the B&G equipment (including the performance stuff B&G offers)  and am buying the remaining pieces I need from him.  I'll be doing all the installation and setup work, but will have access to B&G technical expertise through them.  I am willing to accept some risk since I have the technical ability to integrate and troubleshoot stuff.

Link to post
Share on other sites
  • 2 weeks later...
On 10/18/2018 at 3:38 PM, WHK said:

I can't answer the warranty question since I am buying the pieces from different sources and installing myself.  I have a local dealer / installer who sells the B&G equipment (including the performance stuff B&G offers)  and am buying the remaining pieces I need from him.  I'll be doing all the installation and setup work, but will have access to B&G technical expertise through them.  I am willing to accept some risk since I have the technical ability to integrate and troubleshoot stuff.

Warranty if you buy and install yourself is fine, you just need to register the stuff.

Part of the value of buying through a competent dealer vs online discounter, is that you get that as part of service, along with consultation and suggestions that are well worth the modest cost above discount supplier. 

 

Link to post
Share on other sites
On 10/31/2018 at 11:30 AM, tomcheney said:

I totally disagree with this as a general statement. Depends on the sensor. 

 

 

On a heavy, low powered boat such as mine, the fractions of a second lag built into NMEA 2k for paddlewheels and wind sensors may not matter much, and depending on your network and traffic it may only matter at certain times. 

Link to post
Share on other sites
9 minutes ago, LionessRacing said:

Warranty if you buy and install yourself is fine, you just need to register the stuff.

Part of the value of buying through a competent dealer vs online discounter, is that you get that as part of service, along with consultation and suggestions that are well worth the modest cost above discount supplier. 

 

Very timely post.  I met with the local dealer today and just pulled the trigger to buy the remaining items through him. 

We reviewed my network diagram and assumptions on wind data throughput from the GND10.  It turns out B&G has a newer wind sensor (WS10) that has a 5 hz update rate over N2K, so the 11 hz rate from the GND10 should be good.  I will take some N2K bus measurements when I get it installed on the hard with the H5000 Hercules.  The dealer is passing my info on to the B&G tech folks so I can provide my results to them.  Proof of the install can't really occur until I splash in the spring and start calibration.  Best case is it all works fine.  Moderate case is it's as good as an unprocessed wind sensor (e.g. like basic Triton package).  If it doesn't work out, in the long term I can replace the wind sensor with B&G analog MHU at a later date. 

  • Like 1
Link to post
Share on other sites
On 11/6/2018 at 10:21 PM, LionessRacing said:

On a heavy, low powered boat such as mine, the fractions of a second lag built into NMEA 2k for paddlewheels and wind sensors may not matter much, and depending on your network and traffic it may only matter at certain times. 

Maybe my message was a little flippant. 

In reality the architecture isn’t much different. Fundamentally there’s a device reading an analogue signal and then that device puts it onto a digital bus for the processor to process. 

Inside something like an H5000 box you have hardware reading analogue inputs and connected to the main processor via an i2c bus (or similar). NMEA 2000 is orders of magnitude faster than 0183 and is comparable to the connection between the A2Ds and the CPU. Delay from these inputs should be minimal.

Often in a cheaper lower rate sensor that only sends data at 1/2/5Hz the sensor tries to do some filtering. Bad filtering just introduced phase lag. This is the problem you’re describing. 

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

Maybe my message was a little flippant. 

In reality the architecture isn’t much different. Fundamentally there’s a device reading an analogue signal and then that device puts it onto a digital bus for the processor to process. 

Inside something like an H5000 box you have hardware reading analogue inputs and connected to the main processor via an i2c bus (or similar). NMEA 2000 is orders of magnitude faster than 0183 and is comparable to the connection between the A2Ds and the CPU. Delay from these inputs should be minimal.

Often in a cheaper lower rate sensor that only sends data at 1/2/5Hz the sensor tries to do some filtering. Bad filtering just introduced phase lag. This is the problem you’re describing. 

In any control system, phase lag is a consideration. I don’t know how fast the IMOCA boats going 20 kts update their rudders, on my ancient B-40, the steering gets livelier in a following seaway. B&G did make a big deal about analog paddles and wind vane connected directly to H5k, one presumes they believed that raw data sent on dedicated channels was better than remote processing. (Though of course if you have multiples paddles one has to go to an expansion box..) Since the compass & accelerometers are running on N2K perhaps it’s irrelevant. 

Link to post
Share on other sites

It's not irrelevant at all. To claim that is to assert that you understand nothing about how to assess the bounds on errors.

Where multiple sources of data are independently sampled, averaged, and delivered outside of a global framework that strives to minimize errors, the results are really quite poor compared to a system in which they are sampled, averaged and delivered in a thoughtful, synchronized manner using math and algorithms designed to eliminate rather than randomize error.

Most high-end instrumentation implementations do not use CAN-based data in the wind triangle for this reason.

 

  • Like 1
Link to post
Share on other sites

Apart from most of 

13 hours ago, Moonduster said:

Most high-end instrumentation implementations do not use CAN-based data in the wind triangle for this reason.

Really? You mean other than most of the TP52s, IMOCAs, Americas Cup boats, F50s etc? All using CAN.

Link to post
Share on other sites

The list is much, much longer than that. Js, 72s, Maxis/Super maxis, half the Wally fleet, most Ultims and on and on.

Said differently:

The compass is arguably the most important sensor on the boat; anyone who can afford a non-CAN solution uses a non-CAN solution. The same is true of Wind and Boat Speed.

And finally:

I'm not arguing that a club racer should buy a Hydrins or that they'd gain 10 seconds on the race course if they did.

I'm arguing that when you say that a CAN-based compass (or CAN-based Wind or Boat Speed) doesn't degrade wind triangle calculations, you're wrong. The effect is non-trivial and easily demonstrated.

 

Link to post
Share on other sites

An update now that the boat is hauled.  I have been working with the local B&G dealer who installs H5000 stuff and have ordered the bits I don't already have from them.  Based on discussions with them and the B&G tech folks, I'm going to leave the Garmin Nexus gWind Race and Garmin GND10 to provide uncorrected wind bearing and speed to the H5000 Hercules via the N2K bus.  The data rate and lag should be sufficient for the H5000 to do performance processing.  If it turns out this doesn't work out, I'll install a B&G analog MHU later to replace the gWind.

Here is what I'm in the middle of installing / upgrading:

  1. Sensors: Existing Airmar analog paddle wheel on centerline, existing Precision 9 Compass, Depth/Temp Airmar N2K version DT800 (new), Wind - Garmin Nexus gWind / GND10 for N2K output
  2. Mast Displays: Three Triton 2 displays mounted in Scanpod enclosure (new - replace Nexus XL displays)
  3. Spray Hood: Two Triton 2, H5000 Graphic, H5000 Pilot Controller (new except an existing Triton 2 I have - replace 2 Nexus Multicontrol, Nexus Wind display, & T41 Triton)
  4. Helm: Zeus 3 - 12"  in Navpod enclosure (new - replaces Raymarine e125), existing Standard Horizon CMP31 remote mic for radio
  5. Autopilot: H5000 Pilot with RF300 rudder sensor (new - replaces Raymarine S1G with wireless remote and Raymarine rudder sensor). Using existing Raymarine linear drive
  6. Navstation: H5000 Hercules Processor (new replaces Nexus FDX server).  The following existing equipment remains - Raymarine e97 MFD with SR50 Satellite receiver (all on N2K network) with Raymarine HD418 radar mounted on removable pole that is inserted near stern. Standard Horizon GX2150 Matrix VHF radio with AIS receiver connected via Actisense NGW-1 converter to N2K bus , Fusion RN70 Stereo (N2K),  Actisense NGT-1 and NGW-1 converteres from N2K to USB on docking station.
    Microsoft Surface Pro 4 with docking station (existing) - All Ethernet stuff (except the Radar and Raymarine e97) is connected to existing Linksys Router that offers DHCP server and both wired or wireless connectivity. 
  7. Internet connectivity is via hotspot on cell phone to router and normally only used for email, downloading software updates and downloading gribs. If I wanted to in the future, external connectivity could easily be upgraded.  

The B&G stuff I'm installing is a combination of equipment that came off another boat that was decommissioned (Hercules Processor, H5000 Graphic, H5000 Pilot Computer & Controller) and stuff I'm buying new (Zeus 3, four Triton 2 Displays).  I've taken off the old Nexus displays and have run the N2K bus cable to anywhere it wasn't already run.  The spray hood has been fit with a piece of Starboard to cover the old instrument holes and new holes cut for the new displays.  I'm waiting for the new B&G equipment that I ordered last week. 

I should get everything installed and cabled over the next couple of weeks.  I'll start setting up the sources and default displays with the boat on the hard.  I can also format and upload my existing polars to the H5000 Hercules.  It won't be until I splash in the spring that sea trials and calibration can happen, but all basic connectivity and display checks can be done.  I'm running OpenCPN on the PC and depending how my budget is doing, may buy Expedition later this winter.

Things to be evaluated after the system is used for the 1st season:

  • How good is the WInd and corrections using the Garmin gWind? 
  • Are the sensors in the Precision 9 Compass for roll and pitch good enough for wind correction and autopilot performance, or is a seperate B&G motion sensor needed?
  • With the Zeus 3 and H5000 pilot control on the spray hood, is the B&G wireless remote needed? Some cruising time and short handed races will determine if this is a later upgrade.

Here is a photo of the spray hood with the Starboard masked so the new holes can be cut.  The old Starboard bezel is in the background:

joVIiwD1lTfafV2DKmSnGzCfgmHDl34wnM7hmrMp

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

An update now that the boat is hauled.  I have been working with the local B&G dealer who installs H5000 stuff and have ordered the bits I don't already have from them.  Based on discussions with them and the B&G tech folks, I'm going to leave the Garmin Nexus gWind Race and Garmin GND10 to provide uncorrected wind bearing and speed to the H5000 Hercules via the N2K bus.  The data rate and lag should be sufficient for the H5000 to do performance processing.  If it turns out this doesn't work out, I'll install a B&G analog MHU later to replace the gWind.

 

Things to be evaluated after the system is used for the 1st season:

  • How good is the WInd and corrections using the Garmin gWind? 
  • Are the sensors in the Precision 9 Compass for roll and pitch good enough for wind correction and autopilot performance, or is a seperate B&G motion sensor needed?
  • With the Zeus 3 and H5000 pilot control on the spray hood, is the B&G wireless remote needed? Some cruising time and short handed races will determine if this is a later upgrade.

 

 

3

Impressed. My upgrade to B&G is sadly on hold. Hopefully, things will change for the better soon.

One thing: Can you hold the helm and access the Pilot Control at the same time? If you need to get around the helm and move forward you will need to rethink the pilot. Keep the pilot control where it is, but add a remote control. 

The Garmin GWind is good - I have seen some reports that is is not working well with H5000 when using GND10.

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

One thing: Can you hold the helm and access the Pilot Control at the same time? If you need to get around the helm and move forward you will need to rethink the pilot. Keep the pilot control where it is, but add a remote control. 

Moogle - There will be a Zeus 3 at the helm which is able to control the autopilot. I put the H5000 pilot control on the spray hood so I can control the helm while under the dodger for deliveries, and when sailing short handed to control the helm from the cockpit for tacks and gybes.

  • Like 1
Link to post
Share on other sites

How do you plan to determine either of these:

  • How good is the WInd and corrections using the Garmin gWind
  • Are the sensors in the Precision 9 Compass for roll and pitch good enough for wind correction and autopilot performance, or is a seperate B&G motion sensor needed?

 

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

How do you plan to determine either of these:

  • How good is the WInd and corrections using the Garmin gWind
  • Are the sensors in the Precision 9 Compass for roll and pitch good enough for wind correction and autopilot performance, or is a seperate B&G motion sensor needed?

 

Moonduster - empirically is the only way I can do this without any formal test setup or another calibrated system for A-B comparison.  Observation of wind strip charts and performance of the autopilot due to wave motion are two things that come to mind.  I have to believe that the  B&G algorithms are good and if good data is input to the processor, the wind strip charts would be more stable due to motion than the uncompensated data.  Similarly the performance of the pilot can be assessed under various conditions.  I know it isn't a perfect test environment but my goal is to improve the overall system performance so I may rely on it.

Link to post
Share on other sites

This is among the most difficult things to test without a solid reference. Wind tunnels help. Two-boat testing, too.

Otherwise, it's a bit like driving a car with soggy shocks - the only way you really notice is by replacing them and then it's a real surprise.

 

Link to post
Share on other sites

Wind tunnel testing?  Absolutely out of the question.....

I will have the ability to do two boat testing with another J/109 and have done so when testing new sails.  The issue is how to get a good reference as you note. 

Link to post
Share on other sites

If you are trying to assess TWD & TWS as one outcome of correctly compensated Wind, heel/pitch & speed instruments, an anchored boat in a low wave/current realm would be one "reference".  (time to get the committee boat a recorder for their data...) 

Match up time stamps and see if you are correcting out to their values. 

J109 should be able to tack and resume target speed in less than 60 seconds above 6 kts TWS. 

Link to post
Share on other sites

I noticed the addition of the H5000 motion sensor enabled the apparent wind angle to update far quicker after completing a tack.  There is information from the sensor the H5000 is using that it doesn't (or should) have from the Precision 9. 

On a past boat, a mast motion correction for apparent wind capability  (via an Airmar PB200) provided a tremendous improvement in autopilot performance for the Raymarine S1G in my last boat, the autopilot being particularly lousy sailing in gusts downwind with a spinnaker. With the mast motion compensation the filling and collapsing of the spinnaker no longer caused the boat to zig zag as the autopilot received errant wind direction information from the mast swaying. I would imagine the same benefit comes to the B&G autopilot although I have not tested it (I use an old Simrad autopilot that is off-network). 

 

 

  • Like 1
Link to post
Share on other sites
  • 2 months later...

@WHD; I am pretty much in the same situation. Nexus server still going strong, however the Nexus wind and compass died the last, and I have installed the GND 10 to receive from NEMA 2k. The three multi XLs are still in good condition, and I want to keep them as long as they still work, but replace the server. Did you try to drive the Multi XLs from the GND 10? I'm considering the same but unsure if it can be done or if you still need the server to drive the displays. 

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

@WHD; I am pretty much in the same situation. Nexus server still going strong, however the Nexus wind and compass died the last, and I have installed the GND 10 to receive from NEMA 2k. The three multi XLs are still in good condition, and I want to keep them as long as they still work, but replace the server. Did you try to drive the Multi XLs from the GND 10? I'm considering the same but unsure if it can be done or if you still need the server to drive the displays. 

An answer to your question and provide an update on my upgrade. 

  • Did you try to drive the Multi XLs from the GND 10? No I did not.  You should be able to do that, but if you rely on the Nexus Server to apply corrections, you'll need to do that externally (e.g. on te N2K side).  Since the GND10 only passes heading and not roll and pitch, you can't do higher performance wind corrections. This is the main reason I decided to finally pull the plug on the Nexus system and convert to the B&G based solution.

I decided to replace almost everything and found some used B&G H5000 components that I got for a good price to help control costs. The following equipment was removed from the boat and will be sold, as soon as I get a chance to take pictures, set prices and put up for sale .  I have manuals for all the equipment removed.

The stuff mentioned in post #21 above has been installed and integrated, with the exception of the H5000 Hercules processor that is at B&G to repair a dead NMEA 2000 interface.  I should get that back by the end of January to start playing with everything on the hard.  Everything is "talking & listening"  the way it should over the N2K bus.  Dockside testing of the Autopilot has been completed and configuring the displays for data sets I want to see is done.  The default selections will be updated as soon as the H5000 Hercules is back in the system so it can be used as the corrected data source. 

Here are some install pictures:

Spray Hood with two Triton 2 display, H5000 Graphic and H5000 Pilot Control

ZvtLfumKIogqNlVvmW-i73tEIy7cb3XAHHq06qEF

Mast with three Triton 2 displays in Scanpod Enclosure

nQBgH7TWYLM8BpCQ2boPvs0UyFZ8adLdpkvuqA7C

Helm mounted Zeus 3 12" in Navpod Enclosure

Ofdtg-sLUhvCgvXja6zigR0Mrg0gFDhH4jLxDMTt

H5000 Pilot Computer mounted in aft stbd area (same location as Raymarine S1G since wiring is reused and N2K cable added)

CgdZY4A-TPval_uK68zpr1yZo1GOx1-fM3nIJHjg

RF300 Rudder Angle Sensor Installed parallel to the linear drive.

PRujVo74FkKZxGQCdWRxbNsPvZOJIZQS_1FvTAcg

H5000 3D Motion Sensor mounted under Stbd Seatee inboard on forward bulkhead (about even with mast at waterline level)

nRz4Rn3p6Ag6mXZvd-AdOAe-fUdVtx0EKGGjkiC6

Precision 9 Compass in forward stbd locker (same location that Raymarine heading sensor was mounted for S1G autopilot).  This is about 12" from the 3D motion sensor

34-oa0ZNncPCRGf0rEUjsFwHcJZHQDHqW14u3hXD

H5000 Hercules Processor inside Nav Station panel where Nexus Server used to be (it is currently out to have a defective N2K port repaired)

9NQTAHGtcKI9y9RCk9ib05XWDFl_AgCgeDGhehiz

  • Like 1
Link to post
Share on other sites

I wouldn't put the 3D sensor so low, it can better discern mast motion compared to other boat motion at the end of the highest lever possible (inside the boat). I put my 3D sensor high up in the forward portion of the stbd locker as my H5000 lives in that locker. A better location would be even higher, like on the bulkhead behind the light fixture in our J109 front cabin.  No need to have proximity to the compass.

Link to post
Share on other sites
  • 2 weeks later...
On 1/17/2019 at 7:36 PM, b393capt said:

I wouldn't put the 3D sensor so low, it can better discern mast motion compared to other boat motion at the end of the highest lever possible (inside the boat). I put my 3D sensor high up in the forward portion of the stbd locker as my H5000 lives in that locker. A better location would be even higher, like on the bulkhead behind the light fixture in our J109 front cabin.  No need to have proximity to the compass.

I checked with B&G and they indicated it is good where I have it near the waterline.  It is measuring angular changes, not linear motion. The angular changes are the same no matter how far from the rotation point the sensor is located.

I also just listed a bunch of removed equipment on the J/109 forum.

  • Like 1
Link to post
Share on other sites
  • 1 year later...

It's been a while since I posted any system updates.  I just finished converting my hand drawn block diagrams to a nice instrument network diagram using the free Digikey Schemeit program. It isn't a true CAD schematic but does allow me to have an easy reference when I need to troubleshoot or change something. Here is a link to the page with the pdf file on my website.  In the upper section, scroll down and click on the link J/109 Vento Solare Instruments. It is the pdf format for the image displayed below.

 

Vento-Solare-Instruments-2Dec2020.png

Link to post
Share on other sites

@WHK I am really impressed by your diagram. Very inspiring. 

When onboard our boat, I keep all the instruments turn on all the time, even when not sailing. At one point I had the need to get the boat power usage down to a minimum. The draw on the instruments is not bad. 

I am thinking of installing a switch where I turn off all the outside instruments. Why do I want to do that:

- Get the power consumption down to an absolute minumum.

- Allow me to check the wind, weather, position down below.

- Turn the cockpit outside instruments off, no need to light up the cockpit and more when we are in a marina or at anchor. 

Have you done that? OR am I over-engineering it? Got a mix of Nexus and B&G.

 

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

@WHK I am really impressed by your diagram. Very inspiring. 

When onboard our boat, I keep all the instruments turn on all the time, even when not sailing. At one point I had the need to get the boat power usage down to a minimum. The draw on the instruments is not bad. 

I am thinking of installing a switch where I turn off all the outside instruments. Why do I want to do that:

- Get the power consumption down to an absolute minumum.

- Allow me to check the wind, weather, position down below.

- Turn the cockpit outside instruments off, no need to light up the cockpit and more when we are in a marina or at anchor. 

Have you done that? OR am I over-engineering it? Got a mix of Nexus and B&G.

 

Mogle - I find the reefer is the largest load when I'm cruising.  I manage that by bringing a cooloer and stashing the food & drinks with some ice in that for the day.  The reefer is turned off during the day and stays cold.  When I run the engine, I'll turn the reefer back on.

Similar when racing.  I can turn off one of the chart plotters (Helm or Nav) or both if desired.  The chart plotters are the GPS navigation source and also used with my H5000 for performance stuff.  If needed, I can use the AIS transponder GPS or VHF internal GPS.  The AIS transponder is pretty low power when in the silent mode if I want to use that and conserve power.  I also keep the PC & docking station powered up when I'm racing since it runs Expedition.  In race mode, I only have a single 105 amp-hour Group 31 AGM battery for house, and a Group 24 AGM start battery that is isolated.  When cruising I can throw in a second Group 31 AGM on the house side.

Depending how much you're trying to squeeze out of the battery, your approach isn't unreasonable.  I assume you've got the Nexus Server and/or GND10 which means the RS485 bus will need to be powered if you want to monitor wind.  The Nexus instruments and displays are pretty low draw. You'll find the chart plotter is more power hungry than the rest of the instruments.

Link to post
Share on other sites

I have an unused switch on my dc panel connected to 10 gauge that I run to my j109 fwd cabin for powering the tail end of my n2k backbone. This allows me to choose if my 4 mast displays are powered, saving power when off and eliminated a 1.1 of a 1.3 vdc voltage drop the displays suffer vs being  powered through the original length of n2k cable. The original problem I was after was the displays were the first to flicker out when the battery gets low. With that portion of the backbone isolated it no longer pulls down the voltage on the compass, motion sensor, depth sensor and fwd mounted zg100 gps, none of which I would want complaining for voltage.

The modification included 10ga ground. Be careful not to create a ground loop.

Link to post
Share on other sites
1 hour ago, b393capt said:

I have an unused switch on my dc panel connected to 10 gauge that I run to my j109 fwd cabin for powering the tail end of my n2k backbone. This allows me to choose if my 4 mast displays are powered, saving power when off and eliminated a 1.1 of a 1.3 vdc voltage drop the displays suffer vs being  powered through the original length of n2k cable. The original problem I was after was the displays were the first to flicker out when the battery gets low. With that portion of the backbone isolated it no longer pulls down the voltage on the compass, motion sensor, depth sensor and fwd mounted zg100 gps, none of which I would want complaining for voltage.

The modification included 10ga ground. Be careful not to create a ground loop.

Interesting on the voltage drop.  I used these Ancor backbone cables that have larger 18ga power conductors and haven't had the voltage dropout issue. The regular drop cables use 22ga for all conductors and could contribute to the voltage drop. Just looked at the power specs for the displays.  

  • B&G Triton 2 is rated at 3 LEN and 1.4W typical
  • B&G 20/20 HV is rated at 2 LEN and 0.8W typical
Link to post
Share on other sites

Ancor make a power isolator for breaking the backbone into different segments for power. It's just that an isolator, i.e. it doesn't inject power, so a power injector is needed somewhere on each side of the isolator. Maretron's injectors, on the other hand, treat both sides separately and so also break up the backbone into different power segments all in one go.

What I don't like about Ancor's power injectors is that they are mixed gender, i.e. they don't follow the Maretron convention (both sides female). The advantage of the Maretron convention is that the terminators are all male, rather than needing a female terminator at one end and a male terminator at the other.

Link to post
Share on other sites

The lens really add up with all the other sensors and 3 displays are h5000 color graphic plus a triton 2. I used the maretron powertap to double power provision, and yes, the gender reverse was a pain. I use the maretron micro cable through my boat (22 gauge?) a small upgrade from simnet. Running a 10 ga for power only got me most of the way, I also had to run 10 ga ground from midship battery compartment to fwd accessories after discovering that picking up ground from a 16 ga wire to assesories in the fwd cabin had a 0.4vdc power drop picking up ground from behind dc panel over that 16ga. Voltage drop is now under 0.2vdc.

 

 

  • Like 1
Link to post
Share on other sites
  • 2 months later...
On 12/3/2020 at 4:56 PM, WHK said:

It's been a while since I posted any system updates.  I just finished converting my hand drawn block diagrams to a nice instrument network diagram using the free Digikey Schemeit program. It isn't a true CAD schematic but does allow me to have an easy reference when I need to troubleshoot or change something. Here is a link to the page with the pdf file on my website.  In the upper section, scroll down and click on the link J/109 Vento Solare Instruments. It is the pdf format for the image displayed below.

 

Vento-Solare-Instruments-2Dec2020.png

Hi, Interesting to see that you have kept the gWind Race transducer and using this data in the H5000 system. Is this working properly and are you getting the heel and yaw compensation? Is it working as well as the Nexus system did when it was still working? There should be two versions ( uncompensated from GND10 and compensated from H5000 processor) of the wind data on the N2k network. How do all the units on the network know which one to use?
We are in a similar situation with our j/109, we have some new electronics and some old electronics and want to make a smart upgrade without replacing everything.

Link to post
Share on other sites

Hello Per.  I am using the H5000 compensated data with the B&G H5000 motion sensor. Yes it is all working together..  The H5000 (Hercules Processor) has many options available for input selection and the processed output sent to the displays.  I also have Expedition that is used for logging the data too, but try and load all the settings into the H5000.

Link to post
Share on other sites
  • 5 months later...

Here is an August 2021 update with information about the Garmin gWind Race in the system.  Back in June I had a problem with wind indication being intermittent and finally dying. I thought it was the GND10 and replaced it with a spare, but the problem continued.  One week before Block Island Race Week I bit the bullet and installed the B&G WS320 Wireless Wind system just so I could get something working.  The most difficult task was drilling the mounting holes on top of the mast in a Bosun's chair.  Integration and setup was easy as I paired the sensor before going up the mast.

Fast forward a few months and a friend with a Nexus system asked if I had a spare Garmin gWind wind sensor to check on his boat because his wind died.  I let him borrow my gWind Race and he confirmed the sensor worked.  He ordered a new sensor and I ordered a new cable.  This week I was able to pull through a messenger line using the old cable and then pulled the new cable through the mast.

Conclusion: The Garmin gWind race with the GND10 attached to the H5000 Processor provides better performance than the B&G WS320 Wireless Wind interface.  The apparent wind angle shows 26 degrees on the gWind Race versus 35 degrees on the WS30.  The 26 degrees agrees with my J/109 polars.  I attribute the difference to the increased sensor height getting away from disturbed air.  The WS320 data rate spec is 5 Hz.  The GND10 outputs wind at an 11 Hz rate, which is adequate for H5000 processing.

The B&G WS700 series wind sensors provide the additional height above the mast, but if you already have a gWind Race, there is no need to spend the extra money.

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

Garmin gWind race

Slight diversion but have you ever had issues with the gWind race AWA "sticking" in low wind speeds (<5 TWS)? On my recently installed one sometimes the top part will not rotate to the wind direction (but the propellor section still turns giving me AWS).

Link to post
Share on other sites
11 minutes ago, Snowden said:

Slight diversion but have you ever had issues with the gWind race AWA "sticking" in low wind speeds (<5 TWS)? On my recently installed one sometimes the top part will not rotate to the wind direction (but the propellor section still turns giving me AWS).

No - sounds like you have a bearing problem.

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

No - sounds like you have a bearing problem.

Thanks - I had feared as much.

I guess the procedure is wash with soapy water and go back to Garmin if that doesn't work? I assume these are not user serviceable.

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

Thanks - I had feared as much.

I guess the procedure is wash with soapy water and go back to Garmin if that doesn't work? I assume these are not user serviceable.

There is an old post on the forum about disassembly and servicing various parts, but generally these are not designed for us meager users to disassemble them.

Link to post
Share on other sites

I’ve got a stupid question, but since all the resident experts are already on this thread, I figure it’ll be an easy answer. I’ve currently got Nexus instruments and am adding a Zeus 3. I’ll be switching to all B&G next year, but holding off for a season. Am I correct that a GND10 is all I need to be able to send wind, speed, and depth to the Zeus via NMEA?

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

I’ve got a stupid question, but since all the resident experts are already on this thread, I figure it’ll be an easy answer. I’ve currently got Nexus instruments and am adding a Zeus 3. I’ll be switching to all B&G next year, but holding off for a season. Am I correct that a GND10 is all I need to be able to send wind, speed, and depth to the Zeus via NMEA?

Yes - that is correct.

Link to post
Share on other sites
On 9/19/2018 at 11:37 PM, Moonduster said:

The primary value in high-end sailing instrumentation is wind triangle calculations that result in reliable TWD and TWS. From these, you get VMG, Time-to-kill and lay line feedback.

How are you getting Time-to-kill (aka time to burn / time to line)??  

It's clear that the H5000 calculates this, but I can only get my Zeus3 display to show a moving start-line on the Race panel. Which is pretty useless when you need to be heads up during the start, not interpreting a line. All I want is a single number - seconds late / seconds early, displayed on the mast.

Link to post
Share on other sites
  • 1 month later...
3 hours ago, fnav said:

@WHK Did you keep the Nexus log transducer or go for a new B&G transducer?

I kept the old Nexus 43mm analog transducer that is installed on the centerline.  Works fine.

Link to post
Share on other sites
On 10/16/2021 at 9:07 PM, WHK said:

I kept the old Nexus 43mm analog transducer that is installed on the centerline.  Works fine.

Ok, good to hear! Connected and powered from the H5000 CPU "Pulse input"?

This means there is no need for new thru-hull fittings when migrating from 43 mm Nexus transducers? Just a new Garmin GDT43 depth-transducer with NMEA2K adapter!

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

Ok, good to hear! Connected and powered from the H5000 CPU "Pulse input"?

This means there is no need for new thru-hull fittings when migrating from 43 mm Nexus transducers? Just a new Garmin GDT43 depth-transducer with NMEA2K adapter!

Correct on pulse input.  For depth I ended up replacing the Nexus with an Airmar DT812PV-235-N2 235kHz Plastic Smart Sensor 12-Degree Tilt NMEA 2000.  It fit in the same housing and connects to the N2K bus.

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

Correct on pulse input.  For depth I ended up replacing the Nexus with an Airmar DT812PV-235-N2 235kHz Plastic Smart Sensor 12-Degree Tilt NMEA 2000.  It fit in the same housing and connects to the N2K bus.

Thanks for very useful info, migrating from Nexus and i also have the gwind Race/GND10 combo. Have you seen any documentation on the actual difference going with NMEA2K gwind vs direct wired B&G MHU?

Link to post
Share on other sites
1 minute ago, fnav said:

Thanks for very useful info, migrating from Nexus and i also have the gwind Race/GND10 combo. Have you seen any documentation on the actual difference going with NMEA2K gwind vs direct wired B&G MHU?

No - other than when I checked with B&G they indicated that their wireless unit output at a 5hz rate.  I measured the GND10 output and it sent the Nexus wind at an 11hz rate so provides adequate sampling per B&G requirements.

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...