SimNet Network Wiring and DeviceNET - Moderated Discussion Areas

  ContinuousWave
  Whaler
  Moderated Discussion Areas
  ContinuousWave: Small Boat Electrical
  SimNet Network Wiring and DeviceNET

search | FAQ | profile | register | author help

Author Topic:   SimNet Network Wiring and DeviceNET

fno posted 04-14-2014 07:21 PM ET (US)
I have two SIMRAD NSS-8 displays that include an integral GPS antenna and built in sounder capability that is supposedly on par with Simrads BSM-1 stand alone sounder module.
The installation instructions and templates are excellent. They are flush mounted--cuz I can. The purists would have had a heart attack at this point. My one regret--and I share the blame--is that I did not get the correct NEMA-2000 cabling in order for the two NSS-8's to talk to each other and the other goodies I will get over time. What I found: SimNET, which is Simrad's network hardware, does not connect directly to any devices. A Micro [DeviceNET Micro] to Simnet adapter cable is needed to go from the head unit to the SimNET backbone. I am wondering if the Micro connections are a duplicate of Maretron's--the Cadillac of NMEA_2000 network hardware. If so, I am thinking to just set my network hardware up with Micro connectors from Maretron.
jimh posted 04-15-2014 10:14 AM ET (US)
The cabling and network accessories used with Simrad NMEA-2000 devices is called SimNet (or perhaps SIMNET or SimNET or some other form), and is distinctly different from the usual NMEA-2000 wiring based on the recommended DeviceNET Micro connectors. The advantage of the SimNet wiring accessories is the connector diameter on the cable end connectors is much smaller, allowing pre-made cables with connectors to be passed through bulkheads via openings that are smaller than would be required for the more common DeviceNET Micro connectors.
If using a mixture of devices on NMEA-2000, one has to decide what form of connectors to use for the backbone wiring, and then use adaptors to accommodate the devices with different connectors.
The SimNet wiring is really no different than the DeviceNET wiring in the manner in which it provides the network backbone, terminations, and power node. There are really only two differences:
--the SimNet cable end connectors are smaller, as mentioned above, and
--the SimNet wiring appliances for network to drop cable do not tend to be able to daisy chain together, like the DeviceNET network-T's can. The have to be connected together with a SimNet cable.
jimh posted 04-15-2014 10:36 AM ET (US)
I haven't found a very good Simrad webpage showing their SimNet accessories, so I have to direct readers to a vendor to see them:
http://www.defender.com/simrad-simnet.jsp
In the SimNet wiring, the extension cables have the same gender connector on both ends. This is distinct from DeviceNET cables, which tend to have the opposite gender connectors at the two ends of a cable. In SimNet wiring, if you need to extend a cable a "joiner" accessory is used. In DeviceNET, you can daisy-chain cables, that is, you just join two cables together to make a longer cable.
In SimNET, the connectors on devices or backbone wiring junctions are of the opposite gender from the cable connectors. This seems to be how SimNet was able to make their cable connectors so small in diameter: the larger mating connector is always on the device, never on the cable.
If you want to connect a SimNet device to a DeviceNET backbone, use the Simrad #24005729 adaptor. This has the normal SimNet cable connector and a male DeviceNET connector:
http://www.defender.com/product3.jsp?path=-1%7C344%7C2028705%7C2028879& id=1013580
If you want to connect a DeviceNET device, say a Lowrance HDS, to a SimNet network backbone, use the Simrad #24006199. This has the normal SimNet cable connector and a female DeviceNET connector:
http://www.defender.com/product.jsp?path=-1%7C344%7C2028705%7C2028879& id=2260596
jimh posted 04-15-2014 10:53 AM ET (US)
It seems to me that installation of the more common DeviceNET wiring as the network backbone will tend to be more advantageous for this reason: many non-Simrad devices you will buy will have a DeviceNET cable and Network-T included with them. This means you can connect the new device to a DeviceNET-wired backbone without having to buy anything else (in most cases).
If you have a SimNet-wired network backbone, for each new non-Simrad device you add it will be necessary to purchase an appropriate adaptor cable to connect DeviceNET device to SimNet backbone (i.e,, Simrad #24006199, about $30), and then to have an open port on a multi-port expansion box, for example, a SimNet seven port joiner (#24006298), or buy a new single Tee-joiner (#24005860, about $23) and a short SimNet cable (#24005829, about $30) to add a new port to the SimNet.
This makes the incremental cost for adding non-SimNet devices to the SimNet backbone at least $30 and as much as $83 if there is not an open port. In contrast, the cost to add a new DeviceNET device to a DeviceNET backbone may be near zero, as the devices often provide the necessary hardware.
Unless you really love those small diameter connectors or plan to have almost exclusively SimNet devices, it is probably better to make the backbone network wiring conform to DeviceNET.
jimh posted 04-15-2014 11:40 AM ET (US)
The most common and perhaps most advantageous installation of SimNet on a small boat will use a configuration as follows:
--the network backbone will be more or less limited to one seven-port SimNet joiner accessory;
--the power node cable will have an integral termination and will be connected to an end port of the seven-port joiner; this saves a port being used for just power or just termination;
--only one plug-in terminator is used, on the other end of the seven-port joiner from the power cable;
--five ports of the seven-port joiner will be open for attachment of SimNet devices using the standard SimNet cables.
Since on a small boat it is unlikely that one will need more than five NMEA-2000 devices, this arrangement will accommodate most installations. Simrad sells a starter kit with this arrangement in mind, called SIMKIT-1, about $89:
http://www.defender.com/product3.jsp?path=-1%7C344%7C2028705%7C2028882& id=1387911
jimh posted 04-15-2014 12:37 PM ET (US)
For information on Maretron, see the separate thread on that topic:
http://continuouswave.com/ubb/Forum6/HTML/003571.html
fno posted 04-15-2014 01:50 PM ET (US)
Jimh, thanks for extensive write up of the available information describing Simnet from Simrad and also the detailed information in the associated Maretron Micro article. Is it not true that Simrad and some others are using a communications protocol adopted by [NMEA] that is, in essence a CAN open communications network? CAN is an open communications protocol that was created by Mercedes-Benz in conjunction with some German university research projects. It's initial function was to enable the communications of multiple computers/processors in a car. It has found favor in industrial and construction environments as it is very reliable and the data words and sentences can be customized to the desired application such as NEMA has done. CANBus, the hardware protocol, is also a fairly reliable long range network that is suitable for ships and industry. I am not sure DeviceNet is the correct terminology, other than it may be a manufacturers method to differentiate themselves. My facts may be off on some of this, so I look forward to any clarifications others may have.
In any case, it appears that NEMA-2000 however it is implemented by a manufacturer is hands down a huge improvement over the confusing NEMA-0183 (non)standard. NEMA-0183 seems intent on forcing boaters to spend one or two more weekends with their head inside their consoles or Internet deciphering the mystery when that time would be much better spent on the water with a cold beer nearby.
jimh posted 04-15-2014 03:57 PM ET (US)
You can find good background on NMEA-2000 from their website, from a white paper collection at
http://www.nmea.org/content/nmea_standards/white_papers.asp
The electrical signaling and bus communication protocols of NMEA-2000 are probably similar to CAN bus. I think this is mentioned in one of the early technical descriptions. This makes good sense, because there is a global automotive industry using CAN bus, and there are dedicated microchips for interfacing a host computer to the CAN bus.
The data carried on the NMEA-2000 network between devices conforms to the NMEA protocol. I don't think CAN bus provides for things like RUDDER ANGLE and SPEED THROUGH WATER.
DeviceNET is another implementation used in factory automation. This gave rise to a standard connector and cable arrangement. NMEA-2000 seems to have borrowed those cables and connectors. Again, that seems like a good idea, as there is a big market for industrial networks, probably larger than recreational boat networks, or at least larger in c.2000 when the NMEA-2000 standard was getting going.
jimh posted 04-15-2014 04:18 PM ET (US)
For more than you will probably ever want to know about DeviceNET, download their open standard, available for free, from the network oversight organization, at
Planning and Installation Manual
DeviceNet Cable Systems
http://www.odva.org/portals/0/library/Publications_Numbered/ PUB00027R1_Cable_Guide_Print_Copy.pdf
It is rather funny to me to see NMEA charging thousands of dollars to get information on their standard, which uses DeviceNET, and the DeviceNET people--ODVA.ORG--freely distributing their technical standards on-line in PDF format.
jimh posted 04-15-2014 04:20 PM ET (US)
MOLEX, a giant manufacturer in the connector business, has a whole line of DeviceNET connectors. See
http://www.molex.com/molex/products/family?channel=products& chanName=family&key=devicenet_industrial_networks
jimh posted 04-16-2014 02:36 PM ET (US)
Re the use of other standards in NMEA-2000: this excerpt is from a white paper on NMEA-2000:

quote:
The electrical characteristics of the physical layer are dictated by the following:
--Media access uses CAN as defined by ISO 11898, Road Vehicles - Interchange of Digital Information, Controller Area Network (CAN) for High-speed Communication.
There is no doubt that the physical layer of NMEA-2000 is built upon CAN bus standards.
The connector is also mentioned in this same white paper:

quote:
The connector selected for the NMEA 2000 backbone is a 5-pin type used in industrial networks and is available from multiple sources (including Molex, Turck Inc., Methode Components, and Daniel Woodhead Company). The connector is available as a 3-port “T” connector, cable-end connector, bulkhead-mount connector and special configurations with internal termination resistors.
Source: http://www.nmea.org/Assets/2000-explained-white-paper.pdf
In a second white paper, more information is give on the connectors:

quote:
CABLES AND CONNECTORS Two methods are provided for connecting to the network backbone cable: a standard connector or barrier strips. These connections are used for connecting segments of backbone cable together, for connecting terminations at the two ends of the cable, for connecting the network power source, and for connecting nodes. The drop cable, the short cable running from the backbone connection to the node equipment, may connect to the equipment anyway the manufacturer chooses. It is the connections at the backbone that are controlled by the NMEA 2000® standard.
This excerpt is quite interesting. For example, we look at the drop cable needed to connect the NMEA-2000 port of an Evinrude E-TEC engine to a NMEA-2000 network. Just as allowed in the standard, the connector at the engine is not the standard NMEA-2000 connector. The manufacturer, Evinrude, has used the option to "connect the equipment anyway the manufacturer chooses," and Evinrude is using an Amp Seal-Tite connector with four pins. Of course, on the network end of their drop cable is a standard NMEA-2000 connector.
Source: http://www.nmea.org/Assets/final_rtcm_2002_white_paper.pdf
Note that the standard for the NMEA-2000 connector are fully defined in the NMEA-2000 standard, but no one can see that standard without buying a copy of it. Even if I bought a copy of the standard, which would cost me over $1,000, I could not disclose the standard to you. So I am afraid I cannot cite anything specific from NMEA-2000 about their connector standard, except what they have already published in their white papers.
NMEA has disclosed more about their standard connector in this excerpt:

quote:
The NMEA Standards Committee searched for a specification that would more than meet the rugged marine requirement for durability and safety. The Open Devicenet Vendor Association’s physical layer cable and connector specification was chosen since it was a proven and robust specification. The requirements for the cable and connector shall meet the Open Device Net Association (ODVA) Volume Three Edition 1.1 as referenced in the NMEA 2000 Standard. The ODVA Specification is a standard for cables and connectors for automotive, robotic and other terribly rough environments, includes an IP67 rating, standardized pin layout, and passes all of the required tests detailed in the NMEA 2000 Standard.
Source: http://www.nmea.org/Assets/ 20090423%20rtcm%20white%20paper%20nmea%202000.pdf
jimh posted 04-16-2014 02:51 PM ET (US)
I think you can get rather far upstream in the manufacturing flow by going to this Japanese connector manufacturer, DDK Ltd of Japan. Check out their DeviceNet Micro series connectors:
http://www.ddknet.co.jp/English/products/circular/cm02/CM02_ENG.pdf
jimh posted 04-16-2014 03:38 PM ET (US)
To get back to Frank's initial comment about going with Maretron cables and fittings for the NMEA-2000 network, I want to add one little caution:
If you use the Maretron power node cable, it has its own T-Connector for inserting into the backbone. This T-Connector is different than most: it has two female connectors for the backbone. This is done so that there is not an exposed pin with the network power on it. However, this means that for terminating this network you will need two terminators, both male gender. If you buy the Maretron starter kit you'll get the right terminators. If you buy this power node for insertion into an existing network, you are going to wind up with one terminator having the wrong gender.

Hop to:

Contact Us | RETURN to ContinuousWave Top Page

Powered by: Ultimate Bulletin Board, Freeware Version 2000
Purchase our Licensed Version- which adds many more features!
© Infopop Corporation (formerly Madrona Park, Inc.), 1998 - 2000.

Author	Topic: SimNet Network Wiring and DeviceNET
fno	posted 04-14-2014 07:21 PM ET (US) I have two SIMRAD NSS-8 displays that include an integral GPS antenna and built in sounder capability that is supposedly on par with Simrads BSM-1 stand alone sounder module. The installation instructions and templates are excellent. They are flush mounted--cuz I can. The purists would have had a heart attack at this point. My one regret--and I share the blame--is that I did not get the correct NEMA-2000 cabling in order for the two NSS-8's to talk to each other and the other goodies I will get over time. What I found: SimNET, which is Simrad's network hardware, does not connect directly to any devices. A Micro [DeviceNET Micro] to Simnet adapter cable is needed to go from the head unit to the SimNET backbone. I am wondering if the Micro connections are a duplicate of Maretron's--the Cadillac of NMEA_2000 network hardware. If so, I am thinking to just set my network hardware up with Micro connectors from Maretron.
jimh	posted 04-15-2014 10:14 AM ET (US) The cabling and network accessories used with Simrad NMEA-2000 devices is called SimNet (or perhaps SIMNET or SimNET or some other form), and is distinctly different from the usual NMEA-2000 wiring based on the recommended DeviceNET Micro connectors. The advantage of the SimNet wiring accessories is the connector diameter on the cable end connectors is much smaller, allowing pre-made cables with connectors to be passed through bulkheads via openings that are smaller than would be required for the more common DeviceNET Micro connectors. If using a mixture of devices on NMEA-2000, one has to decide what form of connectors to use for the backbone wiring, and then use adaptors to accommodate the devices with different connectors. The SimNet wiring is really no different than the DeviceNET wiring in the manner in which it provides the network backbone, terminations, and power node. There are really only two differences: --the SimNet cable end connectors are smaller, as mentioned above, and --the SimNet wiring appliances for network to drop cable do not tend to be able to daisy chain together, like the DeviceNET network-T's can. The have to be connected together with a SimNet cable.
jimh	posted 04-15-2014 10:36 AM ET (US) I haven't found a very good Simrad webpage showing their SimNet accessories, so I have to direct readers to a vendor to see them: http://www.defender.com/simrad-simnet.jsp In the SimNet wiring, the extension cables have the same gender connector on both ends. This is distinct from DeviceNET cables, which tend to have the opposite gender connectors at the two ends of a cable. In SimNet wiring, if you need to extend a cable a "joiner" accessory is used. In DeviceNET, you can daisy-chain cables, that is, you just join two cables together to make a longer cable. In SimNET, the connectors on devices or backbone wiring junctions are of the opposite gender from the cable connectors. This seems to be how SimNet was able to make their cable connectors so small in diameter: the larger mating connector is always on the device, never on the cable. If you want to connect a SimNet device to a DeviceNET backbone, use the Simrad #24005729 adaptor. This has the normal SimNet cable connector and a male DeviceNET connector: http://www.defender.com/product3.jsp?path=-1%7C344%7C2028705%7C2028879& id=1013580 If you want to connect a DeviceNET device, say a Lowrance HDS, to a SimNet network backbone, use the Simrad #24006199. This has the normal SimNet cable connector and a female DeviceNET connector: http://www.defender.com/product.jsp?path=-1%7C344%7C2028705%7C2028879& id=2260596
jimh	posted 04-15-2014 10:53 AM ET (US) It seems to me that installation of the more common DeviceNET wiring as the network backbone will tend to be more advantageous for this reason: many non-Simrad devices you will buy will have a DeviceNET cable and Network-T included with them. This means you can connect the new device to a DeviceNET-wired backbone without having to buy anything else (in most cases). If you have a SimNet-wired network backbone, for each new non-Simrad device you add it will be necessary to purchase an appropriate adaptor cable to connect DeviceNET device to SimNet backbone (i.e,, Simrad #24006199, about $30), and then to have an open port on a multi-port expansion box, for example, a SimNet seven port joiner (#24006298), or buy a new single Tee-joiner (#24005860, about $23) and a short SimNet cable (#24005829, about $30) to add a new port to the SimNet. This makes the incremental cost for adding non-SimNet devices to the SimNet backbone at least $30 and as much as $83 if there is not an open port. In contrast, the cost to add a new DeviceNET device to a DeviceNET backbone may be near zero, as the devices often provide the necessary hardware. Unless you really love those small diameter connectors or plan to have almost exclusively SimNet devices, it is probably better to make the backbone network wiring conform to DeviceNET.
jimh	posted 04-15-2014 11:40 AM ET (US) The most common and perhaps most advantageous installation of SimNet on a small boat will use a configuration as follows: --the network backbone will be more or less limited to one seven-port SimNet joiner accessory; --the power node cable will have an integral termination and will be connected to an end port of the seven-port joiner; this saves a port being used for just power or just termination; --only one plug-in terminator is used, on the other end of the seven-port joiner from the power cable; --five ports of the seven-port joiner will be open for attachment of SimNet devices using the standard SimNet cables. Since on a small boat it is unlikely that one will need more than five NMEA-2000 devices, this arrangement will accommodate most installations. Simrad sells a starter kit with this arrangement in mind, called SIMKIT-1, about $89: http://www.defender.com/product3.jsp?path=-1%7C344%7C2028705%7C2028882& id=1387911
jimh	posted 04-15-2014 12:37 PM ET (US) For information on Maretron, see the separate thread on that topic: http://continuouswave.com/ubb/Forum6/HTML/003571.html
fno	posted 04-15-2014 01:50 PM ET (US) Jimh, thanks for extensive write up of the available information describing Simnet from Simrad and also the detailed information in the associated Maretron Micro article. Is it not true that Simrad and some others are using a communications protocol adopted by [NMEA] that is, in essence a CAN open communications network? CAN is an open communications protocol that was created by Mercedes-Benz in conjunction with some German university research projects. It's initial function was to enable the communications of multiple computers/processors in a car. It has found favor in industrial and construction environments as it is very reliable and the data words and sentences can be customized to the desired application such as NEMA has done. CANBus, the hardware protocol, is also a fairly reliable long range network that is suitable for ships and industry. I am not sure DeviceNet is the correct terminology, other than it may be a manufacturers method to differentiate themselves. My facts may be off on some of this, so I look forward to any clarifications others may have. In any case, it appears that NEMA-2000 however it is implemented by a manufacturer is hands down a huge improvement over the confusing NEMA-0183 (non)standard. NEMA-0183 seems intent on forcing boaters to spend one or two more weekends with their head inside their consoles or Internet deciphering the mystery when that time would be much better spent on the water with a cold beer nearby.
jimh	posted 04-15-2014 03:57 PM ET (US) You can find good background on NMEA-2000 from their website, from a white paper collection at http://www.nmea.org/content/nmea_standards/white_papers.asp The electrical signaling and bus communication protocols of NMEA-2000 are probably similar to CAN bus. I think this is mentioned in one of the early technical descriptions. This makes good sense, because there is a global automotive industry using CAN bus, and there are dedicated microchips for interfacing a host computer to the CAN bus. The data carried on the NMEA-2000 network between devices conforms to the NMEA protocol. I don't think CAN bus provides for things like RUDDER ANGLE and SPEED THROUGH WATER. DeviceNET is another implementation used in factory automation. This gave rise to a standard connector and cable arrangement. NMEA-2000 seems to have borrowed those cables and connectors. Again, that seems like a good idea, as there is a big market for industrial networks, probably larger than recreational boat networks, or at least larger in c.2000 when the NMEA-2000 standard was getting going.
jimh	posted 04-15-2014 04:18 PM ET (US) For more than you will probably ever want to know about DeviceNET, download their open standard, available for free, from the network oversight organization, at Planning and Installation Manual DeviceNet Cable Systems http://www.odva.org/portals/0/library/Publications_Numbered/ PUB00027R1_Cable_Guide_Print_Copy.pdf It is rather funny to me to see NMEA charging thousands of dollars to get information on their standard, which uses DeviceNET, and the DeviceNET people--ODVA.ORG--freely distributing their technical standards on-line in PDF format.
jimh	posted 04-15-2014 04:20 PM ET (US) MOLEX, a giant manufacturer in the connector business, has a whole line of DeviceNET connectors. See http://www.molex.com/molex/products/family?channel=products& chanName=family&key=devicenet_industrial_networks
jimh	posted 04-16-2014 02:36 PM ET (US) Re the use of other standards in NMEA-2000: this excerpt is from a white paper on NMEA-2000: quote: The electrical characteristics of the physical layer are dictated by the following: --Media access uses CAN as defined by ISO 11898, Road Vehicles - Interchange of Digital Information, Controller Area Network (CAN) for High-speed Communication. There is no doubt that the physical layer of NMEA-2000 is built upon CAN bus standards. The connector is also mentioned in this same white paper: quote: The connector selected for the NMEA 2000 backbone is a 5-pin type used in industrial networks and is available from multiple sources (including Molex, Turck Inc., Methode Components, and Daniel Woodhead Company). The connector is available as a 3-port “T” connector, cable-end connector, bulkhead-mount connector and special configurations with internal termination resistors. Source: http://www.nmea.org/Assets/2000-explained-white-paper.pdf In a second white paper, more information is give on the connectors: quote: CABLES AND CONNECTORS Two methods are provided for connecting to the network backbone cable: a standard connector or barrier strips. These connections are used for connecting segments of backbone cable together, for connecting terminations at the two ends of the cable, for connecting the network power source, and for connecting nodes. The drop cable, the short cable running from the backbone connection to the node equipment, may connect to the equipment anyway the manufacturer chooses. It is the connections at the backbone that are controlled by the NMEA 2000® standard. This excerpt is quite interesting. For example, we look at the drop cable needed to connect the NMEA-2000 port of an Evinrude E-TEC engine to a NMEA-2000 network. Just as allowed in the standard, the connector at the engine is not the standard NMEA-2000 connector. The manufacturer, Evinrude, has used the option to "connect the equipment anyway the manufacturer chooses," and Evinrude is using an Amp Seal-Tite connector with four pins. Of course, on the network end of their drop cable is a standard NMEA-2000 connector. Source: http://www.nmea.org/Assets/final_rtcm_2002_white_paper.pdf Note that the standard for the NMEA-2000 connector are fully defined in the NMEA-2000 standard, but no one can see that standard without buying a copy of it. Even if I bought a copy of the standard, which would cost me over $1,000, I could not disclose the standard to you. So I am afraid I cannot cite anything specific from NMEA-2000 about their connector standard, except what they have already published in their white papers. NMEA has disclosed more about their standard connector in this excerpt: quote: The NMEA Standards Committee searched for a specification that would more than meet the rugged marine requirement for durability and safety. The Open Devicenet Vendor Association’s physical layer cable and connector specification was chosen since it was a proven and robust specification. The requirements for the cable and connector shall meet the Open Device Net Association (ODVA) Volume Three Edition 1.1 as referenced in the NMEA 2000 Standard. The ODVA Specification is a standard for cables and connectors for automotive, robotic and other terribly rough environments, includes an IP67 rating, standardized pin layout, and passes all of the required tests detailed in the NMEA 2000 Standard. Source: http://www.nmea.org/Assets/ 20090423%20rtcm%20white%20paper%20nmea%202000.pdf
jimh	posted 04-16-2014 02:51 PM ET (US) I think you can get rather far upstream in the manufacturing flow by going to this Japanese connector manufacturer, DDK Ltd of Japan. Check out their DeviceNet Micro series connectors: http://www.ddknet.co.jp/English/products/circular/cm02/CM02_ENG.pdf
jimh	posted 04-16-2014 03:38 PM ET (US) To get back to Frank's initial comment about going with Maretron cables and fittings for the NMEA-2000 network, I want to add one little caution: If you use the Maretron power node cable, it has its own T-Connector for inserting into the backbone. This T-Connector is different than most: it has two female connectors for the backbone. This is done so that there is not an exposed pin with the network power on it. However, this means that for terminating this network you will need two terminators, both male gender. If you buy the Maretron starter kit you'll get the right terminators. If you buy this power node for insertion into an existing network, you are going to wind up with one terminator having the wrong gender.