Instrument wiring

[GTBecker]

I'm installing a chartplotter package (Raymarine A65) in my 26' cruiser. Out of the box, its GPS is continually powered via the display unit, but NMEA out from the A65 requires that it, too, is operating. That's fine if the chartplotter supply is good but, if it blows a fuse or develops a failure the GPS will not be powered, thus no NMEA, thus no DSC position. I've provided independent power to the GPS to cure that. The boat also has a cell phone, via the Cobra MR F300 BT handset, and a handheld VHF in a ditch bag, worst case.

As I was upside down with a cramping neck I did some, ah, critical thinking. It would seem foolish (read: stupid) to put all helm electronics on one fuse, so the question is how to best power and interconnect the GPS, chartplotter, DSC radio and cell to provide best redundancy of both power and position reporting.

I've looked for such a thread but missed it if it is here. Can someone point me to some assistance to think this through?

Thanks.

Tom

 

[gcboat]

Re: Instrument wiring

Hey Tom,

I spoke with a marine elec. tech a couple of weeks ago concerning - almost -
your same issue on my boat. I might be going off in the wrong direction for you, if so just consider it my advanced age.
Anyhow what he considered doing was to run two complete different sets of electric ( via a two switch bus bar ) to each head unit. With the main units having the redundancy if needed. The plan sounded good at the time but the more I've gotten to think about it there seem to be more questions than answers.
I guess it basically boils down to two complete and separate power sources.
Backup plan so to speak.
Since my boat is still "under construction" I can get by with this.
As I mentioned I still haven't grasped his entire concept yet.

 

[pvanv]

Re: Instrument wiring

As a marine instrument tech, as well as a EE, I feel that we can easily over-complicate things, or at least over-think them.

It's always good practice to have a separate fuse (or breaker) for each unit. That way, should one unit blow a fuse, the rest keeps running. The fuse sizing should be large enough for the unit, yet small enough to blow before the wiring is threatened. Keep in mind that the purpose of the fusing is not to protect the unit per se, but rather to protect the wiring. Many times a failed unit (already bad) is what blows the fuse. Typically the manufacturer of the unit will either supply a fuse, or recommend the appropriate amperage for it.

These units are relatively low power draw (unlike a power windlass or winch), so I don't see any need for expensive or elaborate buss systems -- just separate wiring and fusing. You can use a common ground, which will help avoid the dreaded "ground loop" and its associated RF interferences, but keep in mind that the ground will support all the units, and so should be sized (heavier) accordingly. The rule of thumb for short wiring runs is: 16awg for up to 10 amps, 14awg for up to 15a, and 12 for up to 20a. So, if you are common-grounding 3 units, add up the amps for each, and that will tell you how heavy a ground wire to use.

There a fly in the ointment here, even with independent fuses: Since the plotter can power the GPS, if for some reason the primary GPS fuse were to blow, and the GPS was wired to be capable of drawing power from the plotter, then the "bad" GPS might also overload the plotter's supply -- and blow that fuse as well. This could also work in vice-versa: If the plotter were to blow its primary fuse, yet be wired to be capable of drawing power from the GPS, a blown plotter could potentially zap the GPS fuse. So if you're looking for the most failsafe method of allowing one unit to die, while keeping the others running, don't interconnect them so that they can power each other.

 

[GTBecker]

Re: Instrument wiring

EE: that's one of the problems. I am one, too, so I tend to both over-engineer (to convince myself I have the best solution) and then try to simplify the implementation while retaining its integrity. In this case, I want GPS data to get to three loads, ideally regardless of other device or supply failures.

The problem has just become a little more complex since I've now found that the data from the Raymarine RS12 is not NMEA level (+12v high), but is TTL (about +4v) - and the DSC radio won't accept TTL - so I need to do level conversion, too.

The A65 chartplotter also resets while cranking due to low supply voltage. I think I'll provide for that with a small AGM battery at the helm, via a relay on the start switch, to carry the electronics through the few seconds of cranking, and a trickle-charge circuit during running for the AGM. $160 for a similar commercial device seems excessive.


Tom

 

[pvanv]

Re: Instrument wiring

Sounds like you have two distinctly different issues here.

I think you may find that the raymarine stuff is using what is called SeaTalk, instead of nmea. Seatalk is a semi-proprietary protocol, which, in addition to being a different voltage level, and having those ray-only connectors, does not need to adhere to the nmea "word" standards. Seatalk is fine for connecting the ray modules to each other, but not to industry-standard nmea units, such as the dsc vhf. I have a similar situation with my simrad autopilot -- while it can read nmea, in order for it to steer to the wind angle (it is, after all, a sailboat), it requires the proprietary simrad wind sensors, not just any units. See if ray has a module to provide nmea 0183 or preferably nmea 2000 from seatalk, for the dsc vhf.

If cranking causes the instrument voltage to drop too low, then you have a separate issue. You either need to have the instruments on a different battery (house battery instead of the cranking battery) during starts, or the wiring needs an upgrade, possibly as simple as adding a capacitor reserve to the instrument feed. Yes, you can provide a "mini AGM house battery", just for the instruments, but I believe you can get a more solid (more reliable, without additional relays, etc.) start/charge bank setup for the same or less cash. You shouldn't need to add an additional battery for each and every sensitive load.

 

[GTBecker]

Re: Instrument wiring

Actually, the RS12 data is 9600 bps TTL NMEA, I find. RNS is not able to discover that, apparently (at least, I can't get RNS to recognize it).

On the battery droop, it is only the A65 that can't tolerate it. Isn't a small 12v AGM the functional equivalent of a bank of caps (excepting bypassing of supply noise, which apparently isn't the problem)? Both provide the current through a cranking droop, right? Agreed that the relay is another potential failure - but relays are among the most reliable electronic devices.

My two Group 27s are on an A-AB-B switch. I usually run on one, based on the day of the month; on odd days I use A. I have never encountered a problem with this until the A65 installation, which balks less if I start with both.

Tom

 

[pvanv]

Re: Instrument wiring

GT, nmea is not 9600 bps, ever. You have your hands on a proprietary system. Probably a flavor of seatalk. The nmea 0183 standard is 4800 bps, the 0183HS (high speed) standard is 38.4k bps, and the newer nmea2000 is 250k bps. Raymarine probably has a module to give "real" nmea for your dsc vhf. Probably the easiest and most reliable way to go, since seatalk is not required to use strictly nmea sentences, so if you convert the ttl signal voltage, you will still need to buffer the data stream and regenerate it at the proper baud rate. Even then, you may end up confusing the nmea interface on the vhf, because of the potential for non-standard sentences. So get the raymarine nmea module, and save a lot of headaches.

As for the voltage droop during cranking, if you have more than one battery bank, it is considered standard practice to dedicate a deep cycle bank for house loads (such as instruments), and separate that from the cranking battery, (with the 1-2-both switch) except for charging, or emergency start situations. That simple, reliable, and cheap setup should get you through.

If you'd like, you can add a battery-isolating module to your separated banks. That will let you charge both banks from your alternator, regardless of the 1-2-both setting, (as long as it's not set to "off") yet isolate the house bank from the effects of the starter draw. I'm not a big fan of those isolators (though they do work reliably), because they have a voltage drop associated with them, so if you want to use one and still get a full charge, the charging regulator needs to take that drop into account. Consequently, if you have twin engines and/or a shore power charger and/or a charging generator, each of the charging regulators would then need to be set up to account for the isolator drop. So, properly installing an isolator can add significant complexity to the charging system.

 

[GTBecker]

Re: Instrument wiring

GT, nmea is not 9600 bps, ever....

Yes, I understand. The data, nevertheless, is NMEA data; all of the sentences are standard. I see nothing proprietary in the data - but the speed and TTL level is non-standard, of course.

If I continue to go that route, I can build my own converter; I'm a heavy microprocessor user (ZBasic modules based on Atmels, mostly), so putting a two-com-port level-and-speed converter together is not difficult. Enclosing it well would be most of such a project. That would be another device, though, whether Raymarine's or mine; it'd be better to just use another GPS source, true NMEA, that I can distribute.

The RS12 is an odd one (and its WAAS doesn't work, anyway; it is hardcoded for the moved birds and doesn't see the replacements), and it will not process DBR RTCM corrections, which I prefer to WAAS, despite Raymarine's initial encouragement to try that. Alas.

Thanks for the sounding board.


Tom

 

[V8BoatBuilder]

Re: Instrument wiring

The A65 outputs NMEA 0183, just hook up the VHF to the A65's output, and leave the RS12 connected directly to the A65.

Fuse the A65 and VHF independently, but power them off of a common bus.

Don't overcomplicate it. Every junction/splice/converter/device is another thing that can fail away from the dock. KISS.

As a Raymarine tech, I've seen more active GPS heads fail than chartplotters.

 

[GTBecker]

Re: Instrument wiring

FYI, here is the data from the Raymarine RS12 GPS, 9600 N81 TTL, on yellow/shield:

$GPGGA,215713,2633.3526,N,08201.5338,W,1,09,01.0,00002.1,M,-026.4,M,,*75
$GPGSA,A,3,02,05,10,12,15,24,26,29,30,,,,02.1,01.0,01.8*02
$GPGSV,4,1,15,02,35,084,51,04,00,104,00,05,12,219,42,08,00,073,00*73
$GPGSV,4,2,15,10,46,027,53,12,13,207,39,15,55,168,52,18,11,246,00*70
$GPGSV,4,3,15,21,10,300,00,24,46,330,53,26,62,131,52,27,03,066,00*7B
$GPGSV,4,4,15,29,55,307,54,30,17,243,49,35,,,00,,,,*78
$GPRMC,215713,A,2633.3526,N,08201.5338,W,000.0,259.2,221208,,,A*66
$GPVTG,259.2,T,,M,000.0,N,000.0,K,A*01
$GPZDA,215714,22,12,2008,,*45
$PSGSA,4,26,02,10,12,30,05,24,29,15,,,,02.1,01.0,01.8,00344,21571308,5*5D
$GPGGA,215714,2633.3526,N,08201.5338,W,1,09,01.0,00002.1,M,-026.4,M,,*72
$GPGSA,A,3,02,05,10,12,15,24,26,29,30,,,,02.1,01.0,01.8*02
$GPGSV,4,1,15,02,35,084,51,04,00,104,00,05,12,219,40,08,00,073,00*71
$GPGSV,4,2,15,10,46,027,53,12,13,207,40,15,55,168,52,18,11,246,00*7E
$GPGSV,4,3,15,21,10,300,00,24,46,330,53,26,62,131,52,27,03,066,00*7B
$GPGSV,4,4,15,29,55,307,54,30,17,243,49,35,,,00,,,,*78
$GPRMC,215714,A,2633.3526,N,08201.5338,W,000.0,259.2,221208,,,A*61
$GPVTG,259.2,T,,M,000.0,N,000.0,K,A*01
$GPZDA,215715,22,12,2008,,*44
$PSGSA,4,26,02,10,12,30,05,24,29,15,,,,02.1,01.0,01.8,00344,21571408,5*5A
$GPGGA,215715,2633.3526,N,08201.5338,W,1,09,01.0,00002.1,M,-026.4,M,,*73
$GPGSA,A,3,02,05,10,12,15,24,26,29,30,,,,02.1,01.0,01.8*02
$GPGSV,4,1,15,02,35,084,51,04,00,104,00,05,12,219,40,08,00,073,00*71
$GPGSV,4,2,15,10,46,027,53,12,13,207,39,15,55,168,52,18,11,246,00*70
$GPGSV,4,3,15,21,10,300,00,24,46,330,53,26,62,131,52,27,03,066,00*7B
$GPGSV,4,4,15,29,55,307,54,30,17,243,49,35,,,00,,,,*78
$GPRMC,215715,A,2633.3526,N,08201.5338,W,000.0,259.2,221208,,,A*60
$GPVTG,259.2,T,,M,000.0,N,000.0,K,A*01
$GPZDA,215716,22,12,2008,,*47
$PSGSA,4,26,02,10,12,30,05,24,29,15,,,,02.1,01.0,01.8,00344,21571508,5*5B

There are ten sentences per second, hence 9600 bps. No data is sent by the A65 to the RS12 at any time.

Here's NMEA from the A65, 4800 N81, on yellow/brown:

$ECRMC,220623,A,2633.353,N,08201.534,W,0.0,259.0,221208,004.0,W,A*04
$ECVTG,259.0,T,263.0,M,0.0,N,0.0,K,A*3B
$ECRMC,220624,A,2633.353,N,08201.534,W,0.0,259.0,221208,004.0,W,A*03
$ECVTG,259.0,T,263.0,M,0.0,N,0.0,K,A*3B
$ECGLL,2633.353,N,08201.534,W,220624,A,A*4C
$VWMTW,20.4,C*14
$VWVHW,,T,,M,0.00,N,0.00,K*54
$VWVLW,0.7,N,0.77,N*7B
$ECZDA,220624,22,12,2008,,*50
$ECRMC,220625,A,2633.353,N,08201.534,W,0.0,259.0,221208,004.0,W,A*02
$ECVTG,259.0,T,263.0,M,0.0,N,0.0,K,A*3B
$ECRMC,220626,A,2633.353,N,08201.534,W,0.0,259.0,221208,004.0,W,A*01
$ECVTG,259.0,T,263.0,M,0.0,N,0.0,K,A*3B
$ECGLL,2633.353,N,08201.534,W,220626,A,A*4E
$VWMTW,20.4,C*14
$VWVHW,,T,,M,0.00,N,0.00,K*54
$VWVLW,0.7,N,0.77,N*7B
$ECZDA,220626,22,12,2008,,*52
$ECRMC,220627,A,2633.353,N,08201.534,W,0.0,259.0,221208,004.0,W,A*00
$ECVTG,259.0,T,263.0,M,0.0,N,0.0,K,A*3B
$ECRMC,220628,A,2633.353,N,08201.534,W,0.0,259.0,221208,004.0,W,A*0F
$ECVTG,259.0,T,263.0,M,0.0,N,0.0,K,A*3B
$ECGLL,2633.353,N,08201.534,W,220628,A,A*40
$VWMTW,20.4,C*14
$VWVHW,,T,,M,0.00,N,0.00,K*54
$VWVLW,0.7,N,0.77,N*7B
$ECZDA,220628,22,12,2008,,*5C

That's nine sentences per two seconds.

The EC talker identifier is not recognized by one position logger I tried - and one of my own devices won't either; my code looks for GPRMC, and I'd bet the logger does, too. FWIW.

Tom

 

[pvanv]

Re: Instrument wiring

Agreed with V8. I've seen more Ray GPS heads kick the bucket than other Ray units. Maybe a different GPS receiver (with good WAAS) is the easiest/cheapest way to go.

 

[GTBecker]

Re: Instrument wiring

Since it's come up, what NMEA GPS sensor w/WAAS and an RTCM input do you see is the most reliable in your experiences? I am, after all, ultimately seeking reliable position reporting.

Do they fail from exposure? Temperature extremes? Should the sensor be mounted other than out on a rail?

Tom

 

[GTBecker]

Re: Instrument wiring

West Marine gave me a great deal on a Raystar 125 yesterday, so my DBR RTCM issue is now solved and everything seems to be talking. I need to improve the DBR antenna ground today so I cannot yet evaluate the difference between DBR and WAAS corrections, but the RS125 does accept DBR data - and immediately removes the WAAS birds from the position solution when it detects it - so RTCM data trumps WAAS if it is provided.

Tom