Sending data to SeaTalk (1): Circuit Design?

[GHA]

Could we hold the drift while the discussion of SeaTalk is still active please.

Not really active any more though.

Anyway, it's relevant to your option 3, middle road. Be aware that max232 boards might be noisy & how could the the Knauf design be improved?



Bump

 

[JohnGC]

From experience of decoding waveforms it's a good idea to keep edge speed up within reason. My longest SeaTalk cable is about 20m, so there will be some HF loss.

Unfortunately, while I understand the issues you raise, I don't have enough experience in circuit design to figure out a circuit. I can do digital, but once you get into transmission lines, even digits become analogue! However, it did occur to me that there may be a suitable line-driver IC that could be used to reduce the component count.

Catching up after a few days away.........

I would use the "official" version. It can be simplified in a couple of ways.

If you just need to use the +12V SeaTalk supply to power the interface, then you don't need D11 or D1.
That suggests you don't need C74 but I would connect it to D8's anode.

The symbols used for inductors are not ones I've seen before (I suspect it's hard to draw curved lines in their CAD package) but my guess is they are RF bead types. Often schematic diagrams don't have enough details to accurately specify Ls unless the part numbers are displayed; unfortunately that's the case here. That means I can't determine why L3 and L11 are both required but I'd just use a single part, I doubt it's critical.

You could remove the -ve edge enhancer, that would eliminate;
RN1a, TR1b, C3, R5, R4, R3, TR1a

Alternatively...........

I can't think of a line driver that would do the same job, but you could use a couple of open-collector/drain comparators.

By the time you've added misc resistors, PSU and protection components you'd have about 20 or 25 devices.

Given that the "official" version has only a few more inexpensive components, they are cheap, the PCB/stripboard layout won't be critical and the circuit is tested - why bother?

 

[sailordel]

I know this is a really old conversation. I came across it because I was thinking of trying to use some of my old seatalk equipment with a new NMEA 2000 installation. Raymarines stng-st1 interface has some issues interfacing with a powered st1 bus so I was thinking of building an isolator on the st1 side and wanted to know how the st1 drive works the diagrams provided by Nigel were very useful to me.

I have since decided to do away with Raymarine finally and forever. I will no longer accept proprietary connection schemes or protocols and am prepared to spend a few more bucks to replace the old equipment. It seems to me that Raymarine appear to have made it deliberately difficult to interoperate between their legacy st1 equipment and NMEA 2000 by not providing galvanic isolation in this interface. I have spent 15 years fighting half implemented protocol interfaces from Raymarine (e.g. limited nmea 183 - st1 translation in the st 60 graphic, autopilots and autopilot heads) this seems to have been designed to get me to spend more money on Raymarine equipment or keep me pure Raymarine.

I am an electronics engineer and in case anyone else is thinking of doing anything like this here is my analysis of the Raymarine circuit.

1. Inductors in the line. Really good idea and really cheap components, in a situation where you have induced spikes caused by starter motors etc cheap insurance.

2. Receive side circuit TR3(a,b) and associated components this is worth duplicating as is. The resistor chain and TR3b set where the switching point is in the 0-12V range, this is important for best noise immunity. Both transistors function as a level shifter to get down to your controller input voltage.

3. TR2(a,b) and associated components. This is the low level driver. It is a non inverting buffer and also provides current limiting. This is necessary because at some point some incompetent installer is going to short the yellow line to the red line. In Knauf's design this will connect 12V straight across his driver transistor. RIP. Again worth duplicating.

4. TR1(a,b) and associated components. This circuit turn on briefly when the output goes high and as others have said this shapes the rising edge. To describe st1 as driven low pulled up high is a little simplistic. In a situation where you have not a lot of pullup (few instruments) and a lot of capacitance (lots of wire) this circuit is necessary to drive the bus high to meet timing requirements. In my installation where I have only a few metres of wire and lots of instruments I don't need it. You can use this as your decision criteria.

In conclusion despite my earlier harsh words for Raymarine this is a nice piece of design and is intended to meet the full st1 spec I would recommend duplicating the input buffer and drive low circuits the need for the positive edge enhancer depends on your system. All of the components are in the cents range and the only difficulty is arranging them and assembling them on veroboard or similar.