LED Anchor Light

[Maine Sail]

Originally Posted by hellosailor

Lopolight, ROFL.

They're selling smart systems that notify the bridge when a light is due for replacement, and 300-1000 euros are common prices on their list. Although, the price list doesn't specify which lights have the brains onboard.

And, their lights are not sold for the US market in any case.

Nice exercise in how to make something simple (a light bulb) into something way more complex, and arguably frail.

Looplights are sold in the US and have been for many years. They appear to be very well made BUT I have replaced a number of failed lights and currently have a dead one sitting on my desk... It seems Evans has had some too. They claim to have made changes, and do stand behind the product very, very we'll, but I have lost count as to how many failures I have seen. I have also noted them as being quite noisy RF wise.

Loop seems like a good company, are responsive and stand behind the product but I am arguably a tad gun shy these days. I have hopes they can work out these issues but they are still there after multiple years...

Our YC launches went through at least six, maybe more, before we replaced them with another brand. They would have sent us more replacements but we simply got tired of replacing them...

50,000 hours, anyone seen that on any LED yet..?

[goboatingnow]

Some might find this an interesting article as well as some of the links in that article to other useful articles


http://www.digikey.com/us/en/techzon...of-fading.html

Dave

[Lake-Effect]

Originally Posted by Saltyhog

To accommodate this, PWM controllers are used and controlled where hours on are counted and as the hours go up, so does the duty cycle and therefore the RMS current.

Seriously, do any of the LED bulbs/fixtures in the marine market do this?

Re linear vs switching regulators - of course a switching regulator will be more efficient, but the linear reg + LED is still alot more efficient than the comparable incandescent bulb.

[conachair]

So regarding RF noise would there be anything to be gained by having a driver down below feeding just the LED ' s at the masthead? Or would that be worse?

[hellosailor]

"I would not trust an LED navigation light to be within specification over time if there is not some allowance made for brightness half life."
There is a very simple way to correct for that degradation. The industry standard, as I recall, is that LED's degrade no more than 50% at 50,000 hours. So you simply start with twice the light that is required, and you know that at 50,000 hours you will still be legal. Or, you do as trucks and busses do in the US: They normally have 12-24 LEDs in each rear light, overvolted and unregulated. They know several will burn out in the course of a year, and then they simply replace the entire assembly as routine depot maintenance. Which is cheaper and more effective than using prime LEDs for that purpose.

Counting the hours? No more effective than routine maintenance, in the greater picture. Nice tweak, but sooner or later the LEDs should be replaced anyhow. From Mainesail's comments, apparently all that technology is just a bit too lightly engineered.

Salty-
Lopo's web site has no US pricing on it, so I took it on faith that they aren't selling to the US market. Even if someone else is bringing them in. Maybe Jamestown bought the US distribution rights.

Cona- Great idea, but then the mast wiring itself becomes a huge antenna, not a good thing. And, in order to regulate properly you'd need to know the exact power loss in that wiring, which would also increase over time. Better to install a servomtoro jukebox at the masthead with a remote control that swapped in different resistors to....Oh, wait, that would be the kind of thing only BMW would do.<VBG>

(There's a reason engineers are sometimes not allowed out of the laboratory.)

[conachair]

Originally Posted by hellosailor

Cona- Great idea, but then the mast wiring itself becomes a huge antenna,

So is it just the dirty output voltage which causes interference? Not knowing a huge amount about these things. Ta



And how does the wiring resistance increase over time?

[goboatingnow]

Theres no engineering reason why you couldn't put the power supply to the LEDs at the electrical panel. This would enable you to produce a good quality clean DC voltage and send it up the mast to a linear regulator in the fitting. done all the time ( see power over Ethernet for example)

The wire will NOT become an antenna. Most switching noise from RF fittings is Air coupled into the antenna, because its happens to be beside the anchor light. Sperate the two by a few feet and most problems go away. Its not difficult to make switchers quiet, its just its an extra filter that get "cost reduced" out of the circuit .


dave

[Paul Elliott]

Originally Posted by hellosailor

Cona- Great idea, but then the mast wiring itself becomes a huge antenna, not a good thing. And, in order to regulate properly you'd need to know the exact power loss in that wiring, which would also increase over time. [...]

Well, speaking as an engineer, it would probably be easier to EMI-filter the mast wires since size and weight aren't at such a premium down at deck-level. Also, you can more or less ignore the wire loss, since you would want to be running constant-current up to the masthead LEDS.

Still, I think that approach is overcomplicated.

Originally Posted by conachair

So is it just the dirty output voltage which causes interference? [...]

And how does the wiring resistance increase over time?

EMI is generally caused by both the switching and control currents running around on the circuit board, and by any switching current that flows through the power wiring. Bypass capacitors and filter inductors (usually ferrites) can do an effective job of cleaning up the power wiring noise, and the quiet switcher designs already include these filters in the light assembly. Any switcher needs some bypass capacitors at the power input just to function, but these may not be sufficient to kill the EMI.

The circuit-board current-loops really have to be minimized by proper design and layout. Short of putting a shield around the unit, there's not much the boat owner can do (and a shield isn't such a bright idea for a nav-light).

The wiring EMI can be bad at all frequencies, but can be killed with external filtering. The circuit-board EMI usually isn't as much of an issue at SSB frequencies since the dimensions are too small for effective radiation. However, especially with a VHF antenna nearby, it can be quite bad at VHF frequencies.
Wiring resistance changing? I dunno, except for corrosion maybe...

[hellosailor]

The wiring resistance will change as it corrodes from moisture, unless someone uses more expensive tinned wire. And it might change as the wire physically changes, since stranded wire hanging vertically will become longer and thinner as it suffers from gravity. Not as badly as coax would, but it should still change. Then of course we can assume the connections will be perfect and they will never increase in resistance as they corrode, can't we? <G>

Whether the interference is RFI or EMI, if there's fifty feet of wire running up the mast, and it is carrying pulsed DC, that sounds like a great potential for the pulsing to create interference. Maybe into poorly shielded coax, maybe out into the universe in general, but after all, isn't that the same thing as a spark-gap transmitter? Pulsing DC over the wires?

Might or might not be a problem, might or might not be simple to fix. In any case you couldn't sell a generic "mount this driver at your panel" solution, because you'd have no idea how much current was going to make it up the "generic" mast on the "generic" wiring run to the nav light. Could you? That's the stopper.

[conachair]

Originally Posted by hellosailor

The wiring resistance will change as it corrodes from moisture, unless someone uses more expensive tinned wire.

Ta. I was along the same lines then.

How many people use untinned wire up the mast more than once?

[Paul Elliott]

Originally Posted by hellosailor

Whether the interference is RFI or EMI, if there's fifty feet of wire running up the mast, and it is carrying pulsed DC, that sounds like a great potential for the pulsing to create interference. Maybe into poorly shielded coax, maybe out into the universe in general, but after all, isn't that the same thing as a spark-gap transmitter? Pulsing DC over the wires?

Well there you go: "pulsing DC". The power cable should not be carrying pulsing DC, at least not pulses with any significant harmonic content. That's the job of the input filtering in the light assembly; to filter the current pulses so the power cable is carrying something close to pure DC. Yes, there are strong pulses on the circuit board in the light, but in a good design the bypass capacitors and any series inductors keep the pulsing contained within the fixture.

EMI or RFI, it's the same thing. I think when we started to be concerned about interference to other non-radio electronic equipment the term EMI (Electro Magnetic Interference) superseded RFI (Radio Frequency Interference). They are pretty interchangeable.

[goboatingnow]

What pulsed DC. It's a smoothed output.

[Paul Elliott]

Originally Posted by goboatingnow

What pulsed DC. It's a smoothed output.

The DC input should be smoothed. The output to the LEDS may be smoothed constant-current, or it may be some flavor of Pulse-Width Modulation (PWM). It depends on the circuit design.

Of course this is all about switching regulators. A linear regulator should have no EMI issues.

[Auspicious]

Originally Posted by estarzinger

so I guess my question is . . . . anyone know for sure (eg has actually tried it) of any led tricolor lights (bulbs that fit an aqua signal 40, or complete fixtures) that will not create noticeable RF noise on the AIS and vhf 16 freqs?

I've been through a few solutions on Aquasignal series 50. All were noisy. I'm back to an incandescent in my tricolor. A Marinebeam LED in the anchor light isn't too bad and I don't mind the RF noise so much at anchor.

Originally Posted by Paul Elliott

Remember, we aren't just interested in VHF emissions, but also in the SSB frequencies.

Lets not forget sat phone and GPS frequencies.

[estarzinger]

Originally Posted by Auspicious

I've been through a few solutions on Aquasignal series 50. All were noisy. I'm back to an incandescent in my tricolor.

Thanks . . . I guess the solution for me will be to run the led deck lights rather than the tricolor. It's 3 lights but still less power than one incandescent.

I will just have to change the wiring around a little so the steaming light is on a seperate circuit from the deck nav lights (right now they are all combined on a "motoring" circuit)