A low cost LED lamp for your anchor light - some test results

The nominal 2 mile visibility can be achieved with an Led drawing less than 1w. Not a great imposition on the electrical system.
It is not important, for an anchor light, that it is visible at 2 miles, but this standard ensures the light is bright enough to stand out from shore lights etc. IMHO it is a sensible standard a luminare that is less bright can be easily be missed against a background of other lights
 
Yes, but it's a cr4p way to drive LEDs! Although it limits the current OK, it doesn't boost the voltage like a Buck converter, see here.

What's the benifit of boosting the voltage? Is it efficiency? Messing about here i make the l200 to be about 70%, not great. But not the end of the world. And a big benifit in that having so few bits i can understand it ;)
 
this is really helpful, thank you!

i am also exploring changing all navigation and interior lights over to LED... i recently purchased a pack of 50x 5mm white LEDs rated 12v including all required resistors, from ebay. promptly delivered two days later and only 2.99, free postage! the ultimate goal is to run the whole system using only solar power. getting there little by little. will post schematics and photos once project is worthy of sharing!! :)

anybody else exploring reducing overall power draw on charging system?
 
this is really helpful, thank you!

i am also exploring changing all navigation and interior lights over to LED... i recently purchased a pack of 50x 5mm white LEDs rated 12v including all required resistors, from ebay. promptly delivered two days later and only 2.99, free postage! the ultimate goal is to run the whole system using only solar power. getting there little by little. will post schematics and photos once project is worthy of sharing!! :)

anybody else exploring reducing overall power draw on charging system?

Are you proposing using separate resistor controlled 5mm LEDs for your interior lights and your navigation lights?
 
hello adrian, yes. i'll mount groups of LEDs (inc resistors) soldered onto veroboard and create LED 'clusters' for each interior light, as required.

my electronics knowledge is self taught - i use electronics in my creative projects and tend to work it out as i go along. so first i'll tackle the interior lighting system. then, move onto exploring naviagation lighting options. currently i'm in the middle of building my own pushpit-mounted solar panel. once thats complete, in place and hooked up, i'll turn to the interior lights project.

any thoughts, input re interior lighting (LED) configs + considerations?
 
What do the knowledgable ones think about something like this for a white mast head. Heatsink doesn't look like it would work but that can't be too hard to make up.

Or something similar, lots on ebay these days..
 
Why would you bother when you can buy high quality MR16 LED lamps for about a fiver?

Or even the ultra-cheap ones which started the thread! (£1.49 for 2 inc pp from China). Mine have been going for over a week continuously on test and don't seem particularly likely to burst into flames. Same supplier has all sorts of LEDS at similar prices. (plus at least one with that strange ebay scam (?) of an absurd price where a couple of extra '0's have mysteriously crept in).

Vic
 
Care needed in selecting resistor values

any thoughts, input re interior lighting (LED) configs + considerations?[/QUOTE]

I can understand that you enjoy putting a project together, I think most people on this forum do.

It's now better understood (I hope) that resistor controlled lamps have some significant drawbacks when the supply voltage varies. Lets assume the voltage will vary from say 12V up to 15V under certain battery charging conditions. Your dilemma is what voltage do you size the resistor for.

I suggest you use very bright high output LEDs and size the resistors for 15v. When the voltage is less than 15V your light output will be less but if you have over-designed the light output by adding extra LEDs the required light output may be about right at the lower voltage. If on the other hand you design your resistor for 12V and the voltage rises, the temperature rises and working life of your LEDs reduces significantly. In the extreme heat can rise such that it produces smoke...something you really don't want.

I note you are proposing to use what are really LEDs designed as panel indicators rather than for area lighting. The downside of these devices is that cooling can only be achieved by conduction through the little tinned copper connection leads. I suggest that's not the best way to cool a device. So you really must not overdrive the device.

I would assemble an appropriate lamp and run it through some tests at the expected voltage range and make a personal assessment of the light output. Most modern multi-meters have a thermistor probe you could use to monitor the temperature rise.

The ideal LED device for lighting is really a 5050. This is a very high performance 3 chip LED. Unfortunately its an SMD and therefore much more challenging to both solder and cool effectively. The surface it sits on has to be used as a heat sink therefore good thermal contact is essential.
Regards
 
Or even the ultra-cheap ones which started the thread! (£1.49 for 2 inc pp from China). Mine have been going for over a week continuously on test and don't seem particularly likely to burst into flames. Same supplier has all sorts of LEDS at similar prices. (plus at least one with that strange ebay scam (?) of an absurd price where a couple of extra '0's have mysteriously crept in).

Vic

Vic

Would you please give us details of your test i.e. voltages used and any voltage/current/power readings. I'm still waiting for my ones to be delivered, despite ordering at the same time as you!

When they arrive I will publish my test results too as a comparison. Maybe you could start a new thread?

Regards
 
this is really helpful, thank you!

i am also exploring changing all navigation and interior lights over to LED... i recently purchased a pack of 50x 5mm white LEDs rated 12v including all required resistors, from ebay. promptly delivered two days later and only 2.99, free postage! the ultimate goal is to run the whole system using only solar power. getting there little by little. will post schematics and photos once project is worthy of sharing!! :)

anybody else exploring reducing overall power draw on charging system?

Hello wasabee
Most LED (the diode itself) drop about 2 volts or a bit less at typically about 30 ma so each 12v bulb will have a resistor of around 300 ohms in series with the diode. This means that the resistor will waste 300milliwatts while the diode itself consumes just 60 milliwatts. Now this waste might not bother you because the LED give a lot of light.
However consider that if you fitted 4 diodes in series with a 130 ohm resistor you get 4 diodes lit for the same current as one of your diodes. Indeed you could run 5 LED diodes off 12v with just 65 ohm resistor and same current.
However there is a conundrum as all these calculations are based on 12v supply. That is fine for a battery not being charged. You will find however that at 14v supply the 5 diodes would go to aprox double the current through the LEDs so destroy them. With 4 LED diodes current would rise from 30ma to 45ma probably destroying the LEDs. With 3 LEDs in series you need a 200 ohm resistor you lose as much power in the resostor as the 3 diodes but at 14 volts the current only rises to 40 ma. Still too much for an LED but then possibly survivable.
You can see that with you one diode and resistor of 3000 ohms the current will rise from 30ma to 36ma which is much more acceptable.
In a similar manner 1 diode with 300 ohm resistor will not drop much current when volts fall to 11v. But a string of 5 diodes will be down to half current at 11V.
An LED is not like an incandescent or resistive lamp. The diode will draw no current until the voltage reaches the forward voltage of the diode. Once that voltage is reached the current rises rapidly so 3v on a 2v LED will carry a huge current and destroy itself. Hence the series resistor or electronic current control that is necessary.
Now in my case with just solar charging there is no problem the nicad battery is pretty much always 12v. But this may not be the case for you.
I have found 2 different styles of shed or cupboard LED light that takes 3 AAA batteries.These seem to have about 12LED diodes but in pairs in series. So about 4v mabe less dropped in diodes from nominal 4.5 v batteries. These have cost around 2.5 squid each and run really well on 12v when 3 are wired in series. However again they would not go well at14v or even 11v. They do seem to be a low cost way to light a cabin.
I would suggest for Nav lights you buy P and S nav lights with LED they are about 25squid a pair here and comply with requirements a stern light LED will cost about 25 squid also leaving you confident you are legal. On the other hand an anchor light is fairly easy to cobble up out of LED diodes.
good luck olewill
 
Last edited:
Would you please give us details of your test i.e. voltages used and any voltage/current/power readings.

Adrian,

I have been running them constantly now for a week on a battery maintained at 12.8v because I first want to eliminate any inherent early-life failure which would obscure results on such a small sample. (the float battery so that the PSU does not get the blame). I will then run at 14v for an extended period to simulate typical on-charge conditions; and finally at 14.8v, though limited to an hour or so as this is typical of boost chargers.

The bulbs are drawing 60mA at 12.8v. My perception of the light output at that voltage is that it would be quite adequate as an anchor light in many situations, but probably not meeting official specs.

I do have a 'proper' LED anchor-light replacement bulb, and quite frankly it is SO bright that I find it embarrassing in a busy anchorage ...... it lights up all the boats around! (hung in the rigging).

Yes, once we have useful longer-term data we should start a new thread.

Vic
 
No it isnt, at least not according to Sterling. And my boost charger set at 14.4v on a perfectly ordinary set up sometimes runs for a lot longer than an hour.

I don't want to continue the point, because it tends to lead to heated arguments! All I can say is that MY Sterling regulator, set for flooded lead-acid batteries, boosts to 14.8v every time it starts up and maintains that for 1hr, no matter what the state of charge to start with.

I have not tried it set for sealed batteries and must do so out of interest. Maybe MY regulator is faulty - who knows? Whatever, I am perfectly happy with the way it does its job:)

Vic
 
I've just ordered these http://cgi.ebay.co.uk/ws/eBayISAPI.dll?ViewItem&item=270813532596 - as I'll be running them (one of them, anyhow) on the end of an 8m long wander lead, and not running the engine, I think a simple connection via a 12v cigarette plug will suffice, without worrying about current regulation.

The idea about the meat paste jar is an interesting one - I seem to remember that Shippams used to supply fish paste in a ribbed jar that could be a good approximation for a fresnel lens, but I haven't seen them about recently, so I'll probably use an old herb/spice jar instead.
 
Adrian,

I have been running them constantly now for a week on a battery maintained at 12.8v because I first want to eliminate any inherent early-life failure which would obscure results on such a small sample. (the float battery so that the PSU does not get the blame). I will then run at 14v for an extended period to simulate typical on-charge conditions; and finally at 14.8v, though limited to an hour or so as this is typical of boost chargers.

The bulbs are drawing 60mA at 12.8v. My perception of the light output at that voltage is that it would be quite adequate as an anchor light in many situations, but probably not meeting official specs.

I do have a 'proper' LED anchor-light replacement bulb, and quite frankly it is SO bright that I find it embarrassing in a busy anchorage ...... it lights up all the boats around! (hung in the rigging).

Yes, once we have useful longer-term data we should start a new thread.

Vic

Vic,

That sounds good to me.

My lamps have not arrived as yet, how long did yours take after ordering?

I will dismantle one lamp to see what components it has and will repeat your tests on the other progressively increasing voltage to 15.5v taking both thermal and current readings.

Regards
 
Top