Anchor light

[LP]

Does anybody have any experience with this product?

http://www.bebi-electronics.com/owl.html

or this one

http://www.bebi-electronics.com/beka.html

It's 1/3 to 1/4 the price of anything I've looked at.

[Trevlyns]

I may be in a different part of the world, but even at those prices, it still seems expensive. How about a full set of conventional navigation lights for 24 UK pounds – search further and you can find a masthead light for about 6 pounds.
Go to e-bay and search for “navigation lights” or just try this link http://cgi.ebay.co.uk/ws/eBayISAPI.d..._promot_widget

[RANCHI OTTO]

Special lights for an....UFO?

[LP]

The key here is power comsumption. I'm looking for the lowest possible current draw because right now, the only way to charge my battery charged is to plug it in to shore power. LED's are like having "supercharged" energizer bunnies.

[Frosty]

A few people here make thier own LED anchor lights. Its not that difficult.

They use the original bayonet fitting from the old bulb and solder a few white LED's (say 10 ) in a circle.

From what I have seen they seem ok , they leave them on all the time.

The LED's here are about 30p--you need a tiny resistor to reduce it to 3v.

I use a similar approach when making warning lights for bilge pumps etc. You can leave them on forever.

[LP]

Jack,

Sounds like a fun project. Do you try to find high output LED's?

[Michael Homsany]

Er, sorry for being intrusive, but I saw the link come thru on our web site.

For Mr. Frost, home grown LED lights aren't that hard to make up and are a fun rainy day project, but not quite as straight forward as it may seem. There was a thread I wrote up for the ybw board a whiles back, here it is quickly.

LED's are current devices. To size a resistor for the device, use this formula:
(Vs-Vf)/I, where Vs is the supply voltage, Vf is the forward voltage drop for the LED and I is the current you want to run the LED at. For example, on an average, with a typical so-called ultra-bright white LED you have:

(12.75-3.5)-.020, so the resistor would need to be 462.5 ohms

There are a number of issues here, first, your batteries likely only hit 12.75 volts by accident either going up or going down in voltage. Once you exceed 12.75 volts, you will start to exceed the maximum design current of the device, so, to be on the safe side figure 14.4 for the Vs so that for a single device, 545 ohms.

Now, thats a fair bit of heat that is going to be wasted, so we can get sexy and wire three LED's in series, so that the Vf will be 3.5*3 or 10.5 volts (batteries shouldn't get this low anyways, so we are ok. On top of that, the 'knee' on this type of LED isn't as sharp as others). We now get (14.4-10.5)/0.02=195 ohms per series of 3 LED's.

There are a number of reasons why *selling* lights made like this isn't a good idea, long-life, inconsistent colour, inconsistent brightness, etc. are some of the trade-offs (with tar and feathers being the penalty!), but it can still be a fun project.

I have up a couple of pages on our site http://www.bebi-electronics.com/regulator.html, http://www.bebi-electronics.com/testing.html and http://www.bebi-electronics.com/specs.html which deal further with the technical aspects of LED's and the 'challenges' putting them to use in a marine environment.

[LP]

Michael,

How would a replacement on a a D-cell flashlight work? 3 volts is fairly close to the 3.5 that you are targeting. I think maybe Jack was talking about that type of a set-up. Perhaps, I'm wrong because he states that he is reducing the voltage to 3 volts. On a flashlight type set-up, I would think that you could wire wire any number in parallel depending on desired current limitations and the number of LED's required to gain the desired visibility.

I need to read more on LED's. Ay what voltage(actually amperage) does the output drop off?

Judging from this site (http://www.maxim-ic.com/appnotes.cfm/an_pk/1021) a charge pump regulator is used even on flashlight type LED's.

I'm off to learn more.

[Michael Homsany]

First on the 3 'd' cells and a parallel circuit, yes, it would work, but you would still need a little 'r' for each device.

On the LED's, they are rated in the range of Vf=3-4 volts. Our own testing has shown that even for large lots of devices (10k), there is still quite a range, which is why we test and sort. Unless you use 1% resistors and match each resistor to each individual LED, the devices with the lowest Vf will pull the most current, will be the brightest, and will fail first. The light would also dim as the battery voltage dropped.

If you used 3 'd' cells, then used a a pair of resistors on a spdt centre-off switch, you could size the first resistor for an average Vs at 4.5v and the current at .02*number of devices, the second when the voltage of the batteries dropped to 3v for a compromise.

Again, if your making it for yourself, who cares? The only person you have to sell it to is yourself (kind of like a fishing lure in that respect, the only thing IT has to catch is the fisherman!).

There is another issue in that the Vf varies based on current thru each device. If you reduce the current from max to say 0.01, the Vf also has a drop. This is frankly a good idea, as the phosphor reflective life is a function of current and temp. On the other hand, the light output is more or less linear to current.

You DO need to be wary of the published claims on the light intensity (mcd figure) on the devices, esp. those out of China. Some of the data sheets show the truth (ie; 10000 mcd is max, not typical, with a very wide spread), others don't. We gave up even on the higher quality Chinese devices because of these issues. On the other hand, yes, you can get them for 30p each.

Charge pumps, aka dc-dc buck, boost, buck/boost converters, are how flashlights *should* operate (few actually do!), it's how you can get ahead (sort of) the forward voltage drops. On the other hand, the converter does chew up energy from the battery to boost the voltage, then regulate the current. If we (Bebi Electronics) were to make flashlights, that's what we'd do. Since we don't, there has been a business decision made to use a linear regulator as it's cost is less dear and the efficiency differences are trivial, especially where it is most vital, when you are in a discharge mode of ships power (>12.75 v).

Hope all of this helps. I have tried my best to only comment on the product (LED's) in general, not to give a sales pitch. There is just a lot of mis-information about and failures of the devices are often judged to be a result of the technology, not the implementation.

Best regards,
Michael