My bad I thought you were going to hard wire it and "snip" the solar lead or 110v lead off.. You'll be fine..
Cheap LED radio inteferrence?
- Thread starter Roger Long
- Start date
- Status
My bad I thought you were going to hard wire it and "snip" the solar lead or 110v lead off.. You'll be fine..
Depending on the soldering iron wattage used and length of time, yes they can be temp sensitive. The are definitely polarity sensitive.
As others have answered - no, that seems to be a pretty basic linear regulator, which means... no RF noise. As also mentioned, the worst issue will probably be weathering, so anything you can do to protect the board and wiring from the elements will probably keep it running longer.
(and I still think it's a cool idea
)
Here's my light:
(Sorry for the poor photo - cell phone - I left my camera on the boat)
The forward extension puts the light unit out of sight on top of the mast. The LED is fastened to an aluminum bolt with a piece of heat shrink tubing.
I don't know yet if I'll actually put it up. I never felt the need for on since I do pretty well by feel after a half century of sailing but it might help guests. Anyway, it was fun to play with on a rainy day.
(Sorry for the poor photo - cell phone - I left my camera on the boat)
The forward extension puts the light unit out of sight on top of the mast. The LED is fastened to an aluminum bolt with a piece of heat shrink tubing.
I don't know yet if I'll actually put it up. I never felt the need for on since I do pretty well by feel after a half century of sailing but it might help guests. Anyway, it was fun to play with on a rainy day.
Doesn't your anchor light illuminate the windex?
Yeah, it's illegal lighting, but then so is what you're proposing (while underway anyway.)
Also, another option is that I think Davis actually makes an attachment for it's Windex to do exactly what you're proposing.
When we're racing we just keep a tiny pencil point flashlight handy (with a red lens on. This also allows us to check telltales and sail shape.
Yeah, it's illegal lighting, but then so is what you're proposing (while underway anyway.)
Also, another option is that I think Davis actually makes an attachment for it's Windex to do exactly what you're proposing.
When we're racing we just keep a tiny pencil point flashlight handy (with a red lens on. This also allows us to check telltales and sail shape.
I love the ideas and ingenuity from everyone on this subject and everyone has their own way of doing things so here's my take on this. I did an overnight long distance solo race last year and I wore a headlamp. It allowed me to direct light wherever I was looking so I could see the windex, sails and telltales. If you're worried about your night vision, you can get one that can switch to red. I didn't notice that my night vision was affected at all.
Just a couple of pennies for you to ponder.
Just a couple of pennies for you to ponder.
we did the same, before our 1st overnight race, I bought everyone their own petzel headlamp ... it worked really well.
Turning your anchor light on is illegal. I doubt that the little LED could be considered an illegal light because it would be too small to ever be mistaken for a navigation light. No one would ever make a point of approach decision based on that little gleam.
The standard Windex light requires another set of wires up the mast and is a lot brighter. I don't need a lighted windvane badly enough to do that.
You didn't see link to my project that he is trying to duplicate did you? I used the Davis windex light, replaced the bulb with an LED and powered it off of a solar lawn light fixture.
Also there is no regulation that says you cannot have other lights on your boat at night than the required lights, just that the required lights be functional, turned on, and visible. If additional lighting was illegal, how do cruise ships get away with all the lights they have?
Too Bright!
I've been wondering if the little LED would be bright enough so I put the Windex light in the bedroom closet last night figuring I could also see how long it was lasting if I woke up about 3 or 4 as I often do.
Wow. I could have read in bed by it. Barbara had to get up and close the door because it was too bright for her to sleep and I could clearly see it shining around the door edges. If anything, it is too bright for the purpose. I don't want it to be distracting. I just want to be able to barely see the Windex on a very dark night.
Can LED's be dimmed by lowering the voltage? A small resistor in the circuit might get it down to the right level. It possibly could make it last longer.
Anyone have an idea of a resistance value to start with?
I've been wondering if the little LED would be bright enough so I put the Windex light in the bedroom closet last night figuring I could also see how long it was lasting if I woke up about 3 or 4 as I often do.
Wow. I could have read in bed by it. Barbara had to get up and close the door because it was too bright for her to sleep and I could clearly see it shining around the door edges. If anything, it is too bright for the purpose. I don't want it to be distracting. I just want to be able to barely see the Windex on a very dark night.
Can LED's be dimmed by lowering the voltage? A small resistor in the circuit might get it down to the right level. It possibly could make it last longer.
Anyone have an idea of a resistance value to start with?
Thats what you get for over spending on the unitRoger Long said:
... The $2 one i got is still a little brighter than necessary, but not too much. I tried it at night and found it would do a reasonable job illuminating things up to a few feet away, but no more. Also the lens cover from the Davis light disperses the light a little more than the naked bulb.Roger and FourPoints
I stand corrected. I am sure by the letter of the law you are correct.
And no, I didn't look at the link provided by FourPoints.
I stand corrected. I am sure by the letter of the law you are correct.
And no, I didn't look at the link provided by FourPoints.
LEDs can be dimmed by varying the voltage supplied to them. To figure out what you need I'd recommend a trimming pot (adjustable resistor) to experiment with. When you get the brightness right then measure the pot and use a resistor of that rating.
Alternately you can go to Radio shack and buy another LED and wire it in series with the current one. They will both light at reduced brightness. Repeat as needed.... not an elegant solution though.
or
get some 10 ohm resistors and try one and see if that floats the boat. If it is still too bright add another till nirvana is achieved.
If you wanted to do it with math you would need to know Vf and If for the LED. Since you bought the thing as a unit you would have to do some measuring to find out what the voltage drop across the LED is (Vf) and what the current flow is at Vf. If you have a multimeter this is pretty simple. Parallel for current and series for voltage.
V=IR is the formula, you are looking for R(led) so R=Vf/If
That would be the total resistance the LED supplies to the circuit. That should give you an idea of the size resistor you will need. Resistances in series are additive. You should get something around 70 ohms as a R(led) value.
Alternately you can go to Radio shack and buy another LED and wire it in series with the current one. They will both light at reduced brightness. Repeat as needed.... not an elegant solution though.
or
get some 10 ohm resistors and try one and see if that floats the boat. If it is still too bright add another till nirvana is achieved.
If you wanted to do it with math you would need to know Vf and If for the LED. Since you bought the thing as a unit you would have to do some measuring to find out what the voltage drop across the LED is (Vf) and what the current flow is at Vf. If you have a multimeter this is pretty simple. Parallel for current and series for voltage.
V=IR is the formula, you are looking for R(led) so R=Vf/If
That would be the total resistance the LED supplies to the circuit. That should give you an idea of the size resistor you will need. Resistances in series are additive. You should get something around 70 ohms as a R(led) value.
Bill,I think you have the right concept but there's a number of significant problems with this description:
- (as already mentioned) to generalize: you put the probes across something ("meter in parallel") for a voltage reading, and you break the circuit and put the meter probes between the break ("meter in series") to measure current.
- the forward voltage drop of an LED (aka voltage "across" it) is relatively constant. Depending on the LED type and colour, this can be between 1.6v to just under 4v. LED intensity is determined by how much current goes through it (typical values are between 1 mA and 20 mA for small LEDs, to 50 mA or more for high-intensity LEDs with heatsinks. Even if you vary the current from dim through to almost meltdown, the actual voltage measured across the LED doesn't change much.
The basic rule of thumb for calculating the proper resistor is as follows:
Say you have a basic small white LED, and you know that forward voltage drop across the LED (Vf) is 2.1v, and current at the brightness you want is 15 mA (0.015 A). (You either get these values from the LED specs, or you simply try the LED with a battery and some resistors (values ranging between 500 and 5000 ohms).
To power one of these LEDs with boat DC (say 12v worst-case) you can now plug into ohms law like so:
- the voltage across the resistor will be; Vsupply - Vf = 12 -2.1 = 9.9v
- desired current is 15 mA (0.015 A)
... then the resistor value will be 9.9/0.015 = 660 ohms. A close off-the-shelf value is 680, that would be just fine in this application.
If you put 2 such LEDs in series then the voltage across the resistor will be 12 - (2 x 2.1) and then the resistor value would be 520 ohms. 560 is probably the closest off-the-shelf value.
Yes if the battery voltage was at 14.5 due to a charger being on, then the LED would have more current... so for the case of a single LED and 680 ohm resistor
I = E/R = (14.5 - 2.1)/680 = about 18.2 mA
so you'd have to confirm that the chosen LED can withstand this. Most can.
Nice little overview and calculator here.
Hold the phone...
In both mine and yours, there is a 4-pin device that is a small flyback switching regulator. On mine the part number is LX801, and I was able to confirm the part and how it works. I also stuck an oscilloscope and frequency counter on it, and confirmed that it oscillates at around 400 to 450 kHz.
(spec sheet as doc. How's your Mandarin? )
In your picture, this device is near the upper edge and it's just a black bar, since it's viewed from the top. Looking harder, I can just see, to its right, a legend on the PC board showing a 4-terminal device labelled U1. A clue I missed.. And the thing that looks like a resistor is actually an inductor.
Anyway, the good news is that this still shouldn't bother your marine VHF reception at all. It didn't bother my cheap Cobra handheld, even with squelch off and the antenna right against the light.
For a more real-world test you can always charge one up, wait til dark, and run it up the mast on a halyard.
* * *
Anyways... now that I've dissected one, I can see that it would be relatively simple to modify it to reduce the brightness, or to change to an amber or red LED. These things are pretty cool, especially at $2 apiece
Today I bought some cheapie garden solar lights ($20 for 10 at Home Depot), and took them apart... and I now have to retract my earlier assertion that these cheap lights don't have any high-frequency regulating components.This is the one I'm thinking of using:
Do you see anything that looks like a high frequency switching power supply there?
In both mine and yours, there is a 4-pin device that is a small flyback switching regulator. On mine the part number is LX801, and I was able to confirm the part and how it works. I also stuck an oscilloscope and frequency counter on it, and confirmed that it oscillates at around 400 to 450 kHz.
(spec sheet as doc. How's your Mandarin?
)In your picture, this device is near the upper edge and it's just a black bar, since it's viewed from the top. Looking harder, I can just see, to its right, a legend on the PC board showing a 4-terminal device labelled U1. A clue I missed.
. And the thing that looks like a resistor is actually an inductor.Anyway, the good news is that this still shouldn't bother your marine VHF reception at all. It didn't bother my cheap Cobra handheld, even with squelch off and the antenna right against the light.
For a more real-world test you can always charge one up, wait til dark, and run it up the mast on a halyard.
* * *
Anyways... now that I've dissected one, I can see that it would be relatively simple to modify it to reduce the brightness, or to change to an amber or red LED. These things are pretty cool, especially at $2 apiece
If you figure out the resistance to cut the brightness about 50%, let me know. It might save me buying a variable resistor.
I assume the added resistance would make the battery last longer. Might there be a chance though that the voltage could drop below some critical threshold so that the LED actually goes dark earlier?
I've thunk it out a bit more. Charged NiCad batteries only put out about 1.2 v, but LEDs have a forward voltage drop of typically 2 volts. So, of course some sort of voltage boost would be necessary, and so that's job 1 of the flyback circuit.
So, yes I believe that lowering the required current would extend the operating time.
I will try some tests with resistance values and different LEDs and get back to you, hopefully before Monday.
So... here are some basic ways one could reduce the LED brightness (and extend the 'on' time) of a cheap solar garden light.
First, here's the circuitboard of the cheap lights I bought yesterday:
It's got the 4-lead LX801 flyback IC I mentioned yesterday.
(Note also that it has 2 LEDs and a switch. This particular model has both a white LED and a funky color LED that, by itself, cycles through red-green-blue. Trippy.... The LEDs are switch-selected)
I put an oscilloscope across the white LED, and saw pulses at about 450 kHz.
I then removed both LEDs, added some leads to the white LED position and tried some basic circuits for current reduction:
First (FIG 1) I simply added some resistance. 2000 ohms (2k) was enough to drop the brightness by a bit more than half (very rough guess). I recommend to anyone to first try different values (or use a 2k or 5k pot or trimmer like I've done) to find the preferred brightness. With this circuit there are also pulses across the LED.
I also tried a bit more elaborate circuit that includes a capacitor for some filtering (FIG 2). I added a small signal diode (1N4148; just about any silicon diode will work) to only pass positive pulses, and then a small capacitor (.05uf) across the LED. This about eliminated all pulses at the LED.
I didn't find any interference with any of these circuits on my cheapie VHF handheld, but if you are going to extend the LED leads and are concerned about pulse radiation, use the circuit of FIG 2, building the resistor-diode-capacitor circuit inside the light/solar panel enclosure, and simply extending the LED. Since there's no pulses across the capacitor + LED, the extending wires won't be radiating the pulses.
Roger, I hope this helps...
First, here's the circuitboard of the cheap lights I bought yesterday:
It's got the 4-lead LX801 flyback IC I mentioned yesterday.
(Note also that it has 2 LEDs and a switch. This particular model has both a white LED and a funky color LED that, by itself, cycles through red-green-blue. Trippy.... The LEDs are switch-selected)
I put an oscilloscope across the white LED, and saw pulses at about 450 kHz.
I then removed both LEDs, added some leads to the white LED position and tried some basic circuits for current reduction:
First (FIG 1) I simply added some resistance. 2000 ohms (2k) was enough to drop the brightness by a bit more than half (very rough guess). I recommend to anyone to first try different values (or use a 2k or 5k pot or trimmer like I've done) to find the preferred brightness. With this circuit there are also pulses across the LED.
I also tried a bit more elaborate circuit that includes a capacitor for some filtering (FIG 2). I added a small signal diode (1N4148; just about any silicon diode will work) to only pass positive pulses, and then a small capacitor (.05uf) across the LED. This about eliminated all pulses at the LED.
I didn't find any interference with any of these circuits on my cheapie VHF handheld, but if you are going to extend the LED leads and are concerned about pulse radiation, use the circuit of FIG 2, building the resistor-diode-capacitor circuit inside the light/solar panel enclosure, and simply extending the LED. Since there's no pulses across the capacitor + LED, the extending wires won't be radiating the pulses.
Roger, I hope this helps...
The value of resistor you need is dependent on the battery voltage. I'm thinking that x thousand ohm resistors would be a 12ish volt system.
LED do have an extinguishing voltage but it is VERY LOW. They basically dim to nothing visible
LED do have an extinguishing voltage but it is VERY LOW. They basically dim to nothing visible
- Status