resistance values for LED's

I expect a decent capacitor at the current regulator outlet could overcome the radio transmission. If you are running switch mode currents through a heap of lighting cable distributed around the boat it will be very noisy electrically. Keeping it confined to a metal box in a power supply should reduce the problem. Shielded cable might also help but I would try a decent size capacitor first.

The Maxim datasheet I looked at states it is switch mode operation with buck and boost circuitry, able to operate from 9V to 40V input and up to 28V output at set current of 1.33A.


Rick W

 

The circuit above can only be turned on or off with a switch.

To make it dimmable I use a pic driving a FET with PWM. The pic and it's low standby power regulator (<100uA) draws around 2mA (a 78 series regulator alone have a 10mA quiescent current !)

If you switch the Led's off, you can sleep the pic and current drops below 100uA.

You can use ie a 4093 and a few components which is cheap to make PWM with as well, which you can adjust with a pot.

 

Unfortunately this will not work. The problem is not on the output voltage, that should be fairly smooth. The harmonic frequencies emitted comes from the high frequency resonance of the inductor - if you connect a cap over the inductor you begin to short the inductor and it cannot transfer the energy to the LED's and the driver gets shorted as well, which may damage it.

 

Fanie
I am suggesting the capacitor go across the load terminals so the voltage at the LEDS is kept constant and they are fed a steady current rather than high frequency current. The Maxim is a constant current source. Providing there is an inductor between the capacitor and the switching MOSFET the voltage variation is all across the inductor if a capacitor is used. No variation in voltage beyond the capacitor.

The current regulator should be short circuit proof. It should happily work with the output terminals bridged; the current held constant at 1.33V. It should be able to work with 1 to maybe 8 LEDS in series and maintain high efficiency.

I have not gone through the Maxim spec in detail but none of the above is difficult to achieve electronically.

Rick W

 

Hi Rick, I understood what you meant.

Dimmable LED's does not worry about frequency or being switched on or off. It does not shorten their lives.

All step up inverters uses a recommended value lo ESR cap on their output, which gives ample smoothing. If measured with a scope you will find a relative small ripple. This ripple is used by the regulator feedback to switch on and off faster or slower which will pump more or less power to the cap.

Changing the cap value to a large value may have the circuit stall, less pumps because it cannot react fast for the now slower ripple and the power will drop.

If the cap value is too small the control may not be able to reach the high switching frequency, or the related components may be driven beyond spec.

A current regulator does not maintain efficiency from ie 1 to 8 LED's. 1 LED produces 1/8th of the light at the same current 8 leds does, but in the case of the 1 LED the current regulator has to absorb the current through it x the voltage across it - which is a relative large loss.

 

led lighting

Frostie,

I am a newbie on this forum, as you might know, asking for help with an old trimaran I have aquired which thanks to the members have recognised and for that a big thanks to one and all.
I seam to have inadvertantly managed to drag this post severley of topic from your original question and I can but appologise.

It is interesting that there is so much knoweldge here that quite honestly I am a bit lost.
I appreciate what you have said Billydoc, BUT like most situations, its all a bit late to start ripping cable out and replacing with shielding.
I have seen at the London boat show, led lamps that will fit into a standard lampholder, these also have switch supplies to them and are very noisy. My friend has bought a couple for comparison and they where just as bad.This evening I did phone to asjk him if anything was mentioned about interfearance. He did not recall anything and unfortunatly the box had been scrapped.
I suppose my main reason was to flagg up to those of us that do not have the expertise to build suppresor circuits ,that LED lighting is NOT the holey grail. BUT in comparison with standard incandescent lamps, that LED's with a resistor, altho not perfect, seam to be more efficiant.
I would love to know comparisons of an LED light lux polar diagram and a standard festoon lamp.



Oh and buy the way, I use lemon scented oil in my lamps, as it helps to keep the mosquitoes away and I also quite like the smell

Kim

 

The advantage of a switching regulator is that it is only switching when power is required, the rest of the time it is supposed to be off and drawing very little power. Almost like a power on demand kind of thing. If you drive a single LED then the switcher will be more efficient than it's 1 LED linear circuit.

Linear circuits have the problem that when you have voltage drops to compensate for, ie when a single LED is driven from 12V, the losses are big and is usually noticable in the form of heat, unless of course you use huge components(no noticable heat), but the losses still remain.

Switchmodes are not as efficient as most people think. A swicher at 90% is very good.

If you use the full supply voltage in a linear circuit, it is 100% efficient. The trick is to get as close as possible.

 

Kim, if you have mosquitoes in your boat you travel too slow

Compact Fluerescents are more efficient than LED's, but they are difficult to dim. LED's are easy to dim. Just try to manuever a boat at night with bright lights at the helm. Also, if you do not need bright lighting then LED's are quite nice, especially the ones I mentioned earlier.

I can draw the circuit for a LED light /dimmer if someone would like to try it, it is not complicated at all. Really

 

Fanie
If the units are creating electrical noise it will be more than a small ripple. Irrespective a clocked regulator will not be dependent on the smoothing capacitor for its switching frequency.

I am using high frequency switching on my PMSM with radio control within a few inches of a GPS. All work fine. So it is possible not to be bothered by the noise. The controller has a switched DC to DC link and is not even enclosed in metal. These are serious currents by comparison with what LEDs require.

In the attached photo you can see the proximity of things. The battery is good for about 100A at 22V, It is taped on and one cable tie under the frame in front of the GPS. The controller is the thing with the two small electrolytic capacitors just in front of the power guage. There is also a switch mode voltage regulator hanging out the side that supplies 5V to the radio control receiver. The RC aerial is the grey wire running back to the GPS mounting.

All works fine without upsetting the GPS. In fact this GPS takes in other signals from heart monitor and cadence pulser so very noisy electrically.

I think you would be trying hard to make the noisiest possible electrical environment to upset a GPS. If you string lighting cables with pulsed currents all around it then it may be possible.

As far as efficiency goes the performance of the switch mode will not change much with the number of LEDs connected in series. They have relatively low switching losses.

The light output for LEDs is roughly 10 times for the same power as incandescent. This is a good article comparing light output:
http://sound.westhost.com/articles/incandescent.htm

I would be looking carefully into any reported electrical noise problems before throwing out the idea of using LEDs. These are becoming very common now.

Rick W

 

Rick, your PMSM radio control transmits at a specific frequency, it is designed for that. The GPS receives at a different frequency, so they don't effect one another.

Inside radio's you will see a host of tin boxes, they are there to shield unwanted harmonics into the air that will affect other appliances.

LED driver circuits will not pass CE testing with their emissions unless it is shielded. It does not have to be large currents to create problems, your car remote draws minute current from it's battery, yet if you keep transmitting the next guy in the parking lot cannot open his car because of the interference.

They use the same principle in banks. They transmit a range of noises in a relative small area so that crooks cannot use their cell phones there.

 

Fanie
That is why I am saying keep the noise in the box and smooth the signal to the LEDs.

These things are becoming widely used. They will not be permitted to radiate interference. It is not a difficult problem to fix.

Rick W

 

The circuits.

LED Driver is an on/off only.

LED Dimmer does the same but you can control the brightness as well which is nice to have.

These are the least complicated circuits I have for this. From here on it begins to become a bit involved

Both circuits can be built by anyone, nothing to it, really. There are no emission levels and if you have the correct amount of LED's in series it will come out quite efficient. None of the components are critical. You may want to keep the electronics out of water, especially sea water.

 

Hi Rick,

You will be surprised Interference can be a bugger, you have no idea.

 

Fanie
How do you know this????

Rick W

 

Fanie
The attached is taken from the Maxim website. You will see they do consider EMI. They give some basic controls. Switching frequency can be preset.

There is a small filter capacitor but the MOSFET drives the LEDS directly so would need an inductor to run with a large filter capacitor.

Any fluoro is a switchhing device and we live with these.

My view is that LEDs should not be dismissed because of noise problems. Nor should PWM drivers as they offer very flexible voltage range for both input and outputs.

Rick W