Absurd Led lightning system

I just riped apart a chinese LED flash light to discover the following absurd circuit.

**broken link removed**

I just can't understand why the heck did the chinese put that 3K9 ?

Probably to keep the led from getting zapped by reverse-polarity static electricity when you walk across the carpet in leather slippers and touch the switch on the flashlight...

but the static electricity doesnt flow... so how come it could effect the led ??

Pherhaps the chinese thought like MikeMI but forgot to consider ash20's point.

You can easily build up a static charge (where I live the relative humidity averages ~5% in the Winter) that will jump a 1/4" (7mm) just by walking across synthetic or wool carpet. That will jump across the flashlight's open switch contacts just fine.

LED's are quite sensitive to static damage.

ash20 said:

but the static electricity doesnt flow... so how come it could effect the led ??

Click to expand...

It doesn't until it finds a path. Why do you think ESD is such a big deal?

i've seen both static as well as induced voltage (a plastic drawer full of LEDs were about 15ft away from an operating tesla coil, and the whole drawer full were flickering, and didn't work later when tested) damage LEDs. max reverse voltage of an LED is usually about 5V. since a reverse biased LED junction acts like an open circuit, that 5V can accumulate quickly in the presence of a static field (or a strong RF field) once a discharge of static breaks down the reverse potential of the LED. the LED is shorted. since 5V is a relatively low threshold, the source of the static charge doesn't even need to come in contact with any part of the LED, as the large field gradient is enough to induce a large voltage across the junction from a considerable distance. that's why static sensitive components are best protected in a conductive plastic bag (the metallized plastic or carbon coated ones being the best, and the pink and blue ones not really very good). the bag acts as a faraday cage where the field gradient inside the shield is zero. in pracice i usually solder a 1N4148 or 1N4004 reversed across an LED (the 4004 i use in applications where a reverse voltage could be connected by mistake). with a diode reversed across the LED junction, the reverse voltage across the LED will always be less than 0.7V. the 2.9k resistor is probably cheaper than a diode, but might not be as effective as a diode in the role of protecting the LED.

thanks unclejad613. So, You mean to say, If I walk accross synthetic carpet, then I (a conductor) will accumulate static charge. When I tuch the flash lights plastic housing near the led, My charge can create enough electric potential gradient so as to destroy the LED, although the plastic housing don't allow it to flow to the led.
I didn't know, statics where that much powerful.

But, I don't understand how a Diode (or resistor) would protect the LED from static fields.
I am thinking the potential field created by static charge would be the same whether you place a diode (or resitor) or not.

Just for the learning purpose here.

the diode is reverse polarity from the LED, therefore the diode will conduct at a forward voltage of 0.7V across the reverse biased LED junction. the LED will conduct at between 1.4V and 3.6V (depending on what color the LED is), while the diode is reverse biased. therefore whichever polarity the static charge is, there is never an opportunity for it to exceed the reverse breakdown voltage of either the LED or the diode and cause one of them to short. the current from the static charge bypasses either reverse biased device through a forward biased junction of the other device. this type of protection is also useful for what i call the "offset sniffer" which i use for primarily determining if an amplifier has DC offset. it's especially useful in multichannel amplifiers, where one channel having offset will shut down the whole amplifier one or two seconds after the power has been turned on. a DMM usually doesn't seem to respond fast enough to get an accurate reading of whether there is offset on a channel, because when the power is turned on most amplifiers take a second or so to stabilize at 0V, and the sampling time of most DMMs is a second or so. so i use the offset sniffer to go down the row of channel outputs to find the channel causing the problem. it's also useful for verifying the presence and polarity of rail voltages, and can also be used connected to a dummy load to visually monitor that there is audio present on the dummy load (useful for burn testing amplifiers with complaints of intermittant audio)(with an AC signal both LEDs glow). in this circuit the 1N4004s not only act as "selector switches" to allow one LED or the other to light, but also protect whichever LED is reverse biased from too much reverse voltage. i have used this for checking amplifiers with up to +/-75V rails.

Ok, I got it using diode, but do you think the resister can do the job of the diode?

in many cases where static alone is the concern, the resistor can do the job. , although is still prefer the diode method, since the max reverse voltage will be clamped to a known value (0.7V). with the very small values of current available from a static source, the resistor works ok because it keeps the LED wires at nearly the same potential in the presence of a static field, but doesn't waste very much current when power is applied (about 1mA is wasted in the resistor). the diode wouldn't waste any current at all because being reversed biased, would act as an open circuit.

Thanks a ton. And sorry chinese !

the complete explanation of Mr. unclejed613 is very informative indeed...
the resistors can also be used as he said but they will just limit the current and i think we don't know the voltage of static discharge.. so we might be helpless here since ESD could vary and so resistor might prove helpless as it will always drop same voltage and if ESD is high enough, led will fuse (or whatever it is called)... but in case with diode, " there is never an opportunity for it to exceed the reverse breakdown voltage of either the LED or the diode "... so diodes are a better option

diodes are the better option, but are also more expensive for the short-term. after all of the warranty replacement costs, the money saved by using resistors, really wasn't saved, but it sure looks good on the manufacturer's books to say "we saved $50,000.00 by using resistors instead of diodes", and the warranty return costs come later.....when nobody is really paying attention anymore