jason9

modified 7 years ago

EC LED Parasitic Capacitance Simulation vs True Parasitic Capacitance Simulation

216

01:40:47

EC simulates LED parasitic capacitance such that the LED lights up when the current flows through the parasitic capacitor even if there is minimal actual voltage, whereas a true LED will only light up upon sufficient voltage being applied and current beginning flowing through the LED crystal rather than seeming to flow through it via its parasitic capacitance.

On the left are two circuits showing the LED unrealistically lighting up from current flowing through the parasitic capacitor. In the CCS example, the parasitic capacitor starts discharged but as it charges up from the current flowing it causes the LED to light up even when insufficient voltage is supplied and no current flows through the LED crystal itself. In the AC example, an ultra high frequency low voltage AC source causes some current to flow through the parasitic capacitor unrealistically causing the LED to light up even though the voltage only ever reaches up to 100mV, far too little to make current flow through the LED crystal itself.

On the right, the LED parasitic capacitor has been removed and an external capacitor added. The LED only lights up when current flows through the LED crystal itself and not when insufficient voltage is supplied and current is only flowing through the external capacitor.

published 7 years ago

hurz

7 years ago

Jason, what is your proof the capacitance is in parallel to the crystal. I see it as part of the crystal caused by its junction capacitance! So you have to model it in series. http://everycircuit.com/circuit/6178759723188224

hurz

7 years ago

Here another try to model a LED http://everycircuit.com/circuit/5221040031793152

jason9

7 years ago

Why model it in series? Won’t that block DC?

jason9

7 years ago

I looked at your second link. It implies that electricity flows through the crystal itself, but without much voltage across the crystal. How does that happen?

jason9

7 years ago

The way I see the parasitic capacitance being generated is this: the wires connect to either end of the crystal, but at low voltages, the crystal has a high resistance, so it acts like a resistor and you can imagine simple air replacing it. Then, the two ends of the wires almost touch, but are separated by a small gap of air (technically crystal, but both are high resistance here, so it shouldn’t matter much). This is exactly how a capacitor is described, only without a plate on the end of either conductor to increase the area and therefore capacitance. Then, there is the crystal at the two terminals of the capacitor. This looks like parallel to me.

jason9

7 years ago

I looked at your first link. Interesting idea. However, doesn’t this imply the current passing through the high resistance crystal?

jason9

7 years ago

In your first link, where current does pass through the always conductive regions of the crystal, it never crosses the PN junction, seeming to instead skip over it through the junction capacitance. So, it would not produce light, as the electrons never cross the junction, and the model simplifies down to what’s shown in this circuit.

BillyT

7 years ago

Some nice constructive discussion here, pity we couldn't have a lot more.

hurz

7 years ago

The capacitance coming form the junction and not, or at least not in this model, from pinning cuz the from of the LED is out of interesst i guess. But you are right, how does it glow if the capacitance is building just an electric field across the junction and no electrons are doing its quanten job? good question, hmmm

ViolationMad

7 years ago

As far as i know the junctiom capacitance is in series to the crystal of the LED. Also this capacitance should not be linear and change with the applied voltage because of the change in the active region due to the electric field. Also there should be a prarallel capacitance createt by the leads and the bonding wires, which hold different pontetials and thus create an electric field aswell. Anyway this capacitance should be extremle low. However due to this non linear effect of the junction, special diodes (varactor diodes) were used to create FM. This should use the same principal but much smaller in normal diodes. When you are more intressted just look for the varactor diode wiki page. @BillyT your comment actually provokes this to become unconstructiv as its not related to the content at all...

BillyT

7 years ago

@ViolationMad, the first bit of your comment gave me pause to think, I never Previously thought of the capacitance reducing or maybe even disappearing. If it indeed goes away completely it throws a whole different light on the Leds operation. We will have to agree to disagree on the last bit of your comment ...

hurz

7 years ago

Capacitance diodes are used in RF for filter tuning and VCOs. They have capacitances around 30pF and need voltages to tune them down to 2pF from 33V. Newer diodes can be tuned with even 5V max.

hurz

7 years ago

@BillyT, your comment sounds like its new to you that we have technical talks on everycircuit. If so, then you have the wrong friends on everycircuit. Ignore the trolls and concentrate on electronics

ViolationMad

7 years ago

@hurz But im just talking about the effekt itself. There is a positiv charged side, a negativ charged side and a neutral area inbetween. Also the current is unsignificant small as long as its not saturated. So this should create an electric field and thus it should have some capacitance, shouldnt it? When a voltage is applied the active region changes its size and this should also change the diodes cpacitance. The effect of it maybe not comparable to to capacitiv diodes but should also take place at every other pn junction.

jason9

7 years ago

Your comment right after @BillyT’s first comment is hard to understand. Could you please write it again with better English so that I know what you’re trying to say?

jason9

7 years ago

@ViolationMad, that is precisely the way it works as according to wikipedia.

hurz

7 years ago

So we all agree the capacitance within EC must be in series to the crystal and is a variable junction capacitance. That would explain why Leds at higher frequency glow even with voltages below the junction treshold. This should be verified irl. BTW if i remember correct the LEDs i have measured a while ago irl had a capacitance of 80pF and more. Pinning is probably about 5pF the rest is the junction capacitance.

jason9

7 years ago

But how can LEDs glow with current through the capacitance? No energy is actually dissipated in a capacitor, and thus no light can be emitted. Anyway, the electrons never cross the band gap emitting light in the process, since it just builds up an electric field capacitively.

ViolationMad

7 years ago

I have never heard of such an effect in real life so this maybe just the simulation. I guess that the sim simplifies the led calculations and brightness down to the current through the led, which should also light it up then, when the series capacitiv reactance becomes low enough due to higher frequency. Otherwise im not sure if the p and n substrates will act like a normal cap (i actually doubt it), because there are effects like quiecent current in diodes that could lead to different results. On the other hand i have heard about something thats called ion-current of caps. I dont really remember tho, so ill have to look for the episode of EEVblog which explained this phenomenom. I will probably test some leds with my function gen later but i dont really expect something to happen.

hurz

7 years ago

I also think its just a simulation effect and irl it wont work. EC just takes the current and makes proportional light out of it. @ViolationMad, sure a p n junction is a capacitor and is one possibility in integrated circuits to have caps. As we already said they can also be used as tuneable capacitors.

BillyT

7 years ago

I have been running this backward and forward in my mind, hurz's comment on varactor diodes made me question whether there was DC current flowing thru the varactor while the capacitance was being changed. It took a bit of searching but I eventually found an item on alternatives to varactors, it stated that Leds could be used to a limited extent, the capacitance limited by manufacturing processes and up until the point that the device starts to avalanche. That leads me to ask, is this also the point where the LED also starts to turn on, and also because an AC voltage flows through a capacitor, could this contribute to that phenomena.

zorgrian

7 years ago

What

add comment in editor